Handbook of Phytoalexin Metabolism and Ac

Handbook of
Phytoalexin
Metabolism
and Action
BOOKS IN SOILS, PLANTS, AND THE ENVIRONMENT
Soil Biochemistry, Volume l , edited by A. D. McLaren and G. H. Peterson

Soil Biochemistry, Volume 2, edited by A. D. McLaren and J. SkujigrS
Soil Biochemistry, Volume 3, edited by E. A. Paul and A. D. Mclaren
Soil Biochemistry, Volume 4, edited by E. A. Paul and A. D. Mclaren
Soil Biochemistry, Volume 5, edited by E. A. Paul and J. N. Ladd
Soil Biochemistry, Volume 6,edited by Jean-Marc Bollag and G. Stotzky
Soil Biochemistry, Volume 7, edited by G. Stotzky and Jean-Marc Bollag
Soil Biochemistry, Volume 8, edited by Jean-Marc Bollag and G. Stotzky
Organic Chemicalsin theSoil Environment, Volumes l and 2, edited by C.

A. 1. Goring and J. W. Hamaker
Humic Substances in the Environment, M. Schnitzer and S. U. Khan
Microbial Life in the Soil: An Introduction, T. Hattori
Principles of Soil Chemistry, Kim H. Tan
Soil Analysis: Instrumental Techniques and Related Procedures, edited by
Keith A. Smith
Soil Reclamation Processes: Microbiological Analyses and Applications,
edited by Robert L. Tate Ill and Donald A. Klein
Symbiotic Nitrogen Fixation Technology, edited by Gerald H. Elkan
Soil- WaterInteractions: Mechanisms and Applications, edited by Shingo
Iwata, Toshio Tabuchi, and Benno P. Warkentin
Soil Analysis: Modern Instrumental Techniques, Second Edition,edited by
Keith A. Smith
Soil Analysis: Physical Methods, edited by Keith A. Smith and Chris E.
Mullins
Growth and Mineral Nutrition of Field Crops, N. K. Fageria, V. C. Baligar,
and Charles Allan Jones
Semiarid Landsand Deserts: Soil Resource and Reclamation, edited by J.
SkujigrS
Plant Roots: The Hidden Half, edited by Yoav Waisel, Amram Eshel, and
Uzi Kafkaf i
Plant Biochemical Regulators, edited by Harold W. Gausman
Maximizing Crop Yields, N. K. Fageria
Transgenic Plants: Fundamentals andApplications, edited by Andrew Hiatt
Soil Microbial Ecology: Applications in Agricultural and Environmental
Management, edited by F. Blaine Metting, Jr.
Principles of Soil Chemistry: Second Edition,Kim H. Tan
Water Flow in Soils, edited by Tsuyoshi Miyazaki
Handbook of Plant and Crop Stress, edited by Mohammad Pessarakli
Genetic Improvementof Field Crops,edited by Gustavo A. Slafer
Agricultural Field Experiments: Design andAnalysis, Roger G. Petersen
Environmental Soil Science, Kim H. Tan
Mechanisms of Plant Growth andImproved Productivity: Modern Ap-
proaches, edited by Amarjit S. Basra
Selenium in the Environment, edited by W. T. Frankenberger, Jr., and Sally
Benson
Plant-Environment Interactions, edited by Robert E. Wilkinson
Handbook of Plant and Crop Physiology,edited by Mohammad Pessarakli
Handbook of Phytoalexin Metabolism andAction, edited by M. Daniel and
R. P. Purkayastha
Soil- WaterInteractions: Mechanisms andApplications, Second Edition,Re-
vised and Expanded, Shingo Iwata, Toshio Tabuchi, and Benno P.
Warkentin
Stored-Grain Ecosystems,edited by Digvir S. Jayas, Noel D. G. White, and
William E. Muir
Agrochemicals from Natural Products, edited by C. R. A. Godfrey
Additional Volumes in Preparation
Seed Development and Germination,edited by Jaime Kigel andGad Galili
Nitrogen Fertilization in the Environment, edited by Peter Edward Bacon
Phytohormonesin Soils: Microbial Production andFunction, W. T. Franken-

berger, Jr., and Muhammad Arshad
This Page Intentionally Left Blank
Handbook of
Phytoalexin
Metabolism
and Action
edited by
M.Daniel
The MaharajaSayajirao University
of Baroda
Vadodara, Gujarat, India
R. P. Purkayastha
University of Calcutta
Calcutta, West Bengal, India
Marcel Dekker, Inc. New York. Basel Hang Kong

Library of Congress Cataloging-in-Publication Data
Handbook of phytoalexin metabolism and action / edited by M. Daniel,

R. P. Purkayastha.
p. cm. - (Books insoils,plants, and theenvironment)
Includes bibliographical references and index.
ISBN 0-8247-9269-6
1. Phytoalexins-Metabolism. 2. Phytoalexins-Physiological
effect. 3. Plants-Disease and pest resistance. 4. Phytoalexins-
-Metabolism-Research. 5. Phytoalexins-Physiological effect-
-Research. 6. Plants-Disease and pest resistance-Research.
I.Daniel, M., Dr. 11. Purkayastha,R.P. 111. Series.
QK898.P66H35 1994
581.2’9-dc20 94-22878
CIP
The publisher offers discounts on this book when ordered in bulk quantities.
For more information, write to Special Sales/Professional Marketing at the
address below.
This book is printed on acid-free paper.
Copyright 0 1995-by MARCEL DEKKER, INC. All Rights Reserved.
Neither this book nor any part may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying, micro-
filming, and recording, or by any information storage and retrieval system,
without permission in writing from the publisher.
MARCEL DEKKER, INC.

270 Madison Avenue, New York, New York 10016
Current printing (last digit):

1 0 9 8 7 6 5 4 3 2 1
PRINTED IN THE UNITED STATES OF AMERICA

Preface
Ever since Mullerand Borger proposedthe concept of phytoalexins in 1940,

these compounds have fascinatedplant pathologists engaged in unraveling
the mysteries of disesase resistance in plants. More than 40 years later,
Bailey and Mansfield, in their excellent treatise Phytoalexins, brought. to-
gether a number of students of phytoalexins to share their experiencesand
views on these compounds. That book gave a fillip to research on these
defensive chemicals, and much has happened since then, especially in the
area of molecular biology of phytoalexin synthesis and gene regulation.
With the great relevance of these studiesto agriculture, microbiology, bio-
chemistry, biotechnology, forestry, and horticulture, more and more re-
search laboratories are laying greater emphasison investigating phytoalex-
ins. The relevant researchliterature has increased tremendously duringthe
past two decades, and an overwhelmingly large wealth information
of exists
scattered in journals on various subjects. It is almost impossible to review
all of themand, therefore, an attempt is made hereto summarize the latest
developments and to present the state of the art of phytoalexin research
worldwide.
The approach here is that of a comprehensive handbook, but of the
vast number of higher plants around us, we have focussed on specific,
economically important species with respect to their phytoalexins, thereby
L
avoiding broad generalizations. For a specific crop, the leading investiga-
tors were asked to summarize all the available data on that plant species,
giving special emphasisto the work done in the authors’ laboratory. New
analytical techniques that have been developed as well as the latest unpub-
lished results were also included in each paper. Thus each chapter contains
muchnew information. Described are several important crops, such as
barley, pea, soybean, cotton, groundnut (peanut), mustard, Citrus, grape,
iii
iv Preface
Phaseolus, Vigna, Dioscorea, tobacco, alfalfa, redclover, potato, sweet

potato, tea, pepper, sesame; medicinal plants, such as Adhatoda, Trian-
thema, Withania, and Tylophora; and various tree species, such as teak,
Cassia,Morinda,Eucalyptus,Syzygium,Mangvera,Anogeissus,Mad-
huca, Heterophragma, Spathodea, Ziqyphus, and Carvia. The topics cov-
ered are: the role of phytoalexins in disease resistance, artificial induction
of phytoalexins, nature of elicitors, biosynthesis, detoxification, enzymatic
conversion, mode of action, and molecular approaches such as cloningof
cDNA-encoding elicitors, receptor sites in membranes, receptor substances,
cellular responsesand their regulation, signal moleculesand systems, signal
transduction and location, and regulation of disease resistance genes. A
number of new phytoalexinsare reported together with their structures and
properties.
We hope that the researchers in plant pathology, biochemistry, micro-
biology,molecularbiology,biotechnology, natural products chemistry,
chemotaxonomy, and plant breeding,will find this book greatly useful. The
methodology provided for each crop should encourage readers to devise
their own experiments for research or teaching. This book will serve as
an excellent reference for teachers and students of plant pathology and
biotechnology. It is also hoped that this book will expose the need and
promises of phytoalexin research to the scientific community and initiate
many new researchers into the exciting field of plant-microbe interactions.
We are grateful to our contributors, who complied with our demands
valiantly and courteously. We wish to thank our publisher, Marcel Dekker,
Inc., especially Christine Dunn, Production Editor, and Russell Dekker,
Editor-in-Chief, for their encouragementand expert guidance, which led to
the prompt publicationof this book.
M. Daniel
R. P. Purkayastha
Contents
Preface iii
Contributors ix
the Past 50 Years

1. Progress in Phytoalexin Research During 1
R. P.Purkayastha
2. Phytoalexins and Host Specificity in Plant Diseases 41

Hachiro Oku and Tomonori
Shiraishi
3. Elicitor and the Molecular Basesof Phytoalexin Elicitation 61

Masaaki Yoshikawa
4. Soybean Phytoalexins:Nature, Elicitation, Modeof Action,

and Role 69
Jack Paxton
5. Cellular Biochemistryof Phenylpropanoid Responsesof

Soybean to Infection by Phytophthora sojae 85
Terrence Lee Graham
6. Culture Darkening, Cell Aggregate Size,

and Phytoalexin
Accumulation in SoybeanCell Suspensions Challenged
with Biotic Agents 117
Robert M.Zacharius, William F. Fett, and PrakashG. Kadkade
7. Phytoalexins as a Factor in the Wilt Resistanceof Cotton 129

M. H. Avazkhodjaev,S. S. Zeltser, H. V. Nuritdinova,
and Raviprakash G. Dani
V
vi Contents
8. Cotton (Gossypium hirsutum)Strategies of Defense Expression:

Accumulation of UniqueAspergillusflaws Control Substances 161
H,J. Zeringe, Jr.
9. Sesquiterpenoid Phytoalexins Synthesized in
Cotton Leaves
and Cotyledons Duringthe Hypersensitive Responseto
Xanthomonas campestrispv. malvacearum 183
Margaret Essenberg and MargaretL. Pierce
10. Chemistry, Biology,and Role of Groundnut Phytoalexins

in Resistanceto Fungal Attack 199
P. V. Subba Raoand Richard N. Strange
11. Phytoalexins fromthe Crucifers 229
Thierry Rowel, Albert Kollman,and Marie-HdItke Balesdent
12. Scoparone (6,7-Dimethoxycoumarin),a Citrus Phytoalexin

Involved in Resistanceto Pathogens 263
Uzi Afek and Abraham Sztejnberg
13. Stilbene Phytoalexinsand Disease Resistance inVitis 287

Wilhelm Dercks,L. L. Creasy, and C.J. Luczka-Bayles
14. Mode of Toxic Actionof Vitaceae Stilbeneson Fungal Cells 317

Roger Pezetand Vincent Pont
15. Inducible Compounds in

Phaseolus, Vigna,
and Dioscorea Species 333
S. A. Adesanya and M. F. Roberts
16. Involvement of Phytoalexins inthe Response of

Phosphonate-Treated Plants to Infection by
Phytophthora Species 375
P.Saindrenan and DavidI. Guest
17. Phytoalexins in Forage Legumes: Studies
on Detoxification by
Pathogens and the Role of Glycosidic Precursors in Roots 391
V. J. Higgins, Dallas K. Bates, and J. Hollands
18. Induction of Phytoalexin Synthesis Medicago

in sativa
(Lucerne)-Verticillium albo-atrumInteraction 405
Christopher J. Smith,J. Michael Milton,
and J. Michael Williams
Contents vii
19. Stereoselective Synthesis of Spirovetivane-Type Phytoalexins 445

Chuzo Iwata and Yoshiji Takemoto
20. Enzymic Conversionof Furanosesquiterpene in Ceratocystis

fimbriata-Infected Sweet Potato Root Tissue 467
Masayuki Fujita, Hiromasa Inoue, K. Oba, and Ikuzo Uritani
21. Defense Strategies of

Tea (Camellia sinensis) Against
Fungal Pathogens 485
B. N. Chakraborty, Usha Chakraborty, and A. Saha
22. Effects of Age-Related Resistanceand Metalaxyl on

Capsidiol Production in PepperPlants Infected with
Phytophthora capsici 503
Byung Kook Hwang
23. Induced Chemical Resistance in Sesamum indicum

Against Alternaria sesami 525
R . K. S. Chauhan and B. M. Kulshrestha
24. Phytoalexins and Other Postinfectional Compounds

of
Some EconomicallyImportant Plants of India 533
M. Daniel
25. Possible Roleof Phytoalexin Inducer Chemicals in

Plant Disease Control 555
Asoke Kumar Sinha
Index 593
Contributors
S. A.Adesanya,Ph.D. Department of Pharmacognosy,Faculty of

Pharmacy, Obafemi Awolowo University, Ile-ife, Nigeria
Uzi Afek, Ph.D. Research Scientist, Departmentof Postharvest Science of
Fresh Produce, The Volcani Center,Bet Dagan, Israel
M. H. Avazkhodjaev, D S c . Professor, Department of Plant Immunology,
Institute ofExperimentalBiologyof Plants, AcademyofSciencesof
Uzbekistan, Tashkent, Uzbekistan
Marie-HBhe Balesdent, Ph.D.Pathologie VCgCtale, Institut National dela
Recherche Agronomique, Versailles, France
Dallas K. Bates, Ph.D. Associate Professor, Department of Chemistry,
Michigan Technological University,
Houghton, Michigan
B. N. Chakraborty, Ph.D. BotanyDepartment,University of North
Bengal, Darjeeling, West Bengal,India
Usha Chakraborty, Ph.D., F.P.S.I. Centre for Life Sciences, University of
North Bengal, Darjeeling, West Bengal,India
R. K. S. Chauhan, Ph.D. School of Studies in Botany, Jiwaji University,
Gwalior, MadhyaPradesh, India
L. L. Creasy, Ph.D. Professor of Pomology, Department of Fruit and
Vegetable Science, Cornel1 University,
Ithaca, New York
Raviprakash G. D a d , P h . D . Scientist, Plant Breeding, Division of Crop
Improvement, Central Institute for Cotton Research, Nagpur, Maharashtra,
India
ix
X Contributors
M. Daniel, Ph.D. Reader, Department of Botany, Faculty of Science, The

Maharaja Sayajirao University of Baroda, Vadodara,Gujarat, India
Wdhelm Dercks, Ph.D. Professor, Fachbereich, Gartenbau, Fachhochschule
Erfurt, Erfurt, Germany
Margaret Essenberg,Ph.D. Regents Professor, Department of
Bio-
chemistry and MolecularBiology,OklahomaAgriculturalExperiment
Station, Oklahoma State University, Stillwater, Oklahoma
William F. Fett, P h D . Research Plant Pathologist, Department of Plant
Science and Technology, Eastern Regional Research Center, Agricultural
Research
Service, U.S. Department of Agriculture, Philadelphia,
Pennsylvania
Masayuki Fujita, Ph.D. Associate Professor, Department of Bioresource
Science, Faculty of Agriculture, Kagawa University, Miki-cho, Kita-gun,
Kagawa, Japan
Terrence Lee Graham, Ph.D. Associate Professor, Department of Plant
Pathology, The Ohio State University, Columbus,Ohio
David I. Guest, Ph.D. Senior Lecturer, School of Botany, University of
Melbourne, Parkville, Victoria, Australia
V. 3. Higgins,Ph.D. Department of Botany,Universityof Toronto,
Toronto, Ontario, Canada
3. Hollands, B.Sc. Department of Botany, University of
Toronto, Toronto,
Ontario, Canada
ByungKookHwang,Ph.D. Professor, Department of Agricultural
Biology, Korea University, Seoul, Republicof Korea
Hiromasa
Inoue,
Ph.D. Assistant Professor, Laboratory of Oral
Bacteriology, Kyushu Dental College, Kitakyushu,
Japan
Chuzo Iwata, Ph.D.Faculty of Pharmaceutical Sciences, Osaka University,
Osaka, Japan
Prakash G. Kadkade, M.Sc., Ph.D. Vice President and Chief Scientist,
Phyton Inc., Ithaca, New York
Albert Kollman Pathologie Vbgktale, Institut Nationale de la Recherche
Agronomique, Versailles, France
B.M.Kulshrestha,Ph.D. Department of Botany, K. R.G.College,
Gwalior, MadhyaPradesh, India
Contributors xi
C.J.Luczka-Bayles,M.S. AssistantDirector,FlowCytometry and

ImagingFacility,Cornel1Center for AdvancedTechnologyinBio-
technology, Cornel1 University,Ithaca, New York
J. MichaelMilton,Ph.D. SeniorLecturerinBotany,Department of
Biological Sciences, University of Wales, Swansea, Wales
H. V. Nuritdinova, Ph.D. Institute of Experimental Biology of Plants,
Academy of Sciences of Uzbekistan, Tashkent, Uzbekistan
K. Oba, Ph.D. Faculty of Domestic Science, Nagoya Women’s University,
Nagoya, Japan
Hachiro Oku, Ph.D. Professor Emeritus, Laboratory of Plant Pathology
and GeneticEngineering,CollegeofAgriculture,OkayamaUniversity,
Okayama, Japan
Jack Paxton, P h D . Associate Professor, Departmentof Plant Pathology,
University of Illinois,Urbana, Illinois
Roger
Pezet,
Ph.D. Staff Researcher,
Phytopathology/Mycology
Department, Swiss FederalAgriculturalResearch Station ofChangins,
Nyon, Switzerland
Margaret L.Pierce,Ph.D. AssistantResearcher,DepartmentofBio-
chemistry and Molecular Biology, Oklahoma State University, Stillwater,
Oklahoma
Vincent Pout Staff Researcher, Laboratory of Organic Chemistry, Swiss
Federal Agricultural Research
Station of Changins, Nyon, Switzerland
R.P.Purkayastha,Ph.D.,D.I.C. Professor, Department of Botany,
University ofCalcutta, Calcutta, West Bengal,India
M. F. Roberts,Ph.D.,D.Sc. Reader,Department of Pharmacognosy,
School of Pharmacy, London University, London, England
Thierry Rouxel,Ph.D. Pathologie VCgCtale, Institut Nationalde la
Recherche Agronomique, Versailles, France
A. Saha,Ph.D. BotanyDepartment,University of North Bengal,
Darjeeling, West Bengal, India
P . Saindrenan, Ph.D. Scientific Researcher, Department of Plant Pathol-
ogy, Institut deBiolgieMolCculairedes Plantes,C.N.R.S.,Strasbourg,
France
Tomonori Shiraishi, Ph.D. Professor, Laboratory of Plant Pathology and
Genetic
Engineering,
College of Agriculture,
Okayama
University,
Okayama, Japan
xii Contributors
Asoke Kumar Sinha, PhD. Professor, Department of Plant Pathology,

Bidhan Chandra Krishi Viswavidyalaya, Kalyani, West Bengal,
India
Christopher J. Smith, PhD. BiochemistryResearch Group, Schoolof

Biological Sciences, University of Wales, Swansea, Wales
Richard N. Strange, PhD. Department ofBiology,UniversityCollege

London, London, England
P. V. SubbaRao, PhD.* Centre de CoopQation Internationale en

Recherche Agronomique pour le Dheloppement (CIIWD), Montpellier,
France
Abraham Sztejnberg, Ph.D. Associate Professor of Plant Pathology and

Head, Department of Plant Pathology and Microbiology,Faculty of
Agriculture, The Hebrew University of Jerusalem, Rehovot, Israel
YoshijiTakemoto, PhD. Faculty of PharmaceuticalSciences,Osaka

University, Osaka, Japan
IkuzoUritani, PhD.? Faculty ofDomesticScience,NagoyaWomen’s

University, Nagoya, Japan
J. Michael Williams,PhD.. Department of Chemistry, University of Wales,

Swansea, Wales
Masaaki Yoshikawa, PhD. Department of Biology, Faculty of Science,

Hokkaido University, Sapporo, Japan
Robert M. Zacharius, PhD. Principal Consultant, R. M. Zacharius and

Associates, Science Consultants, Highland, Maryland
S. S. Zeltser, PhD. Institute of Experimental Biology ofPlants, Academy

of Sciences of Uzbekistan,Tashkent, Uzbekistan
H. J. Zeringue, Jr. Chemist, Commodity Safety ResearchUnit, Southern

Regional Research Center, Agricultural Research Service,
U.S. Department
of Agriculture, New Orleans, Louisiana
+Currentq f f i l i o n : Department of Biology, University College Lnndon, London, England

?Current qffiliution: Chairman of Board of Trustees, ?chi Konan Gakuen School Corpora-
tion, Konan City, Japan
1
Progress in Phytoalexin Research
During the Past 50 Years
R. P. Purkayastha
University of Calcutta, Calcutta, West Bengal, India
1. INTRODUCTION
Several landmarks have been established in the domain of plant sciences
during the past half centurybut the discovery of phytoalexin is undoubtedly
an outstanding one since it has opened up a new vista in plant pathology.
Muller and Borger (1940) laid a firm foundation for the phytoalexin con-
cept, although in the beginning it did not receive much researchers’ atten-
tion. However, ithas been studied withan increasing awarenessfor the last
two decades but the controversy as to whether phytoalexin alone confers,
resistance to all plant diseases or this is one of the several mechanisms of
disease resistance of plants remains.
A. Some Relevant Research Prior to Phytoalexin Theory Proposed

The response of plants to fungal infection was first recognized by Ward
(1905) in England and subsequently by Bernard (191 1) in France (Cruick-
shank, 1978). The restricted and unrestricted growth of the pathogen on
resistant and susceptible hosts, respectively, clearly indicates
the differential
response of host cultivars to a pathogen. This fundamental discovery of
responsivemechanisms of plants against pathogens probably forms the
basis for the concept that an immunesystemexistsinplants.Wingard
(1928) first observed symptoms of recovery in tobacco ring spot virus-
infected plantsand presumed the infected plantsto have acquired immunity
against disease. In 1933, Chester also consideredthe possibility that plants
react to bacterial or fungal infection in the same way as animals react to
microorganisms or viral infectionsby producing specific antibodies. But he
later realized that since plants have no circulatory system,the possibility of
antibody formation in plants ishighlyimprobable.Wallace (1940) also
1
2 Purkayastha
pointed out that symptoms of recovery in tobacco plants infected with

sugarbeet curlytop virus wasdue to the formation of protective substances
in plants. It was also demonstrated that if the protective substances are
transferred to other plants they may develop passive acquired immunity
against the disease. Perhaps these reports prompted Muller and Borger to
visualize that a functionally similar mechanism comparableto the antigen-
antibody reactionin animals could also be operative in plants.
B. Experiments lead to Phytoalexin Theory and

Modern Phytoalexin Concept
Muller and Borger (1940) tested several varieties of potato tubers against
virulent and avirulent strains of Phytophthora infestans. They observed
that when the cut surfaceof a potato tuber was inoculated withan avirulent
strain of P . infestans and reinoculated after 24 hr with a virulent strain, it
failed to induce any symptom. It strongly suggestedthat the postinfectional
production of an antifungal substance caused inhibition of fungal growth
in host tissues. They have drawn a number of conclusionsfrom their experi-
ments that form the basic postulatesof the phytoalexin theory. It would be
worthwhile to include these conclusions as summarized by Cruickshank
(1963): “l.) Phytoalexin inhibitsthe growth of the fungus inthe hypersensi-
tive tissue and this is formed or activated onlywhen the parasite comes in
contact with host cells;2.) the defensive reaction takes place in living cells
only; 3.) the inhibitory substance may be regarded asthe product of ‘necro-
biosis’ ofthe host cells;4.) phytoalexin is nonspecific in its toxicity towards
fungi; 5.) basic response of resistant and susceptible plants is similar
but the
speed of phytoalexinformation differs; 6.) the defense reactionis restricted
to the tissue colonized bythe fungus and its immediate neighborhood;7.)
the resistant state is ‘acquired’ after attempted infection and is not ‘inher-
ited’; 8.) the speed of host reaction is determined by the sensitivity of the
host cell. This is specificand genotypically determined.” Accordingto the
original definition of Muller and Borger (1940)’ phytoalexin is a chemical
compound produced by living host cells only when theseare invaded by a
parasite and consequently necrobiosis occurs.
Muller (1956) redefined phytoalexins as “antibiotics” that are the result
of an interaction of two different metabolic systems, host and parasite, and
that inhibit the growth of microorganisms pathogenic to plants. The term
phytoalexin (PA) means “warding off compound in plants” (phyton, Gr.
for “plant”; alexin, Greek for “warding off compound”). Various defini-
tions of phytoalexin have been proposed earlier by different researchers.
But considering the present state of knowledge it has become imperative
to modify its original definition. Phytoalexin may now be defined as an
Progress in Research 3
antimicrobial, low molecular weight, secondary metabolite formed de novo

as a result of physical, chemical, or biological stress which resists or sup-
presses the activity of invaders, and its rate of production/accumulation
depends eitheron host genotypes or both host and pathogen genotypes. It
should be borne in mind that the same pathogen produces more PA in
resistant cultivarsthan in susceptible ones. Again, two pathogens produce
different amounts of PA in the same host cultivarand sometimes the host
alone can also producePA due to physical or chemical stress without any
contact with the pathogen or its metabolites. A pathogen may be moreor
less sensitive to PA depending on its detoxifying ability. Phytoalexins are
regarded as a class of compoundsthat are of diverse chemical nature. The
occurrence of similar phytoalexin in different hosts and different phyto-
alexins in the same host have also been reported. Whatever may be the
chemical nature of phytoalexins, the major questions are 1.) whether PA
production is the only mechanism of resistance or immune response of a
host to disease or is one of the multicomponent and coordinated mecha-
nisms of disease resistance, 2.) whether PA is really involved in disease
resistance of all plants,and 3.) whether PA could be successfully exploited
for crop protection by induction or direct applicationor by genetic manipu-
lation. If so, which process would be most effective, simple,durable, non-
phytotoxic, and inexpensive remainsto be workedout.
C. Early Work on Phytoalexins
Muller and Borger (1940) first detected an antifungal substance known as

phytoalexin in a potato tuber which was produced in responseto a fungal
infection. Subsequently they carried out a seriesof experiments (1940-1961)
to establish that phytoalexin was a defensive substance formed in plants
only when they were attackedby an organism. From 1960 onward Cruick-
shank and his associatesat the Division ofPlant Industry, CSIRO, Australia
continued researchon phytoalexins and identified twonew phytoalexins-
pisatin and phaseollin- which they isolated from pea and bean, respec-
tively. A number of paperson phytoalexins werealso published by Uehara
(1958a-c) from the Hiroshima Agricultural College, Japan. In 1963, Klar-
man and Gerdemann at the University of Maryland reported that resistance
of soybean to three Phytophthora species was associated withthe produc-
tion of phytoalexin. In England, Purkayastha and Deverall initiated work
on phytoalexin at the Imperial College of Science and Technology, London.
They (1964, 1965) first demonstrated that the steady growth of Botrytis
fabae and inhibitionof growth ofB. cinerea on leaves of Viciafaba (broad
bean) were due to the production of phytoalexin. Earlier, Gaumann and
Kern(1959a,1959b) at the EidgTechnicalHochschule,Zurichisolated
4 Purkayastha
orchinol, an antifungal substance, from the infected (by Rhizoctonia re-

pens) tuber of Orchis militaris. Similarly, another antifungal substance was
also detected in infected (by Ceratocystisfimbriata) roots of sweet potato
by. Hiura (1943), which was identified by Kubota and Matsuura (1953) as
ipomeamarone. Both ipomeamarone and orchinol were later considered
as phytoalexins. Although antifungal substances were detected in several
infected plants much before the phytoalexin concept came into existence,
these werenot regarded as induced defensive substances.
D. Merits and Demeritsof Phytoalexin Research

It is an undeniable fact that formation of new antimicrobial substances
(may be phytoalexins) in plants as a result of any stress isan indication of
defense response. Hence activation of defense by any biotic or abiotic agent
could be considered asa nonconventional methodof disease control. Both
resistant and susceptible hosts respondto their parasites (biotic agents) but
the speed of response is always much higher in resistant cultivars than
susceptible ones.It suggests that host genotypes react differently to a patho-
gen. A higher rate of phytoalexin production in plants may be deemed as
an additional biochemical parameter for screening disease-resistant germ-
plasm. Some phytoalexin-inducing chemicals (nonphytotoxic)are now be-
ing used to treat seeds or plants (asfoliar spray) for inducing resistance.
Besides, phytoalexinis a vast source of new natural products which could
be exploited for some useful purposes. Some of them are known chemo-
therapeutants. Highphytoalexin-yieldingcultivarmaybeselected for
breeding purposeas a source of resistance cultivar. Biogenetic relationships
in plantsare now determinedon the basis of the chemicalnature of phyto-
alexin, producedby the plant. For instance, Leguminosae (Fabaceae), Sola-
naceae,Compositae, and Convolvulaceaeproduceisoflavonoids,terpe-
noids, polyacetylenes,and furanosesquiterpenoids, respectively.
Several reports of the phytotoxic nature of PA, nondetection of PA in
some infected host plants,rapid inactivation/degradation of PA by fungal
enzymes or field conditions, the problem of identification of PA due to
nonavailability of adequate amounts of material, and very little use of
phytoalexins as chemotherapeutants have discouraged many phytoalexin
workers in recent years. Besides, it is also doubtful as to whether phyto-
alexin alonecontrols plant resistance.
II. PROGRESS IN PHYTOALEXIN RESEARCH: A N OUTLINE

A survey of literature reveals the gradual progress in phytoalexin research
during the past 50 years. At the onset, a brief outline is given below:
1940-1959: 1.) Detection, isolation, characterization; identification of new

antifungal compounds (phytoalexins) and bioassay; 2.) physiological
factors affecting PA production; 3.) testing sensitivity of different par-
asiteshonparasites to PA; 4.) induction of PA; 5.) mode of action; 6.)
relation with disease resistance.
1960-1969: 1-6 continued. 7.) Mechanism of PA induction; 8.) host speci-
ficity and PA; 9.) biosynthesis of PA; 10.) degradation of PA; 11.)
structure determination of PA in same host by different fungi; 12.)
changes in host metabolismdue to PAproduction; 13.) association of
gene withPA production.
1970-1979:1-13 continued. 14.) PAL and PA; 15.) PA production by
bacteria and viruses; 16.) metabolism of PA; 17.) disease control by
PA; 18.) elicitation of PA; 19.) isolation and characterization of elici-
tors; 20.) structure of elicitor; 21.) PA as a taxonomic character; 22.)
PA and immunity; 23.) suppression of PA induction; 24.) mode of
action of PA elicitors; 25.) tissue specific PA; 26.) elicitors of PA
from bacteria; 27.) enzymes and detoxification of PA; 28.) PA-plant
antigens -disease resistance.
1980-1989: 1-28 continued. 29.) Race-specific PA elicitors; 30.) mecha-
nisms for the phytotoxicity of PA; 31.) factors affecting elicitationof
PA; 32.) radioimmunoassay for the PA; 33.) PA chemistry and mode
of action.
1990: 1-33 continued; 34.) Sensitive and rapid assay for antibacterial activ-
ity of PA; 35.) PA production in progeny ofan interspecific cross;36.)
PA in nodules; 37.) antisera raised against resveratrol (a phytoalexin);
38.) gene-encodingenzymesof PA biosynthesis(i.e.,phenylalanine
ammonialyase, 4-coumarate:CoA ligase, chalcone synthase, chalcone
isomerase, stilbene; synthase, 3-hydroxymethyl-3-glutarylcoenzyme A
reductase); 39.) isolation of stilbenesynthasegenes from grapevine
and transfer to tobacco for increasing disease resistance in transgenic
plants.
Extensive reviewson phytoalexins have been published by several work-

ers (Cruickshank, 1963,1978,1980; KuC, 1972,1976; Ingham, 1972,1973,
1982; Deverall, 1972, 1976; Purkayastha, 1973, 1985, 1986; Van Etten and
Pueppke, 1976; Keen and Bruegger, 1977; Harborne and Ingham, 1978;
Keen, 1981; Bailey and Mansfield, 1982; Sequiera, 1983; Yoshikawa, 1983;
KuC and Rash, 1985; Ebel, 1987; Wood, 1986; Mahadevan, 1991; Dixon,
1992). This chapter is intendedto highlight the important.developmentsin
phytoalexin research duringthe past 50 years.
It appears from previous reviews that during the first 20 years phyto-
alexin research was restricted mainly to detection, isolation, identification,
6 Purkayastha
and characterization of a very few phytoalexins,factors affecting PA pro-

duction, sensitivity of microorganisms to PA, and induction of PA for
disease resistance in plants. Since1970 scientists of several disciplines have
realized the importance ofphytoalexins and havedevelopednovelap-
proaches to study these natural products. Papers published up to 1980
could be classified into six broad categories: 1.) isolation of PA from a
number of plant families (morethan 20); 2.) biosynthesis of PA;3.) toxicity
of PA; 4.) detoxification of PA; 5.) role of PA in disease resistance; 6.)
elicitors and elicitationof phytoalexins by fungi, bacteria or their metabo-
lites, viruses, animals, chemical and physical agents. During the last 10 or
12 years a few more aspects have been added to PA research as mentioned
earlier.
111. RECENTRESEARCH ON ELICITATION OF SOME

COMMON PHYTOALEXINS
Elicitation of PA is usually caused by an elicitor, which may beeither biotic
or abiotic. A pathogen metabolitethat stimulates PA production in a host
may be definedas an elicitor accordingto Cruickshank (1980). But the term
specific elicitor has been used for metabolites that stimulate “differential
phytoalexin production on various host cultivars similarto the race of the
fungus that produces them” (Keen, 1975). The presence of PA elicitor in
cell-free extracts of fungi was first demonstrated by Uehara (1959), al-
though the term elicitor was not mentioned. Subsequently, many workers
have reported elicitation of PASby various elicitors.It is interesting to note
that production of similar PA could be induced in a host by both biotic
and abiotic agents. How is it possible? The mechanism is not yet clearly
understood. According to Albersheim et al. (1986), plants can recognize
oligosaccharide fragments of fungal cell walls that are released by enzymes
present constitutively in the cellwallsof plants. Apart from this, fungi
and bacteria can also secrete enzymes capable of releasing oligosaccharide
elicitors from the cell walls of plants. Sometimes injury of plants by mi-
crobes can causethe release of plant enzyme. This enzyme solubilizes frag-
ments of plant cell walls that elicit PA production/accumulation. Albers-
heim et al.(1986) also suggestedthat abiotic elicitors such as heavy metals,
UV light, organic solvents, and freezing are likely to work by activating
plant enzymes that release oligogalacturonide elicitors from the walls of
plant cells. There are several mechanisms by which PA accumulation can
be activated. It is also not unlikely that a similar change may occur inthe
repetitive DNA (KuC, 1987) due to treatment by more than one substance
prior to the production of new messenger RNA. However,the sequence of
events involved in the biosynthesis of PA could be studied at a molecular
level. Several working models have been proposed (Albersheim and Ander-
sonProuty, 1975; Yoshikawa, 1983; Ellingboe, 1982; Purkayastha, 1986;
Gabriel etal., 1988) but all of them remainto be confirmed.
The rate of production of PASis generally higher in case of incompati-
ble host-parasite interaction (Akazawaand Wada, 1961; Cruickshank and
Perrin, 1968; Partridge and Keen, 1976; Purkayastha et al., 1983). It is not
unreasonable to speculate that elicitor molecules of avirulent pathogens are
probably more stimulatory to PA biosynthesis. The degree of stimulation
may depend on several factors such as quantity of elicitors released bythe
organism; speed of release; chemical nature of the elicitor; presence or
absence of receptors inthe host cell membrane, if present; strong or weak
response of receptor;duration of treatment; and environmental conditions.
Several comprehensive reviews pertaining to elicitors of PAS have been
published (Albersheim and Anderson-Prouty, 1975; Callow, 1977; Darvill
and Albersheim, 1984; Keen and Brueggar, 1977; Purkayastha, 1986;
Yoshikawa, 1983; alsoseeSmithet al., Yoshikawa and Paxton in this
volume). Therefore, a brief account of the elicitation of some common
phytoalexins (Fig.1) is given below.
A. Pisatin
The isolates of Fusarium solani which differed in their pathogenicity also
showed differential pisatin-eliciting potential.It was confirmed when their
culture filtrates were tested on pea (Daniels and Hadwiger, 1976). There
was a difference in the concentration of elicitor in the culture filtrates of
isolates. The elicitor was fairly heat-stable and also stable in freezing,but
eliciting activitywas reduced significantly by pronase digestion. This strong-
ly suggests that some of the activities were due to proteinaceous compo-
nents. An interesting observationwas made by Shiraishi et al. (1978) who
detected both elicitor and suppressor of pisatin inthe pycnospore germina-
tion fluid of Mycosphaerella pinoides (pea pathogen) (Crute et al., 1985).
The elicitor and suppressor were highand low molecular weight substances,
respectively. The suppressor usually counteracts the activity of elicitor. The
components of suppressor were identified as low molecular weight peptides
which inhibited pisatin accumulation in pea leaves when inoculated sepa-
rately with nonpathogens likeErysiphe graminis hordei and Stemphylium
sarcinaeforme. As a result pea leaves became susceptible to S. sarcinae-
forme. Yamoto et al. (1986) demonstrated that pisatin could be induced in
pea leaves by elicitors fromMycosphaerella pinoides,M . melonis, and M.
ligulicola. Accumulation of pisatin increased after removal of epidermis
and application of elicitors (high molecular weight compound > 10,OOO Da)
from germination fluid of the fungus. The authors also studied the effect
8 Purkayastha
0
Pisatin
CH; CH3
Phaseollidin Kievitone
*b I
Glyceollin un
Medicarpin
Ho%o>
H0
' 0' CH3 CH2
Maackiain Rishitin
Figure 1 Structures of some common phytoalexins.
of a suppressor (low molecular weight compound< 10,OOO Da) of pisatin

obtained from the germination fluidof M. pinoides which counteractedthe
activity of pisatin elicitors.
Elicitation of pisatin by various abiotic elicitors such as metallic salts
(Uehara, 1963); mercuric chloride, sodium iodoacetate, sodium selenate
(Perrin and Cruickshank, 1965); actinomycin D (2 x lo-' M)(Schwochau
and Hadwiger, 1968); cupric chlorideand ethylene (Chalutzand Stahmann,
1969); and synthetic peptides (Hadwiger et al., 1971) has been reported.
Hadwiger et al. (1976) raised an induced mutant of pea by treatment with
sodium azide which was able to produce more pisatin without extraneous
supply of sodium azide. Elicitation of pisatin by UV light was also noted
I
OH
CH3
Lubimin Capsidiol
Casbene
0
gCH2
CH3 CH3
H2
0
Momilactone A Momilactone B
by Hadwiger and Schwochau (1971). It appears from the above statements

that pisatin production could be elicited in plants by physical, chemical,
and biological agencies.
B. Phaseollin
An elicitor of phaseollin was isolated from the mycelial walls and culture
filtrates of Colletotrichum lindemuthianum,which wasidentified asa poly-
saccharide. The molecular weight varied between 1 million and 5 million
DA, and consisted predominantlyof 3-and 4-linked glucosyl residues( A n -
derson-Prouty and Albersheim, 1975). An amount equivalent to 100 ng of
glucose eliciteda similar responsein the bean tissue.
The regulation system of phaseollin synthesis in cell suspension cultures
of dwarf french bean(Phaseolus vulgaris)was studied by Dixon and Chris-
topher (1979). Considerableamount of phaseollin accumulated when french
bean was treated with an elicitor from the cell wall of C. lindemuthianum.
10 Purkayastha
But the elicitors isolated from the cell walls of P. megasperma var. sojae
and Botrytis cinerea were less effective.
A carbohydrate-rich extracellular component from a raceof C. linde-
muthianum showed a high level ofPA activity on a resistant cultivar “Dark
Red” of kidney bean but not on the susceptible cultivar “Great Northern.”
Other extracellular components were also recognized as elicitors by both
cultivars. It is noteworthy that the two cultivarsof Phaseolus vulgaris dis-
played a differential response to extracellular components. These observa-
tions support the hypothesis that both general and specific mechanisms
exist in race-cultivar interaction (Tepperand Anderson, 1986).
A glucan was isolated from the cell wall extracts of Fusarium oxy-
sporum f. sp. lycopersici (Anderson, 1980a)and a polypeptide (monilicolin
A) from mycelia of Monilinia fructicola (Cruickshank and Perrin, 1968).
Both compounds elicited phaseollin production. Apart from biotic elicitors,
several abiotic agents such as mercuric chloride (Fraile et1980),
al., abscisic
acid, benzylaminopurine, silvernitrate (Stoessel and Magnalato, 1983), ox-
adiazone, herbicides (Rubin et al., 1983; Ricci and Rouse, 1983), ozone
and SO2also elicited phaseollinin P . vulgaris.
C. Clyceollin
An elicitor of glyceollin was isolated from the mycelial cell wall of Phy-
tophthora megaspermavar. sojae by Ebel et al. (1976). This elicitor stimu-
lated the activity of phenylalanine ammonialyaseand also induced glyceol-
lin production in soybeancell cultures. They concluded that the action of
elicitors is not species or variety specificbut is a part of the general defense
response of plants. Ayers et al. (1976a) also isolated a branched 0-glucan
elicitor from both cell walls and culture filtrates of P. megasperma. The
elicitor was considered to be predominant by a 3-3, 6-linked glucosyl resi-
due. The hyphal cell walls of P. megasperma f. sp. glycinea (Pmg) also
showed considerable eliciting activity which was due to three classes of
carbohydrates. The activities of other glucans or glucomannans and man-
nans obtained from other sources were much lower as reported by Keenet
al. (1983). Chitin and Chitosan also actedas elicitors ofPASin other plants
but their activities were very low in the case of soybean cotyledons.
Purkayastha and Ghosh (1983) reported elicitor activity of fresh myce-
lial wall extract of Myrothecium roridum. Spores suspended in mycelial
wall extract, drops placed on leaf surfaces of soybean, and incubated for 48
hr. The results of bioassay test revealedthat the spores suspended in myce-
lial wall extract were more inhibitory than the spores suspended in sterile
distilled waterand incubated on leaf surfaces for a similar period. Mycelial
wall extract induced greater production of glyceollin in soybean leaves.
Soybean leaves treated with only mycelial wall extract and mycelial wall
extract containing conidia ofColletotrichum dematium var. truncata pro-
duced 73.22 and 217 pg/g (fresh wt) glyceollin, respectively,after 48 hr of
incubation (Purkayastha and Banerjee, unpublished).
There is evidencethat oligogalacturonides derivedfrom the pectic poly-
saccharides of plant cell walls can serve as regulatory molecules that induce
glyceollin accumulation in soybean (Davis et al., 1986). This is inconsistent
with the hypothesis that oligogalacturonides play a major role in plant
disease resistance.
An elicitor was also extracted from wounded, frozen cotyledons of
soybean. When extract was heated at 95OC for 10 min its eliciting activity
was lost. It suggests that the elicitor is thermolabile. But addition of calcium
chloridewithelicitorenhancedelicitingactivity.Lyon and Albersheim
(1982) observed maximum accumulation of glyceollin when elicitor was
applied immediatelyafter the cuttingof soybean cotyledons.
Elicitors extractedfrom the cell wallsof Saccharomyces cerevisiaewere
identified asstructural glucans. Theseare able to stimulate glyceollin accu-
mulation in soybean (Albersheim et al., 1978). Specific elicitorsof glyceol-
lin were also detected in the cellular envelops of incompatible races of
Pseudomonas glycinea. However, elicitor activity couldnot be detected in
lipopolysaccharide preparations, exopolysaccharide fractions, or the cul-
ture fluids of various races of P. glycinea. Elicitors were solubilized with
sodiumdodecyl sulfate and then preparations fromfivebacterialraces
excepting one had similar specificity for elicitation of glyceollin in cotyle-
dons of two soybean cultivars (Brueggar and Keen, 1979). These observa-
tions suggest that elicitors are not always race-specific.
Cahill and Ward (1989) compared the release of glyceollin elicitors into
culture fluids of a metalaxyl-sensitive and a tolerant (50- > 500 pg/ml)
isolate of Phytophthora megasperma f. sp. glycinea following addition of
metalaxyl to the culture medium. The elicitor activity was increased mark-
edly by metalaxyl treatment in culture fluidsthe of sensitive isolate butnot
in that of tolerant isolate. Most elicitor activity was detected in fractions
(obtained by gel filtration) correspondingto the second ofthe two carbohy-
drate peaks. These findingsare very interesting because stimulationof elici-
tor activity may cause more accumulation of PA which in turn may stimu-
late host defense responses.
Among the abiotic elicitorsof glyceollin, gibberellic acid, sodium azide
(Purkayastha and Chakraborty, 1985), cloxacilin (Purkayastha and Baner-
jee, 1990), and benzyl penicillin (Purkayastha and Banerjee, unpublished)
are already known.Benzyl penicillin (100 ppm) and cloxacillin (100 ppm)-
treated soybean leaves produced 17.67 pg/g (fresh wt of leaves) and 35.25
pg/g glyceollin, respectively. Ghosh and Purkayastha (1990) demonstrated
12 Purkayastha
that ajmalicine (100 ppm), an alkaloid, could induce glyceollin(84.80 pg/g

fresh wt) insoybeanleaves.Theystated that cadmiumchloride M)is
also an,elicitorof glyceollin (67.52 pg/g) (Ghosh and Purkayastha, 1992).
D. RishitinandLubimin
Elicitation of rishitin formation in potato by a glucan and a lipoglycopro-
tein isolated from the cell wall and cytoplasm of Phytophthora infestans
was noted earlier by Chalova et al.(1976, 1977). The elicitors of terpenoid
PASsuch as rishitinand lubimin, were also obtainedfrom autoclaved soni-
cates of P. infestans, P. parasitica, Pythium aphanidermatum, Achylafla-
gellata, and Aphanomyces euteichesand were able to elicit PA production
in potato tuber slices. Potatoes treated with glucan extractedfrom P . meg-
asperma var. sojae produced a greater amount of rishitin (29 pg/g freshwt)
while glucan produced a lower amount (19.5 pg/g fresh wt) (Cline et al.,
1978).
Elicitors of rishitin could playan important role in regulating defense
reaction in potatoes (Terekhovaet al., 1980). Two potato cultivars [Temp.
(R.I.) and Belorusskii rannii (r)] having two different resistant genes were
inoculated with two races of P. infestans. The tubers were extracted with
acetone and the acetone extract contained more rishitin-inducing substances
than alcohol extract of the same. It was suggested that the intensity of
release of rishitin inducers depended on the races ofP. infestans.
When spores ofP. infestans were killed by freezing, followed by thaw-
ing, only sporangia and cystospores elicited terpeneformation in the slices
of potato. Even dead cystospores were also ableto elicit rishitin and lubi-
min accumulation in potatoes (Henfling et al., 1980). Bostock et al. (1982)
detected two fatty acids (eicosapentaenoic and arachidonic acids) in the
mycelia of P. infestans which elicited sesquiterpenoidPA accumulation in
potatoes.
An interesting observation was made by Metlitskii et al. (1984). They
detected two types of substances in the cell walls of P. infestans, namely,
elicitors and suppressors of defense reaction in potatoes. Elicitors were
extracted from six races differing in genes for virulence in potatoes. Eventu-
ally these elicitors were tested on tuber discs carryingthe resistant gene R,.
The extracts from incompatible races acted as elicitors of rishitin while
those from the compatible races acted as suppressorsof rishitin. Eliciting
activity was found to be 50% higher than the control while the suppressing
activity reached up to 200%. A pronase-sensitive, heat-labile elicitor of
rishitin was obtained from the germination fluid of P . infestans. Rishitin
accumulation in potato tuber slices was observed when treated with this
elicitor. Gel permeation chromatography of germination fluid revealed the
presence of several substances. However, active substances were precipi-

tated with ammonium sulfate (Osman and Moreau, 1985). l
An important role of lipid and nonlipid componentsof mycelial extract

of P. infestans in the elicitation of PA in potato tuber tissue was pointed
out by Bryan et al. (1985). There were two fractions of mycelial extract:
lipid-containingand lipid-free fractions. The most activefraction was com-
posed of carbohydrate, proteins, and lipid. A heat-releasedpreparation of
mycelia containing very low levels of eicosapentaenoic, arachidonic, and
dihomo-y-linolenic acids werefound to be more activethan a lipid extract
of the mycelium. In1986 Woodward and Pegg demonstrated that mycelial
extract and culture filtrates of Verticillium alboatrurn were able to elicit
rishitin production inboth resistant and susceptible isolines oftomato. The
fungus grown in Czapek-dox medium for 33 days produced high molecular
weight substance which elicited larger quantities of rishitin in susceptible
than in resistant plants. But low molecular weight glucan obtained.from
14-day-old culture filtrate of the fungus produced higher 'levels of rishitin
in both resistant and susceptible lines. The nonspecificity of low molecular
weight glucans is thought to preclude their involvement in single-gene resis-
tance mechanisms.
Mycelia and spores of incompatible races of P. infestans and Helmin-
thosporium carbonurn induced rishitin and lubimin production in potato.
The eliciting activity of sporesand mycelia of H . carbonurn was lost when
treated with heat, ethanol, or liquid NZ.But P. infestans retained its elicit-
ing activity evenafter heat or ethanol treatment. Gas chromatography mass
spectrophotometric analysisof mycelial extract of H. carbonurn failed to
detect the presenceofelicitors(arachidonicacid and eicosapentaenoic
acid). The most significant observationwas that inoculation of potato with
a compatible race of P. infestans suppressed'rishitinand lubimin accumula-
tion in responseto subsequent inoculation with an incompatible raceof the
same fungus. According to Zook and KuC (1987), it is unlikely that the
suppression by P. infestans is due to inhibition of the pathway for the
synthesis of rishitinand lubimin inpotato tissue.
In 1987, Rohwer et al. detected accumulation of rishitin and several
structurally related sesquiterpene derivative in potato tuber when treated
with culture filtrate,of P. rnegasperma f. sp. glycinea but the rate of accu-
mulation was rapid in non-host-incompatible interaction, less rapid in host-
incompatible interaction,and slow in compatible interaction. The cause of
differential response .of host and nonhost cultivarsto an elicitor remains yet
to.be elucidated.
Among the abiotic agents mercuric acetate (Cheemaand Haard, 1978)
was found to be an elicitor of potato PAs-rishitin and lubimin, Calcium
and strontium ( S 3 ' ) ions also enhanced rishitin accumulation but not lubi-
14 Purkayastha
min in potato tubers when treated with arachidonic acid (Zook et al., 1987).
The same cations, however, in the presence of poly-L-lysine(PL) increased
lubimin accumulation even greater than rishitin. On the contrary, Mg2+,
which did not affect arachidonic acid, elicited both rishitin and lubimin.
But it inhibitedthe accumulation of PAS after treatment with PL. Zook et
al. (1987) suggested that the mobilization of calcium may play a central
regulatory role inthe expression ofPA accumulation inpotato tissues after
elicitation.
E. Momilactone
Momilactone (Aand B) and oryzalexin (A,B, C, D) are two knownPASof
rice. Induction of momilactone A in rice coleoptiles and leaf sheaths by
gibberellic acid (GA,)was recorded by Ghosal and Purkayastha (1984).
Since GA is a degraded diterpene (Birch et al., 1958) it may as aactprecur-
sor. Besides, gibberellin-mediated enzyme production may also account for
the elicitation of momilactone biosynthesis in plants. A fungicide known
as 2,2,-dichloro-3,3-dimethylcyclopropanecarboxylicacid (WL 28323) has
been found to activate the natural resistance of rice plants against blast
disease causedby Pyricularia oryzae (Cartwright et al., 1977). The activity
of this fungicide is unique because it doesnot itself stimulate momilactone
production but rather increases the capacity of rice plants to synthesize
more momilactones in response to fungal infection. Sodium azide and x-ray
also elicited momilactone in rice
(Purkayastha and Ghosal, unpublished).
F. Other Phytoalexins
A high molecular weight elicitor of casbene (a castor PA) was isolated from
3-day-old culture of Rhizopus stolonifer. The purified fraction showed
eliciting activity which contained both protein and carbohydrates. Heat
treatment (at 6OoC or higher) for 15 min inactivated the elicitor. Higher
concentration (2 x lo-* M), however, increased (about 14-fold) casbene
synthetase activity in extractsof treated split seedlings (Stekoll and West,
1978).Victorin,ahost-specifictoxinobtainedfrom Helminthosporiurn
victoriae, was tested for its ability to elicit the oat PA, avenalumin. This
toxin elicitedPA in Avena sativaas reportedby Mayama et al. (1986).
An elicitor detected in the culture filtrate of Chaetomium globosum
was found to be heat-labile. It is interesting to note that the elicitor activity
was correlatedwith the activityofpectinolyticenzymesin the culture
filtrate of the said fungus.The culture filtrates inducedthe release of heat-
stable elicitors from carrot cell homogenates. The eliciting activities of cul-
ture filtrates of Botrytis cinerea, Fusarium moniliforme, and Helmintho-
sporium oryzae were also recorded earlier (Amin et al., 1986). Elicitation
of capsidiol formation in fruits of Capsicum annuum L. by copper sulfate

(0.1 M),sodium nitrate (0.1 M), and chloramphenicolwas demonstrated by
Watson and Brooks (1984).
In 1986, Kessmann and Barz found that polymeric compounds ob-
tained from Ascochyta rabiei could elicit PA production in cotyledons of
Cicer arietinum. Even wounding could also induce PA. Accumulation of
isoflavones biochanin A, formonetin, and their glucoside conjugates was
detected in infected cotyledons. Wounding caused additional accumulation
of the pterocarpan PAS, medicarpin and maackiain.
W. STRUCTURE OF ELICITOR
Although a large numberof biotic and abiotic elicitors of PAS have been
reported by previous workers, very little work has beendone so far on the
structure of elicitors. Albersheim et al.
(1986) presented the structure of
elicitor-active heptaglucosidesand identified the structure of active elicitor
of PA. The structure is given in Fig.2.
G+ 6G+ 6G+ 6G+6 G 4

3 3
4 4
G G
G+ 6G+6G*6G+6G+ G 6G- 6G+6G+6G+

3G
+
: 1G
3
4
G
G-6G
+6G+
G + 6G + 6G+6G4 6G+
G-*6G+6G
7 3
+
G
W
G G
Figure 2 Structures of eight hepta-@-glucosidealditols. Structureshownin box
(right top) was active as an elicitor of PA accumulation. The seven other hepta-8-
glucoside alditolswere inactive as elicitors, even at the concentration25 times more
than thatof the active elicitor (Albersheimet al., 1986).
16 Purkayastha
V. DETOXlFlCATlONlDECRADATlONOF PHYTOALEXIN
A N D ROLE O F ENZYMES
Generally, PAS have been considered as a possible means by which plants
attempt to defend themselves against parasiticattack. But some pathogens
or strains of a pathogen are moderately or highly tolerant to these toxic
compounds (phytoalexins) while others.are sensitive to them. It has been
conclusively demonstratedin a few diseases that virulent parasites canme-
tabolize these toxic compounds and hence it is presumedthat these parasites
could have their own detoxification mechanisms. The study of such mecha-
nisms of parasites is important because 1.) pathogenicity of an organism
may depend on its ability to detoxify the host's toxic compounds, 2.) con-
trol of detoxification mechanism may help to enhance disease resistanceof
host plants,and 3.) genes for PA detoxification might be usedthe in patho-
gens for biocontrol of weeds in the crop field as suggested by Van Etten et
al., 1989; (seealso Higgins et al. inthis volume).
Several pathogensof pea could metabolize PASsuch as pisatin, maacki-
ain, and medicarpin. For instance, Nectria hematococca can detoxify PAS
as follows:
terocarpans Pisatin
(pea) N. haematococca
erocarpans
es pea)
Maackiain
Medicarpin
(chick
]
(chick pea)
Nectria
haematococca
[ la-hydroxydienones
rabiei
Pisatin demethylase,an enzyme of N. haematococca (an ascomycetous

fungus), is a microsomal cytochrome P450 monooxygenase (Matthews and
Van Etten, 1983). Thisenzymecatalyzespisatindetoxification.Usually
Pda genes of N; haematococca code for cytochrome P450 isozymes. All
ascospore progeny of N. haematococca that werehighly or moderately
virulent on pea were Pda+ and therefore tolerant to pisatin (McIntosh et
al., 1989). A phenotype which was referred to Pda- showed lack of
' a s
ability to demethylate pisatin, such isolate of N. haematococca was sensi-

tive to pisatin and also nonpathogenic on pea (Van Etten et al., 1980).
Evidence suggests that N. haematococca DNA fragment that confers the
ability to demethylate pisatin containsthe structural gene for cytochrome
P450. Apart from N. haematococca, Ascochyta pisi, and Fusarium oxy-
sporum f. sp. pisi can also degrade pisatinby demethylation (Fuchset al.,
1980; Sanz Plater0 and Fuchs, 1978). However, P. pisi and Rhizoctonia
solani differ significantly in their rate of demethylation and in virulence.
These findings indicatethat each parasite requiresa specific detoxification
system for its pathogenicity. Thereare reports that some pathogens cannot
metabolize the PAS of their own hosts. Therefore, PA metabolism cannot
be strictly correlated with host specificity (VanEtten et al., 1989).
Fusarium solani f. sp. phaseoli can detoxify all four PAS ofbean
(Phaseolus vulgaris),namely kievitone, phaseollin, phaseollidin, and phase-
ollin isoflavan. An extracellular enzymatic system is involved in conversion
of kievitone to kievitone hydrate (Fig. 3). Generally, kievitone hydrate is
apparently mediatedby a single enzyme designated as kievitone hydratase.
Detoxification of kievitone occurs by hydration of the isopentenyl side
chains (VanEtten et al., 1989).
A few pathogens (F. solani f. sp. phaseoli, Colletotrichum lindemuthia-
num) and nonpathogens (Saptorianodorum,Stemphyliumbotryosum)
have been foundto detoxify phaseollin. The initial metabolites of phaseol-
lin producedby the aforesaid organisms are shown below.
Phaseollin
F. solani f. sp.
phaseoli
la-hydroxyphaseollone C. Iinde
(van den Heuvel rnuthianum
et al., 1974)
6a-Hydroxyphaseollin S.
(Burden etal., 1974) nodorum
12.13-Dihydro- S.
dihydroxyphaseollin botryomm
(Bailey etal., 1977)
Phaseollin isoflavan
(Higgins et al., 1974)
Detoxification of potato phytoalexin lubimin byGibberella pulicaris(ana-

morph: Fusarium sambucinum) has also been reported by Desjardins et al.
(1989). The pathwayof lubimin metabolism has been proposed,the begin-
ning of which is shown as follows:
18 Purkayastha
H0
by Fusarium
-S
’
OH O ’ OH OH 0 /
te Kievitone Kievitone
-
H o q $
Lubimin
by Phytophthora
15-dihydrolubimin
H0 a
Copsenone
Copsi d io1
by Botrytig
cinemaP
’ 1-1 0
Figure 3 Detoxification of phytoalexins by different fungi.
1 lubimin -cyclodehydroisolubimin
LubiminL 2-dehydro-
lubimin
Lubimin-sensitiveisolate of G. pulicaris metabolizedlubiminrelatively

slowlyandproducedonly15-dihydrolubiminandisolubimin.Boththe
products were toxic to theisolate but not cyclodehydroisolubimin. The
reduction of the sesquiterpenoid PA lubimin has been demonstrated using
Phytophthora capsici, Glomerella cingulata, and F. sulphureum by Ward

and Stoessl (1977). The transformation of PAS by reductive reaction has
been studied inthe metabolism of wyerone derivatives byB. cinerea and B.
fabae. Both these fungi are able
to reduce the ketones wyeroneand wyerone
epoxide to the respective alcohols wyerol
and wyerol epoxide:
B. fabae
wyerone
reduced acidWyerone
(Vicia faba)
B. fabae
l epoxide Wyerone
B. cinerea and B. fabae
wyerone
Ward and Stoessl(l977) reported that capsidiol (pepperPA) was degraded

to capsenone in shake culturesof B. cinerea and F. oxysporum f. sp. vasin-
fectum but degradation was not detected in the cultures of Phytophthora
capsici or Monilinia fructicola.
B. cinerea
Capsidiol
(Capsicum fmtescens)
Ipomeamarone, a sweet potato PA, was degraded by Botryodiplodia

theobromae and also by B. cinerea. It was detected when this phytoalexin
was addedto the mycelial suspension of these fungi and incubated for 24 hr
at 25OC. The degraded product was not identified.
Phytoalexin degradation/detoxification and fungal pathogenicity have
been discussed in detail by several workers (Van Etten et al., 1982, 1989;
Purkayastha, 1985; Boominathanetal., 1986; Mahadevan, 1991). Al-
though an apparent correlation exists betweenthe higher rate of PA degra-
dation and greater pathogenicity of a species, it is not applicable in all
cases. Any significant change in any factor(s) involved in the degradation
process may alter this relationship. The enzymes associated with the degra-
dation process may be intracellular or extracellular, constitutive or induc-
ible, single or multiple. Usually more than one step may be involved in
complete degradation of a toxic PA to a nontoxic product. These aspects
are yet to be studied in detail in cases of newly reported PAS. Degraded
products of several knownPAShave not been identified so far.
20 Purkayastha
VI. PHYTOALEXIN A N D PLANT ANTIGENS RELATED

TO DISEASE RESISTANCE
Purkayastha (1973) first attempted to show a relationship among PAS,

plant antigens,and disease resistanceon the basis of available information.
The objectives are to determine whether both PAS and plant antigens are
associated with disease resistanceand, if so, whether both couldbe consid-
eredasreliablebiochemicalparameters for screeningdiseaseresistance
germplasm in vitro. Since production of both PAS and plant antigens are
genetically controlled (based on evidence), it wasthought judicious to select
these two stable parameters.After about 20 years of research it has become
more or less certain that a relationship exists among the three. In 1978,
Cruickshank stated that the net rate of PA accumulation was greater in
resistant cultivars than in susceptible ones. He also suggestedon the basis
of the consistency of the cultivar data that PA concentrations for individual
plants withina progeny are genetically inherited and that selections could be
made on this basis. There is substantial evidence from several gene-for-gene
systems that PAS accumulate in high quantitiesafter inoculation of plants
with avirulent but not with virulent parasite races (Keen,1982). It suggests
that both host and pathogen genes are involved in both susceptible and
resistant disease reactions. Recentlyit was demonstrated conclusively that
isolation of stilbene synthase genes (responsible for synthesizing the stil-
bene-type PA resveratrol when attacked by pathogens) from grapevineand
transfer of these genes to tobacco could increase its resistanceto Botrytis
cinerea infection. This increased disease resistance in transgenic plants is
due to an additional foreign phytoalexin (Hain etal., 1993). It suggests that
foreign PA in a plant may also confer resistanceto disease.
The involvement of antigens in disease reaction of plants was also
reported by several workers. They found antigenic similarity in susceptible/
compatible host-parasite interaction while antigenic disparity in resistant/
incompatible interactions. The main reason for considering PA and plant
antigens together is that both are believed to be involved in the disease
reaction and probably production of PA depends on the interaction of
host-parasite proteins or specific antigens. The dimer model proposed by
Ellingboe (1982) also presented the similarview. The dimers are viewed as
being formedby the primary protein productsof the R genes and avr genes,
directly binding oneanother in a protein-protein interaction. Alternatively,
one of them might bind the gene or mRNA transcript ofthe other (Gabriel
and Rolfe, 1990). Thedimer formation was attributed to incompatible
interaction (Ellingboe, 1982). The formation of dimer was also supported
by Gabriel et al. (1988) who proposed a separate model known as the ion
channeldefensemodel.Whatevermaybe the interpretation of various
models, the two basic factors common to all cases are that 1.) both host
and pathogen genes or their products are involved in disease reaction and
2.) more or lessPA accumulates in host tissues after infection. The involve-
ment of antigen (protein) in disease reaction was also suspected when one
of the host antigens was found to be missing after chemical induction of
resistanceinsusceptiblehostcultivars (Chakraborty and Purkayastha,
1983; Purkayastha and Banerjee, 1990).
A number of previous workers presented conclusive evidence that resis-
tant cultivars of different host species produced more PAS in response to
fungal infection than the susceptible ones. Resistant cultivar of soybean
Harosoy 63 produced seven times morePA than susceptible cultivarHaro-
soy when plants were inoculated with Phytophthora megaspermavar. sojae
(Klarman,1968).Keen et al. (1971) alsoreportedthat PA (6a-hydroxy-
phaseollin) accumulated more rapidly (20-50 times faster) in the hypocotyls
of Harosoy 63 than in Harosoy when inoculated with that pathogen. The
resistant tomato plants also produced more rishitin (phytoalexin)than the
susceptible plants (Tijamosand Smith, 1974)after inoculation with Verticil-
lium alboatrum. The leaves of highly resistant sugarbeet showed greater
amounts of 0-vulgarin than susceptible ones when infected with Cercospora
beticola (Johnson et al., 1976). Besides, monogenically resistant cowpeas
accumulated 10-fold or more kievitone than the near-isogenic susceptible
cultivar (to Phytophthora vignae) (Partridge and Keen,1976). In 1981,
Vaziri et al. noted that the rate of medicarpin production was higher in
resistant cultivarof alfalfa following inoculation with P. megasperma f. sp.
medicaginis. A number of resistant rice cultivars, such as Mahsuri, Rupsail,
and Bad Kalamkati, produced greater amountof PA momilactone A than
the susceptible cultivars (Jaya, IR-8, CR-126-42-1) when inoculated with
Acrocylindrium (= Sarocladium) oryzae (Purkayastha et al., 1983). In a
separate investigation, Purkayastha and Chakraborty (1983) demonstrated
that UPSM-19, a resistant (to Macrophomina phaseolina) cultivar of soy-
bean, was able to produce significantly higher amounts of glyceollin than a
susceptible cultivar Soymax. The rapid accumulation of PA in resistant
cultivar ILC 3279 of Cicer arietinum after inoculation withAscochyta ra-
biei was also noted by Weigard et al. (1986). The average production of
medicarpin in the resistant cultivar was found to be 20 pmol/g fresh wt,
while it was only 5 pmol/g in susceptible cultivars. Kessmann and Barz
(1987) prepared cell suspension cultures of C.arietinurn from both resistant
cultivar ILC3279 and susceptible cultivar ILC1429 and inoculated withA .
rabiei. These two cultures differed in their accumulation of medicarpin and
maackiain. In view of the above facts it can be presumed that PAis associ-
ated with disease resistance of many plants. Hahn et al. (1985), however,
stated that phytoalexins maynot be solely responsiblefor the initial inhibi-
22 Purkayastha
tion of Phytophthora megaspermaf. sp. glycinea in the resistant cultivarof

soybean.
Involvement of PAS in plant disease resistance has been demonstrated
by many workers but whether the absence of common antigens between
host and parasite is also an important factor in disease resistance has not
been ascertained. In this chapter, however, an attempt has been made to
summarize the work done so far on host-pathogen antigens pertaining to
disease resistance/susceptibility of plants with a view to evaluate how far
the validity of “common antigens’’ or “matching proteins” concept is appli-
cable in determining basic compatibility and the absence of same in basic
incompatibility. PAS and plant antigens have always been considered sepa-
rately by the researchers. No attempt was made to work on both aspects
together and hence it was not possible to show any relationship between the
two. Sincethe gene-for-gene concept of Flor(1956) is now well recognized,
it is not inconceivable that host genes responsible for the production of
matching proteins (with parasite) may also contain a PA suppressor gene($.
If a suppressor gene is eliminatedor inactivated by any physical or chemical
treatment, the susceptible plant may become temporarily resistant and pro-
duces/accumulates more PA. Sinceboth resistant and susceptible cultivars
contain PA-producing genes,the rate of production in susceptible cultivar
is very low may be due to the presence of PA suppressor gene. Again, the
same cultivaris able to produce morewhen treated witha suitable chemical
and inoculated with the same pathogen. Recently, it wasobserved in a
number of cases that chemical induction of resistance in susceptible host
cultivars causes two changes: 1.) production of more phytoalexin and 2.)
missingof a specific host antigen. The gene which was responsible for
the missing antigen (protein) could be a PA suppressor gene. It would be
interesting to trace the missing antigenand its relationship,if any, with the
rate of PA production.
In 1948, Fedotova recognizedthe significance of antigenic relationship
between host and parasite in disease susceptibility of plants. Subsequently,
Dineen (1963) reported that susceptibility of host increases with antigenic
similarity while resistanceof the host is characterized by disparity of anti-
genic determinants. This concept is applicable in vertebrate animalsbut not
plants accordingto Charudattan and DeVay (1972). However, experimental
evidence suggeststhat the common antigen relationshipis not an unnatural
phenomenon and cannot be entirely ruled out on the basis of a few excep-
tions. An interesting observation was made by Doubly et al. (1960), who
found that a specific antigen in eachof the four races of Melampsora Iini
was commonly shared by only those lines of flax (Linum usitatissimum)
that weresusceptible to a particular race. This was later confirmed by
Peterman (1967). During the last three decades the common antigen rela-
tionship has been detected ainnumber of host-parasite systems (Table 1).
In viewof the findings summarized in the table it is reasonable to
speculate that common antigenic substances underlie compatibility of host
and parasite albeitthere are exceptions. For instance,Phytophthora infest-
ans shared antigens in common with its hostspotato and tomato and also
with its nonhost tobacco but not with nonsolanaceous species (Palmerly
and Callow, 1978).
More than 100 host pathogenhon-pathogen combinations including
several cultivars of soybean, rice, groundnut, pigeon pea, jute, and bean
and their respective pathogens and nonpathogens were examined. The re-
sults of immunodiffusion and immunoelectrophoretic tests reveal that there
is no common antigenic relationship between hostsand nonpathogens (45
combinations tested) while more than 50% combinations exhibited cross
reactive antigens between known hosts and pathogens (67 combinations
tested) irrespectiveof virulent and avirulent strains.No cross-reactive anti-
gen was detectedin immunodiffusion test using resistant hosts (10 cultivars)
and their parasitesbut it was detected at a very low titer by enzyme-linked
immunoassay in case ofa resistant soybean cultivarUPSM-19. All known
susceptible cultivars, however, showed common antigenic relationship (28
combinations tested). Absence and presence of common antigens in immu-
nodiffusion test could be regarded an as indication of resistant (incompati-
ble) and susceptible (compatible) reactions, respectively. The absence or
failure to detectcommonantigensmaybedue to severalreasons.The
method of extraction of antigens as well as the age of plant tissues and
culture of microbes have a marked influenceon the yield of antigenic sub-
stance and may account for the failure to detect (DeVay and Adler, 1976).
Besides, avirulence of pathogen, resistance of host, or low titer of antigen/
antiserum of host/pathogen or due to any other technical error. The degree
of susceptibility or resistance of a cultivar or virulence or avirulence of a
pathogen cannot always be accurately determined on the basis ofthe results
of immunodiffusion or immunoelectrophoretic tests because ina number
of cases no correlation could be established between the two. The common
antigen concept could be related with parasitism to some extent but not
with pathogenicity. This viewwas also expressed earlier by a group of
workers. It is necessary to mention that a threshold titer of both host and
pathogen antigens isessential for compatible interaction. The examples
cited above clearly indicate that host-pathogen antigens have a role in
disease reaction. On the contrary, PA accumulation/production is also
involved in disease reaction of plants. Therefore, it is not unnatural to
speculate that a relationship exists among them. Both factors should be
24 Purkayastha
Table 1 Presence/Absence of Common Antigenic Determinants Between

Compatible Hosts andParasites
Presence/
absence of
common
Host Ref.Parasite antigens
1 4
Arachis hypogea Macrophomina + Purkayastha andGhosal

(Groundnut) phaseolina (1 987)
Avena sativa Ophiobolus + Abbot (1973)
graminis
Coffea arabica Hemileia + Alba et al. (1983)
(Coffee) vastatrk
Citrullus Fusarium + Abd-El-Rehim etal.
vulgaris semitectum (1971)
(Watermelon)
Corchorus Colletotrichum + Bhattacharyya and
capsularis corchori Purkayastha (1985)
(Jute)
Glycine max Macrophomina + Chakraborty and
(Soybean) phaseolina Purkayastha (1983)
Colletotrichum + Purkayastha and
dematium var. Banerjee (1990)
truncata
Myrothecium + Ghosh and Purkayastha
roridum (1990)
Gossypium Verticillium + Charudattan andDe Vay
hirsutum dahliae (1972)
(Cotton) Fusarium + Venkataraman
oxysporum et al. (1973)
f. sp. Abd-El-Rehim
vasinfectum et al. (1988)
Fusarium solani + -do-
Xanthomonas + DeVay etal. (1967)
malvacearum
Helianthus Agrobacterium + DeVay et al.(1970)
annuus tumefaciens
(Sunflower)
Ipomoea batatus Ceratocystis + DeVay etal. (1967)
(Sweet potato) firnbriata
Linum Melampsora lini + Doubly etal. (1960)
usitatissimum
(Flax)
Table 1 Continued
Presence/
absence of
common
Host Ref.
Parasite antigens
1 2 3 4
Medicago sativa Corynebacterium - Carroll et al. (1972)

(Alfalfa) insidiosum
Oryza sativa Acrocylindium + Purkayastha and Ghosal
(Rice) oryzae (1985)
( =Sarocladium
oryzae)
Triticum aestivum Puccinia - Johnson (1962)
(Wheat) graminis
Zea mays Ustilago maydis + Wimalajeewa and DeVay
(Maize) (1971)
Solanum Phytophthora + Alba and DeVay (1985)
tuberosum infestans
(Potato)
Hordeum Erysiphe + Heide and Smedegaard
aestivum graminis Petersen (1985)
(Barley) f. sp. hordei
Lycopersicum Phytophthora + Palmerly and Callow
esculentum infestans (1978)
(Tomato)
Trifolium repens Rhizobium + Dazzo and Hubbek (1975)
(Clover) trifollii
Cajanus cajan Fusarium udum + Purkayastha et al. (1991)
(Pigeon pea)
Gossypium Fusarium + Kalyanasundaram et al.
arboreum vasinfectum (1975)
(Cotton)
Cotton roots Thielaviopsis + Guseva etal. (1979)
basicola
Abelmoschus F. vasinfectum + Balasubramanian and
esculentus Kalyanasundaram
(a member of (1978)
cotton family)
26 Purkayastha
considered together for screening disease resistant/susceptible germplasm

(Table 2).
There is evidence that alteration/regulation of specific host antigen(s)
bychemicalsmayinduceresistanceinplants.Forexample,gibberellic
acid (100 pg/ml) and sodium azide (100 pg/ml) altered antigenic pattern
of a susceptible (to Sarocladium oryzae) rice cultivar Jaya (Ghoshal and
Purkayastha, 1987); sodium azide (100 pg/ml) also changed the antigenic
Table 2 Rate of Phytoalexin Production/Accumulation and Presence/Absence of

Cross-Reactive Antigensin Different Host-Parasite Systems
Phyto-
Presence/
alexin
Host pg/g absence
Parasite fresh
Reaction
cultivars CRA" of wt Ref.
1
Acrocylindrium Rice (leaf Momilac-

ovzae sheath) tone A
(= Sarocladiun CV.Mahsuri Resistance 19.68 - Purkayastha et
oryzae) CV.Jaya Susceptible 8.64 + al. (1983)
Macrophomina Soybean Glyceollin
phaseolina (roots)
CV.UPSM-19Resistance 421.00 - Purkayastha
CV.Soymax Susceptible 265.00 + and Chakra-
CV.R-l84 Susceptible 279.00 + borty (1983)
Colletotrichum Soybean Glyceollin
dematium (leaves)
.
var truncata CV.UPSM-19 Resistance 176.76 - Chattopadhyay
CV.DS-178 Susceptible 71.10 + (Bandyopad-
CV.PK-327 Susceptible 106.66 + hyay) (1989)
Myrothecium Soybean Glyceollin
roridum (leaves)
CV.UPSM-19 Resistance 441.68 - Ghosh (1990)
CV.DS-74-24-2 Susceptible 135.76 +
M. phaseolina Groundnut Cis-3,5-dimethoxy
(root) stilbene
CV.RSHY-1 Resistance 32.66 - Purkayastha
CV.TMV-2 Susceptible 10.56 + (and Ghoshal
(unpublished)
*CRA = Cross reactive antigens; + = CRA present; - = CRA not detected in immunodiffusion
test.
pattern of a susceptible (to M. phaseolina) soybean cultivar Soymax (Chak-

raborty and Purkayastha, 1987). Recently it was reported that cloxacillin,
an antibiotic, reduced anthracnose disease of soybean (cultivar Soymax)
caused by Colletotrichum dematiumvar. truncata and altered the antigenic
pattern of the susceptible host cultivar (Purkayastha and Banerjee, 1990).
In another investigation, Purkayastha and Ghosh (1991) showed that cad-
mium chloride treatment caused disappearance of a CRA in soybean cul-
tivar.
Higher productionof PA and changes in antigenicpattern after chemi-
cal treatment of a susceptible host suggest that a relationship may exist
among PA, host antigen,and disease resistance. The disappearance of par-
ticular host antigen(s) or proteins in treated plant could be due to the
inactivation of a particular gene@)which is accounted for the biosynthesis
of missing proteins and that gene@) may be a PA suppressor gene. Since
several genesare responsible for PA production, inactivationof genes may
cause lower production of PA. However, this observation is an important
one and provides a very valuable basisfor further investigations of a similar
type of activity involvingother host-parasite systems.
VII. EPILOG
In the past half century phytoalexin has been fully established as one the of
major factors in plant disease resistance. Its application in plant protection,
plant breeding, germplasm screening,and the study of biogenetic relation-
ships among plants is now more or less known. But whether higher rate of
production/accumulation of PA or inability of a parasite to degrade/detox-
ify PA is more important in plant disease resistanceis not clear. Since PA
detoxifying mechanisms of parasites differ in many respects, it is not un-
usual to find differential tolerance of pathogens to a PA. A specific gene
which controls detoxification couldbe considered asa diagnostic character
of that strain of a pathogen. Besides, pathogen’s host range could also be
determined usingthis diagnostic character (Van Etten et al., 1989).
Another important aspect of PA research is the elicitation of PA by
biotic and abiotic agents along with identification and determination of the
structure of elicitor. It is also not clearly understood asto how both biotic
and abiotic agents elicit similarPA in a host plant.It is a matter of specula-
tion that both may cause similar change in DNA sequence and consequently
more ornew mRNA or both could activate enzymes involved inPA biosyn-
thesis. What are the factors affecting elicitor activityof a pathogen? How
long doesan elicitor remain active? an Is elicitor capable of eliciting similar
amounts of PA in differentplant tissues? If not, is it due to any difference
in cell membrane receptors? These questions remain unanswered. A potent
28 Purkayastha
elicitor can activate host’s resistance under certain conditions. Therefore,

effective, stable, and nonphytotoxic elicitors should be identified for use in
disease control. Apart from PA elicitors, genetic control of race-cultivar
specificity and PA synthesis have been discussed earlier by several workers
(Paxton, 1980; Dixon et al., 1983, 1992; Crute, 1985; Bailey, 1987). Gene-
for-gene relationships have been demonstratedor suggested in at least 43
different plant-pathogen interactions (Gabriel and Rolfe, 1990). Both com-
patible and incompatiblehost-pathogeninteractionsdepend on specific
combination of host-pathogen genes. These genes may control both pro-
duction of PA and host-pathogen antigens (proteins/enzymes). Because
there is evidencethat chemical induction of resistance ainsusceptible culti-
var sometimes alter antigenic pattern of host cultivar and also stimulates
PA production. Hence it is expedient to identify the specific antigen(s)
which disappears due to induction of resistance in susceptible host cultivar
and to determine whether it has any relation with PA suppressor gene,
if any, present in the susceptible host. Induction of disease resistance by
alteration/regulation of specific host antigen@) could be a novel approach
to crop protection program. Recent evidence suggests that foreign PA in a
transgenic plant also confers resistance to disease. It is not improbable that
isolation of new PAS from many other plants in future may be more potent
and nonphytotoxic like several other chemotherapeutants. However, PA
research will continueuntilmolecularmechanismofphytoimmunityis
clearly knownto the plant scientists.
REFERENCES
Abbott, L. K.(1973). Taxonomy and host specificity of Ophiobolusgraminis Sacc.,
an application of electrophoretic and serological techniques. PhD thesis, Mon-
ash University, Clayton, Victoria, Australia, p. 200.
Abd-El-Rehim, M. A., Ibrahim, T. A., Michail, S. H., and Fadel, F. M. (1971).
Serological and immunoelectrophoretical studies on resistant and susceptible
watermelon varieties to Fusarium semitectum Berk and Rev. Phytopathology
Z. 71~43-55.
Abd-El-Rehim, M. A.,Abou-Taleb,E. M., and Johamy, A. (1988). Common
antigen@) in cotton to Fusarium oxysporum f. sp. vasinfectum. Journal of
Phytopathology 121:217-223.
Akazawa, T.,and Wada, K.(1961). Analytical studyof ipomeamarone and chloro-
genic acid alterations in sweet potato roots infected by Ceratocystisfimbriata.
Plant. Physiology36139-141.
Alba, A. P. C., and DeVay, J. E. (1985). Detection of cross-reactiveantigens
between Phytophthora infestans(Mont.) de Bary and Solanum species by indi-
rect enzyme linked immunosorbent assay. Phytopathology Z . 112:97-104.
Alba, A. P. C., GUZZO, S. D., Mahlow, M. F. P., and Moraes, W. B. C. (1983).
Common antigens inextracts of Hemileia vastatrixuredinospores and of Cof-

fea arabica leaves and roots. Fitopathologia Brasileira8:473-483.
Albersheim, P., and Anderson-Prouty, A. J. (1975). Carbohydrates, proteins, cell
surfaces and the biochemistry of pathogenesis.Annu. Rev. Plant Physiol. 2 6
31-52.
Albersheim, P., Valent, B. S., Hahn, M., Wade, M., and Cline, K. (1978). Elicita-
tion and sites of formation of phytoalexins and induced resistance. 3rd Int.
Congr. Plant Pathol., Munchen, Germany (Abstr.).
Albersheim, P., Darvill, A. G., Sharp, J. K., Davis, K. R., and Doares, S. H.
(1986). Studies on the role of carbohydrates in host-microbeinteractions. Rec-
ognition in Microbe-Plant Symbiotic and Pathogenic Znteractions (Ben Lug-
tenberg, ed.), NATOAS1 Series, Vol. H4, Springer-Verlag, Heidelberg, pp.1-
13.
Amin, M., Kurosaki, F., and Nishi, A. (1986). Extracellular pectinolytic enzymes
of fungi elicit phytoalexin accumulation in carrot suspension culture. J. Gen.
Microbiol. 132~771-777.
Anderson, A. J. (1980a). Differences in biochemical compositions and elicitor activ-
ity of extra-cellular components produced by three races offungal plant patho-
gen Colletotrichum lindemuthianum. Can.J. Microbiol. 2 6 1473-1479.
Anderson, A. J. (1980b). Studies on the structure and elicitor activity of fungal
glucans. Can. J. Bot. 58:2343-2348.
Anderson-Prouty, A. J., and Albersheim, P. (1975). Host pathogen interactions.
VIII. Isolation of a pathogen synthesized fraction rich in glucan that elicits a
defense responsein the pathogen's host. Plant Physiol.56(2):286-291.
Ayers, A. R., Ebel, J., Finelli, F., Berger, N., and Albersheim, P. (1976a). Host-
pathogen interactions. IX. Quantitative assays of elicitor activity
and character-
ization of the elicitor present in the extracellular medium of cultures of Phy-
tophthora megasperma var. sojae. Plant Physiol.57~751-759.
Ayers, A. R., Ebel, J., Valent, B., and Albersheim, P. (1976b). Host-pathogen
interactions. X. Fractionation and biological activity of an elicitor isolated
from the mycelial walls of Phytophthora megasperma var. sojae. Plant Physiol.
52760-765.
Ayers,A. R., Valent, B., Ebel, J., and Albersheim, P. (19764. Host-pathogen
interactions XI. Composition and structure of wall-released elicitor fractions.
Plant Physiol.52766-774.
Bailey, J. A. (1987). Phytoalexins: a geneticview of their significance. In Genetics
and Plant Pathogenesis(P. R. Day and G. J. Jellies eds.), pp. 233-244. Black-
well Scientific Publications.
Bailey, J. A., and Mansfield, J. W. (1982). Phytoalexins, Blackie, London.
Bailey, J. A., Burden, R. S., Mynett, A., and Brown, C. (1977). Metabolism of
phaseollin by Septoria nodorum and other non-pathogens of Phaseolus vul-
garis. Phytochemistry16:1541-1544.
Balasubramanian, R., and Kalyanasundarm, R. (1978). Electrophoretic and anti-
genic studies of soluble proteins in'Fusariumwilt ofcotton. Proc. Indian Acad.
Sci. Sect.B. 87(8):223-230.
30 Purkayastha
Bernard, N. (1911). Sur la function fungicide des bulbs d’ophrydes.Ann. Sci. Nut.
(Bot.)14221-234.
Bhattacharya, B., and Purkayastha, R. P. (1985). Occurrence of common antigens
in jute and Collectotrichum corchori.Curr. Sci. 54:251-252.
Birch, A. J., Rickards, R. W., and Smith, H.(1958). The biosynthesis of gibberellic
acid. Proc. Chem. Soc., p. 192.
Boominathan,K., Gurujeyalakshmi, G., and Mahadevan, A. (1986). Detoxification
mechanisms of plant pathogens. Vistas in Plant Pathology (A. Varma and J.
P. Varma, eds.), Malhotra, New Delhi, pp. 45-69.
Bostock, R. M., Lane, R. A., and KuC, J. A. (1982). Factors affecting the elicitation
of sesquiterpenoid phytoalexin accumulation by eicosapentaenoic and acidara-
chidonic acidin potato. Plant Physiol.701417-1424.
Brueggar, B. B., and Keen, N.T.(1979). Specific elicitorsof glyceollin accumula-
tion inthe Pseudomonas glycinea-soybean host parasite system.Physiol. Plant
Pathol. 1243-5 1.
Bryan, I. B., Rathmell, W.G., and Friend, J. (1985). The role of lipid and non-lipid
components of Phytophthora infestansin the elicitation of the hypersensitive
response in potato tubertissue. Physiol. Plant Pathol.26331.
Burden, R. S., Bailey, J. A., and Vincent, G. G . (1974). Metabolism of phaseollin
by Colletotrichum lindemuthianum. Phytochemistry 13: 1789-1791.
Cahill, D. M., and Ward, E. W. B. (1989). Effects of metalaxyl on elicitor activity,
stimulation of glyceollin production and growth of sensitive and tolerant iso-
lates ofPhytophthora megasperma f. sp. glycinea. Physiol.Mol. Plant. Pathol.
35(2):97-112.
Callow, J. A. (1977). Recognition, resistance and the role of plant lectins in host-
parasite interactions. Adv. Bot. Res. 4~1-49.
Carroll, R. B., Lukezic, F. L., and Roslyn, G. L. (1972). Absence of a common
antigen relationship betweenCorynebacterium insidiosumand Medicago sativa
as a factor in disease development.Phytopathology 62:1351-1360.
Cartwright, D., Langcake, P., Pryce, R. J., Leworthy, D. P., and Ride, J. P.,
(1977). Chemical activation of host defence mechanism as abasisof crop
protection. Nature 267511-513.
Chakraborty, B. N., and Purkayastha, R. P. (1983). Serological relationship be-
tween Macrophominaphaseolinaand soybean cultivars.Physiol. Plant Pathol.
23:197-205.
Chakraborty, B. N., and Purkayastha,R. P. (1987). Alteration in glyceollin synthe-
sis and antigenicpatterns after chemical induction of resistancein soybean to
Macrophomina phaseolina. Can.J. Microbiol. 332335-840.
Chalova, L. T., Baramidze, V. G., Yurgonova, L. A., D’yakov, Yu, T., Ozeret-
skovskya, 0. L., and Metlitskii, L. V. (1977). Isolation and characterization of
the inducer of protective responses the of potato fromthe cytoplasmic contents
of the Phytophthora pathogen. Transl.Dokl. Akad. Nauk. SSSR 235:1215.
Chalova, L.T.,Ozeretskouskya, 0. L., Yurganova, L. A., Baramidze, V. G . ,
Protosenko, M. A., D’Yakov, Yu, T., and Metlitskii, L. V. (1976). Metabolites
of phytopathogenic fungi as elicitors of defense reactions in plants. Trousb.
Dokl. Akad. SSSR 230:722-725.
Chalutz, E., and Stahmann,M. A. (1969). Induction of pisatin by ethylene. Phyto-

pathology591972-1973.
Charudattan, R., and DeVay, J. E. (1972). Common antigens among varieties of
Gossypium hirsutum and isolates of Fusarium and Verticillium sp. Phytopa-
thology 62~230-234.
Chattopadhyay, Ratna (Mrs. Bandyopadhyay) (1989). Studies on phytoalexin and
plant antigens in relation to resistance of soybean (Glycine may (L.) Merrill)
cultivars to Anthracnose disease. PhD thesis, University of Calcutta, pp. 1-
194.
Cheema, A. S., and Haard,N. F. (1978). Induction of rishitin and lubiminin potato
tuber discs by non-specific elicitors and the influence of storage conditions.
Physiol. Plant Pathol.13:233-240.
Chester, K. S. (1933). The problem of acquired physiological immunity in plants.
Quart. Rev. Biol. 8129-154.
Cline, K.,Wade, M., and Albersheim, P. (1978). Host pathogen interactions. XV.
Fungal glucans which elicit phytoalexin accumulation in soybean also elicitthe
accumulation of phytoalexins in other plants. Plant Physiol.62(6):918-921.
Cruickshank, I. A. M. (1963). Phytoalexins. Annu. Rev. Phytopathol. Z:351-374.
Cruickshank, I. A. M. (1978). A reviewof the roleofphytoalexinsindisease
resistance mechanisms. Pontifcae Academiae Scientiarum Scripta Varia (21)
IV(2):1-61.
Cruickshank, I. A. M. (1980). Defenses triggeredby the invader: chemical defenses.
Plant Disease:An Advanced Treatise, Vol. 5 (J. G . Horsfall and E. B. Cowl-
ing, eds.), Academic Press,London, pp.247-267.
Cruickshank, I. A. M., and Perrin, D. R. (1968). The isolation and partial charac-
terization of monilicolin A, a polypeptide with phaseollin-inducing activity
from Monilinia fructicola. Life Sci. 7~449-458.
Crute, I. R. (1985). The genetic bases of relationships between Microbial Parasites
and their hosts. Mechanisms of Resistance to Plant Diseases (R. S. S. Fraser,
ed.), Martinus Nijhoff, Dodrecht, pp. 80-142.
Crute, R.,DeWit, P. J. G . M., andWade, M. (1985). Mechanismsbywhich
genetically controlled resistance and virulence influence host colonization by
fungal and bacterial parasites.Mechanisms of Resistanceto Plant Diseases(R.
S. S. Fraser, ed.), Martinus Nijhoff, Boston, pp. 197-284.
Daniels, D. L., and Hadwiger, L. A. (1976). Pisatin-inducing components in fil-
trates of a virulent and avirulent Fusarium solani cultures. Physiol. Plant Pa-
thol. 89-19.
Darvill, A. G., and Albersheim, P. (1984). Phytoalexins and their elicitors:a defense
against microbial infection in plants. Annu. Rev. Plant Physiol.35243-275.
Davis, K. R., Darvill, A. C., Albersheim, P., and Dek, A. (1986). Host-pathogen
interactions. XXIX. Oligo-galacturonides releasedfrom sodium polypectateby
endogalacturonic acid lyase are elicitors of Phytoalexins in soybean. Plant
Physiol. 80(2):568-577.
Dazzo, F. B., and D. Hubbek (1975). Cross-reactive antigensand lectins as determi-
nants of symbiotic specificity in the Rhizobium clover association. A. A. A. A. A. A. A. A. A. A. P Mi-
crobiol. 301017-1033.
32 Purkayastha
Desjardins, A. E., Gardner, H. W., and Plattner, R. D. (1989). Detoxification of

potato phytoalexin lubimin by Gibberella pulicaris. Phytochemistry 28431-
437.
DeVay, J. E., and Adler, H. F. (1976). Antigens common to hosts and parasites.
Annu. Rev. Microbiol. 30147-168.
DeVay, J. E., Schnathorst, W. C., and Foda, M. S. (1967). The dynamic role of
molecular constituents in plant-parasite interactions (C. J. Mirocha and I.
Uritani, eds.), Bruce, Minneapolis, pp. 313-328.
DeVay, J. E., Romani, R. J., Monadjem, A. M., and Etzler, M. (1970). Induction
of Phytoprecipitins in sunflower gall tissue in response to infection by Agro-
bacterium tumefaciens. Phytopathology601289(Abstr.).
Deverall, B. J. (1972). Phytoalexins and disease resistance.Proc. R. Soc. Lond. B.
181:233-246.
Deverall, B. J. (1976). Current perspectives in research on phytoalexins. Biochemi-
cal Aspects of Plant-Parasite Relationship(J. Friend and D.R. Threlfall, eds.),
Academic Press, New York, pp. 207-223.
Dineen, J. K. (1963). Antigenic relationships between host and parasite. Nature 197:
268-269.
Dixon, R.A. (1992). Characterised defence response genes. Molecular Plant Pathol-
ogy: A Practical Approach, Vol. 1 (S. J. Gurr, M. J. McPherson, and D. J.
Bowles, eds.), Oxford University Press, Oxford, p. 216.
Dixon, R. A.,and J. L. Christopher (1979). Stimulation of de novo synthesis
of L-phenylalanine ammonia lyase in relation to phytoalexin accumulation
in Colletotrichum lindemuthianum elicitor-treated cell suspension cultures of
French bean(Phaseolus vulgaris). Biochem. Biophys. Acta 586(3):453-463.
Dixon, R. A., Dey, P. M., and Lamb, C. J. (1983). Phytoalexins: enzymology and
molecular biology. Advances in Enzymology and Related Areasof Molecular
Biology (A. Meister, ed.), Wiley, New York, pp. 1-135.
Doubley, J. A., Flor, H. H., and Glagett, C. 0. (1960). Relation of antigens of
Melampsora liniand Linum to resistance and susceptibility. Science 131:229.
Ebel, J . (1986). Phytoalexin synthesis: the biochemical analysis of the induction
process. Annu. Rev. Phytopathol. 24235-264.
Ebel, J., Ayers, A. R., and Albersheim, P. (1976). Host pathogen interactions.XII.
Response of suspension cultured soybean cells to the elicitor isolated from
Phytophthora megasperma var. sojae, a fungal pathogen of soybean. Plant
Physiol. 52775-779.
Ellingboe, A.H. (1982). Genetical aspects of active defence. Active Defense Mecha-
nisms in Plants (R. K. S. Wood, ed.), Plenum Press, New York, pp. 179-192.
Fedotova, T.I. (1948). Significance of individual proteinsof seed in the manifesta-
tion of the resistance of plants to diseases. Trudy Leninnger Inst., Zasheb.
Rast. Sborn. 1:61-71.
Flor, H.H. (1956). The complementary genic systems in flax and flax rust. Adv.
Genet. 8:29-54.
Fraile, A., Garcia, A. F., and Sagasta, E.M. (1980). Phytoalexin accumulation in
bean (Phaseolus vulgaris)after infection withBotrytis cinerea and treated with
mercuric chloride.Physiol. Plant Pathol.169-18.
Fuchs, A., De Vries, F. W., and Sam, M. P. (1980). The mechanism of pisatin
degradation by Fusarium oxysporum f. sp. pisi. Physiol. Plant Pathol.I 6 1 14-
133.
Gabriel, D. W., and Rolfe, B. G. (1990).Working models of specific recognition in
Plant-microbe interactions. Annu. Rev. Phytopathol. 28365-391.
Gabriel, D. W., Loschke, D. C., and Rolfe, B. G. (1988). Gene-for-gene recogni-
tion: the ion channel defense model. Molecular Genetics of Plant-Microbe
Interaction (R.Palacios, D. P. S. Verma, and M. N. St. Paul, eds.), APS
Press, p. 314.
Gaumann, E., and Kern, H. (1959a). Uber die isolierungand den chemischen Nadi-
weis des orchinobs. Phytopathol. 2.35:347-356.
Gaumann, E., and Kern, H. (1959b). Uber chiemische Abwehrreaktionenbei or-
chideen. Phytopath. 2.361-26.
Ghosh, Sumita (1990). Studies on host-parasite interactions with special reference
to Myrothecium leaf spot disease of soybean. PhD thesis, University of Cal-
cutta, p. 222.
Ghosh, S., and Purkayastha, R. P. (1988). Immunoserological studieson soybean-
Myrothecium interactions, 75th Session of the Indian Science Congress,Pune
(Abstr.).
Ghosh, Sumita, and Purkayastha, R. P. (1990). Analysisof host-parasite cross
reactive antigens in relation to Myrothecium-infection of soybean. Indian J.
Exp. Biol.28:l-5.
Ghosh, Sumita, and Purkayastha, R. P. (1992). Elicitation of glyceollin by natural
products and metallic salts.Znt. J. Trop. Plant Dis. 1099-108.
Ghoshal, A., and Purkayastha, R. P. (1984). Elicitation of momilactone by gibber-
ellin in rice. Curr. Sci.53506-507.
Ghoshal, A., and Purkayastha, R. P. (1987). Biochemical responsesof rice (Oryza
sativa L.) leaves to some abiotic elicitors of phytoalexin. Indian J. Exp. Biol.
25:395-399.
Guseva, N. N., Gromova, B. B., and Lanlas, E. S. (1979). Cross-reacting antigens
of pathogens and their host plants. Acta Phytopathol. Acad. Scient. Hung.14
449.
Hadwiger, L. A., Jafri, A., Von Broembseu, S., and Eddy, R., Jr. (1971). Mode
of pisatin induction, increased template activity and dye-binding capacity of
chromatin isolated from polypeptide treated pea pods. Plant Physiol. 5352-
63.
Hadwiger, L. A., and Schwochau, M. E. (1971). Ultra-violet light-induced forma-
tion of pisatin and phenylalanine ammonia lyase, Plant Physiol.42346-351.
Hadwiger, L. A., Sander, C., Eddyvean, J., and Ralston, J. (1976). Sodium-azide-
induced mutants ofpeas that accumulatepisatin. Phytopathology 66629-
630.
Hahn, M. G., Bonhoff, A., and Grisebach, H. (1985). Quantitative localisation of
the phytoalexin glyceollin in
relation to fungal hyphae in soybean
roots infected
with Phytophthora megasperma f. sp. glycinea. Plant Physiol.72591-601.
Harborne, J. B., and Ingham, J. L. (1978). Biochemical Aspects of Plantand
Animal Coevolution.Academic Press, London, pp. 343-405.
34 Purkayastha
Henfling, J. W.D.M., Bostock, R. M., and Kub, J. (1980). Cell walls of Phy-
tophthora infetans contain an elicitor of terpene accumulation in potato tu-
bers. Phytopathology 70(8):772-776.
Heide, M., and Smedegaard-Petersen, V. (1985). Common antigens between barley
powderymildewandtheir relation to resistanceandsusceptibility. Can. J.
l
I Pathol.
Plant 7(4):341-346.
Hein, R., Reif, H. J., Krause, E., Langebartels,R., Kindl, H., Vornam, B., Wiese,
W. Schmetzer, E., Schreier, P. H., Stocker, R. H., and Stenzel, K. (1993).
Disease resistance resultsfrom foreign phytoalexin expression in a novel plant.
Nature361:153-156.
Higgins, V. J., Stoessl, A. and Heath, M. C. (1974). Conversion of phaseollin to
phaseollin-iso flavone by Stemphyllium botryosum. Phytopathology 64: 105-
107.
Hiura, M. (1943). Studies in storageand rot of sweet potato. Sci. Rep. Gifu Agric.
Coll. Jpn.5 0 1-5.
Ingham, J. L. (1972). Phytoalexins and other natural products as factors in plant
disease resistance.Bot. Rev. 38343-424.
Ingham, J. L. (1973). Diseaseresistanceinhigherplants. The conceptofpre-
infectional and post-infectional resistance.Phytopathol. Z. 78314-335.
Ingham, J. L. (1982). Phytoalexins from the Leguminosae. Phytoalexins (John A.
Bailey and JohnW. Mansfield, eds.), Blackie, London, pp.21-80.
Johnson, R. (1962). Serological studies on wheat and wheat stem rust. MS thesis,
University of Saskatchewan, Saskatoonsask,Canada, p. 57.
Johnson, G., Maag, D. D., Johnson, D. K., andThomas, R. D. (1976). The
possible role of phytoalexins in the resistance of sugarbeet (Beta vulgaris) to
Cercospora beticola. Physiol. Plant Pathol. 8:225-230.
Kalyanasundaram, R., Lakshminarashimam,C., and Venkataraman, S. (1975).
Common antigens in host-parasite relationship.Curr. Sci.44(2):55-56.
Keen, N. T. (1975). Specific elicitors of plant phytoalexin production: Determinants
of race specificityin pathogens. Science 18274-75.
Keen, N. T. (1981). Evaluation of the role of phytoalexins. Plant Disease Control
(R. C. Staplesand Gary H. Joenniesson, eds.), Wiley, New York, pp. 155-157.
Keen,N. T. (1982). Specific recognition in gene-for-gene host-parasite systems.
Advances in Plant Pathology, Vol. 1 (P. H. Williams and D. Ingram, eds.),
Keen, N. T., and Brueggar, B. (1977). Phytoalexins and chemicals that elicit their
production in plants. ASC Symp. Ser. 621-26.
Keen,N. T., Sims, J. J.,Erwin, D.C.,Rice,E., and J. E. Partridge (1971).
6-a-Hydroxyphaseollin, an antifungal chemical inducedin soybean hypocotyls
by Phytophthora megasperma var. sojae. Phytopathology 61: 1084-1089.
Keen, N. T., Yoshikawa, M., and Wang, M. C. (1983). Phytoalexin elicitor activity
of carbohydrate from Phytophthora megasperma f. sp. glycinea and other
sources. Plant Physiol.71(3):466-471.
Kessmann, H.,and Barz, W. (1986). Elicitation and suppression of phytoalexin
and isoflavone accumulation in cotyledons ofCicer arietinumL. as caused by
wounding and by polymeric compounds from the fungus Ascochyta rabiei. J.

Phytopathol. 117(4):321.
Kessmann, H., and Barz, W. (1987). Accumulation of isoflavonesand pterocarpan
phytoalexins incell suspension cultures ofdifferent cultivars of chickpea (Cicer
arietinum). Plant CellRep. 6(1):55.
Klarman, W. L. (1968). The importance of phytoalexins in determining resistance
of soybeansto three isolates of Phytophthora. Neth.J. Plant Pathol. Suppl.I:
171-175.
Klarman, W. L., and Gerdemann, J. W. (1963). Resistance of soybeans to three
Phytophthora species due to the production of a phytoalexin. Phytopathology
53~1317-1320.
Kubota, T., and Matsuura, T. (1953). Chemical studieson the black rot disease of
sweet potato. J. Chem. Soc. Jpn. (Pure Chem Sect) 74248-251.
KuC, J. (1972). Phytoalexins. Annu. Rev. Phytopathol. 10207-232.
KuC, J. (1976). Phytoalexins. Encyclopedia of Plant Physiology, Vol. 4, Physiologi-
cal Plant Pathology(R. Heitefuss and P. H. Williams, eds.), Springer-Verlag,
Berlin, pp. 637-652.
KuC, J. (1987). Plant immunization and its applicability for disease control.Innova-
tive Approaches to Plant Disease Control (Ilan Chet, ed.), John Wiley and
Sons, New York, pp. 255-274.
KuC, J., and Rash, S. J. (1985). Phytoalexins. Arch. Biochem. Biophys.232455472.
Lyon, G. D., and Albersheim, P. (1982). Host pathogen interactions. 21. Extraction
of a host-labile elicitor of phytoalexin accumulation from frozen soybean (Gly-
cine mm, cultivar Wayne) stems.Plant Physiol.70(2):406409.
Mahadevan, A. (1991). Post-infectional Defence Mechanisms.Today and Tomor-
row’s Printers and Publishers, New Delhi.
Matthews, D. E., and Van Etten, H. D. (1983). Detoxification of the phytoalexin
pisatin bya fungal cytochrome.Arch. Biochem. Biophys.224(2):494-505.
McIntosh, S. F., Matthews, D. E., and Van Etten, H. D. (1989). Two additional
genes for pisatin demethylation and their relationship to the pathogenicity of
Nectria haematococcaon pea. Annu. Rev. Phytopathol. 27: 143-164.
Mayama, S., Tani, T., Ueno, T., Midland, S. L., Sims, J. J., and Keen, N. T.
(1986). The purification of victorin and its phytoalexin elicitor activity in oat
leaves. Physiol. Mol. Plant Pathol.29(1):1-18.
Metlitskii, L. V., Ozeretskovskya, 0. L., Vasyukova, N. I., Sabirdina, M. S., and
Chalenko, G. I. (1984). Substances in the cell walls of Phytophthora infestans
as inducers and suppressor of defence reactions inpotato. Dokl. Akad. Nauk.
SSSR 274(4): 1020.
Muller, K. O., and Borger, H. (1940). Experimentelle Untersuchungen Uber die
Phytophthora Resistenz der Kartoffel. Arb. Biol.Reichsanst. Land-Forst-
wirtsch, Berlin-Dahlem23:189-23 1.
Muller, K. 0 . (1956). Einige einfache Versuchezum nachweisvow Phytoalexinen.
Phytopath. Z. 22237-254.
Osman, S., and Moreau, R. (1985). Potato phytoalexin elicitors in Phytophthora
infestans spore germination fluids.Plant Sci.41(3):205-209.
36 Purkayastha
Palmerly, R. A., and Callow, J. A. (1978). Common antigens in extracts of Phy-

tophthora infestans and potatoes. Physiol. Plant Pathol. 12241-248.
Partridge, J. E., and Keen, N. T. (1976). Association of the phytoalexin kievitone
with single resistance of cowpeasto Phytophthora vignae. Phytopathology 66:
426-429.
Paxton, J. (1980). A new working definition ofthe term “Phytoalexin.” Plant D&.
64734.
Perrin, D. R., and Cruickshank, I. A.M. (1965). Studies on phytoalexins. VII.
Chemical stimulation of pisatin formation in Pisum sativum L. Aust. J. Biol.
Sci. 18:803-816.
Peterman, M. A. (1967). Relation of antigens in selected host-parasite systems of
Linum usitatissimum and Melampsora h i . MS thesis, North Dakota State
University, Fargo, p. 62.
Purkayastha, R. P. (1973). Phytoalexins: plant antigens and disease resistance. Sci.
Culture 39528-535.
Purkayastha, R. P. (1985). Phytoalexins. Frontiers in Applied Microbiology, Vol.
1. (K. G. Mukherji, N.C. Pathak, Ved Pal Singh, eds.), LucknowPrint House,
pp. 363-407.
Purkayastha, R. P. (1986). Elicitors and elicitation of phytoalexins. Vistas in Plant
Pathology (A. Varma and J. P. Verma, eds.), Malhotra, New Delhi, pp. 25-
44.
Purkayastha, R. P., and Banerjee, R. (1990). Immunoserological studieson cloxa-
cillin-induced resistance of soybean against anthracnose. Zeitschriftfur Pflan-
zenkrankheiten und Pflanzenschutz.J. Plant Dis. Prot. 97(4):349-359.
Purkayastha, R. P., and Chakraborty,B. N. (1983). Immunoelectrophoretic analy-
sis of plant antigens in relation to biosynthesis of phytoalexin and disease
resistance of soybean. Trop.Plant Sei. Res. 1(1):89-96.
Purkayastha, R. P., and Chakraborty,B. N. (1985). Induction of disease resistance
and associated changes in Macrophomina-infected soybean. National Seminar
on Advances in Mycology and Plant Pathology, Punjab University, Chandi-
garh (Abstr).
Purkayastha, R. P., and Deverall, B. J. (1964). A phytoalexin type of reaction in
the Botrytis-infected leaves of bean (Viciafaba L.).Nature 201:938-939.
Purkayastha, R. P., and Deverall, B. J. (1965). The growth of Botrytisfabae and
B. cinerea in leaves of bean (Viciafaba L.). Ann. Appl. Biol.56:139-147.
Purkayastha, R. P., and Ghosh, S. (1983). Elicitation and inhibition of phytoalexin
biosynthesis in Myrothecium-infected soybean leaves.Indian J. Exp. Biol. 21:
216-218.
Purkayastha, R. P., and Ghoshal, A. (1985). Analysis of cross-reactive antigens of
Acrocylindrium otyzae andrice in relation to sheath rotdisease. Physiol. Plant
Pathol. 27:245-252.
Purkayastha, R. P., and Ghosal, A, (1987). Immunoserological studieson root rot
of groundnut (Arachis hypogeaL.). Can. J. Microbiol. 33547-651.
Purkayastha, R. P., and Ghosh, Sumita (1992). Heavy metal salt inducing disease
resistance and altering specific antigen of soybean leaves. Znt. J. Trop. Plant
Dis. 10:131-142.
Purkayastha, R. P., Ghoshal, A., and Biswas, S. (1983). Production of momilac-

to sheath rot disease. Curr. Sci.
tone associated with resistance of rice cultivars
52(3):131-132.
Purkayastha, R. P., Ghosal, A., Garai, M., and Ghosh, S. (1991). Cross-reactive
antigens as determinants of susceptibility of pigeon pea to Fusarial wilt. In
Botanical Researchesin India (N.C. Aery and B. L. Chaudhury, eds.), Himan-
shu, Udaipur, pp. 508-513.
Ricci, P., and Rousse, G. (1983). Absence of induction of phytoalexins and resis-
tance in carnation by the selective herbicide oxadiazone. Acta Hort. 141:109-
113.
Rohwer, F., Fritzmeier, K. H., Scheel, D., and Hahlbrock, K. (1987). Biochemical
reaction of different tissues of potato (Solanum tuberosum)to zoospores or
elicitors from Phytophthora infestans.Accumulation of sesquiterpenoid phyto-
alexin. Planta 170(4):556-561.
Rubin, B., Penner, D., and Saettler, A. W. (1983). Induction of isoflavonoid pro-
duction in Phaseolus vulgaris leaves by ozone, sulfur dioxide and herbicide
stress. Environ. Toxicol. Chem. 2(3):295-306.
Sanz Platero, de M., and Fuchs, A. (1978). Degradation of pisatin, an antimicrobial
compound producedby Pisum sativum. Phytopathol. Mediterr.1214-17.
Schwochau, M. E., and Hadwiger, L.A. (1968). Stimulation of pisatin production
in Pisum sativum by Actinomycin D and other compounds. Arch. Biochem.
Biophys. 126731-733.
Sequiera, L. (1983). Mechanisms of induced resistance in plants. Annu. Rev. Micro-
biol. 3751-79.
Shirashi, T., Oku, H., Yamashita, M., and Ouchi, S. (1978). Elicitor and suppressor
of pisatininduction in spore germination fluid of pea pathogen Mycosphaerella
pinoides. Ann. Phytopathol. Soc. Jpn. 44:659-665.
Stekoll, M., and West, C. A. (1978). Purification and properties of an elicitor of
castor bean phytoalexin from culture filtrates of the fungus Rhizopus stoloni-
fer. Plant Physiol.(Bethesda) 61~38-45.
Stoessel, P., and Magnalato, D. (1983). Phytoalexins in Phaseolus vulgaris and
Glycine maxinduced by chemical treatment, microbial contamination and fun-
gal infection.Experientia39153-154.
Tepper, C. S., and Anderson, A. J. (1986). Two cultivars of bean display a differen-
tial response to extracellular components from Colletotrichum lindemuthia-
num. Physiol. Mol. Plant Pathol.29(3):411-420.
Terekhova, V. A., Mustafa, M., and D’Yakov, Yu. T. (1980). Effect of metabolites
of potato tubers on the induction of the phytoalexin, rishitin by the fungus
Phytophthora infestans. Biol. naukc! (Mos.)2:83-87.
Tjamos, E. C., and Smith, I. M. (1974). The role of phytoalexins in the resistance
of tomato to Verticilliumwilt. Physiol. Plant Pathol.4249-259.
Uehara, K. (1958a). On the production of phytoalexin by the host plant as a result
of interaction between the rice plant and the blast fungus (Pyricularia oryzae
Cav.). Ann. Phytopathol. Soc. Jpn. 23:127-130.
Uehara, K. (1958b). On some properties of phytoalexin produced as result of the
interaction between Pisum sativumL. and Ascochytapisi Lib. I. on the activity
38 Purkayastha
affected by ultra-violet irradiation and on some physicochemical properties of

Phytoalexin. Ann. Phytopathol. Soc. Jpn. 23:230-234.
Uehara, K. (1958~).On the phytoalexin ofthe soybean pool in reactionto Fusarium
sp., the causal fungus of pod blight. I. Some experiments on the phytoalexin
production as affected by host plant conditions and on the nature of the phyto-
alexin produced.Ann. Phytopathol. Soc. Jpn. 23:225-229.
Uehara, K. (1959). On the phytoalexin production of the soybean pod inreaction to
Fusarium sp. the causal fungus ofpod blight. Ann. Phytopathol. Soc. Jpn. 2 4
224-228.
Uehara, K. (1963). On the production of phytoalexin by metallic salts. Bull. Hiro-
shima Agr. Coll. 2:41-44.
Van Etten, H. D., and Pueppke, S. G. (1976). Isoflavonoid phytoalexins.Biochemi-
cal Aspects of Plant Parasite Relationships (J. Friend and D. R. Threlfall,
eds.), Academic Press,New York, pp. 230-289.
Van Etten, H. D., Matthews, P. S., Tegtmier, K. J., Dietert, M. F., and Stein, J. I.
(1980). The association of pisatin tolerance and demethylation with virulence
on pea inNectria haematococca. Physiol. Plant Pathol. 16257-268.
Van Etten, H. D., Matthews, D.E., and Smith, D.A.(1982). Metabolism of
Phytoalexins. In Phytoalexins (J. A.Bailey and J. W. Mansfield,eds.),
Blackie, Glasgow, pp. 181-217.
Van Etten, H. D., Matthews, D. E., and Matthews, P. S. (1989a). Phytoalexin
detoxification: importance for pathogenicity and practical implications.Annu.
Rev. Phytopathol. 22143-164.
Van Etten, H., Matthews, D., Matthews, P., Miao, V., Moloney, A., and Straney,
D. (1989b). A familyof genes for phytoalexin detoxificationin the plant patho-
gen Nectria haematococca. Molecules in Plants and Plant-Microbe Interactions
(B. Lugtenberg, ed.), Springer-Verlag, Berlin,pp. 219-228.
Van den Heuvel, Van Etten, H. D., Serum, J. W., Coffen, D. L., and Williams,
T. H. (1974). Identification of a-hydroxyphaseollone, a phaseollin metabolite
produced by Fusarium solani.Phytochemistry 13:1129-1 13 1.
Vaziri, A., Keen, N. T., and Erwin, D. C. (1981). Correlation of medicarpin pro-
duction with resistant to Phytophthora megasperma f. sp. medicoginin alfalfa
(Medicago sativa)seedlings. Phytopathology 11(2):1235-1238.
Venkataraman, S., Lakshimnarasimhan, C., and Kalyansundaram, R. (1973). Anti-
genic determinants of host pathogen specificity, 2ndInt. Congr. Plant Pathol.
958 (Abstr.).
Wallace, J. M. (1940). Evidence of passive immunization of tobacco, Nicotiana
tabacum from the virus of curly top. Phytopathology 30673-679.
Ward, E. W. B., and Stoessl, A. (1972). Post-infectional inhibitors from plants. 111.
Detoxification of capsidiol, an antifungal compound from peppers. Phytopa-
thology 62~1186-1187.
Ward, E. W. B., and Stoessl, A. (1977). Phytoalexins from potatoes: evidence for
the conversion of lubiminto 15-dihydrolubimin by fungi.Phytopathology 67:
468-471.
Ward, H. M. (1905). Recent researches on the parasitism of fungi. Ann. Bot. 191-
54.
Watson, D. G., and Brooks, C. J. W. (1984). Formation of capsidiol in Capsicum

annuum fruits in response to non-specific elicitors. Physiol. Plant Pathol. 24:
331-337.
Weigard, F., Koster, J., Wettziers, H. C., and Barz, W. (1986). Accumulation of
phytoalexins and isoflavone glucosides in a resistant and a susceptible cultivar
of Cicer arietinum during infection with Ascochyta rabiei. J. Phytopathol.
115(3):214-221.
Wimalajeewa, D. L. S., and De Vay, J. E. (1971). The occurrence and characteriza-
tion of a common antigen relationship between Ustilago maydisand Zea mays.
Physiol. Plant Pathol.1523-535.
Wingard, S. A. (1928). Hosts and symptoms of ring spot: a virus disease of plants.
J. Agric. Res. 37(3):127-153.
Wood, R. K. S. (1986). Elicitors-What of? Vistas in Plant Pathology (A. Varma
and J. P. Verma, eds.), Malhotra, New Delhi, pp. 13-24.
Woodward, S., and Pegg, G . F. (1986). Rishitin accumulation elicited in resistant
and susceptible isolines of tomato by mycelial extracts and filtrates from cul-
tures of Verticillum albo-atrum. Physiol.Mol. Plant Pathol.29(3):337-347.
Yamoto, Y., Oku, H., Shirashi, T., Ouchi, S., and Koshizawa, K. (1986). Nonspe-
cific induction of pisatin and local resistance in Pea leaves by elicitors from
Mycosphaerellapinoides, M. melonis and M. lignilicola and the effect of sup-
pressor fromM. pinoids Jr. Phytopathology 117(2):136.
Yoshikawa, M. (1983). Macromolecules, recognition and the triggering of resis-
tance. Biochemical Plant Pathology (J. A. Callow, ed.), John Wiley t Sons,
New York, pp. 267-298.
Zook, M. N., and KuC, J. A. (1987). Differencesinphytoalexinelicitation by
Phytophthora infestansand Helminthosporium carbonumin potato. Phytopa-
thology 77(8):1217.
Zook, M. N.,Rush, J. S., and KuC, J. A. (1987). A role for Ca” in elicitation of
rishitin and lubimin accumulation inpotato tuber tissue. Plant Physiol. 84(2):
520-525.
2
Phytoalexins andHost Specificity
in Plant Diseases
Hachiro Oku and Tomonori Shiraishi
College of Agriculture, Okayama University, Okayama, JapanI
1. INTRODUCTION
One of the main interests in host-parasite interaction has been the role of
phytoalexins in plant diseases for about five decades. Reviews by Ingham
(1970), Cruickshank et al. (1971), and KuC (1972) show that the role of
phytoalexins in disease resistance rather
is diverse dependingon host-para-
site combinations. However, the roles were found to be more important
than generally believed when we conducted preciseand careful experiments.
Here we describe our results and discuss on the role of phytoalexins in host-
parasite specificity inplant diseases.
In addition, taking the rate of phytoalexin accumulation as an indica-
tion of a defense reaction, we found that some pathogenic fungi secrete
factors which suppress defense reactions of hosts. The factors, which are
called suppressors, are demonstrated to be responsiblefor the host-specific
pathogenicity of these fungal plant pathogens.
The mechanisms of elicitationand suppression of phytoalexin biosyn-
thesis and associated defense reactions are also described inthis chapter.
II. PHYTOALEXINS IN OBLIGATE PARASITISM

It had been widely believed that phytoalexins may not be involved in the
host-parasite relation of obligate parasitic diseases. This concept comes
from the idea that obligate parasitismis a kind of symbiosisand substances
harmful to parasites, such as phytoalexins, should not be produced by host
plants.
However, the possibility of phytoalexin formation in downy mildew of
tobacco was provided by Cruickshank and Mandryk (1960) and Shepherd
and Mandryk (1962,1963). Leath and Rowel1 (1970) suggested that a phyto-
alexin might be responsible for the inability of Puccinia graminis to infect
41
42 Oku and Shiraishi
corn leaves, although they didnot show any direct evidencefor its produc-
tion. Bailey and Ingham (1971) reported that phaseollin was detected in
bean leaves inoculated withUromyces appendiculatusonly when the leaves
responded with cellular browning, but 'they did not describe the role of
phaseollin inthe resistance of bean against rust fungus. Klarman and Ham-
merschlag (1972) extracted hydroxyphaseollin from soybean leaves which
had been infected with tobacco necrosis virus.
Oku et al. (197%)first presented evidencethat phytoalexin may partici-
pate in resistance of barley to powdery mildew diseaseand then proved the
role of pisatin in resistanceof pea against powdery mildew fungus (Oku et
al., 1975a). Later on, many reports appeared indicating that phytoalexin
may play some roles in resistance of obligate parasitic diseases. We will
describe our results on barley and pea powdery mildew diseases.
A. PhytoalexinActivity in Barley Powdery Mildew
Previously, Ouchi et al. (1974a,b) found that the preliminary inoculation

with an incompatible race of Erysiphe graminis induced resistance to a
primarily compatible race and that the preliminary inoculation with the
compatible race rendered barley leaves accessible to primarily incompatible
races. On the basis of these results we proposed the hypothesis that the
primary recognition of invading microbes as compatible or incompatible
may be controlled by an all-or-nothing type of gene action (Ouchi et al.,
1974b). During the analysis of these race-cultivar interactions, we found
that the growth of the secondary hyphaeof a compatible raceis markedly
inhibited on the resistance-induced barley leaves. This fact strongly sug-
gested the participation of phytoalexin-like substance in the resistance to
powdery mildew disease. Therefore, we conducted experiments and eluci-
dated that an antifungal activity was induced in barley leaves by infection
with powdery mildew fungi (Oku et al., 1975b,c).
The primary leaves of 8- to 10-day-old seedlings of barley cultivars
having different compatibilities with race 1, Kobinkatagi, HES4, Nos. 21
and 241 were densely inoculated with fresh conidia of E. graminis f. sp.
hordei race 1 with a soft hair brush. The inoculum density was adjusted to
give more than five conidia per epidermal cell. The inoculated plants were
incubated at 2OoC for an appropriate time and the leaf surface was rubbed
with a wet cotton ball to remove the fungal materialas much as possible in
order to negate a possible spore density effect. The adaxial surface of the
inoculated leaves was injured slightly by pressing gently with the tip of a
glass capillary. Then the droplets of deionized water were placed on the
injured parts of the leaves and incubated 2OoC for 4 hr to extract the
phytoalexin. The droplets were then collected by a glass capillary. The
Host Specificity in Plant Diseases 43
droplets similarly collected from noninoculated leaves served as control.

These diffusates (0.025 ml) were placed in small agarose blocks (25 m m 2
x 2 mm) on glass slides and allowed to diffuse into agarose for 3 hr at
room temperature.The antifungal activitywas tested by placing the conidia
of E. graminis hordei race 1 on diffusate-impregnated agarose blocks. After
incubation at 2OoC for 16 hr, the percentage of germination was determined
on the basis of conidia germinated with a characteristic germ tube. Percent
inhibition was calculated as a ratio of germination in the diffusate from
inoculated leavesto that in the exudates from noninoculated control.
The results of the time course studyfor the production of phytoalexin
activity are indicated in Fig.1.
Thus, the phytoalexin activity was found in barley powdery mildew
disease. The isolation and characterization of phytoalexin from powdery
mildew barley hasnot yet been accomplished, but the active factor responsi-
ble for this antifungal activity should be termed as phytoalexin Muller’s
in
definition (Muller, 1956) because it could not be detected in noninfected
healthy leaves.
Figure 1 Comparison of phytoalexin induction in some cultivar-race interactions

of barley powdery mildew disease. A = Kobinkatagi-race 1 (reaction type 4); B =
No. 21-race 1 (reaction type 3); C = No. 241-race 1 (reaction type 2); D = H. E.
S. Crace 1 (reaction type0). Percent inhibition was calculatedas follows:
( 070 germination in exudates from inoculated leaves
l o o - 070 germination in exudates from noninoculated leavesx loo)
B. Pisatin in Powdery MildewedPea

Further evidence for the production of phytoalexin in obligate parasitic
disease was provided by Oku et al. (1975a). They densely inoculated young
pea seedlings grown in a growth chamber with conidia of Erysiphe pisi
and extracted an antifungal principle from diseased seedlings. The active
substance was purified in crystalline form and identified as pisatinby physi-
cochemical analysis. Oku et al. obtained6 mg of purified pisatin from 40 g
fresh wt of powdery mildewed pea seedlings.
111. PRODUCTIVITY OF TWO PHASES OF PHYTOALEXINS A N D

THEIR RESPECTIVE ROLES IN DISEASE RESISTANCE
As evident from Fig. 1, phytoalexin activity was detectable 8-20 hr after
inoculation in incompatible combinations of barley powdery mildew dis-
ease. The activitywas proportional to the degree of incompatibility between
host cultivar and parasite. We considered the activity at this stage to be
first-phase phytoalexin. Phytoalexin activity first became detectable 8 hr
after inoculation, coinciding with the time taken to induce resistance and
also with the time of fungal penetration into host epidermal cell. Phyto-
alexin activity was also detected in the later stages of the pathogenesis of
barley powderymildew around the fungal colonies formed on the leaves of
the compatible host (Table 1). Phytoalexin activity at this stage was called
second-phase phytoalexin.
The first-phase phytoalexin seems to play a role in determining host-
parasite specificity because the activity was produced in parallel with the
degree of incompatibility. This fact also indicates that first-phase phyto-
alexin might be produced as a result of recognition by the host cell of a
Table 1 Accumulation of Phytoalexin in Tissue

Surrounding Fungal Coloniesof Erysiphe gaminis
Formed on Compatible Cultivar"
Age of colonies
(days)
Percent
inhibitionb
0 W
15 17
25 41
'Tested with Kobinkatagi-Race 1combination.
bEstimated from germ tube growth of Cochliobolus miya-
beanus.
'Significantly different at95% level of confidence.
Host Specificity
Diseases in Plant 45
parasite to be rejected. If so, the production of first-phase phytoalexin

should be suppressed in accessibility-induced leaves and increased resis-
tance-induced barley leaves.
A barley cultivar, Kobinkatagi, was first inoculated with a compat-
ible (E. graminis hordeirace 1) or incompatible race(E. graminis triticirace
tZ), incubated for 48 hr, and the inoculum removed and then inoculated
with a second race.The effect of the inoculation on the phytoalexin produc-
tion in response to the second fungus was compared (Fig. 2) (Oku et al.,
1975b).
The production of first-phase phytoalexin in incompatible interaction
was suppressed when the barley leaves were primarily inoculated with a
compatible race. Conversely, the preliminary inoculation withan incompat-
ible race conditioned the leaves to produce the same level of phytoalexin on
subsequent inoculation regardless of the compatibility of the challenger
race.
Similarly, two phasesof phytoalexin production were found in pisatin
accumulation when inoculated with compatible or incompatible pathogens
on pea leaves (Oku et al., 1975a; Shiraishi et al., 1977).
That is, pisatin was
detectable 12-15 hr after inoculation with the incompatible powdery mil-
dew fungus E. graminis. The inoculation of resistant pea cultivarto pow-
dery mildew disease (Resistant Stratagem, withE. pisi or E. graminis) also
induced first-phase pisatin.
INOCULATION PERCENT INHIBITIONOF CONIDIAL GERMINATION

1ST 2ND 20 40
NONE - RACE 1
RACE 1 - NONE
RACE 1 - RACE 1 I
RACE 1” T2 I I
NONE - T2
T2 - NONE
T2 - T2 I
T2 - RACE 1
Figure 2 Effect of accessibility and resistance induction on phytoalexin produc-

tion in barley leaves. Accessibility (or resistance) was induced in barley leaves (CV.
Kobinkatagi) by inoculation with race 1 (or t2), and phytoalexin production was
estimated 12 hr after secondinoculation.
46 Shiraishi Oku and
In contrast, the inoculation of pea leaves with compatible pathogens,

E. pisi or Mycosphaerella pinodes, did not elicit first-phase pisatin but
induced second-phase pisatin (21 hr after inoculation).
Here the absence of first-phase pisatin in pea leaves seems to baffle our
understanding, if we consider that pisatin is only an antifungal substance,
because both pathogens E. pisi and M. pinodesare highly tolerant to pisatin
(Oku et al., 1975a; Shiraishi etal., 1978a). In other words, these fungi are
unnecessary to suppress the pisatin production because their growthis not
inhibited by pisatin.
These results lead tous an idea that pisatin hasanother function besides
its antifungal activity,.and we examined the effectof artificial administra-
tion of pisatin at early stages of infection by these fungi on the establish-
ment of infection.
As indicated in Fig. 3, the infectionof pea leaves byE. pisi was strongly
inhibited when pisatin was administered within 15after hr inoculation even
at concentrations which do not exert any effect on conidial germination,
i.e., 30 or 100 ppm. Administration of pisatin at 16 or 19 hr postinoculation
however, did not give any significant effect on infection establishment by
E. pisi (Oku et al.,1976).
Similarly, M. pinodes is highly tolerant to pisatin as assessed by the
inhibitory activity to spore germination or germ tube elongation (EDSO=
500 ppm). This fungus, however, could not establish infection on pea if a
very lowconcentration (less than 50 ppm) of pisatinwas administered artifi-
cially into thesporeinoculum.The perforation ofcellophanesheet
by germ tube of M. pinodes was also inhibitedby the same concentration
of pisatin that inhibited infection on the host, showing that infection-
inhibiting activity of pisatin is rather a direct action to the pathogen than
the host-mediated action (Shiraishi etal., 1978a).
These results obtained by experiments with E. pisi and M. pinodes
suggest that if we direct our attention to the inhibitoryeffect of phytoalex-
ins on the infection establishment of a pathogen, their role in determining
host-parasite specificity should be much more important than has been
considered (Christensonand Hadwiger, 1973; Kirhly et al., 1972; Pueppke
and VanEtten, 1974). The results also suggest that some substances which
were synthesized during the infection processbut have no antibiotic activity
may play a significant part in determining host-parasite specificity.In this
connection, Berard et al. (1972) reported that a diffusate from the incom-
patible interaction of bean anthracnose protected the bean plant from the
compatible pathogen though the diffusate had no antibiotic activity. In
fact, we detected a substance whichhasnoantifungalactivitybuthas
infection-inhibiting activity in pea leaves1 hr after treatment with elicitor
from M. pinodes. This substance, an infection inhibitor in our terms, is
b
a
6 8 10 12 13 15 16 19
Time after inoculation to ,pisatin application (hr)
Figure 3 Effect of pisatin on infection establishment of E. pisi on pea leaves.

After the lower epidermis of half leaves was removed, the opposite surface was
inoculated with E. pisi, and the stripped mesophyll tissue was brought into direct
contact with(A) 100 ppm, (B) 30 ppm, or (C) 10 ppm pisatinsolution at the
above time intervals. Infection frequencywas estimated 48 hr after inoculation. (a)
represents a significant difference and (b) a nonsignificant difference from control
respectively at p = 0.05.
not yet be characterized but the molecular weightis 329 as determined by

mass spectrography (Yamamoto etal., 1986).
The inoculation of the pathogenic fungi E. pisi and M. pinodes on pea
leaves did not induce pisatin biosynthesis until the infection was completed,
unlike the nonpathogens E. graminis and Stemphylliumsarcinaeforme,
which induced pisatinto a detectable level by 12 hr after inoculation. These
facts strongly suggest that the pathogenic fungi havean ability to suppress
to
the first step of the defense reaction, hence pisatin production, in order
avoid the inhibitory action of pisatinon penetration.
Table 2 Antifungal Activity of Second-Phase

of Erysiphe
Phytoalexin Against Several Races
graminis
rcent Race
Race 1 98.ab
Hr 74
Hr 4 97.O
E. graminis triciti t2 97.0
'Refer to Fig.1.
bNosignificant difference at95% level of confidence.
The role of second-phase phytoalexin may be diverse according to the

sensitivity of pathogenic fungus against phytoalexin produced bythe host
plant.
Table 2 shows the sensitivity of several races of E. graminis to the
phytoalexin solution obtained fromthe surrounding barleyleaf tissue colo-
nized by powdery mildew fungus. Thus, all races of E. graminis are simi-
larly sensitive to barley phytoalexin. In these cases second-phase phyto-
alexinmightpossiblyberesponsible for the resistanceagainstcolony
development. As a matter of fact, the colony of barley powdery mildew
fungus usually grows very rapidly in the early stage of pathogenesis and
gradually levelsoff as the incubation timeis prolonged, finally stopping the
growth at the later stage. The phytoalexin accumulationwas most promi-
nent at the tissues surrounding the colony that ceased to grow. In this
connection, Pierre (1971) suggested that phaseollin and another unidenti-
fied phytoalexin induced in bean plants may play a role in restricting the
size of the lesions produced in bean leaves. The same is true of medicarpin
in alfalfa (Higgins, 1972) and also of rishitin in potato tubers (Sat0 et al.,
1971).
In contrast, second-phase phytoalexins have nothing to do with the
development of lesions formedby the pathogens which are tolerant to the
host phytoalexin.
As described above, pea powdery mildew fungus is highly tolerant to
pisatin. Therefore, once infection is established,the fungus colonizes freely
on pea plant tissues, even though high levels of pisatin accumulate.
Heavy inoculation withE. pisi causes wilting of inoculated pea leaves.
In inoculated leaves, approximately300 pg/g fresh weight of pisatin accu-
mulates 4 days after inoculation (Oku et al., 1975a), when wilting begins.
This fact suggests that pisatin is the main cause of wilting because sucha
powerful toxin witha wilting effectcannot be produced by the pathogenic
Host Specificity
fungus which is a typical obligate parasite. In fact, 300 pg/ml of pisatin

causes complete burst of isolated protoplast from mesophyll layers of pea
leaves (Shiraishi etal., 1975). Thus, the main cause of wilting in pea leaves
heavily infected with powdery mildew fungus may notthe be fungus itself,
but pisatin producedby the host leaves affectingthe plasma membrane of
the same host cells. In other words, the wilting by this disease may be said
to be a self-destruction effect resulting
from a defense mechanism.
W. MECHANISM OF SUPPRESSION OF DEFENSE REACTION

INCLUDING THE ACCUMULATION OF
FIRST-PHASE PHYTOALEXIN
As described, the pathogenic fungus seemed to have some mechanisms to

suppress defense reactions including the accumulation of first-phase phyto-
alexins. Therefore, we searched for substances which suppress the defense
reaction of host usinga pea-M. pinodes system.
Since first-phase phytoalexins accumulate at an early stage of patho-
genesis, our search has been focused on the pycnospore germination fluid
of the pathogenic fungus M. pinodes. As the result, we obtained both
elicitor and suppressor of pisatin biosynthesis from spore germination fluid
(Shiraishi et al., 1978b). That is, dialyzation or ultrafiltration of germina-
tion fluid gave an elicitor for pisatin synthesis in high molecular weight
fraction and suppressor inlow molecular weight fraction. The elicitor was
-
found to be a polysaccharide (Mw 70,000). At least two kinds of suppres-
sors are contained in spore germination fluid and they are glycopeptides.
Asshownin Table 3, suppressor counteracts the activity of elicitor
Table 3 Effect of Fractions Prepared from Spore Germination

Fluid ofMycosphaereiia pinodeson Pisatin Induction in Pea
Leaves
Pisatin accumulated*
Treatment
with leaves)
fresh
bg/g of
water Distilled 0
Concentrated
fluid
germination 0
molecular
weight
High fraction 27.2
molecular
weight
Low fraction 0
Mixture of low and high molecular
weight fraction (1 : 1) 0
‘Pisatin was determined after

24 hr incubation at2OOC.
Table 4 CorrelationBetween the Leguminous Host Range of Mycosphaerella

pinodes and Induction of Susceptibility in Those Plants to Alternaria alternataby
the Suppressor (F3 from M. pinodes
Degree of colonization by
Legume A . alternaria A . alternaria
species M. pinodes (15B) (15B) + F5
Arachis hypogaea 0 0 0
Glycine may 0-1 0 0
Lespedeza buergeri 2 0 2
Lotus corniculatus 0 0 0
L. bicolor 0 0 0
Medicago sativa 1 0 1
Millettia japonica 2 0 l
Pisum sativum 4 0 4
Trifolium pratense 1 0 1
T. repens 0 0 0
Viciafaba 0 0 0
Vigna sinensis 0 0 0
Number indicates the amount of infection hyphae; from nothing (0) to abundant formation
(4).
(Shiraishi et al., 1978b). Inthe presence of suppressor, eight species of pea

nonpathogens established infection on pea leaves. Among these,Alternaria
alternata 15B, an avirulent isolateof pear pathogen,grew and colonized on
pea and formed conidia on suppressor-treated pea leaves 4-7 days later.
Further, as shown in Table 4, Alternaria 15B could infect five species of
leguminous plants to which M. pinodes was pathogenic in the presence of
the suppressor, but not on other plant species. Inother words, the specific-
ity of the biological activity ofthe suppressor coincided with the host range
of the producer fungus,M. pinodes (Oku etal., 1980).
Thus, it was found that suppressoractsnotonlyassuppressor of
pisatin biosynthesis but also as the determinant of pathogenicity of M.
pinodes by suppressingthe other defense reactions.
The same type of suppressors were found in the spore germination fluid
of M . melonis and M. ligulicola (Oku et al., 1987).
Since the suppressor causes no visible injury to pea leaf tissue and
isolated protoplast, it can hardly bea host-specific toxin.
A mannan glycoprotein isolated from culture filtrate of Phytophtora
megasperma f. sp. glycinea was reported to suppress the glyceollin accumu-
lation in soybean ina race-specific manner (Ziegler and Pontzen, 1982).
Host Specificity
As the mode of action, we recently found that the suppressor inhibits

the plasma membrane ATPase (Yoshioka al., et 1990). The inhibitory activ-
ity of the suppressor from M. pinodes is nonspecific invitro, i.e., active to
plasma membrane ATPases prepared from pea, bean, cowpea, soybean,
and barley, contradicting the result shown in Fig.4. However, the activity
was found to be specificat the tissue level (Shiraishi et al.,1991) by electron
microscopy accordingto the method of Hall et al.(1980). The principle of
this method is that when the ATPase in the tissue is active, the phosphate
released from added ATP is precipitated as lead base by addition of lead
nitrate, but not when ATPase is inactive. As shown in Fig. 5, vanadate
Relative ATPase activity

(%to water control)
0 50 l00
I 1
0 5 0 mY KN03
m100 uY NaN3
Pea
j 0 1 BY Na3V04
D O p p m suppressor
Cowpea
Soybean
Kidney bean
Barley
. ., ,I. . .1 .~ ..._.
.,. . . . ..
I
Figure 4 Effect of inhibitorsandsuppressorfrom Mycosphaerella pinodes on

ATPases prepared from plasma membrane fractions
of various plants.
Water Na 9 V04 Suppressor
m
a,
a
m
a,
ah
Figure 5 Effect of orthovanadate and suppressor from M.pinodes on ATPase

activity of various plants invivo detected by lead precipitation procedures (see text).
inhibits ATPase nonspecifically but the suppressor inhibits ATPase only of

pea.
Further, this specific inhibitionof ATPase was also found at the inter-
face of host and parasite (Shiraishi etal., 1991). That is, as shown in Fig.
6,
M. pinodes inhibits ATPase of pea until 6 hr after inoculation but M.
ligulicola, a chrysanthemum pathogen, does not.
Since the suppressor inhibits the decline of pH of droplets of elicitor
solution or water placed on pea leaves, the suppressor may inhibit the
. .
Figure 6 Evidence that pea plasma membrane ATPase is suppressed by the pea
pathogen M. pinodes but not by the nonpathogen M. ligulicola at host-parasite
interfaces.
proton pump ATPase of pea plasma membrane, hence the activity of pea
cells to defend. The inhibitory activity of suppressor on pea plasma mem-
brane ATPase in vivo was found to be temporary.
V. PARTICIPATION OF PROTEIN KINASE IN ELICITATION

OF THE DEFENSE REACTION
Signal transduction in the elicitation of a defense reaction is a matter of
worldwide interestfor physiological plant pathologists.
Sincealmost all elicitorssecretedbypathogens are highmolecular
weight substances, it is hardly consideredthat elicitors directly activatethe
gene for resistance. Therefore, many scientists believe thereto be receptors
in plant cell surfacefor elicitors and several scientists demonstrate the bind-
ing of elicitors to receptors @bel and Grisebach, 1988; Yoshikawa et al.,
1983). As the result of interaction between elicitor and receptor, some signal
substances which directlyor indirectly activatethe gene for resistance may
be producedand so activate; hence defense reactions.
We have recentlyfound that protein kinasemay playan important role
in the process of elicitationof defense reaction in pea plant (Shiraishi et al.,
1990). Verapamil, a Ca” channel blocker, and K-252a, a strong inhibitor
of protein kinase, inhibit pisatin accumulation in pea epicotyl which had
been treated with elicitorfrom M. pinodes. Since LaCl, and EGTA didnot
inhibit pisatin accumulation and, further, since verapamil inhibits pisatin
accumulation even if applied 6 hr after the elicitor treatment, when the
pisatin biosynthetic pathway has already been activated, it can hardly be
considered that Ca” plays a role as second messenger for signaling of
pisatin biosynthesis in pea tissue induced by elicitor. On the other hand,
K-252a inhibits pisatin accumulation when applied to pea epicotyl only
before the elicitor treatment. This fact suggests that signal transduction
occurs very rapidly after the elicitor treatment and, further, that protein
kinase may play a key role inthe signal transduction for pisatin biosynthesis
because K-252a inhibitsmarkedly the invitrophosphorylation ofpea
plasma membrane proteins.
Rapid and transient phosphorylation of specific proteins by treatment
with fungal elicitors was reported in parsley cell suspension cultures by
Dietrich et al.(1990), and they also suggestthat protein phosphorylationis
involved inthe signal transduction process following elicitor treatment.
As for the substrate of protein kinase, Harrison et al. (1991) suggested
that the phosphorylation of regulatory proteins bindingto silencer region
may regulate the expression of CHS gene in bean because the DNA binding
activity ofthe proteins is lost by treatment with alkaline phosphatase. Datta
and Cashmore (1989) reported that the phosphorylationof nuclear protein
binding to promoters of certain photoregulatedgenes decreasesthe binding
activity. Kobayashi (1991) found, using his system of barley and Erysiphe
pisi, a non-host resistance combination, that the rearrangement of cyto-
plasmic strand, mainly composed of actin filaments, plays a very important
rolein the expressionofdefensereaction. That is,severalcytoplasmic
strands composed of actin filaments are rearranged and appeared underthe
appressoria when inoculated. However, we found that K-252a completely
inhibits the appearance of actin filaments. This isnot direct evidence,but it
is possible to assume that the phosphorylation of actin by protein kinase
may be one of the processes for the expression of defense reaction not only
in pea but also in barley.
VI. R E G U L A T I O N O F EXPRESSION O F GENES FOR

PHYTOALEXIN BIOSYNTHESIS
Much information is available to indicate that the infection of plants by

plant pathogens induces various kindsof isozymes which are not detected
in healthy plant tissues, and that the newly synthesized isozymes are the
result of net protein synthesis. Thesefacts suggest that the genes encoding
new proteins are activated by infection.
Recent progress of genetic engineering enables these phenomenato be
understood at the gene level. Suspension-cultured plant cells are generally
used for this type of experiment because the signal molecules distribute
simultaneously to each plant cell. However, in some plants or organs the
al.
resistance is reportedto be expressed at the tissue or organ level but not at
the cellular level.For example, Tomiyamaet (1967) clarified that potato
tuber tissue needs more than 10 cell layersin order to keep resistance against
late blight disease.
Anyway, the feature of activation of genes encoding phytoalexin bio-
synthesis was examined using cultured bean cells by several scientists. Iso-
lated bean cells cultured in the dark were treated with elicitor prepared
from Colletotrichum lindemuthianurn, and the amount of mature mRNA
accumulated in the cellswasdeterminedby northern blot hybridization
analysis with cDNA of phenylalanine ammonia-lyase (PAL), chalcone syn-
thase (CHS), and chalcone isomerase (CHI), which are key enzymes for
isoflavonoid phytoalexin biosynthesis. Results show that all mRNA (PAL,
CHS, and CHI) accumulate very rapidly, reaching a maximum level 3 hr
after treatment with elicitor, and then decrease to the original level (Ed-
wards et al., 1985; Hahlbrock and Scheel, 1981, Lamb et al., 1989; Mehdy
and Lamb, 1987; Ryder et al., 1984).
We studied the effect of elicitor and suppressor isolatedfrom the pea
pathogen M. pinodes on the defense reaction of pea with respect to the
pisatin biosynthesis at enzyme (PAL and CHS, which are key enzymes for
pisatin biosynthesis) and their gene activation, using pea epicotyl tissues
(Yamadaetal., 1989). Pea epicotylgrownin the dark has a minimal
amount of PAL and CHS. However, treatment of etiolated pea epicotyl
tissue with elicitor activates the accumulation of PAL and CHS mRNAs
within an hr, followed by an increase in enzyme activity, and then pisatin
biosynthesis. Concomitant presence of suppressor with elicitor results the in
delayof the gene transcriptions for 3 hr and increase of PAL enzyme
activity for 6 hr. As the result, pisatin accumulationis delayed for 6-9 in
pea epicotyl. Thusit was demonstrated that the pisatin biosynthesis is acti-
vated in pea epicotyl tissues by treatment with elicitorat the gene level,and
the suppressor of the pathogenic fungus suppresses the expression of genes

responsible for the defense reaction.
It should be noted that our results are in close agreement with the in
situ hybridization experiment ofthe accumulation of potato PAL mRNA
reported by Cuypers et al., (1988). They found a remarkable difference in
the timing of PAL mRNA accumulation between compatible and incom-
patible interactions of late blight disease. A marked increase in accumula-
tion of PAL mRNA was observed3 hr after inoculation withan incompati-
blerace at the infection site, but it was 6 hr after inoculation with a
compatible race. The coincidence of a 3-hr delay in the accumulation of
PAL mRNA in compatible interaction with our result of suppressor-treated
pea epicotyl suggests the ideathat some factors such as suppressors might
get involvedin the compatibility of potato late blight disease.
In our experimental results, the expression and suppression of PAL
and CHS genes are coordinately regulated by treatment with elicitor and
suppressor. Coordinate activation of transcription of PAL and CHS genes
is also reported in bean suspension cultured cells as well as bean hypocotyl
by treatment with elicitor (Edwards et al., 1985; Hahlbrock and Scheel,
1989; Lamb et al., 1989; Mehdy and Lamb, 1987; Ryder et al., 1984).
VII. EPILOG
In this chapter, the importance of phytoalexinsin resistance mechanisms of
plant was described from biological and biochemical perspectives.
There isa large amountof information on the mechanism of resistance
of plants against invading microorganisms but little is known about the
pathogenicity of pathogens, especially how pathogens escape the defense
reaction of host plant to infect and colonize these plants. Host-specific
toxins are certainly pathogenicity factors, but they have been found in
limited genera of pathogenic fungi and are hardly considered to be the
determinant of pathogenicity in obligate parasites.
The most important description in this chapter maybe the finding of
suppression of defense reaction and the mode of action. Though the known
data on suppressors are limited at the present time, the biological evidence
shows that many more suppressors will befound in the future to be determi-
nants of pathogenicity.
Another major interest on our part is to clarify the series of molecular
events inthe process of defense reaction elicitation.
Progress in these basic fieldsof plant pathology may shed lightfor the
development ofnew control measure of plant diseases without polluting the
environment.
REFERENCES
Baily, J. A., and Ingham, J. L. (1971).Phaseollin accumulation in bean(Phaseolus
vulgaris) in response to infection by tobacco necrosis virus and rust Uromyces
appendiculatus. Physiol. Plant Pathol.1:451-456.
Berard, D. F., KuC, J., and Williams, E. B. (1972).A cultivar specific protection
factor from incompatible interactions of green bean with Colletotrichum linde-
muthianum. Physiol. Plant Pathol.2:123-128.
Christenson, J. A., and Hadwiger, L. A. (1973). Induction of pisatin formation in
pea foot region by pathogenic and nonpathogenic clones of Fusarium solani.
Cruickshank, I. A. M., and Mandryk, M. (1960). The effect of stem infection of
tobacco with Peronospora tabacinaAdam on foliage reactionto blue mold. J.
Aust. Inst. Agric. Sci.26:369-372.
Cruickshank, I. A. M., Briggs, D., and Perrin, R. (1971). Phytoalexins as determi-
nants of disease reaction in plants. J. Indian Bot. Soc. Golden Jubilee 50A:
1-11.
Cuypers, B., Schmelzer, E., and Hahlbrock, K. (1988). In situ localization of rap-
idlyaccumulatedphenylalanineammonia-lyasemRNA around penetration
sites of Phytophthora infestansin potato leaves. Mol. Plant-Microbe Interact.
1:157-160.
Datta, N., and Cashmore, A. R. (1989). Binding of pea nuclearprotein to promot-
ers of certain phosphorylated genes is modulated by phosphorylation. Plant
Cell 1:1069-1077.
Dietrich, A.,Mayers, J. E., and Hahlblock, K. (1990). Fungalelicitortriggers
rapid, transient, and specific protein phosphorylationin parsley cell suspension
cultures. J. Biol. Chem.265:6360-6368.
Ebel, J., and Grisebach, H. (1988). Defense strategies of soybean against the fungus
Phytophthoru megasperma f. sp. glycinea: a molecular analysis. Trends Bio-
chem. Sci. I3:23-27.
Edwards, K.,Cramer, C. L., Bolwell, G. P., Dixon, R. A., Schuch, W., and Lamb,
C. J. (1985). Rapid and transient induction of phenylalanine ammonia-lyase
mRNA in elicitor treated bean cells. Proc. Natl. &ad. Sci. USA 82:6731-6735.
Hahlbrock, K., and Scheel, D.(1989). Physiology and molecular biology of phenyl-
propanoid metabolism. Ann. Rev. Plant Physiol.40347-369.
Hall, J. L., Browning, A. J., and Harvey, D. M. R. (1980). The validity of the lead
precipitation technique for the localization of ATPase activity in plant cells.
Protoplasma 104:193-200.
Harrison, M. J., Lawton, M. A., Lamb, C.J., and Dixon, R. A. (1991). Character-
ization of a nuclearprotein that binds to three elements withthe silencer region
of a bean chalcone synthase.Proc. Natl. Acad. Sci. USA, 882515-2519.
Higgins, V. J. (1972). Role of phytoalexin medicarpinin three leaf spot diseases of
alfalfa. Physiol. Plant Pathol.2289-300.
Ingham, J. L. (1970). Phytoalexins and other natural products as in plant disease
resistance. Bot. Rev.38343-423.
Kirdy, Z., Berna, N., and Ersek, T. (1972). Hypersensitivity as a consequence,not

the cause, of plant resistance to infection. Nature 239456-458.
Kobayashi, I.(1991). Recognition of Erysiphepisi by barley coleolptilecells and the
induction of inaccessibility,PhD thesis, Okayama University, pp. 26-42.
Klarman, V. L., and Hammerschlag, F. (1972). Production of the phytoalexin,
hydroxyphaseollin, in soybean leaves inoculated with tobacco necrosis virus.
Kut, J. (1972). Phytoalexins. Ann. Rev. Phytopathol. 10207-232.
Lamb, C. J., Lawson, M. A., Dron, M., and Dixon. R. A. (1989). Signal transduc-
tion mechanism for activation of plant defenses against microbial attack. Cell
56215-224.
Leath, K. T., and Rowell, J. B.(1970). Nutritional and inhibitory factors in the
resistance ofZea maysto Puccinia graminis. Phytopathology 601097-1 100.
Mehdy, M. C., and Lamb, C. J. (1987). Chalcone isomerase cDNA cloning and
mRNA induction by fungal elicitor, wounding and infection. EMBO J. 6 1527-
1533.
Muller, K. 0. (1956). Einige Einfache Versuche mm Nachweis von Phytoalexinen.
Phytopathol. Z. 22237-259.
Oku, H., Ouchi, S., Shiraishi, T., and Baba, T. (1975a). Pisatin production in
powdery mildewed pea seedling.Phytopathology 651263-1267.
Oku, H., Ouchi, S., Shiraishi, T., Baba, T., and Miyagawa, H. (1975b). Phyto-
alexin production in barley powdery mildew as affected by thermal and biologi-
cal predispositions.Proc. Jpn.Acad. 51:198-201.
Oku, H., Ouchi, S., Shiraishi, T., Komoto, Y., and Oki, K. (1975~).Phytoalexin
activity in barley powdery mildew.Ann. Phytopathol. Soc. Jpn. 41: 185-191.
Oku, H., Shiraishi, T., and Ouchi, S. (1976). Effect of preliminary administration
of pisatinto pea leaf tissueson subsequent infection by Erysiphepisi DC. Ann.
Phytopathol. Soc. Jpn. 42597-600.
Oku, H., Shiraishi, T., and Ouchi, S. (1987). Role of specific suppressorsin patho-
genesis ofMycosphaerella pinodes. Molecular Determinants of Plant Diseases
(S. Nishimura, C. D. Vance,and N. Doke, eds.), Japan Scientific Press,Tokyo/
Springer-Verlag, Berlin, pp. 145-156.
Oku, H., Shiraishi, T., Ouchi, S. Ishiura, M., and Matuseda, R. (1980). A new
determinant of pathogenicity in plant disease. Naturwissenschqften 62310-
311.
Ouchi, S., Oku, H.,Hibino, C., and Akiyama, I. (1974a). Induction of accessibility
and resistance inleaves of barley by some racesof Etysiphe graminis. Phytopa-
thol. Z. 7924-34.
Ouchi, S., Oku, H., Hibino, C., and Akiyama, I. (1974b). Induction of accessibility
to a nonpathogen by preliminary inoculation with a pathogen.Phytopathol. Z.
79142-154.
Pierre, R. (1971). Phytoalexin induction in beans resistant or susceptible to Fu-
sarium and Thielaviopsis. Phytopathology61:332-337.
Pueppke, S. G., and VanEtten, H. D.(1974). Pisatin accumulation and lesion
development in peas infected withAphanomyces euteiches,Fusarium solani f.
sp. pisi, or Rhizoctonia solani. Phytopathology 64: 1433-1440.
Host
PlantSpecificity in Diseases 59
Ryder, T. B., Cramer, C. J., Bell, J. N., Robbins, M. P., Dixon, R. A., and Lamb,
C. J. (1984). Elicitor rapidly induces chalcone synthase mRNA in Phaseolus
vulgaris cells at the onset of the phytoalexin defense response. Proc. Natl.
Acad. Sci. USA, 815724-5728.
Sato, N., Kitazawa, K., and Tomiyama, K. (1971). The role of rishitin in localizing
the invading hyphae of Phytophthora infestans in infection sites at the cut
surfaces of potato tuber. Physiol. Plant Pathol.1:289-295.
Shephard, C. J., and Mandryk, M. (1962). Auto inhibitors of germination and
sporulation in Peronospora tabacina Adam. Trans. Br. Mycol. Soc. 45:233-
244.
Shephard, C. J., and Mandryk, M. (1963). Germination of conidiaof Peronospora
tabacina Adam. 11. Germination in vivo.A u t . J. Biol. Sci.1677-87.
Shiraishi, T., Araki, M., Ysohioka, H., Kobayashi, I., Yamada, T., Ichinose, Y.,
Kunoh, H., and Oku, H.(1991). Inhibition of ATPase activity in pea plasma
membranes in situ by a suppressor from a pea pathogen, Mycosphaerella pi-
nodes. Plant Cell Physiol. 32:1067-1075.
Shiraishi, T., Hori, N., Yamada, T., and Oku, H.(1990). Suppression of pisatin
accumulation by an inhibitor of protein kinase. Ann. Phytopathol. Soc. Jpn.
56261-264.
Shiraishi, T., Oku, H., Isono, M., and Ouchi, S. (1975). The injurious effect of
pisatin on the plasma membrane of pea. Plant Cell Physiol.16939-942.
Shiraishi, T., Oku, H., Ouchi, S., and Tsuji, Y. (1977). Local accumulation of
by powdery mildew fungi.Phytopa-
pisastin in tissues of pea seedlings infected
thol. Z. 88131-135.
Shiraishi, T., Oku, H.,Tsuji, Y.,and Ouchi, S. (1978a). Inhibitory effect of pisatin
on infection process of Mycosphaerella pinodes on pea. Ann. Phytopathol.
Soc. Jpn. 44641-645.
Shiraishi, T.,Oku, H., Yamashita, M., and Ouchi,S. (1978b). Elicitor and suppres-
sor of pisatin induction in spore germination fluid of pea pathogen, Mycos-
phaerella pinodes.Ann. Phytopathol. Soc. Jpn.44:659-665.
Tomiyama, K., Sakai, R., Sakuma, T., and Ishizaka, N. (1967). The role of poly-
phenols in the defense reaction in plants induced by infection. The Dynamic
Role of Molecular Constituentsin Plant-Parasite Interaction (C. J. Mirocha
and I. Uritani, eds.), Bruce, St. Paul, MN, pp. 165-182.
Yamada, T., Hashimoto, H., Shiraishi, T., and Oku, H. (1989). Suppression of
pisastin, phenylalanine ammonia-lyase mRNA and chalchone synthase mRNA
accumulation by a putative pathogenicity factor from the fungus Mycosphaer-
ellapinodes. Mol. Plant-Microbe Interact.2~256-261.
Yamamoto, Y., Oku,H.,Shiraishi, T., Ouchi, S., and Koshizawa, K. (1986).
Non-specific induction of pisatin and local resistance in pea leavesby elicitors
from Mycosphaerellapinodes, M. melonis and M. ligulicola and the effect of
suppressor fromM. pinodes. J. Phytopathol. 112136-143.
Yoshikawa, M., Keen, N. T., and Wang, M. C. (1983). A receptor on soybean
membranes for a fungal elicitor of phytoalexin accumulation. Plant Physiol.
73:497-506.
Yoshioka, H., Shiraishi, T., Yamada, T., Ichinose, Y.,and Oku, H.(1ggO). Sup-
pression of pisatin production and ATPase activity in pea plasma membrane

by orthovanadate, verapamil and a suppressor fromMycosphaerella pinodes.
Plant Cell Physiol.3 8 1139-1 146.
Ziegler, E., and Pontzen, R. (1982). Specific inhibition of glucan-elicited glyceollin
accumulation in soybeans by an extracellular mannan-glycoprotein of Phy-
tophthora megasperma f. sp. gljcinea. Physiol. Plant Pathol. 20321-332.
3
Elicitor and the Molecular
of Phytoalexin Elicitation
Masaaki Yoshikawa
Hokkaido University, Sapporo,Japan
1. INTRODUCTION
Disease resistance in many plant-fungal pathogen interactions has been
suggested to be due to inducible production of antibiotic low molecular
weight compounds, i.e., phytoalexins (Keen, 1981). The induction of phy-
toalexins in infected plants is presumed to be mediated by an initial recogni-
tion processbetweenplants and pathogens whichinvolvesdetectionof
certain unique molecules of pathogen origin, termed elicitors, by recogni-
tional receptor-like molecules in plants, thereby setting off a cascade of
biochemical events leading ultimatelyto phytoalexin accumulation (Yoshi-
kawa, 1983; Yoshikawaand Masago, 1982). However, detailed mechanisms
involved in each biochemical process leading to the phytoalexin production
are poorly understood. The main subject of this chapter is the molecular
basis ofthe elicitation of glyceollin,a phytoalexin (Yoshikawa et al., 1978a)
produced by the expression of monogenic resistance in soybean (Glycine
max) to incompatible races of Phytophthora megasperma f. sp. glycinea
(Pmg).
II. ELICITORSOFPATHOGEN ORIGIN THAT FUNCTION IN

GLYCEOLLIN PRODUCTION IN VIVO
Isolated mycelial walls of many fungi possess potent elicitor activity to
inducephytoalexinaccumulationinplants.There are, however,several
unresolved questions regardingthe in vivo involvement ofthe wall-associ-
ated elicitors. A major detracting argument arises from the observation
that active elicitor moieties can only be extracted from fungal wallsby
severe treatments such as autoclaving or exposure to acids or alkalines,
which are unlikely to exist in biological environment. This raisesthe ques-
tion of how normally insoluble elicitor molecules on or in the fungal walls
may come in to contact with the corresponding recognitional molecules in
61
62 Yoshikawa
plant cells during natural infection processes. Although unnaturally ex-

tracted elicitors are frequentlyusedinmanytypesofstudiesincluding
biological and structural analysis, the possibility exists that such elicitors
may not be those functioning in phytoalexin elicitation in fungus-infected
plant tissues (Yoshikawa, 1983).
Our studies (Yoshikawa et al., 1981) with the soybean-Pmg system
indicated that highly active carbohydrate elicitors of glyceollin production
werereleased into a soluble form from insolublemycelialwallsof the
fungus by a factor contained in soybean tissues. These studies provideda
new insight for the in situ production of soluble elicitors which could be
more efficiently recognized by plant cells. The elicitor release occurred as
rapidly as 2 min after incubation of mycelial walls or actively growing
hyphae with soybean tissues, suggesting that the process may beimportant
as the earliest plant-pathogen interaction leading to the induction of gly-
ceollin production.
The factor capable of releasing elicitors was purified from soybean
cotyledons to apparent homogeneity and shown to possess &1,3-endoglu-
canase activity (Keen and Yoshikawa, 1983). Identity of the factor to the
glucanase was further supported by the facts that &1,3-endoglucanase puri-
fied from the bacterial culture of Arthrobacter Zuteus possessed similar
elicitor-releasing activityand that antibodies raised againstthe highly puri-
fied soybean glucanase inhibited both the elicitor-releasing and glucanase
activity to similar extents. Furthermore, pretreatment of soybean tissues
with the antibodies partially inhibited glyceollin accumulation otherwise
induced by mycelial walls, suggestingthat the elicitor release indeed occurs
in natural infection processes and the released elicitors are responsible for
glyceollin elicitation inthe infected tissues.
111. ACTIVITY AND STRUCTURE OFTHE

CLYCANASE-RELEASED ELICITORS
The glucanase-released Pmg elicitors active in glyceollin elicitation were
heterogeneous in size, ranging from 1,0oO to more than 100,0oO Da when
evaluated by gelfiltration. Activity of different size fractionsthe ofreleased
elicitors to induce glyceollin accumulation in soybean tissues wereat least
10-100 times higher, based on weight concentrations, than the previously
reported elicitors extracted by autoclaving as well as acidand alkaline treat-
ments, and the released elicitors accountedfor more than 90% of the total
elicitor activity ofthe native mycelial walls.
Tentative structures of the released elicitors were deduced by use of
sugar, 13C-NMR,and enzymatic analysis (Fig. 1). Hepta-P-D-glucopyrano-
side (G7), the smallest elicitor-active molecule obtained by acid hydrolysis
Elicitation
Underlying
Molecular Events 63
Released carbohydrate
Cell wall bound elicilor (I) Elicitor-active (a) Elicitor-inactive (m)
8-1.6 m.6 (various chain length)
8-1.3
1%:
C-$-C * - C-$-"
8-1.3f
'Host
p
+
B-1.3-endoglucanase
G-C-C..
B-i.3 C
....
"-E-"
B-1.3(~
B-1,3(?
C
E
2 Fungus
Fraclion number
Figure 1 Tentatively proposed structures of elicitors bound to cell walls of Phy-

tophthora megasperma f. sp. giycinea and its released forms due to attack by soy-
bean 8-1,3-endoglucanase. Insertedfigure (A) is an elution profile of total carbohy-
drates released by soybean glucanaseon Sephadex G-100.8-1,dGlucans of various
chain length are bound to cell walls through p-1,3 side chains (I). Upon infection,
&1,3 side chains are attacked by endo-type soybean/3-1,3-glucanase, resulting inthe
release of elicitor-active &1,6 chains of various chain length with side chains of
/3-1,3-linked one- or two-glucose moieties [II, correspond to fraction H in (A)] and
di- or trimer of &1,3-glucans [111, correspond to fraction L in (A)] derived from the
endoglucanase attack of 8-1,3 side chains of cell wall-bound elicitors. Small arrows
indicate the sites for glucanase attack.
of Pmg cell walls (Sharp et al., 1984), was chemically synthesized and was
also used for structural comparison. "C-NMR and sugar analysis indicated
the presence of @-1,6-and @-1,3-glucose linkagesfor various size fractions
of the released elicitors. The released elicitors, as well as G7, were not
decomposed by @-1,3-endoglucanase, suggestingthe main chain to be 0-
1,6-linked.'In contrast, @-1,3-exoglucanase partially degraded both the elic-
itors resulting in complete loss of elicitor activity. These results indicate
that the released elicitors are composed of @-1,6-linked main chains of
different length and originally bound to fungal cell walls by @-1,3-linked
side chains. Elicitors are thus released, upon infection, dueto the attack of
the side chains by the host @-1,3-endoglucanase, leaving one or two of
@-1,3-linked glucose moieties to each side chain ofthe released elicitors.
64 Yoshikawa
W. MOLECULAR CLONING OF cDNA ENCODING THE

ELICITOR-RELEASING FACTOR, P-1,3-ENDOGLUCANASE
IN SOYBEAN
Soybean ,f3-1,3-endoglucanase thus appears to be a key host component
involved in the earliest soybean-Pmg interaction leading to the induction of
a plant defense reaction, by releasing elicitor-active carbohydrates from
mycelial walls. Wetherefore cloned and characterized cDNA encoding soy-
bean P-1,3-endoglucanaseto further elucidate the role of this enzyme inthe
expression of disease resistance (Takeuchial., et 1990).
Several cDNA clonesfor the glucanase gene were obtained by antibody
screening of aX gtl 1 expression library prepared from soybean cotyledons.
Hybrid-selected translation experiments indicated that the cloned cDNA
encoded a 36-kDa precursor protein product that was specifically immuno-
precipitated with ,f3-1,3-endoglucanase antiserum. Nucleotide sequence of
three independent clones revealed a single uninterrupted open reading frame
of 1041 nucleotides, corresponding to a polypeptide of 347 residue long.
The primary amino acid sequence of &1,3-endoglucanase as deducedfrom
the nucleotide sequencewas confirmed by direct amino acid sequencing of
trypsin digests of the glucanase, indicating that the cloned cDNAs were
indeed those for the soybean glucanase. The soybean P-1,3-endoglucanase
exhibited 53% amino acid homologyto a &1,3-glucanase cloned from cul-
turedtobaccocells and 48%homology to a&(1,3-1,4)-glucanase from
barley.
E. coli cells expressing the cloned full-length cDNA (pEG488) synthe-
sized a protein positive to the glucanase antiserum which, upon a solubiliza-
tion and reconstitution, possessed both the P-1,3-endoglucanase and elici-
tor-releasing activities, firmly establishing that both activities were due to
/3-1,3-endoglucanase. Furthermore, tobacco plants transformed by Agro-
bacterium tumefaciens carrying pEG488 also synthesized a glucanaseanti-
serum-positiveprotein.Ourpreliminaryexperimentsindicated that the
transgenic tobacco plants with higher levels of P-1,3-endoglucanase activity
enhanced disease resistance to several fungal tobacco pathogen.
l V. A SPECIFIC RECEPTOR ON SOYBEAN

MEMBRANES
FOR THE CLUCANASE-RELEASED ELICITORS
We previously demonstratedthat soybean membranes containeda specific
binding site for intercellular &1,3-glucanof Pmg, mycolaminaran, which
possessed weak elicitor activity (Yoshikawa et al., 1983). Further study was
made to examine whether soybean membranes contain receptors specific
for the glucanase-released elicitors which appear to play a crucial role in
Molecular Events
Elicitation
Underlying 65
the elicitation of glyceollin accumulation in the fungus-infected soybean

tissues.
A direct binding assay between the I4C-labeled released elicitorsand the
isolated soybean membranes was used. Total binding of the I4C-labeled
released elicitors ina concentration-dependent manner, suggesting the exis-
tence of specific binding sites. A Scatchard plot of the binding data dis-
closed the presence of a single class of binding sites having a Kd value of 4
x 10" M and approximately 20 binding sites per cotyledon cell. Preincu-
bation of soybean membranes above 5OoC or with proteases abolishedthe
specific binding. Higher binding activity was observed with a membrane
fraction rich in plasma membrane-associated ATPase obtained by aqueous
two-polymer-phasesystem.Theseresultsindicate the existence of heat-
labile proteinaceous binding sites specific for released elicitors on soybean
membranes, presumably plasma membranes. Furthermore,the binding ac-
tivity of several carbohydrates obtained from Pmg and other sources or
from chemical modification of the glucanase-released elicitors correlated
with their elicitor activity to induce glyceollin accumulation in soybean
cotyledons. Experiments are now being conducted to solubilize and isolate
the binding site.
VI. TRANSCRIPTIONAL ACTIVATION OF THE GENES FOR

CLYCEOLLIN BIOSYNTHETICENZYMES
Our previous study (Yoshikawa et al., 1978b) using transcriptional and
translational inhibitors indicatedthe glyceollin productionwas mediated by
de novo mRNA and protein synthesis. Further study directly measuring
mRNA levels withthe use of P-labeled cDNAs for phenylalanine ammonia-
lyase (PAL), chalcone synthase (CHS),and chalcone isomerase (CHI) dem-
onstrated that the glucanase-released elicitors as well as the native fungal
cell walls induced the gene transcription of each enzyme involved in glyceol-
lin biosynthesis within1-2 hr after elicitor application. Concomitantto the
induction of glyceollin biosynthesis, glyceollin-degrading activity present in
soybean tissues was reduced after elicitor treatment. It may therefore be
deduced that these two biochemical actions of the elicitors, the induction
of glyceollin synthesis and the reduction of glyceollin turnover activity,
synergistically lead to the massive glyceollin accumulation in the elicitor-
treated soybean tissues.
A s summarized in this chapter, research in the last several years has
disclosed the outline of a sequence of biochemical events that appear to
participate in invocation of disease defense reactions (Fig.2). The processes
probably involve pathogen-associated molecule (elicitor)-plant receptor in-
teraction, formation of signal transmitting substances, and gene activation,
66 Yoshikawa
HOST C E L L
..
signal
transmission n
::-f”“
\\
I Bound elicitor 1 New i

messenger RNA
Reduced I
phytoalexin I
Phytoalexin
turnover
ac.tivity \ synthesizing
enzyme
l /
1 Phytoalexin
I
i .//’
Released synthesis
elicitor
Phytoalexin,--’
Receptor
accumulation
Figure 2 Schematicrepresentationofpossiblesequenceofbiochemicalevents
leading to plant defense reactions. The scheme exemplifies phytoalexin accumula-
tion mechanism in the Phytophthora megasperma f. sp. glycinea-soybean interac-
tion, but it may be applicable to other plant-pathogen systems. Contact of incom-
patible fungal races with host cells results in rapid release of phytoalexin elicitors
from fungal cell wall surface, due to attack by &1,3-endoglucanase constitutively
present in host cells. The released elicitors then interact with the complementary
receptors on plant plasma membrane. This interaction generates second messengers
which transmit the signal to the nucleus where de novo transcription is invoked.
The resulting new messenger RNA leads to the synthesis of enzymes involved in
phytoalexin biosynthesisand the phytoalexin is formed. Levels of phytoalexin accu-
mulation are accelerated by simultaneous inhibition of the phytoalexin-degrading
system, which may also result from the elicitor-receptor interaction. Compatible
fungal races may either possess elicitorsthat, upon release, cannot interact withthe
host receptors, or produce “suppressors” that interfere with the elicitor-receptor
interaction or inhibit one ofthe subsequent host metabolic processes leadingto the
phytoalexin accumulation.
resulting in de novo biosynthesis of enzymes responsible

for production of
phytoalexins or other defense substances. At present the schemeis still
hypothetical in may respects, however,and requires further evaluation of
each process before molecular details the
in expression of disease resistance
can be fully understood.
Molecular Events Underlying Elicitation 67
REFERENCES
Keen, N. T. (1981). Evaluation of the role of phytoalexins. Plant Disease Control
(R. C. Staples, ed.), John Wiley and Sons, New York, pp. 155-177.
K&n, N. T., and Yoshikawa, M. (1983). /3-1,3-Endoglucanase from soybean re-
leases elicitor-active carbohydrates from fungus cell walls. Plant Physiol. 71:
460-465.
Sharp, J. K., McNeil, M., and Albersheim, P. (1984). The primary structure of one
elicitor-active and seven elicitor-inactive hexa(P-D-glucopyranosy~)-D-g~ucitols
isolated from the mycelial walls of Phytophthora megasperma f. sp. glycinea.
J. Biol. Chem.25911321-11336.
Takeuchi, Y., Yoshikawa, M., Takeba, G., Tanaka, K., Shibata, D., and Horino,
0. (1990). Molecular cloning and ethylene induction of mRNA encoding a
phytoalexin elicitor-releasing factor, /3-1,3-endoglucanase, in soybean. Plant
Physiol. 93:673-682.
Yoshikawa, M. (1978). Diverse modes of action of biotic and abiotic phytoalexin
elicitors. Nature 275546447.
Yoshikawa, M. (1983). Macromolecules, recognition, and the triggering of resis-
tance. Biochemical Plant Pathology(J. A. Callow, ed.), John Wiley and Sons,
Yoshikawa, M.,Keen,N. T., and Wang, M.C.(1983). A receptor on soybean
membranes for a fungal elicitor of phytoalexin accumulation. Plant Physiol.
73~497-506,
Yoshikawa, M.,and Masago, H. (1982). Biochemical mechanismof glyceollin accu-
mulation in soybean. Plant Infection: The Physiological and Biochemical Basis
(Y. Asada et al., eds.), Japan Sci. Soc. Press, Tokyo/Springer-Verlag, Berlin,
pp. 265-280.
Yoshikawa, M., Matama, M., and Masago, H. (1981). Release of a soluble phyto-
alexin elicitorfrom mycelial walls of Phytophthora megasperma var. sojae by
soybean tissues.Plant Physiol.621032-1035.
Yoshikawa, M., Yamauchi, K., and Masago, H. (1978a).Glyceollin: its role in
restricting fungal growth in resistant soybean hypocotyls infected with Phy-
tophthora megasperma var. sojae. Physiol. Plant Pathol. 12:73-82.
Yoshikawa, M., Yamauchi, K., and Masago, H. (1978b). De novo messenger RNA
and protein synthesis are required for phytoalexin-mediated disease resistance
in soybean hypocotyls.Plant Physiol.61:314-317.
Yoshikawa, M., Yamauchi,K., and Masago, H. (1979). Biosynthesis and biodegra-
dation of glyceollin by soybean hypocotyls infected with Phytophthora meg-
asperma var. sojae. Physiol. Plant Pathol.14:157-169.
4
Soybean Phytoalexins: Elicitation,
Nature, Mode of Action, and Role
Jack Paxton
University of Illinois, Urbana, Illinois
1. INTRODUCTION
I was pleased to be asked to write a chapter on my research and to discuss
where phytoalexin research should go inthe future. Some of my thoughts
along these lines were discussed previously (Paxton, 1988b).
The objectives of my research are to understand how a plant “recog-
nizes” a pathogen or pest, and what it does to defend itself after it recog-
nizes the pathogen or pest. Armed with this information, we can improve
disease control in soybeansand other crops. Through better understanding
of plant disease, gained by study of a model system, scientists should be
able to increase crop yields by preventingcrop losses inan environmentally
sound and sustainable way.
Soybeans, Glycine max [L.]Merr., are a major world crop. The value
of U.S. production alone in 1991 was over $10 billion. Phytophthora root
rot is a serious disease of soybeans and Phytophthora sojae Kauf. and
Gerd. (Hansen and Maxwell,1991)(previouslyknownas Phytophthora
megasperma Dreshs. f. sp. glycinea Kuan and Erwin) causes losses esti-
mated at 3% of this crop on average each year(Plant Pathology Extension,
University of Illinois, 1990). Some farmers’ soybean crops can be decimated
by this disease while others escape unscathed, as seen by above-ground
symptoms. Evidence is accumulating, however, that a large toll istaken on
soybean roots. This root loss is translated into yield losses which go unre-
ported because they are considered normal. The plant replaces roots that
have beenrotted off at the expense of seed production.
Research inthe summer of 1991 by Morris Huckand me (unpublished),
using a minirhizotron, confirmedthat root loss duringthe growing season
in soybeans is substantial. But generally this root lossgoesunobserved
because of problems in observing roots (Paxton, 1974). Kittle and Gray
(1982), establishedthat soybean yields could be reduced by pathogens by at
69
70 Paxton
least 26%, basedon fumigation of normal soilsand use of foliar sprays to

control a number of fungal diseases.
Phytophthora root rot of soybeans was chosen as a model system to
study because of the value of the crop and the seriousness of the disease
(Paxton, 1983). These factors led to the development by plant breeders of
several cultivarsand near-isolines of soybeansthat differ from oneanother
in onedominant gene, Rps, for resistance to races ofthe pathogen.
When Istarted my research on this disease a new cultivar of soybeans,
Harosoy 63, hadjust been released by ProfessorR. L. Bernard. He created
this cultivar by crossing Harosoy with Blackhawk (which was resistant to
Phytophthora root rot) and then back-crossing the progeny with Harosoy
for six generations to give a cultivar which was agronomically identical to
Harosoy but now contained the Rpsl gene, which conferred resistance to
Phytophthora root rot. Little didwe know at the time that the single domi-
nant gene strategy of disease resistance would lead to the discovery of
numerous races of the pathogen, several of which could attack this and
other single, dominant genes that were subsequently found and used in
soybean breeding for disease resistance. With this in mind, let’s explore
what we have learnedabout this disease.
II. ELICITORS
Previous work on elicitors of phytoalexins was summarized recently (Pax-
ton, 1989). These compounds often arise from pathogen cell wallsand seem
to be specifically recognized by the plant cell membrane.
An important insight from my laboratory was Frank’s discovery that
fungi release elicitors, then called inducers, of phytoalexin production in
soybeans (Frank and Paxton, 1971). We were the first to recognize that P .
sojae cultures release compounds capable of inducing phytoalexin produc-
tion in soybean plants. This elicitor appeared to be a glycoprotein with a
molecular weight of about 10,OOO. We found elicitor activity in extracts of
Venturia inaequalis, Rhizoctonia solani, Diplodia zeae, and Helminthospo-
rium turcicum, suggestingthat elicitors of phytoalexin productionare com-
mon in fungi. We also found that plants may have the abilityto stimulate
the production of these elicitors. For this research we developed a useful
cut-cotyledon bioassay that has been widely used in subsequent elicitor
studies.
Ayers et al. (1976a,b) explored the composition of elicitors released
from purifiedP. sojae cell wallsafter autoclaving. Theyfound at least four
fractions, with varying composition and ability to elicit defense responses
on soybean cotyledons. They suggested that the elicitor activity of each
fraction resides in the 3 and 3,6 highly branched glucan component of the
Soybean Phytoalexins 71
fraction. They also suggested that the mannosyl residues, which represent
about 1% of the undegraded glucan, participate in the activity of this
molecule.
Keen, in one of the first studies looking for specific elicitors, which
would explain race specificityin fungal pathogens, presented evidence that
culture filtrates of race 1 of P. sojae contained elicitors that differed in
some respects from race 3 of this fungus. Although hewas not able to
purify and identify this specific elicitor, he speculatedthat absence of this
elicitor may be involved in the difference between race 3 and race 1 reac-
tions of soybean cultivars (Keen, 1975). Keen and LeGrand (1980) later
reported the isolation of specific elicitors extractedfrom the cell walls ofP.
sojae with an alkali treatment. Surfaceglycoproteins on P. sojae were
suggested to play the role of race-specific phytoalexin elicitors. Isolated cell
walls of P.sojae were extracted with0.1 N NaOH at O O C . The high molecu-
lar weight glycoproteins in this extract elicited glyceollin accumulation in a
race-specific fashion. The glycoprotein contained only glucose and man-
nose as sugars. And while its activity was diminished by boilingor pronase
treatment, the elicitor activity was destroyed by periodate treatment. This
suggested that the carbohydrate portions are important for elicitor activity.
A s they point out, this observation is clouded by considerable variation in
replicates, a high concentration requirementfor activity, and a lack of total
specificity.
The specificity in this case was not very clear-cut sincethe differences
between the levels of glyceollin induced by elicitorpreparations from viru-
lent and avirulent raceswere small. This work remainsto be confirmedbut
provides an interesting lead toward finding a very useful type of compound.
We suggested that this type of compound could be very useful in plant
breeding efforts (Paxton and Jacobsen, 1974).
Race-specific moleculesthat protect soybeansfrom P.sojae were found
by Wadeand Albersheim (1979). These appearedto be glycoproteinsfound
in the incompatible races of the pathogen but not the compatible races.
Introduction of this glycoproteinfraction into wounds 90 min before subse-
quentinoculation was sufficient to protectsoybeanplants.Subsequent
work by Desjardins etal. (1982) indicated that glycoprotein fractions from
both compatible and incompatible racesof P. sojae could protect soybean
plants from subsequent infection by this fungus. Variability in
the bioassay
unfortunately prevented further purification and identification of these
components.
Bonhoff and Grisebach (1988) found that laminarin and polytran N
also are effective elicitors of glyceollin, comparable to the glucan elicitor
isolated from P. sojae by Sharp and coworkers (1984). Interestingly, on cut
soybean cotyledons,the reverse wastrue in that laminarin was lesseffective
72 Paxton
inelicitingglyceollinaccumulation. Furthermore, whentheypretreated

soybean roots with laminarin, they increased resistance of seedlings against
P. sojae. Digitonin, when added with the laminarin, further increased this
elicitation of glyceollin accumulation in intact roots. Soybeanroots deposit
callose whentreated with digitonin but not when treated with various glu-
cans.
An interesting observation was made by Yoshikawa et al. (1981) that
soybean tissuescontain an enzyme which released elicitor from P.sojae cell
walls. This suggests a role for such enzymes in “recognition” of potential
pathogens when they come in contact with soybean tissue, by specifically
releasing fragments that are free to diffuse to binding sites on the cell
membranes. At a recent Congress ofPlant Pathology in Kyoto, Yoshikawa
reported that he has succeeded in releasing very active elicitor fragments
from the cell walls ofP.sojae with a P-1,3-endoglucanase. These fragments
appear to be slightly largerthan the acid-hydrolyzed fragments studiedby
Sharp et al. (1984), and they were 10-100 times more activein elicitationof
glyceollin in soybeans.
Transformation of this elicitor signalinto gene activation also has re-
ceived attention. Chappell et al. (1984) saw a rapid induction of ethylene
biosynthesis in cultured parsley cells that had been treated with P. sojae
elicitor. Ethylene production is a common response of plant cellsto stresses
such as pathogen attack. Parsley cell cultures respond by production of
ethylene within 1 hr of treatment with this elicitor. Hauffe et al. (1985)
showed that cultured parsley cells respond to this elicitor by producing
S-adenosyl-L-methionine:bergaptol and S-adenosyl-L-methionine:xantho-
toxol O-methytransferases. Kuhnet al. (1984) demonstratedthat treatment
of parsley cells with the P. sojae elicitor induced phenylalanine ammonia-
lyase and 4-coumarate:CoA ligase mRNAs. Farmer (1985) demonstrated
that 5 pg/ml of P. sojae elicitor suppressed lignin accumulation intreated
soybean suspension cultures. Phenolics, normally polymerizedinto lignin,
were probablydiverted into glyceollin and other compounds. This also
suggests that elicitor recognition is a common phenomenon in triggering
various defense responses in plants. The importance of phenylalanine am-
monia-lyase to this process in soybeans is suggested by the work of Simcox
and Paxton (1984).
It should be noted, however, that ethylene maynot have a direct role in
the elicitationof glyceollin accumulation in soybean plants (Paradies et al.,
1980).
Ziegler and Pontzen (1982) found an extracellular invertase inP.sojae
which was capable of inhibiting glucan-elicited glyceollin accumulation in
soybeans. This glycoprotein suppressor was active at 0.3-10 pg/ml and
was race-specific. The invertase would only inhibit elicitation of glyceollin
accumulation on soybean cultivarsthat could be attacked by

the race which
was the source of the invertase. This work has not been confirmed and
remains an area worth exploring.
111. PHYTOALEXINS
Phytoalexins were first defined byMuller and Borgerin1941 and our
understanding of them as progressed to a more recent definition (Paxton,
1981). They are compounds producedde novo in plantsafter microorgan-
ism attack or other stress and they play a role in plant disease and insect
resistance. The precise role of these compounds ishard to determine as is
true for finding and assaying manyantifungal compounds (Paxton, 1991c;
Vaillancourt and Paxton, 1986).
Glyceollin is the phytoalexin identified from soybeans and occurs as a
series of isomers. Five isomers have been identified but their activity has
not been studied. Some species of Glycine appear to produce almost exclu-
sively one isomer and various tissues of soybean produce predominantly
different isomers(Keen et al., 1986). Why this is the case isnot understood
but certainly is worth investigating.
A phytoalexin, PA,,was identified in soybeans (Keen and Paxton,
1975)but has yetto be characterized.It is a yellow compound witha strong
absorbance at 492 nm at pH 7, which shifts to 430 nm at pH 2. It may
represent an oxidation product of glyceollin, but that has yet to be deter-
mined.
My laboratory also discovered phytoalexin accumulation incorn (Lim
et al., 1970). Despite efforts by Simcox and other graduate students, this
elusive compoundor compounds hasnot been identified.
Several metabolic processesare activated when a plant is attacked by a
pathogen (Paxton, 1991a). Phenylalanine ammonia-lyase is one ofthe first
enzymes to be activated. This enzyme has been studied extensively, partly
because the assay for it is relatively simple.
It is my opinion that crucial parts of plants, such as seeds, accumulate
compounds that can serve as primary deterrents to pathogen and insect
attack. Failing in this role, these compounds can then be converted to much
more toxic compoundsthat serve as secondary defense. Phytoalexins, then,
would be a secondary defense, and they are toxic to plant cells as well as
pathogens and insects. Among the compounds naturally accumulated in
soybeans, precursors of glyceollinare common, especially isoflavanoids in
seeds. Isoflavanoids serve as precursors for glyceollins in soybean tissues
and studies show these isoflavanoids to be held as complex conjugates in
different concentrations in different tissues in soybeans (Graham, 1991;
Graham and Graham, 1991).
74 Paxton
IV. ROLE IN DISEASE AND INSECT RESISTANCE
The roleof phytoalexins inplant disease and pest resistance has been very
hard to determine with scientific precision. Regardless of this difficulty,
considerable evidence has accumulated that phytoalexins playan important
role in soybean disease resistance. Someof the first evidence in soybeans
was that the removal of phytoalexins can causea resistant plant to become
susceptible (Klarmanand Gerdemann, 1963). Later we found that addition
of a phytoalexin back into the plant can make a susceptible plant become
resistant (Chamberlainand Paxton, 1968). The phytoalexins accumulate in
the plant after inoculation witha timing appropriate to inhibition of patho-
gen growth (Frankand Paxton, 1970). Another approach was our discovery
that prior inoculation witha non-pathogen will make the susceptible plant
become resistant to subsequent inoculation with what would have beena
pathogen (Paxton and Chamberlain, 1967; Svoboda and Paxton, 1972).
More evidencefor the role of phytoalexins and an explanation of the impact
of environment on plant disease wasthe finding that environmental condi-
tions could affect phytoalexin accumulation (Murch and Paxton, 1979). We
found that anaerobic conditions, such as would occur around soybean roots
during flooding, drastically depressed the soybean plant’s ability to accu-
mulateglyceollin.Thiscouldexplain why flooding is so important for
disease development in the field.
Other environmental effects that influence glyceollin accumulation in
soybeans are salinity and temperature (Murch and Paxton, 1980a, 1980b).
All of these environmental effects on glyceollin accumulation in soybeans
relate nicely to field conditions of wet, cold soils which are ideal for Phy-
tophthora root rot development (Murch and Paxton, 1980~).Calcium ion
seems to play an important role in elicitation and therefore disease resis-
tance aswell (Stab and Ebel, 1987).
One of the best proofs of the role of phytoalexins in plant disease
resistance comes fromthe work of Schafer et al. (1989). They transferred a
gene for pisatin demethylase (pisatin is a phytoalexin in peas) into a corn
pathogen and converted it to a pea pathogen as well. This clearly suggests
that pathogens have to have the ability to deal with phytoalexins of their
host in orderto be pathogens. Our work would suggest that this is true for
soybean Phytophthora root rot as well (Paxton, 1982).
It also is probable that a pathogen must have an extensive set of en-
zymes to be a pathogen. This could help explain the common occurrence of
loss of virulence in pathogens held in culture. Some enzymes necessary for
pathogenicity probably are not necessary for survival as saprophytes on
artificial media.
V. MODE OF ACTION
The mode of action of biologically active compounds is generally of interest

to scientists. It is surprising that more has not been done on the mode of
action of phytoalexins. Our workon glyceollin suggeststhat glyceollin has
at least twoimportant modes ofaction, at physiological levels in plants. We
discovered that glyceollin can inhibit electrontransport at site 1 in soybean
and corn mitochondria (Boydston etal., 1983). Later we also showed that
glyceollin can inhibit ATPase in plasma membranes of both plant and
pathogen (Giannini et al., 1988a,b). Tonoplast vesicles werealmost twice as
sensitive to glyceollin as plasma membrane vesicles. Cell membranes are
most likelya site for both elicitation and toxic action of phytoalexins (Pax-
ton, 1979).
Glyceollin is toxicto plant cells as well as pathogensand insects (Bhan-
dal et al., 1987; Hart et al., 1983; Fischer et al., 1990b). This suggests that
these compounds have evolved asa general stress response and may serve
to protect the plant by creating very toxic environments at the cellular level.
VI. DISCUSSION
My initial studies onthe nature of disease resistance in walnuts introduced
me to compounds which can accumulate in plants and be stored in relatively
nontoxic forms such as glycosides (Paxton and Wilson, 1965). The plant
cell is armed with glycosides whichare hydrolyzed rapidlyto the aglycone,
which is often more toxic. This happens when the cell is damaged and
protects the plant fromattack, much likeour white blood cells produce free
radicals to help destroy foreign organismsour in bodies.
Phytoalexins were discovered in soybeans by Dr. Gerdemann, shortly
after the first phytoalexin was characterized in peas byPerrin and Bottom-
ley (1961)in Australia.
Gerdemann’s story of serendipity deservesto be retold here. Dr. Gerde-
mann was a mycologist and as such realized that Phytophthora species
require thiamine to grow. He reasoned that one explanation for resistance
of soybean varietieswas that the resistant cultivars had tissues deficient in
thiamine and the fungus therefore couldn’t grow. As a typical mycologist
without a large research budget, he set upa simple experiment in which he
introduced a thiamine solution into a soybean stem via a string wick that
had been threadedthrough the stem with a lacing needle (Fig. 1). When he
then inoculated the plant in the wick wound, he found that the plant was
susceptible. Being a good scientist, he also ran a control experiment in
which just plain waterwas introduced into the wound via the wick,and the
Paxton
--*
76
soybean plant,<- /wick
Figure 1 Wick experiment.
result was that the inoculated plantwas again susceptible. It appeared that
it was not something hewas putting into the plant that made it susceptible,
but rather something he was taking out of the plant. When he tested the
exposed tip of the wick for materials that would inhibit the growth of
Phytophthora, he found that nothing.inhibitory to the growth ofthe patho-
genwas on the tip from the uninoculated plant. But the tip from the
resistant inoculated plant contained compounds inhibitory to the growth of
the pathogen, when addedto media with the fungus.
During my first sabbatical, spent with Dr. Cruickshank in Australia,
we explored the accumulation of phytoalexins in plant cells and found that
the process could occur withoutprofound changes in the cell or cell death
(Paxton et al., 1974). The plant cells, however, were often subsequently
killed by phytoalexins, whichare toxic to the plant cell as well as the fungal
or bacterial pathogen (Bhandal etal., 1987). There stillis a need for study
of cell death and how it relates to phytoalexin accumulation. Pectic sub-
stances released from cell wallson cell death can elicit phytoalexin accumu-
lation and may bea signal to the plant that repair processesare needed.
l Workwith
Dr.
Kogan found that phytoalexins could
deter
insect attack
on soybeans, just as they play a role in protecting soybeans from plant
pathogens. Physiologically significant levels of glyceollin deter insect feed-
ing by casual feedersbut not true pests of soybeans, as would be anticipated
(Hartet al., 1983; Fischer etal., 199Ob). If true soybean pests were deterred
by glyceollin it would suggest that the compound did not occur at feeding
sites. The true soybean pest must havea method of blocking accumulation
at the feeding siteor excluding or detoxifying the compound. This pointis
confirmed inthe work of Schafer et al. (1989).
VII. EPILOG
The study of disease resistance of soybeans by the researchers mentioned

abovehasopenedmanydoors to better understanding plant disease in
general and therefore to improved control of plant diseases. These studies
might even leadto a better understandingof how pollenation is controlled
in plants (Fett et al.,1976).
One exciting discovery we made is the presence of elicitors of phyto-
alexin accumulation in cultures of the pathogen. These compounds have
been partially identified and appear to be small carbohydrates. It appears
that the plant produces enzymes, P-1,3-glucanases, which cleave the cell
wall of the pathogen to release fragmentsthat can be boundto receptors in
the plant cell wall (Schmidt and Ebel, 1987). One concern isthat the frag-
ments used are released from fungal cell walls by autoclaving them under
acid conditionsand therefore may be artifacts.
Since theyare carbohydrates, elicitors could be stable compounds with-
out environmental detriment.Yet these compoundsmay hold the key to an
interesting and valuable methodof controlling plant diseases and pest at-
tack. Since the plant responds to these compounds in the same way it would
to a pathogen, the plant can be “vaccinated” against subsequent attack or
protected from attack by having its surface covered with the elicitor (Pax-
ton, 1973). When the pathogen gains ingress through a wound or direct
penetration, the plant could be “notified” of the invasion and mount its
inducible defense mechanisms. An example of the possibility of this ap-
proach is our work with Polytran L (Bhandal and Paxton, 1991) and the
demonstration that elicitors canact on many different plants to cause accu-
mulation of phytoalexins (Clineet al., 1978).
Another aspect worth exploringis the use of these elicitors to protect
plants against insect attack (Kogan and Paxton, 1983; Paxton, 1991b). It
appears that the same compounds interfere with insect feeding, and this
feeding would logically introduce the elicitors into plant cells.
Much remains to be understood as to exactly what the true elicitor is,
how it binds to a presumed receptor, what secondary signal is then sent to
the plant cell nucleus to activate genes required for the accumulation of
phytoalexins, and how these phytoalexins are then accumulated in the infec-
tion site to inhibit growth of the pathogen. This provides interesting leads
to novel controls for plant diseasesbyusing stable, natural elicitors to
evoke the plant’s normal defense mechanisms (Paxton, 1988a).
These mechanisms undoubtedly include the accumulation of phytoalex-
ins that also have been described above.It is my opinion that several com-
pounds that could have very important roles in plant disease resistance
remain to be discovered. Some -of these are compounds that have been
78 Paxton
variously described as nonexistent since they are so chemically reactive as

to bedestroyedwhentheplantcells are crushed to extract them from
plant tissue. These compoundsare highly reactiveand generally would not
survive the tituration of plant tissues. More studied examples are -superox-
ide ion (Afanas’ev, 1989) and nitric oxide (Snyder and Bredt, 1992).
Another possibility is that compounds that can be formed the in process
of extracting plant tissues. These compounds are not there until the re-
searcher damages the tissue and releases enzymes or compartmentalized
compounds that normally would not mix. My thesis research (Paxton and
Wilson, 1965) depended on this phenomenon. The British chemist Daglish
discovered that walnutscontainedlargequantities of hydrojuglone 4-
glucoside but not juglone, as had been previously thought. This reduced
quinone was kept reduced and made more water-soluble by the glucose
attached in the 4 position. When cells are ruptured, as occurs in harvestor
pathogen or insect attack, 8-glucosidases are released which rapidly hydro-
lyze the compound. The aglycone is then readily oxidized to the very reac-
tive naphthoquinone juglone. It is juglone which is very toxicto fungi and
bacteria, and stainsthe hands brown when walnutsare dehulled.
I feel that there are several compoundsthat can be called phytoalexins
in soybeans which have yet to be identified. These compoundsare chemi-
cally reactive enough that they are not seen in crude extracts of soybean
plants and more sophisticated chemical methods will be required to study
them. One such method is electron paramagnetic resonance, and unpub-
lished studies with Linn Belford and Robert Clarkson in the Chemistry
Department at the University of Illinois suggest that free radicals (that
would fill the above criterionof active compounds very difficultto chemi-
cally isolate)do indeed accumulate in inoculated soybean stems.
Synergism of these compounds in planta is yet another area that re-
mains to be investigated. Phytoalexins are generally tested for activity on
thin-layer chromatography plates or petri plates in relatively inert growth
media. Inthe plant, with enzymesand active metabolism, these compounds
undoubtedly interactwith other compounds in many different and unantici-
pated ways. Some interesting work we have done suggests this is the case
with several soybean compounds that can deter insect feeding (Fischer et
al., 1990a).
Although phytoalexins show some tendenciesto be chemically related
in taxonomically related plants, this is not always the case (Kumar et al.,
1984). Phytoalexins are not always closely related to disease resistance in
soybeans (Paxton and Chamberlain, 1969).
Biological control is another area that needs much moreattention. We
have reliedtoo much inthe past to kill the pathogen withthe most powerful
chemical we can without endangering ourselves. It turns out that in the
process we have exposed ourselvesto hazards that were probably unneces-

sary (Duncan and Paxton, 1981). Biological control as pointed out by re-
cent research can involve establishing native organisms on the plant that
preclude the attack of pathogens simplyby occupying a niche that is needed
by the pathogen or producing something that makes the pathogen less fit
for survival or less fit as a pathogen.
REFERENCES
Afanas’ev, I. B. (1989). Superoxide Ion, Vols. 1 and 2. CRC Press, B o a Raton,
FL.
Ayers, A. R., Ebel, J., Valent, B., and Albersheim, P. (1976a). Host-pathogen
interactions. X. Fractionation and biological activity of an elicitor isolated
from the mycelium walls ofPhytophthora megasperma var. sojae. Plant Phys-
iol. 22760-765.
Ayers, A. R,,Valent, B., Ebel, J., and Albersheim, P. (1976b). Host-pathogen
interactions XI. Composition and structure of wall-released elicitor fractions.
Plant Physiol.52766-774.
Bhandal, 1. S., and Paxton, J. (1991). Phytoalexinbiosynthesisinduced by the
fungal glucan, Polytran L, in soybean, pea and sweet pepper tissues. Journal
of Agricultural and Food Chem. 39~2156-2157.
Bhandal, I. S., Paxton, J. D., and Widholm, J. M. (1987). Effect of Phytophthora
megasperma f. sp. glycinea culture filtrate and cell wall on glyceollin produc-
tion and cellviabilityincellsuspensionculturesofsoybean (Glycine may
[Merr.]). Phytochemistry 262691-2694.
Bonhoff, A., and Grisebach, H.(1988).Elicitor-induced accumulation of glyceollin
and callose in soybean roots and localized resistance against Phytophthora
megasperma f. sp. glycinea, Plant Sci.54~203-209.
Boydston, R., Paxton, J. D., and Koeppe, D.E. (1983). Glyceollin: a site specific
inhibitor of electrontransport in isolated soybean mitochondria. Plant Physiol.
72151-155.
Chamberlain, D. W., and Paxton, J. D. (1968). Protection of soybean plants by
phytoalexin. Phytopathology 581349-1350.
Chappell, J., Hahlbrock, K., and Boller, T. (1984). Rapid induction of ethylene
biosynthesis in cultured parsley cells by fungal elicitor and its relationship to
the induction of phenylalanine ammonia-lyase, PIanta 161~475480.
Cline, K., Wade, M., and Albersheim, P. (1978). Host-pathogen interactions. XV.
Fungal glucans which elicit phytoalexin accumulationin soybean also elicitthe
accumulation of phytoalexins in other plants. Plant Physiol. 62918-921.
Desjardins, A., Ross, L. M., Spellman, M. W., Darvill, A. G., and Albersheim, P.
(1982).Host-pathogen interactions. XX. Biological variation in the protection
of soybeansfrom infection by Phytophthora megasperma f. sp. glycinea. Plant
Physiol. 69:1046-1050.
Duncan, D. R.,and Paxton, J. D.(1981).Trifluralin enhancement ofPhytophthora
root rot of soybean. Plant Dk. 6.5~435-436.
80 Paxton
Farmer, E. (1985). Effects of fungal elicitor on lignin biosynthesisin cell suspension

cultures of soybean.Plant Physiol.78338-342.
Fett, W. F., Paxton, J. D., and Dickinson, D.B.(1976). Studies on the self-
incompatibility response ofLilium longiflorum.Am. J. Bot. 63:1104-1108.
Fischer, D., Kogan, M., and Paxton, J. (199Oa). Deterrency of Mexican bean beetle
(Coleoptera: Coccinellidae) feeding by free phenolic acids. Journal of Entomo-
logical. Sci.25:230-238.
Fischer, D. C., Kogan, M., and Paxton, J. (199Ob). Effect of glyceollin, a soybean
phytoalexin, on feeding by three phytophagous beetles (Coleoptera: Chryso-
melidae, Coleoptera: Coccinellidae): dose vs. response. Environmental Ento-
mol. 191278-1282.
Frank, J. A., and Paxton, J. D. (1970). Time sequencefor phytoalexin production
in Harosoy and Harosoy 63 soybeans. Phytopathology 60315-318.
Frank, J. A., and Paxton, J. D. (1971). An inducer of soybean phytoalexinand its
role in the resistance of soybeans to Phytophthora rot. Phytopathology 61:954-
958.
Giannini, J. L., Briskin, D. P., Holt, J. S., and Paxton, J. D. (1988a). Inhibition
of plasma membrane and tonoplast H + transporting ATPases by glyceollin.
Phytopathology 781000-1003.
Giannini, J. L., Holt, J. S., Paxton, J., and Briskin, D. P. (1988b).Glyceollin
effects on Phytophthora megasperma f. sp. glycinea plasma membraneH+ and
Ca+/ conductance (Abstr). Phytopathology 78: 1503.
Graham, T. L. (1991). Flavonoid and isoflavonoid distribution in developing soy-
bean seedling tissuesand in seed and root exudates. Plant Physiol.95594-603.
Graham, T. L., and Graham, M. Y. (1991). Glyceollin elicitors induce major but
distinctly different shifts in isoflavonoid metabolism in proximal and distal
soybean cell populations. Mol. Plant Microbe Interact.4:60-68.
Hansen, E. M., and Maxwell,D. P. (1991).Speciesof the Phytophthora meg-
asperma complex. Mycologia 83:376-38 1.
Hart, S. V., Kogan, M., and Paxton, J. D. (1983). Effect of soybean phytoalexins
on the herbivorous insects Mexican bean beetle and soybean looper. J. Chem.
Eco~.9657-672.
Hauffe, K. D., Hahlbrock, K., and Scheel, D. (1986). Elicitor-stimulated furano-
coumarin biosynthesis in cultured parsley ce1ls:S-adenosyl-L-methionine:ber-
gaptol and S-adenosyl-L-methionine:xanthotoxolO-methyltransferases, 2.Nu-
turforsch. 41c:228-239.
Keen, N.T. (1975). Specific elicitors of plant phytoalexin production: Determinants
of race specificity in pathogens?, Science 18274-75.
Keen, N. T., and LeGrand, M. (1980). Surface glycoproteins: evidence that they
may function as the race specific phytoalexin elicitors of Phytophthora meg-
asperma f. sp. glycinea, Physiol. Plant Pathol.12175-192.
Keen, N.T., and Paxton, J. D. (1975). Coordinate production of hydroxyphaseollin
and the yellow-fluorescent compound PAL in soybeans resistant to Phytoph-
thora megaspermavar. sojae. Phytopathology65:635-637.
Keen, N. T., Lyme, R. L., and Hymowitz, T. (1986). Phytoalexin production as a
Soybean 81
chemosystematic parameter withinthe genus Glycine. Biochem. System. Ecol.

14481486.
Kittle, D. R., and Gray, L. E. (1982). Response of soybeans and soybean pathogens
to soil fumigation and foliar fungicide sprays.Plant Dis. 68213-215.
Klarman, W. L., and Gerdemann, J. W. (1963). Resistance of soybeans to three
Phytophthora species due to the production of a phytoalexin.Phytopathology
53~1317-1320.
Kogan, M., and Paxton, J. D. (1983). Natural inducers ofplant resistance to insects.
Plant Resistance to Insects (P. A. Hedin, ed.), American Chemical Society,
Kuhn, D., Chappell, J., Boudet, A., and Hahlbrock, K. (1984). Induction of phe-
nylalanineammonia-lyase and 4-coumarate:CoA ligase mRNAs in cultured
plant cells by UV light or fungal elicitor. Proc. Nut. Acad. Sci. USA 81:1102-
1106.
Kumar, S., Shukla, R. S., Singh, K. P., Paxton, J. D., and Husain, A.(1984).
Glyceollin: a phytoalexin in leaf-blight disease of Costus speciosus. Phytopa-
thology 741349-1352.
Lim, S. M., Hooker, A. L., and Paxton, J. D. (1970). Isolation of phytoalexins
from corn with monogenic resistance to Helminthosporium turcicum. Phytopa-
thol0g.Y 601071-1075.
Murch, R. S., and Paxton, J. D. (1979). Rhizosphere anaerobiosis and glyceollin
accumulation in soybean.Phytopathol. 2.9691-94.
Murch, R, S., and Paxton, J. D.(1980a). Rhizosphere salinity and phytoalexin
accumulation in soybean. Plant and Soil54:163-167.
Murch, R. S., and Paxton, J. D. (1980b). Temperature and glyceollin accumulation
in Phytophthora-resistant soybean. Phytopathol. Z. 92282-285.
Murch, R. S., and Paxton, J. D. (1980~).Environmental stress and phytoalexin
accumulation in soybean. Bull. Soc. Bot. France Actual. Bot. 122151-155.
Paradies, I., Konze, J. R., Elstner, E. F., and Paxton, J. D.(1980).Ethylene:
indicator but not inducer of phytoalexin synthesis in soybean. Plant Physiol.
661106-1109.
Paxton, J. D.(1973). Plants “self” recognition may aid disease control. Illinois
Research, Univ. Ill. Agric. Exp.Sta. 15(4):13.
Paxton, J. D. (1974). Phytoalexins, phenolicsand other antibiotics in roots resistant
to soil-borne fungi.Biology and Control of Soil-borne Plant Pathogens (C. W.
Brudhl, ed.), American PhytopathologicalSociety, St. Paul, MN, pp. 185-192.
Paxton, J. D. (1979). Elicitors of incompatible host responses:the role of host cell
membranes. Recognition and Specificity in Plant Host-Parasite Interactions
(J. M. Daly and I. Uritani, eds.), pp. 153-163.
Paxton, J. D. (1981). Phytoalexins-a working redefinition. Phytopathol. Z. IOI:
106-109.
Paxton, J. D. (1982). Degradation of glyceollin byPhytophthora megasperma var.
sojae. Active Defense Mechanisms in Plants (R. K. S. Wood, ed.), Plenum
Press, New York, pp. 343-344.
Paxton, J. D.(1983). Phytophthora root and stern rot of soybean. Biochemical
82 Paxton
Plant Pathology (J. A. Callow, ed.), John Wiley and Sons, London, pp. 19-
27.
Paxton, J. D. (1988a). Fungal elicitors of phytoalexins and their potential use in
agriculture. Biologically Active Natural Products: Potential Use inAgriculture.
American Chemical Society Symp. Ser. No. 380, pp. 109-1 19.
Paxton, J. D.(1988b). Phytoalexins in plant-parasite interactions. Experimental
and Conceptual Plant Pathology. Oxford and IBH, New Delhi, pp. 537-549.
Paxton, J. D. (1989). Fungal elicitors ofplant phytoalexins. Handbook of Natural
Toxins Vol. 6, Marcel Dekker, New York, pp. 439-457.
Paxton, J. D. (1991a). Biosynthesisand accumulation of legume phytoalexins. My-
cotoxins and Phytoalexins(R.P. Sharma, ed.), Telford Press, pp. 485-499.
Paxton, J. D. (1991b). Phytoalexins and their potential role in the control of insect
pests. American Chemical Society Symposium Series No. 449, pp. 198-207.
Paxton, J. D. (1991~).Assays for antifungal activity. Meth. Plant Biochem. 6:33-
46.
Paxton, J. D., and Chamberlain, D. W. (1967). Acquired local resistance of soy-
bean plants to Phytophthora spp. Phytopathology 52352-353.
Paxton, J. D., and Chamberlain, D. W. (1969). Phytoalexin production and disease
resistance in soybeansas affected by age. Phytopathology 59775-777.
Paxton, J. D., Cruickshank, I. A. M., and Goodchild, D. (1974). Phaseollin pro-
duction by live beanendocarp. Physiol. Plant Pathol.4:167-171.
Paxton, J. D., and Jacobsen, B. J. (1974). A new method for screening soybeans
for Phytophthora root rotresistance. Proc. Am. Phytopathol. Soc. 1:67-68.
Paxton, J. D., and Wilson, E. E. (1965). Anatomical and physiological aspects of
branch wilt disease ofPersian walnut. Phytopathology 55:21-26.
Perrin, D. R., and Bottomley, W.(1961). Pisatin: an antifungal substance from
Pisum sativum. Nature191:76-77.
Puckett, J. L., and Paxton, J. D. (1986).Auxin starvation enhancesglyceollin
production in soybeancell suspensions. Agron. Abstr. 6 (Abstr).
Schafer, W., Straney, D., and Ciuffetti, L. (1989). One enzyme makes a fungal
pathogen but not a saprophyte, virulent on a new host plant. Science 246247-
249.
Schmidt, W., and Ebel, J. (1987). Specific binding of a fungal glucan phytoalexin
elicitor to membrane fraction from soybean (Glycine mm). Proc. Natl. Acad.
Sci. USA 84:4117-4121.
Sharp, J. K.,Albersheim, P., Ossowski, P., Pilotti, A., Garegg, P., and Lindberg,
B. (1984). Comparison of the structures and elicitor activitiesof a synthetic and
mycelial-wall-derived hexa @-D-glucopyranosy1)-D-glucitol,J. Biol. Chem.259
11341-11345.
Simcox, K., and Paxton, J. D. (1984). Induced pathogenicity in glyphosate-treated
soybeans. Phytopathology 74: 1271.
Snyder, S. H., and Bredt, D. S. (1992). Biological roles of nitric oxide. Sci. Am.
26668-77.
Stab, M., and Ebel, J. (1987). Effects of Ca on phytoalexin induction by fungal
Arch. Biochem. Biophys.252416-423.
elicitor in soybean cells,
Svoboda, W. E., and Paxton, J. D. (1972). Phytoalexin production in locally cross-

protected Harosoy and Harosoy-63 soybeans. Phytopathology 62:1457-1460.
Vaillancourt, L. J., and Paxton, J. D. (1986). Evaluation of biological assays for
the elicitation of glyceollin in soybeans. Phytopathology 76:1 1 12 (Abstr).
Wade, M., and Albersheim, P. (1979). Race-specific molecules that protect soy-
beans from Phytophthora megasperma var. sojae. Proc. Natl. Acad. Sci. USA
764334437.
Yoshikawa, M., Matama, M., and Masago, H. (1981). Release of a soluble phyto-
alexin elicitorfrom mycelial walls ofPhytophthora megasperma var. sojae by
soybean tissues.Plant Physiol.621032-1035.
Ziegler, E., and Pontzen, R. (1982). Specific inhibition of glum-elicited glyceollin
accumulation in soybeans by an extracellular mannan-glycoprotein of Phy-
tophthora megasperma f. sp. glycinea. Physiol. Plant Pathol.20321-331.
5
Cellular Biochemistryof
Phenylpropanoid Responses of Soybean
to Infection by Phytophthora sojae
Terrence lee Graham
The Ohio State University,Columbus, Ohio
1. INTRODUCTION
A. GeneralBackground
The fungal pathogen Phytophthora megasperma Drechs. f. sp. glycinea
(Hildeb.) Kuan and Erwin (recently redesignatedas P . sojae) can infect all
vegetativesoybeanorgans(Sinclair, 1982). It causessymptomsranging
from rapid pre- and postemergence seedling damping offto slowly spread-
ing lesions on older plant tissues. The soybean-P. sojae association pro-
vides an excellent systemfor the study of molecular aspects of host-patho-
gen interactions. Resistance to the pathogen in soybean is determined by
major dominant Rps genes occurring at seven loci, with several allelic forms
at two of these loci. There are at least 25 known races ofP.sojae which are
characterized by their different specific interactions with these Rps genes
(Schmitthenner, 1985). Where a given Rps gene provides effective resistance
to a given race of P. sojae, the interaction between host and pathogen is
termed incompatible. Wherea given Rps gene provides no resistance to a
given race ofP . sojae, the interaction is termed compatible.
Antibiotic pterocarpan phytoalexins have generally been accepted to
play a role inthe Rps gene-mediated restriction ofthe spread of P . sojae in
incompatible interactions (Keen and Yoshikawa, 1982). Research to date
has ledto the characterization of four isomeric pterocarpan antibiotics now
referred to as glyceollins I-IV (Burden and Bailey, 1975; Lyne and Mul-
heirn, 1978; Lyne et al., 1976; Partridge and Keen, 1977; Sims et al.,1972).
Both the timing and magnitude of the accumulation of the glyceollins differ
markedly in compatibleand incompatible infectionsand are consistent with
the proposed role of the glyceollins in race-specific resistance (Darvill and
Albersheim, 1984; Ebel, 1986; Keen and Yoshikawa, 1982).
85
86 Graham
The biotic elicitors, which may be responsiblefor glyceollin elicitation

in infected tissues, include @l+ 3, @l+ 6-linked cell wall glucans from
P.sojae (Ayers et al., 1976; Sharp et al., 1984) and a-l,4-~-galacturonides
from the plant cell wall (Nothnagel et al., 1983). A synergistic interaction
between the biotic elicitors of host and pathogen origin has been reported
(Davis et al., 1986).
Investigations on the regulation of glyceollin biosynthesis have focused
primarily on three specific enzymesof early phenylpropanoidand flavonoid
metabolism[phenylalanineammonialyase(PAL),chalconesynthase
(CHS) and chalcone isomerase (CHI)]. An early research project suggested
that increased activityof these enzymes maynot be requiredfor the earliest
glyceollinaccumulationininfectedtissues (Partridge and Keen,1977).
However, later studies, employing direct assays of enzymatic activity (Bon-
hoff et al., 1986a,b; Bornerand Grisebach, 1982), radiolabelincorporation
from earlyprecursors(Moesta and Grisebach, 1981;Yoshikawaetal.,
1978; Zahringer et al., 1978), and messenger RNA measurements (Esnault
et al., 1987; Habereder et al., 1989; Schmelzer et al., 1984) suggested that
glyceollin biosynthesis is accompanied by increased transcription and activi-
ties of these enzymes and by transient synthesis the of isoflavone precursor
of glyceollin, daidzein. Recently, race-specific induction of several of the
later enzymes in glyceollin biosynthesiswas also reported in infectedroots
(Bonhoff et al., 1986a,b).
Although P. sojae infectsallsoybeanseedling organs, race-specific
resistance, characterized by rapid and large accumulationsof glyceollin, is
expressed somewhat differently in each individual organ. Moreover, a series
of important studies demonstratedthat the age and/or developmental state
of the specific organ infected by P. sojae as well as environmental condi-
tions, particularly light, strongly influence race-specific accumulation of
the glyceollins (Bhattacharyyaand Ward, 1986a,b; Lazarovits et al.,1981;
Paxton and Chamberlain, 1969; Wardand Buzzell, 1983; Wardand Lazar-
ovits, 1982). Furthermore, specific cell populations within a given organ
may be involved in glyceollin accumulation (Grahamand Graham, 1991b;
Hahn et al., 1985; Yoshikawaet al., 1978).
A few laboratories have begun to examine discrete cellular aspects of
the responses of various plant tissuesto infection or elicitor treatment(for
a recent review, see Graham and Graham, 1991~).For instance, several
reports have elegantly elucidated localized changes in gene expression or
defenseproductaccumulation at thesite of attemptedpenetration by
pathogens (Cuypers et al., 1988; Pierce and Essenberg, 1987; Schmelzer et
al., 1989;Snyder and Nicholson,1990). In severalmorerecentstudies,
changes in the expression of specific genes (Stermer et al., 1990) or dis-
Phenylpropanoid
Responses of Soybean to P. sojae 87
tinctly different accumulationsof defense products (Grahamand Graham,

1991a,b) have been documented in cell populations proximaland distal to
the sites of infection or elicitor treatment. In our own work, using highly
sensitivehigh-performanceliquidchromatography (HPLC) metabolite-
profilingprocedures(Graham,1991a), we havedemonstrated that dis-
tinctly different shifts in the expression of the phenylpropanoid pathways
occur in proximaland distal soybean cell populations in response to the P.
sojae wall glucan elicitor (Grahamand Graham, 1991a,b). As discussed in
more detail below, similar responses occur in tissues infected by incompati-
ble races of P. sojae. These multipleand complementary defense responses
are carefully coordinatedby the host both spatially and temporally to pro-
vide effective resistance.
In this chapter are described some ofthe progress being made in under-
standing the multiplicity of phenylpropanoid defense responses in soybean
and the cellular biochemistry of their regulation. Cellular biochemistry re-
fers to those aspects of the phytoalexin and other phenylpropanoid re-
sponses related to spatial and temporal coordination of cellular response,
cellular specialization in response, cell-to-cell communication, and signal
perception andtransduction.
B. Special Constraints in Cellular Research and Approaches

to OvercomeThem
Since the pioneering studies of Yoshikawa and coworkers (1978) on cellular

aspects of the soybean-P. sojae interaction, it has become increasingly
apparent that a clear correlation of phytoalexin accumulations to the in-
compatible (hypersensitive) response requires that one quantitate phyto-
alexin levels inthe very discrete cell populationsthat are immediately rele-
vant to pathogen containment. Generally these are the cells at or just in
advance of the infectionfront.
Since a natural infection front is often composed of just a few cells
surrounding the point of attempted penetration by the pathogen, a key
problem has been the development of protocols to allow the measurement
of molecular responsesat a cellular level. Methods such as in situ hybridiza-
tion, immunolocalization, andHPLC have been particularly useful in mea-
suring responsesat the mRNA, protein, and product levels, respectively.A
recent review summarizesthe development and use of some of these meth-
ods, some of their limitations,and the insights they are providing us (Gra-
ham and Graham, 1991~).
Even thoughwe have made considerable progress in measuring molecu-
lar.responses in discrete cell populations, very few methods are sensitive
88 Graham
enough to measure responses at a truly cellular level. Those that can be

employed at a single-cell level (e.g., in situ hybridization and immunolocali-
zation) are qualitativeor atbest semi-quantitative. This has led researchers
to employ protocols which increase the size of the cell populations undergo-
ing hypersensitivity.For example, Yoshikawaet al. (1978) circumvented the
problem of sensitivity of detection by creating an artificially large infection
front by uniformly inoculating the surface of a longitudinally wounded
hypocotyl. The use of cell suspension cultures has also been employed to
rapidly subject a large population of small cell clusters to a pathogen or
elicitor. Another approach has been to take advantage of the fact that
under certain conditions some hypersensitive responses progress through
tissues as a very slowly spreading necrotic lesion. For instance, soybean
cotyledons show very sharp and localized necrotic lesions to incompatible
isolates of Phytophthoru sojae under high light intensity, slowly spreading
necrotic lesions under low light intensities, and virtually no race-specific
resistance in the dark (Graham et al., 1990). One can gain a great deal of
information from such a system by studying the responses under varying
conditions and either analyzing cells at a given distance from the point of
inoculation over timeor sampling cell populations at various distancesfrom
the point of inoculation at a given time,or preferably both.
Another constraint to effective cellular research that is there are usually
many tissues in a given organ and these tissues are sometimes themselves
only a single cell layer thick. As pointed out by Barz and Hoesel (1979),
distributions of flavonoids areoften tissue-specific within an organ. Thus,
work with hypocotyls at a cellular level is complicated by the fact that a
pathogen may spread longitudinally within a tissue or radially through a
series of different tissues. Research on roots shares these problems with the
added complicationthat the longitudinal zonewhich is subjectto infection
is often near the root tip and undergoing rapid growth and developmental
changes aswell. Tissue culture again has the potential advantageof greater
cellular homogeneitybut suffers fromthe fact that the cells are often dedif-
ferentiated, they are not in a natural environment, and it is very difficult
to measure cell-to-cell communication ina meaningful way. In our labora-
tory, we began our studies with cell cultures (Pierson and Graham, 1987)
but quickly shifted to the use of cotyledons when we realized the compar-
ative architectural simplicity and cellular uniformity of this organ. Soy-
bean cotyledons are largely composed of tightly aligned columnsof meso-
phyll parenchyma cells of very uniform dimension. As we will describe in
more detail below, these attributes and the tight packing of these cells make
them nearly ideal in measuring cellular responseand cell-to-cell communi-
cation.
A final major constraint to effective cellular research isthe extraordi-
Phenylpropanoid
Responses of Soybean to p. sojae 89
nary complexity ofthe interactions taking place in infected tissues. The fact
that one is trying to observe discrete molecular events in the interaction
between two living systems leads to great difficulties in reproducibility.
Moreover, there are almost certainly multiple signaling processes taking
place, leading to overlapping or possibly conflicting effectson the process
under measurement. These problems are greatly compounded by the fact
that in most systemsthe genetics of eitherthe host or pathogen (orboth) is
ambiguous, a fact that is often overlooked in the literature. For instance,
although near-isogenic isolines of soybean are available carryingthe various
Rps resistance genes, nearly all work on the soybean-P. sojae interaction to
date has used genetically undefined fieldor laboratory isolates of the fun-
gus. Recentefforts at obtaining pure-breeding races of P. sojae have shown
that although such isolates may displaya particular race phenotype, they
often harbor the genes whichare determinant for other race characteristics
as well @hat et al., 1993). Even though the interaction of such an isolate
may show the appropriate gross symptomatic phenotype, it is completely
unknown as to howsuchgeneticheterogeneitymay affect more subtle
interactions measured at a molecular level. Also, different researchers use
different isolatesof a given race or favor the use of different races in their
studies. In addition to heterogeneity in isolates as regards race characteris-
tics, different isolatesof a given race can differ greatly in their aggressive-
ness. The aggressiveness of a particular race can dramatically effect the
spatial and temporal aspectsof interaction with the host at a cellular level
(Graham and Graham, unpublished).
Electrophoretic karyotypingof a large number of field and laboratory
isolates of P. sojae (our results, unpublished) has confirmed the remarkable
genetic heterogeneity ofdifferent isolates of P. sojae. As many as10 differ-
ent chromosomes are found in various P. sojae isolates, but no isolate
carries all of these and some isolates share as few as one or two chromo-
somes in common with others.
In the soybean system, two complementary approaches are being un-
dertaken to circumvent some of these problems associated with examina-
tion of intact infected tissues. First of all, pure-breeding races are being
developed from defined field parentage@hat et al., 1993). These racesand
subsequent variants generated by mutation or transformation will greatly
facilitate the interpretation of infection studies. Second, as described in
detail below,we have determinedthat P. sojae wall glucan elicitor prepara-
tions not only induce allof the multiple phenylpropanoid responses seen in
incompatible infections,but induce them with similarspatial and temporal
coordination. This crucially important finding allows us to examine host
responses to purified elicitor molecules and thus isolate and study very
specific cellular responses of the host.
90 Graham
II. DEVELOPMENTAL REGULATION AND GENETICS OF THE

DISTRIBUTION OF PHENYLPROPANOIDDERIVED METABOLITES
Before one addressesthe role of alterations in phenylpropanoid metabolites
in the defense responses of various organs,it is criticalto thoroughly under-
stand the constitutive distributionof these metabolites within these organs
and the developmental and genetic parameterswhich influence their turn-
over. Only in this manner can experiments on induced changes associated
with infectionor elicitor treatmentbe properly designedand interpreted.
A. Distribution and Developmental Regulation

Various soybean organs and root and seed exudates display characteristic
and complex aromatic metabolite profiles (Graham, 1991b). Over a hun-
dred individual UV-absorbing metabolites can be separated by gradient
HPLC and only a handful of those in any given organ have been identified.
However, the predominant constitutive metabolites characteristic of each
of these organs have been isolated and characterized. These include the
isoflavones (daidzein, genistein, glyceteine, formononetin, and biochanin
A) and their conjugates, and the flavonols (kaempferol, quercetin, and
isorhamnetin) and their conjugates. Generally, the isoflavones are the pre-
dominant metabolites in all young soybean seedling organs and in older
cotyledons, stemand root tissues, and the flavonols are the major metabo-
lites in older leaf tissues (Graham, 1991b).
Although presentat detectable levels in soybean tissues, formononetin
and biochanin A are comparatively minor metabolites. However, large con-
stitutive pools of conjugates of daidzein, genistein(Graham, 1991b), and,
occasionally, glyceteine (Morris et al., 1991)are present in various organs.
The presence of high levels ofconjugates of daidzein isof particular interest
since daidzein isthe first committed metabolite inthe synthesis of glyceol-
lin. Genistein, although not considered a precursor of glyceollin, possesses
antibiotic activity against P. sojae (Rivera-Vargas et al., 1993). There is
little information on the biological activities of glyceteine.
Soybean tissues generally contain low amountsof the free isoflavones
(Graham, 1991b). The only exception is daidzein, which is sometimes pres-
ent in substantial amounts inroot tissues. The malonylglucosyl conjugates
of the isoflavones are the predominant forms found constitutively in all
soybean organs. The glucosyl conjugate of daidzein is present in lower
amounts in all tissues, while the glucosyl conjugate of genistein is found
at substantial levels only in lower hypocotyl sections and the crown. The
malonylated conjugate of daidzein is the predominant metabolite inroots
and hypocotyls, particularly in the root tip, where it reaches the highest
levelsin the seedling. Lower amounts of the malonylated conjugate of
Phenylpropanoid
genistein are also present inthe root and hypocotyl, and both daidzein and
genistein conjugates are secreted by soybean roots. The detection of free
daidzein and genistein in soybeanroot exudates may be the result of enzy-
matichydrolysisduringcollection or processingofsamples(Graham,
1991b). Very much lower amounts of conjugated glyceteine (peak 19.5 of
Graham, 1991b) have since been identified by Morris et al. (1991), present
predominantly inthe lower hypocotyl.
In cotyledon tissues, malonylglucosyl conjugates of daidzeinand gen-
istein are at about equal and relatively high levels. Althoughboth are pres-
ent in very young leaves, the malonylated conjugate of genisteinthe is only
major isoflavone in older leaves (Graham, 1991b).
The seed contains vast storesof the isoflavones (Graham, 1991b).The
relative amounts and distribution in the various parts of the embryo are
very similar to those seen in the counterparts of young germinated seed-
lings, suggesting that there may be very little net synthesis or turnover of
these compounds in the first few days postgermination (Graham, unpub-
lished). The deposition of daidzein and genistein and their conjugates has
also been investigated during seed development and maturation (Graham
and Olah, unpublished). As in seedling tissues, the free isoflavones do not
accumulate to appreciable amounts during seeddevelopment.However,
accumulation of the glucosyl conjugate of daidzein occurs first and then
falls off as the malonylglucosyl conjugate of daidzein accumulates later in
embryo development.The malonylglucosyl conjugate of genistein accumu-
lates later than the malonylglucosyl conjugate of diadzein without prior
accumulation of the corresponding glucosyl conjugate.The timing of these
various accumulationsis interesting and may reflect developmentally pro-
grammed inductionof the corresponding enzymes.
During seed imbibition, the conjugates of both daidzein and genistein
are released in a continuousbut saturable manner (Graham, 1991b). If the
imbibition medium is replaced,the exudate quickly reestablishessaturating
amounts of the isoflavones.Thissuggests that the verylargestoresof
daidzein and genistein present in the seed (and, as noted above, in the
root) may be available in the rhizosphere, where they may play a role as
chemoattractantsfor P.sojae (Morris and Ward, 1992).
Light has a particularly strong effect on the amounts of and distribu-
tion of the isoflavones in soybean tissues (Graham, 1991b). In the dark,
isoflavone levels in the root tips are greatly reduced, while those in the
cotyledon are higher. It is possible that at least part of the root isoflavone
pools may be transported to the roots from the larger stores present in the
cotyledons (Graham, 1991b).
Older soybean leavesare predominated by glycosides of the flavonols
kaempferol, quercetin (Buttery and Buzzell, 1975),and isorhamnetin (LeVan
92 Graham
and Graham, unpublished). The geneticsunderlying the formation of

kaempferol and quercetin and their glycosides has been elegantly described
by Butteryand Buzzel (1975).Parallel studieson isorhamnetin remainto be
done. These various flavonols are not present at readily detectable levels
except inthe leaf (Graham, 1991b).
B. Genetics
Although the genetics of flavonoland flavonol glycosideformation in soy-
bean has been described, there is littleinformation available on the genetics
of isoflavone and isoflavone conjugate formation. However, evidence to
date suggests much less genetic diversity in the nature of the isoflavone
aglycones and their conjugates. We have examined over 100 soybean lines
of various genetic backgrounds and none of these vary markedly in the
amounts or distributions of the isoflavones. Amongthe lines examined are:
1.) the Williams isoline series (Williams, Williams 79, Williams 82) carrying
no Rps resistance genes and the Rps 1c.and l k genes respectively, 2.) the
Harosoy differentials, carryinga wider range ofRps genes integrated back
into the Harosoy background,3.) soybeans from allof the various flavonol
genetic groups describedby Buttery and Buzzel (1975), 4.) the Bradyrhizo-
biurn japonicurn response mutants of the cultivar Bragg (Abbasi et al.,
unpublished),including the nonnodulating mutants nod49 and nod139
(Carroll et al., 1986),and the supernodulating mutants nts382 and nts1007
(Carroll et al., 1985),and 5.) Glycinesojae. Thus the presence and distribu-
tion of the isoflavones isnot influenced by the Rps genes, the genes control-
ling flavonol metabolism, or the genes controlling Bradyrhizobiurn nodula-
tion. The fact that Glycine sojaecontains the same compounds with similar
organ-specific distribution (Abbasi etal., unpublished) underscoresthe lack
of genetic diversity in these compounds within the genus.
111. MULTIPLICITY OF INDUCED PHENYLPROPANOID RESPONSES

IN INFECTED A N D ELICITOR TREATED TISSUESA N D SPATIAL
AND TEMPORAL ASPECTS OF THEIR REGULATION
The isoflavone daidzein is the first committed metabolite inthe biosynthetic

pathway to the soybean phytoalexin glyceollin. As noted above, daidzein
and the related isoflavone genistein were previously thought to be tran-
siently synthesized onlyafter pathogen attack or upon treatment with elici-
tors of the isoflavonoid phytoalexins (Ebel, 1986).The discovery that very
large poolsof these isoflavoneswere present in soybean seedling tissues as
preformed conjugates (Graham etal., 1990) thus raised obvious questions
as to the potential role(s) of the conjugates in disease resistance and/or in
Phenylpropanoid
glyceollin accumulation. Inaddition, the recent discoverythat large deposi-

tions of cell wall phenolics occur in response to infection or elicitor treat-
ment (Graham and Graham, 1991a) suggested a possible complementary
role of these phenylpropanoids in defense as well.
As described below, studies on the relative roles of these various phe-
nylpropanoids in infectedand elicitor-treated tissues confirm that soybeans
maypossess a multitude ofdiscreteyetcomplementarymechanismsof
defense based on the differential control of various alternative phenylpro-
panoid pathways (Fig.1). Intriguingly, different populations of cellsappear
to play distinct roles in the differential deployment of the responses. The
host appearsto coordinate these defense strategies in a highly sophisticated
manner, not only through the regulation of complex metabolic pathways
withcomplementaryfunctions but through the careful orchestration of
IPHENYLALANINE I
- PROXIMAL
., OCH3 LIGNIN/SUBERIN
" -- D I STAL
I CHALCONE/ I
FLAVANONE
Figure 1 Phenylpropanoid defense responses of proximal and distal soybeancells.

94 Graham
cellular communicationand response. In addition, organ- or tissue-specific,

spatial, developmental, and environmental effects on glyceollin accumula-
tion and/or resistance have beenwell documented in the soybean-P. sojae
interaction. Our preliminary results suggest that some of these organ- or
tissue-specific effects may also bedue to specific differences inthe regula-
tion of the various alternative pathways outlined in Fig.1.
A. Responses of Infected Tissues

Although all soybean seedling organs are infected by P. sojae and all of
these tissues show expression of race-specific resistance conditioned by the
Rps genes, we began our infection and elicitor studies with cotyledon tissues
for several reasons (Graham et al., 1990). By far the most valuable aspect
of the cotyledon system, however, is the fact that cotyledons are made
up largely of tightly aligned columns of remarkably uniform mesophyll
parenchyma cells. This attribute greatly simplifiesthe examination of spa-
tial and temporal events relating to the infection front (Graham et al.,
1990) and has provided us witha very powerful tool to examine cell-to-cell
communication following elicitor treatment (Graham and Graham, 1991b).
Our studies with infected cotyledon tissues (Graham et al., 1990) dem-
onstrated that there is a rapid and nearly complete net hydrolysis of both
diadzein and genistein conjugates at the infection front in incompatible
infections between 12 and 24 hr, followed bythe accumulation of glyceollin
during the period 24-36 hr. Both the stoichiometry and timing of the re-
sponses were consistent with a role of the hydrolysis of the daidzein con-
jugatesin the accumulation ofglyceollin.Freegenisteintransientlyac-
cumulated to levels as high as 800 nmol/g tissue between 24 and 48 hr.
Subsequent research showed that genistein is toxic to P. sojae (Rivera-
Vargas and Graham, 1993) at levels well below those released at the infec-
tion front. In compatible infections, the release of both isoflavones from
their conjugates was much later, well after the infection front had pro-
gressed through the tissue, and very low levels of glyceollin accumulated
only after 72 hr (Graham et al., 1990). Thus, in cotyledon tissues carrying
the Rps IC gene for resistance, infection results were consistent with two
race-specific events contributingto the antibiotic containment of P. sojae.
First is the rapid release ofthe antibiotic genistein and glyceollin precursor
diadzein from preformed conjugates, and secondis the somewhat later
formation of the more elaborate antibiotic glyceollin from daidzein. Al-
though limited net de novo synthesis of 5-deoxyisoflavones(total daidzein
and glyceollin) was also seen in this study, suggesting that newly synthesized
daidzein could also be contributing to the glyceollin accumulations (Gra-
ham et al., 1990),the results suggesteda potentially moreimportant role of
Phenylpropanoid
release of daidzein from preformed conjugates in the early incompatible

response. In contrast, net accumulations of daidzein, as its malonylated
conjugate, occurred in tissues ahead of the infection front (Graham, un-
published results), suggesting that de novo synthesis and conjugation was
favored over conjugate hydrolysis in these outlying tissues. These infection
results wereour first indicationsthat distinctly different metabolic switches
were being thrown in different cell populations in the overall incompatible
response. The accumulation of isoflavone conjugates in cells aheadof the
infection front is intriguing and suggests that the signal(s)whichelicit
isoflavone conjugate synthesis might spatially precede thosefor conjugate
hydrolysis and glyceollin accumulation, thus “setting up” the cells in ad-
vance of the infection front to respond witha larger free isoflavone release
(and thus genistein and glyceollin response) ifthe infection front progressed
into this tissue. In essence, this distal cell response could raise the defense
potential of these uninfected tissues.
Subsequent studies inour laboratory have examinedthe potential roles
of the diadzein and genistein conjugates in infected hypocotyl, root, and
leaf tissues as well. As expected,we have found that the design and interpre-
tation of experiments with these organs (especially roots) is much more
complicated dueto the greater difficulty in sampling discrete tissues within
these organs. However, to date these studies generallysupport the cotyle-
don results,including the nethydrolysisofconjugates at the infection
front, the subsequent accumulationof glyceollin, andthe net accumulation
of the isoflavone conjugates aheadof the infectionfront. The relative con-
tribution of the various responses outlined above for cotyledon tissuesmay
be somewhat different in these organs due to differences inthe amounts of
daidzein and genistein in these tissues. On the other hand, based on very
preliminary studies with hypocotyl and leaf tissues, Morris et al. (1991)
suggest that the conjugates play only a “secondary” role in the glyceollin
response to infection, even though the timing of hydrolysis of the conju-
gates in their studiesis actually consistent witha primary role in glyceollin
accumulation. Unfortunately, these studies (Moms et al., 1991) were very
limited in their examination of specific temporal and spatial events, and
those events which were followed were not correlated to the infection front
per se, whichwas never localized.
B. Responsesto the Clucan Elicitor from P. sojae

Results with infected tissues prompted us to reexamine the activity of the
P. sojae wall glucan elicitor which previously was characterized for its
elicitation of glyceollin (Darvilland Albersheim, 1984). We wishedto deter-
mine if some of the isoflavone turnover events seen in incompatible infec-
96 Graham
tions could be triggered by these glucan preparations. Our initial finding,

that the P.sojae wall glucan also induced large accumulations of the isofla-
vone conjugates in soybean tissues, was very intriguing (Grahamand Gra-
ham, 1991b). Even more relevant to the results from infected tissues was
the fact that, while glyceollin was induced in cells proximal to the site of
elicitor treatment, the isoflavoneconjugatesaccumulatedinmassive
amounts in cells as many as 25-30 cells from the elicitor treatment. Thus
it seemed that the glucan induced the two discrete (proximal and distal)
biosynthetic events seen in infected tissues, but not the net hydrolysis of
conjugates observed at the infection front. In these studies, however, we
further reported that the isoflavones also showed net accumulation in prox-
imal cells, the amounts relative to glyceollin depending on the amount of
elicitor, the age of the cotyledons, and the assay conditions.
Thisdifferential control of the relative amounts ofglyceollin and
isoflavone synthesis in proximal cells suggested that we might stillbe moni-
toring a mixed cell responseand prompted us to evaluate cellular events in
proximal cells in more detail. As noted in our initial work (Graham and
Graham, 1991b), the proximal cell layer(four cells thick) was enriched for
“proximal” cells,but might contain some cells showing the “distal” response
as well. This possibility led us to examine even thinner cell layers at various
times and under varying conditions. As we selectively sampled cells more .
and more immediately proximal to the elicitor treatment, we uncovered a
rapid net hydrolysis of the conjugates of both daidzein and genistein within
the first 4 hr (Graham and Graham, unpublished). Thelevels of the malo-
nylated conjugates of daidzein and genistein fall precipitously within this
short period to levels often only a third of their original amount. Following
this net hydrolysis, however, isa later period of renewed synthesis of daid-
zein conjugates. Dependingon the conditions of the assay and the relative
flux into glyceollin, total daidzein levels may partially recoveror even un-
dergo a net accumulation. Genistein conjugates,on the other hand, show
little reaccumulation in this cell layer. In distal cells, conjugates of both
isoflavones show net accumulations with no early hydrolysis.
Thus even the net hydrolysis ofthe isoflavone conjugateswhich accom-
panies the incompatible responseat the infection front is seen in proximal
cells in responseto elicitor, although, for reasons we do not yet understand,
the hydrolysis is more prolongedin infected tissues. Net isoflavone conju-
gate hydrolysis also occurs upon infection of alfalfa tissues by Ascochyta
irnperfecta (Olah and Sherwood, 1973). In these studies,the authors found
that glycosidases of both host and pathogen origin could be detected in
infected tissuesbut that isozymes correspondingto the pathogen predomi-
nated. A similar contribution of both host and pathogen enzymes may be
operating in soybean. Although our results in elicitor-treated tissues suggest
Phenylpropanoid
that host enzymes maycontribute to this response, we have demonstrated

the presence in P. sojae of enzymes capable of rapidly hydrolyzing the
daidzein and genistein conjugates (Rivera-Vargas andGraham, 1993). In-
triguingly, these enzyme activities are associated with the growing hyphal
tip in P. sojae, which places them in a key position to have an impact on
events at the infection front. The P.sojae wall glucan may activate only the
host enzymes, giving rise to a more transient net effect. Since conjugate
hydrolysis occurs much later in the compatible response (well after the
infection front passes through the tissue), either the host or the pathogen
enzymes or both may be temporarily suppressed in compatible infections.
It is intriguing to speculate that control of conjugate hydrolysis at the
infection front by the host and pathogen may play a very central role in
defining the race-specific outcome ofthe interaction between P.sojae and
soybean by contributing to, or failing to contribute to, the net and timely
release of genisteinand daidzein. To decipher the relative roles of hostand
pathogen in conjugate hydrolysis in planta will require that the enzymes of
both host and pathogen origin be characterizedand their regulation exam-
ined in situ during infection.
Although the HPLC profiling protocolsgave usa very complete picture
of the turnover of various soluble aromatic metabolites in discrete cell
populations, they did not allow us to monitor the accumulation of wall-
bound phenolics. Elicitor-induced accumulations of glyceollin in the cut
cotyledon assay are nearly always accompanied by a browning of the cut
cotyledonsurface, an increaseinitshydrophobicity and a remarkable
toughening of the uppermost cell layer. This prompted us to examine the
effects of the P. sojae wall glucan elicitor on peroxidase isozymes and
phenolic polymer deposition. Although wounding alone caused the gradual
induction of a specific group of anionic peroxidases and the subsequent
deposition of both lignin and suberin like polymers, the P.sojae wall glucan
was found to greatly increase the rapidity of this response (Graham and
Graham, 1991a). Within just 4 hr of wall glucan treatment, phenolic poly-
mer levels were over 10 times those in wounded controls. This more rapid
induction of the phenolic polymers was accompanied by correspondingly
earlier inductionof the anionic peroxidases. Althoughthe peroxidases were
also induced substantially in distal cells,
there was very little phenolic poly-
mer deposition, possibly due to a lack of appropriate substrates or suffi-
cient oxygenfor this highly oxidative process.
Of additional interest was the fact that large quantities of monomeric
hydroxycinnamic acids (coumaric and ferulic) aswell as several unidentified
phenolics were incorporated into the cell walls in ester linkages (Graham
and Graham, 1991a). The accumulation of such wall-esterified phenolic
acids occurs in other species as well and has previously been suggested to
98 Graham
play a potential role in disease resistance (Hahlbrock and Scheel, 1989).

Interestingly, the deposition of these acids would most likely be mediated
by a transferase and not a peroxidase. Thus their deposition suggests yet
another response that may be induced by P. sojae wall glucan. Their pres-
ence in the wall is intriguing and opens the possibility that these phenolics
could serve as a store for future, more rapid wall polymer accumulations.
This possibility would be even more interesting if the esterified hydroxycin-
namic acids were induced in distal cells. We have to yetexamine distal cells
for wall-bound phenolic esters.
Events similar to those seen in response to the glucan elicitorare also
induced at the infection front in incompatible infections (Graham and Gra-
ham, unpublished). Thus a verycomplexseries of phenolic depositions
occurs inthe walls of cells proximal to the infection front or siteof elicitor
treatment. Of greatest interest, however,is the very early and massive na-
ture of these responses when comparedto isoflavone hydrolysis or glyceol-
lin accumulation. The wall phenolic responses, taken together, represent
not only a much earlierbut a 17-fold greater commitment of phenylpropane
skeletons than the glyceollin response (Graham and Graham, 1991a). It is
possible that de novo synthesis of phenylpropanoids in proximal cells is
predominantly directedinto the wall defenses. This would explain the low
level of net de novo accumulation of soluble phenolics, including daidzein
in these tissuesand the apparent use of the preformed daidzein conjugates
for glyceollin synthesis.In distal cells,on the other hand, de novo synthesis
is directed toward isoflavone synthesis and conjugation rather than the
phenolic polymers.
The various events described at length aboveare summarized in Fig. 1.
It is clear that P. sojae infection or elicitor treatmentturn on a remarkably
complex array of phenylpropanoid-related responses. These responses, in
turn, are coordinated both spatially and temporally in a highly sophisti-
cated manner by the host. We have seen that this coordination involves a
very precise “gating” of metabolitesinto and out of the various phenylpro-
panoid pools (isoflavone conjugate, wall phenolic,and glyceollin) in several
distinct cellular zones. AlthoughP. sojae wall glucan preparations are ap-
parently sufficient to trigger all of these cellular events, it is obvious that
additional regulation of cellular response istaking place.
W. SIGNAL MOLECULES FOR THEREGULATION

O F CELLULAR RESPONSE
The multiplicity of soybean phenylpropanoid responses to infection or elici-
tor treatment, the sophisticated differential deploymentof these responses
in differentcell populations, and the fact that P. sojae wall glucan prepara-
Phenylpropanoid
tions induce all of the various coordinated responses seen in incompatible

infected tissues make this system a particularly rich onefor studies on signal
perception, signaltransduction, and cell-to-cell communications.The abil-
ity to follow the various responses so clearly in soybean cotyledons has
proven to be a particularly powerful tool for these studies for reasons
alluded to above. A major emphasis of our lab is thusto begin to decipher
the various molecular and cellular components involved in perception, sig-
nal generation, and signal transduction in response to the P. sojae wall
glucan elicitor.
An obviousfirst step in such an effort, however, isto further character-
ize the signals that initiate or conditionthe various responses. This is partic-
ularly important due to the inherent heterogeneity of polysaccharide elicitor
preparations. Questions we are currently addressing include: Arethe vari-
ous wall glucan responses described above due to a single elicitor or are
there multiple elicitors present in the glucan preparations? Are there host
factors, such asthe pectic oligomers,that act as synergistsor coelicitors? If
so, how do these various pathogen-and host-derived primary signals work
together to initiate or propagate the proximal and distal cell responses to the
glucan elicitor?Is there actuallya separate signal for the distal response?If
so, is this generated inthe plant as a result of the proximal responseor is it
released or processed from one of the primary proximal elicitor signals?
Since age and organ developmental state influence the various responses,
what are the roles of cellular growth regulators in the regulation of the
various cellresponses?Light and nutritional status alsohave profound
effects on the responses. Are these later effects simply influencing the gen-
eral physiologyofthecell or do they have more specific and selective
influences on signal perceptionand transduction?
To simplify discussions, we have defined two overall types of regula-
tory processes: 1.)primary elicitoror signal processes which serve to trigger
the phenylpropanoid defense responses, and 2.) secondary molecular or
cellular processes which may serveto condition or modify specific cellular
response(s) to the primary signals. Although these various processes are
highly intertwined,we have begunto gain some insightsinto their nature.
Being an external signal,P.sojae wall glucanper se is clearlya primary
elicitor. However, as we have demonstrated, this elicitor differentially in-
duces a variety of phenylpropanoid responses in different soybean cell pop-
ulations, including isoflavone conjugate hydrolysis, isoflavone conjugate
accumulation,glyceollinaccumulation,wall-esterifiedhydroxycinnamic
acid accumulation, and phenolic polymer deposition. These various activi-
ties could be due to 1.) the presence of more than one primary elicitor
activity in the P. sojae wall glucan preparation, 2.) the generation, during
infection or elicitor treatment, of additional primary or secondary elicitor
100 Graham
signals in situ (e.g., host cell wall oligogalacturonides) which act indepen-
dently or as coelicitors withP.sojae wall glucan, 3.) inherent differences in
the responses of specific cell populationsto the same elicitor,or 4.) differ-
ential conditioning of cellular responses to the elicitor by cellular growth
regulators. As detailed below, preliminary evidence from our laboratory
and others suggeststhat each of these mechanisms mayoperate to achieve
differential control of these pathways in specific situations. What is not
clearishowthesevariousmechanismsofregulationworktogether to
achieve the sophisticated and precise control of the various pathways sug-
gested byinfection and elicitor studies with various tissues.
A. Multiple PrimaryElicitors
As noted above,we have demonstratedthat the classical unfractionatedP.
sojae wallglucan preparation ofAyerset al. (1976) triggers all of the
various phenylpropanoid responsesthat occur in incompatible infections.
Moreover, the responses induced by the wall glucanand incompatible infec-
tions are nearly identical in regardto the precise composition ofthe prod-
ucts and their timingand spatial coordination. This finding underscoresthe
potential importanceof the wall glucan as a primary determinant of defense
gene stimulationinsoybean and providesapointoffocus for future
studies.
The possible existence of multiple primary signals in the classical P.
sojae wall glucan preparation has been documented in several cases. First
of all, by its very nature, the branched /31,3/61,6 glucan itself is heteroge-
neous in both size and position of branch points. Thus, dependingon how
they are generated, elicitor fragments of different sizesand activities canbe
released. Albersheim and coworkers have elegantly demonstrated that the
highest per unit activity in acid hydrolysates is found in a seven-residue
branched 01,3//31,6 glucan (Sharp etal., 1984). However, using enzymatic
hydrolyseswithasoybean/3-1,3-endoglucanase,Yoshikawa(1988)con-
cluded that the highest glyceollin elicitor activity is present in the largest
wall glucan fragments. Thus,it is still nottotally clear as to which subfrac-
tions of the P. sojae glucan possess the most potent glyceollin elicitor activ-
ity inplanta. Nonetheless, the enzymatic release of the glucan elicitorfrom
the pathogen cell wall has been attributed to a now cloned /3-1,3-endoglu-
canase (Takeuchi et al., 1990a,b) and elegant research on the receptor for
highly defined glucan elicitors iswell underway (Cheong and Hahn, 1991;
Cheong et al., 1991; Cosio et al., 1992).
There is also evidence for elicitors with distinctlydifferent activities in
the P. sojae wall glucan preparations. In addition to the glucan elicitor
active in soybean, Hahlbrock and coworkers identified a minor protein-
Phenylpropanoid
aceous componentof the P.sojae wall glucanpreparation which serves asa

specific elicitor of the furanocoumarin phytoalexins in parsley (Parker et
al., 1988). In preliminary experiments, we have discovered that if an auto-
claved P. sojae glucan preparation is filtered through 0.2-pm filters and
thus separated into soluble and insoluble fractions, the insoluble wall glu-
can preparation is a very potent elicitorof isoflavone conjugates, glyceollin,
and phenolic polymers. The soluble portion is a relatively poor elicitor of
glyceollin orthe phenolic polymers,but is an excellent elicitor ofthe isofla-
vone conjugates. Since this soluble elicitor also induces the distal buildup
of the isoflavone conjugates, it may be at least partially responsiblefor the
isoflavone response of soybean tissues to infection or unfractionated P.
sojae wall glucan preparations. Whether this elicitor is a minor, nonglucan,
component of the wall preparation or a smaller, defined fragment of the
p1,3/p1,6glucan is unknown.
Related to the P. sojae wall glucan is the much smaller cytoplasmic
@l ,3//31,6glucan mycolaminaran. Mycolaminaran is alsoa heterogeneous
glucan containing30-36 glucose residues. It exists in both phosphorylated
and nonphosphorylated forms which vary in their relative levels during the
life cycle of Phytophthora. At certain stages in the life cycle, the neutral
form of mycolaminaran can reach levels as high as30% of the dry weight
of the mycelia (Wang and Bartnicki-Garcia, 1973, 1974). Mycolaminaran
has been shown to be a relatively weak elicitor of glyceollin (Keen et al.,
1983). Its effects on the other pathways outlined in Fig. 1 have not been
examined. Structural analogs of molecules often play a role as competitive
inhibitors or alternative ligands. Despite its structural similarity to the P.
sojae wall glucan, the possible role of mycolaminaran in modifying or
complementing the activity of the P. sojae wall glucan has also not been
examined.
Host oligogalacturonides have also been proposed as primary elicitors
or elicitorsynergists(Darvill and Albersheim, 1984; Davisetal., 1986;
Ebel, 1986). Although they could be classified as secondary, in the sense
that they are released only after infection, since they are generated in the
apoplast we will treat them in our discussions as primary cell signals. Like
P. sojae wall glucan, pectic oligomers are by nature heterogeneous. Dis-
tinctly different activities are associated with oligomers of different sizes.
While larger fragments, withan optimal size of12 residues, are most active
as glyceollin elicitors in soybean (Nothnagelet al., 1983), fragments of 2-
10 residues are optimal as elicitorsof protease inhibitors in potato (Ryan,
1981). Recently, the smaller oligogalacturonides (optimal activity of 7 resi-
dues) have also been shown to induce the deposition of lignin in castor bean
cultures (Bruceand West, 1989).
Not surprisingly, pectate degrading enzymes have also been demon-
102 Graham
strated to possess elicitor activity @bel, 1986). The activity of the pectate
lyase from Erwinia carotovora has been particularly well characterized.
This enzyme has been reportedas a glyceollin elicitor and as a synergist to
glyceollin elicitation byP. sojae wall glucan (Davis et al., 1986). Recently,
the discovery of proteinaceous inhibitors of endopolygalacturonidase (Cer-
vone et al., 1989) led to the hypothesis that the expression of these inhibi-
tors in host tissues could lead to limited digestion of host cell wall pectin
fractions and thus the accumulation of the oligogalacturonide elicitor syn-
ergists or to factors which leadto host cell death (Doares et al.,1989).
Preliminary experiments in our laboratory, using a purified pectate
lyase preparation from Aspergillus japonicus, suggests one possible expla-
nation for the synergism exerted by these preparations. Although we ob-
served no changes in the various phenylpropanoids in responseto pectate
lyase alone, treatment of soybean cotyledon tissues with low levels of P.
sojae wall glucan in the presence of pectate lyase ledto an increase in the
relative flux of metabolites into glyceollin and a marked decrease in the
accumulation of isoflavone conjugates.Pectate lyase thus has the effect of
shifting somewhat the balance of response to P. sojae wall glucan specifi-
cally away from the isoflavones and in favor of glyceollin. Since host cell
wall degradation might be expected to occur at the infection front, this
could be an additional factor contributing to the local accumulation of
glyceollin vs. distal accumulationof isoflavones. Although a direct elicitor
synergistic activity of the released pectic oligomers has been proposed, it
seems possiblethat dissolution of the middle lamella of cells in the vicinity
of the infection front may also simply expose more cells to the wall glucan
elicitor, thus facilitating its activity.
B. Secondary Regulation and Conditioning of Cellular Responses
Wound-Associated Competency Factors for Cellular Response

to the P. sojae Wall GIucan Elicitor
Classically, elicitors of the glyceollin response of soybeans have been as-
sayed usinga cut cotyledon assay in which the cut abaxial surface
of excised
cotyledons is exposed to a solution of elicitor (Frank and Paxton, 1971).
Over a period of 24-48 hr the glyceollins are produced and diffuse into the
elicitor droplet, which is then collected for UV or HPLC analysis. A s de-
scribed above, for many of our experiments, we have used a modification
of this assay (Graham and Graham, 1991b) in which the elicitor droplet is
allowed to dry onto the cut surfaceand a column of cells is then harvested
from the cotyledon. The cell column is then thin-sliced into disks and the
disks are extracted separatelyto allow examination of proximaland distal
cell responses. Several years ago we also initiateda minimal-wounding coty-
Phenylpropanoid
ledon infiltration assay in which cotyledons on intact plants are infiltrated

with elicitor by subepidermal injection and cotyledons are later harvested
and extracted for analysis (Lundry et al., 1981). This assay, although not
providing information on proximal and distal cells, allows us to examine
the effects of elicitor under minimal wound conditions.
One of our earliest observations with the cotyledon infiltration assay
was that elicitation of the glyceollin response required much higher levels of
elicitor(cf.Figs.2a and 2b). This was true withthreedifferent/3-1,3/
/3-lY6-linked glucan preparations: P. sojae wall glucan, mycolaminaran,
and yeast wall glucan. Inaddition to the marked increase inthe amount of
-
0 5 10 15 20 25 30 35
(W ELICITOR
CONCENTRATION (pg/mL).
Figure 2 (a) Activities of glucan elicitors in infiltration asssay. (b) Activities of
glucan elicitors in cut
cot assay.
104 Graham
elicitor required for half-maximal elicitation, maximal glyceollin elicitation

was significantly loweredand the assay more clearly differentiated between
the activities of the various elicitors. Further analysis of cells within the
cotyledon infiltration assay demonstratedthat the responses shown in Fig.
2a werecontributed solelyby the tiny fraction of wounded cellsat the point
where the infiltration needle wasinserted.
This result suggested to us that wounding not only greatly stimulates
the glyceollin response but may be a prerequisite for elicitor activity. To
test this hypothesis, we returned to the cut cotyledon assay. When the
wounded surface ofthe cut cotyledon assaywas immediately washed prior
to elicitor application,the glyceollin response was greatly diminished and in
some experiments completely abolished (Fig. 3). If washing was delayedfor
1-3 hr, the response to elicitor increased nearly to the level of unwashed
cotyledons. However, if washing were delayed even further (more than 4
hr), the cells were no longer responsive to elicitor regardless of washing
(Fig. 3). These results suggest that a washable wound factor@)may tran-
siently inducea state of elicitor competency inthe exposed cells.
To provide further evidence for such a factor@),we attempted to re-
store elicitor competencyto washed cotyledons by adding various amounts
of the wound washings back to washed cotyledons. As shown in Fig. 4,
elicitor competency is restored aindose-responsiveand saturable manner.
An obvious questionis whether the wound factor affects onlythe gly-
ceollin responseor also affects isoflavoneand phenolic polymer accumula-
tions. Our preliminary results suggest that at least two wound factors may
be present, one of which greatly increasesthe sensitivity of cells to elicita-
TIME OF WASHING (HR AFTER WOUNDING)
Figure 3 Time courseof establishment of competent state.

105
0 l 2 3
WOUNDEXUDATECO-APPLIEDWITHELICITOR(EQUIVALENTS)
Figure 4 Restoration of elicitor competency.
tion of all the phenylpropanoid responses (i.e., dramatically lowers the

amount of elicitor needed for half-maximal response). The other factor is
required for specifically “gating” released or induced daidzein toward the
formation of glyceollin.
The ability to restore elicitor competencyto washed cells suggestedthat
we could take two complementary approaches to the identification of the
wound factor(& Using restoration of elicitor competency as an assay, we
could purify the wound factor(s) directly or we could test known wound-
associated factors for their activity inrestoration.
Since we found that the amount of the wound factor present and the
effectiveness of washing in removing it was dependent on a number of
physiological parameters of the soybean plants used as a source for the
cotyledons (particularly light, age,and nutritional status), these conditions
must be controlled if the assay isto be used effectively.
Many different wound-associated molecules and possible elicitor syner-
gists have been reported. Some of these are listed in Table1 along with their
effects on restoration of competency to elicitor-incompetent (washed) cells.
Although someof these factors had interesting effects on phenylpropanoid
metabolism (alone or in combination withthe wall glucan), none of them
fully restored either aspect of elicitor competency to elicitor-incompetent
cells. Perhaps of most interest were the lack of restoration activity by the
pectate lyase [known as an elicitor synergist (Davis et al., 1986)] and the
strongly inhibitory activity of jasmonic acid [a signal known to induce
the protease inhibitors in solanaceous plants (Farmerand Ryan, 1992)] on
glyceollin accumulation.
Purification of the competency factor@)has just begun. At least one of
106 Graham
Table 1 Effects of Wound Factors or Potential Coelicitors on Glyceollin

Response in Elicitor-Incompetent Cells”
Factor Effect onphenylpropanoid

soybeanresponses
~
Wound exudate
Strongstimulation
of
glyceollin and phenolic
polymer
accumulations
Inhibition of daidzein and genistein conjugate accumu-
lations
Abscisic
acid Inhibitiongenistein
of conjugate accumulation
(slight
inhibition of glyceollin accumulation)
ACC/ethylene No significant
effect
Glutathione(reduced)Inhibitionofdaidzein and genistein conjugateforma-
tion, stimulation of phenolic polymer accumulation
(slight stimulation of glyceollin accumulation)
Jasmonic
acid
Strong
inhibition
glyceollin
of accumulation
Oxalic
acid
(Slight
inhibition
daidzein
of and genistein
conjugate
and glyceollin accumulations)
Pectate lyase
(Slight
stimulation
daidzein
of and genistein conjugate
accumulation, slight inhibition of glyceollin accumu-
lation)
Salicylic
acid
Stimulation
daidzein
of and genistein conjugate accu-
mulation
Traumatic acid No significant
effect
“Reported here, for simplicity, are those effects only on washed cotyledon cells in the presence
of the P. sojae wall glucan. Full details of other effects of these factors will bereported
elsewhere. Effects in parentheses were minor.
the activities is stableto relatively long-term storageat -8OOC and can be

readily fractionated. Further characterization is underway.
We have thus demonstratedthat wound-associated factors are required
for the competency of soybean cellsto respond to the P. sojae wall glucan
elicitor. Both sensitivity of the cells to elicitor and specific “gating” ofthe
general phenylpropanoid respbnseto glyceollin may be affected. None of
the known wound-associated factors tested fully restored competency to
elicitor-incompetent cells.
We hypothesize that the wound factor(s)may be releasedfrom dead or
dying cells in the hypersensitive lesion and that they may then condition
neighboring cells at the infection front to become competent for elicitor
responsiveness. Once identified thesefactors may shedimportant informa-
tion on the regulation of resistance vs. susceptibility by both host and
pathogen, possiblyat a race-specific level.
Phenylpropanoid Responsesof Soybean to P. sojae 107
Role of Cellular Growth Regulators

in Mediating
or Modulating Cellular Response.
A number of investigations have suggestedthat cellular growth regulators
may alter the expression of defense-related genes. Wereviewhereonly
those experiments which deal directly withthe regulation of phenylpropa-
noid metabolismor with the responses of discrete cell populationsto infec-
tion or elicitor treatment.
Ethylene. Perhaps mostconvincing,thoughstillsomewhatambigu-
ous, are the possible roles that ethylene may play in defense responses.
Ethylene and itsimmediateprecursor,l-aminocyclopropanecarboxylic
acid (ACC) have been implicatedthe in induction of such diverse responses
asphenylalanineammoniumlyase(Chappelletal.,1984),peroxidases
(Abeles et al., 1989), hydroxyproline-rich glycoproteins (Roby etal., 1986;
Rumeau et al., 1988), protease inhibitors (Ryan, 1981), and hydrolytic en-
zymes such as 8-glucanasesand chitinases (Mauchand Staehelin, 1989) and
proteases (Veraand Conejero, 1989). Thus, likeP. sojae wall glucan, ethyl-
ene is associated with a large number of defense responses. Indeed, early
increases in ACC and ethylene have been associated with P. sojae wall
glucan treatment in parsley (Chappell et al., 1984) and soybean (Lambert
and Graham, 1987). As with P. sojae wall glucan, however, in very few
cases is the association of this growth regulatora givenwith response clearly
understood.
As an example, although Chappell et al. (1984) demonstratedthat the
induction of ACC synthaseis one of the earliest responses of parsley tissues
to P.sojae wall glucan (<60 min), they concludethat the accumulation of
ACC alone is not sufficient to account for the subsequent induction of
phenylalanine ammonia lyase. A similar lack of a direct cause-and-effect
relationship of ethylene to defense gene expressionor phytoalexin accumu-
lation, respectively, was found by Mauch et al. (1984) in pea and by Parad-
ies et al. (1980) in soybean.
One ofthe clearest demonstrationsof a role for ethylene in host defense
gene expression is the recent work of Ecker and Davis (1987). These re-
searchers were able to demonstrate ethylene induction of several defense
gene mRNAs only when the uppermost layers of treated tissues were exam-
ined. Thus, as we have demonstrated in soybean, itis vital in research on
host responses to signal molecules to examine the responses of highly dis-
crete cell populations. Otherwise one may be examining a composite or
average of many separate responses. Researchis underway in our labora-
tory which we hope will help better definethe role of ACC and ethylene in
regulation of the various pathways outlined in Fig. 1. Although ACC and
108 Graham
ethylene are early, wound-associated factors, as noted above, our prelimi-

nary studiesdo not implicate themas cellular competencyfactors.
The role of ethylene in the signal process thus remains somewhat un-
clear. However, two recent papers may have particular relevance to their
role in soybean. In an examination of infection and elicitation in soybean
roots, Reinhardt et al. (1991) demonstrated that even though the glucan
elicitor induced glyceollin accumulation in this organ, the rapid burst in
ethylene biosynthesis characteristic of incompatibleinfection was not seen
with elicitor treatment. This may suggest that ethylene is not involved in
responses to the glucan per se but may play another role in defense gene
regulation. That role may in fact be the enzymatic release of elicitorfrom
the pathogen in planta, since Yoshikawa et al. (1990) have shown that
ethylene induces the activity of the host P-1,3-endoglucanases involved in
glucan elicitor release.
Polyamines. A second growth regulator which has been investigated
for its effects specifically on phytoalexin elicitation are the polyamines.
Preliminary work in pea (Hadwiger et al.,1974) and in soybean (Graham,
unpublished) has demonstrated that high levels (>500 PM) of the poly-
amines spermineand spermidine, but not the diamine putrescence, elicitthe
isoflavonoid phytoalexins pisatin and glyceollin, respectively. Although the
concentrations requiredfor elicitation are high, the fact that the polyamines
ate natural plant metabolites prompteda preliminary investigation of their
possible role asinternal messengers in pea (Teasdaleand Hadwiger, 1977).
It was concluded that changes in the concentrations of the polyamines in
response to infection by Fusarium were too small to account directly for
phytoalexin elicitation. However,the possibility that the polyamines might
play a conditioning rather than a messenger role was not investigated.
In preliminary studies (Lambert and Graham, unpublished), we dem-
onstrated that wounding of soybean tissues results in a transient burst of
putrescine (to four times that in control tissues within 12 hr and back to
control levels in 24 hr; Fig. Sa). Accompanying this burst in putrescine are
transient and stoichiometric accumulations of both spermine and spermi-
dine, which peak sharply at 24 hr (Fig. Sa). Treatment of wounded tissues
with P.sojae wall glucan completely suppresses the putrescine and spermine
accumulations (Fig. 5b), but markedly enhances and lengthens spermidine
accumulation over the period 24-48 hr. ThusP.sojae wall glucan treatment
affects wound-associated changes in polyamine levels. Although it is possi-
ble that the polyamines affect cellular competency,the late nature of these
changes and our preliminary experiments on competency restoration with
them (data not shown) do not implicate the polyamines in this function.
Changes in polyamines, however, could conditionor regulate later aspects
of cellular response.
Phenylpropanoid Responsesof Soybean to P. sojae 109
200
0-0 PUTRESCINE
0 24 48
(4 TIME AFTER TREATMENT (hr)
W
2 1
2ooP
150
/A 0-0
0".
A-A
PUTRESCINE
SPERMINE
SPERMIDINE
48
04
0
V
Q
O
--'U-
O
.--
24
-.-. 72
(b) TREATMENT
AFTER
TIME (hr)
Figure 5 (a) Induced changes in polyamines: Williams cotyledons, wounded con-

trol. @) Induced changes in polyamines: Williamscotyledons, wounded and PMG-
glucan-treated.
Cytokinins. Althoughtherehasbeenconsiderableresearch on the

roles of the cytokinins in green islandformation in response to biotrophic
fungi and in the growth abnormalities associated with fasciation and gall-
formingdiseases, there hasbeencomparativelylittleresearch on the
involvement of cytokinins in disease resistance.
However, a particularly clear demonstration of an effect of the cytoki-
nins on phenylpropanoidmetabolismcomes from the workofMiller
(1969). He demonstrated that 1 nM zeatin or 5 pM kinetin in the presence of
an auxin stimulated the productionof large quantities of two unidentified
conjugates of daidzein, called compounds I and 11, in soybean callus or
110 Graham
suspension cultures. Neither cytokinins nor auxins alonetoled the accumu-

lation of these conjugates. We have repeated the observation of Miller in
callus cultures; it is almost certainthat Miller’s compounds correspondto
the glucosylated and malonylated conjugates of daidzein. Thus cytokinins
may be involved in regulation of isoflavone and isoflavone conjugateturn-
over. How this specifically relates to proximal and distal cell events in
elicitor-treated or infected tissues needsto be examined.
V. EPILOG
As noted above,our current research directions center on further character-
ization of the primary signals triggering the various phenylpropanoid re-
sponses and the secondary signals that condition cellular response. Our
immediate objectives are to fractionate and characterize the various possi-
ble primary elicitor activities associated withthe P. sojae wall glucan, my-
colaminaran, and hostoligogalacturonide preparations and the cellular
conditioningactivitiesassociated with thewoundcompetencyfactors.
Complementaryto this are our efforts to more fully characterize the role@)
that cellular growth regulators may play in mediating or conditioning the
responses of different cell populations. Taken together, these research ef-
forts should provide us with important clues as to the signal transduction
processes per se. Cotyledon tissues have proven to be particularly valuable
in such studies due to their relative cellular uniformity and to the much
greater ease of treatment, analysis, and interpretation of results. Since all
of the responses we have elucidated are expressed in cotyledons, wewill
continue to employ them for the initial characterizations of elicitor and
growth regulator responses.
REFERENCES
Abeles,F.B.,Biles,C. L., and Dunn, L. J. (1989). Hormonal regulation and
distribution of peroxidase isozymes in the Cucurbitaceae. Plant Physiol. 91:
1609-1612.
Ayers, A. R., Ebel, J., Valent, B. S. and Albersheim, P. (1976). Host- pathogen
interactions. X. Fractionation andbiologicalactivity of an elicitorisolated
from the mycelial wallsof Phytophthora megasperma var. sojae. Plant Physiol.
57: 760-765.
Barz, W., and Hoesel, W. (1979). Metabolism and degradation of phenolic com-
pounds in plants. The Biochemistryof Plant Phenolics(Swain, T.et al., eds.),
Bhat, R. G., McBlain, B. A., and Schmitthenner, A. F. (1993). Development of
pure lines of Phytophthora sojaeraces. Phytopathology 83:473-477.
Bhattacharyya, M. K., and Ward, E. W. B. (1986a). Expression of gene-specific
Phenylpropanoid
and age-related resistance andthe accumulation of glyceollinin soybean leaves

infected with Phytophthora megasperma f. sp. glycinea. Physiol. Mol. Plant
Pathol. 29 105-11 1.
Bhattacharyya, M. K., and Ward, E. W. B. (1986b). Resistance, susceptibility and
accumulation of glyceollins 1-111 in soybean organs inoculated with Phytoph-
thora megaspermaf. sp. glycinea. Physiol.Mol. Plant Pathol.29227-237.
Bonhoff, A., Loyal, R., Ebel, J., and Grisebach, H. (1986a).Race:cultivar specific
induction of enzymes relatedto phytoalexin biosynthesis in soybean roots fol-
lowing infection with Phytophthora megasperma f. sp. glycinea. Arch. Bio-
chem. Biophys.246:149-154.
Bonhoff, A., Loyal, R., Feller, K., Ebel, J., and Grisebach, H. (1986b). Further
investigations of race:cultivar-specific induction of enzymes related to phyto-
alexin biosynthesis in soybean roots following infection with Phytophthora
megasperma f. sp. glycinea. Hoppe-Seyler’sZ. Biol. Chem.362797-802.
Borner, H., and Grisebach, H. (1982).Enzymeinduction in soybeaninfected by Phy-
tophthora megasperma f. sp. glycinea.Arch. Biochem. Biophys.21265-71.
Bruce, R. J., and West, C. A. (1989).Elication of lignin biosynthesis and isoperoxi-
dase activity by pectic fragments in suspension cultures of castor bean. Plant
Physiol. 91:889-897.
Burden, R. S., and Bailey, J. A. (1975). Structure of the phytoalexin from soybean.
Phytochemistry 141389-1390.
Buttery, B. R.,and Buzzell, R. I. (1975). Soybean flavonol glycosides: identification
and biochemical genetics.Can. J. Bot. 53:219-224.
Carroll, B. J., McNeil, D. L., and Gresshoff, P. M. (1985). Isolation and properties
of soybean Glycine mar (L.)Merr. mutants that nodulate in the presence of
high nitrate concentrations. Proc. Natl. Acad. Sci. USA 82:4162-4166.
Carroll, B. J., McNeil, D. L., and Gresshoff, P. M. (1986).Mutagenesis of soybean
(Glycine mar (L.) Merr.) and the isolation of non-nodulating mutants. Plant
Sci. 47:109-114.
Cervone, F., Hahn, M. G., De Lorenzo, G., Darvill, A., and Albersheim, P. (1989).
Host-pathogen interactions. XXXIII.A plant protein converts a fungal patho-
genesis factor into an elicitor of plant defense responses. Plant Physiol. 90:
542-548.
Chappell, J., Hahlbrock, K., and Boller, T. (1984). Rapid induction of ethylene
biosynthesis in cultured parsley cells by fungal elicitor and its relationship to
the induction of phenylalanine ammonia-lyase. PIanta I61:475-480.
Cheong, J. J., and Hahn, M. G. (1991).A specific high-affinity binding site for the
hepta-@-glucoside elicitor exists in soybean membranes. Plant Cell3: 137-148.
Cheong, J. J., Birberg,W.,Fugedi,P., Pilotti, A., Garegg, P. J., Hong, N.,
Ogawa, T., and Hahn, M.G. (1991). Structure-activity relationships of oligo-
@-glucoside elicitors of phytoalexin accumulation in soybean. Plant Cell3: 127-
136.
Cosio, E. G., Frey,T., and Ebel, J. (1992). Identification of a high-affinity binding
protein for a hepta-@-glucoside phytoalexin elicitor in soybean. Eur. J. Bio-
chem. 2041115-1123.
Cuypers, B., Schmelzer, E., and Hahlbrock, K. (1988). In situ localization of rap-
112 Graham
idlyaccumulatedphenylalanineammonia-lyase MRNA around penetration

sites of Phytophthora infestansin potato leaves. Mol. Plant Microbe Interact.
1:157-160.
Darvill, A. G., and Albersheim, P. (1984). Phytoalexins and their elicitors-a de-
fense against microbial infection in plants. Ann. Rev. Plant Physiol. 35243-
275.
Davis, K. R., Darvill, A. G., and Albersheim, P. (1986). Several biotic and abiotic
elicitors act synergistically in the induction of phytoalexin accumulation in
soybean. Plant Mol. Biol. 623-32.
Doares, S. H., Bucheli,P.,Albersheim, P., and Darvill, A. G.(1989). Host-
pathogen interactions. XXXIV. A heat labile activity secreted by a fungal
phytopathogen releases fragments ofplant cell walls that kill plant cells. Mol.
Plant Microbe Interact. 2:346-353.
Ebel, J. (1986). Phytoalexins synthesis: the biochemical analysis of the induction
process. Ann. Rev. Phytopathol. 24235-264.
Ecker, J. R., and Davis, R. W. (1987). Plant defense genes are regulated by ethyl-
ene. Proc. Natl. Acad. Sci. 845202-5206.
Esnault, R., Chibbar, R. N., Lee, D., Van Huystee, R. B., and Ward, E. W. B.
(1987). Early differences in production of mRNA’s for phenylalanine ammo-
nia-lyase and chalcone synthetasein resistant and susceptible cultivarsof soy-
bean inoculated withPhytophthora megasperma f. sp. glycinea. Physiol.Mol.
Plant Pathol.30293-297.
Farmer, E. E., and Ryan, C. A. (1992). Octadecanoid precursors of jasmonic acid
activate the synthesis of wound-inducible proteinase inhibitors. Plant Cell 4
129-134.
Frank, J. A., and Paxton, J. D. (1971). An inducer of soybean phytoalexinand its
role inthe resistance of soybeanto Phytophthora rot. Phytopathology 61:954-
958.
Graham, T. L. (1991a). A rapid, high resolution HPLC profiling procedure for
plant and microbial aromatic secondary metabolites. Plant Physiol. 9.5584-
593.
Graham, T. L. (1991b). Flavonoid and isoflavonoid distribution in developing soy-
bean seedling tissuesand in seed and root exudates. Plant Physiol.9.5594-603.
Graham, M. Y.,and Graham, T. L. (1991a). Rapid accumulationof anionic peroxi-
dases and phenolic polymers in soybean cotyledon tissues followingtreatment
with Phytophthora megasperma f. sp. glycinea wall glucan.Plant Physiol.97:
1445-1455.
Graham, T. L., and Graham, M. Y. (1991b). Glyceollin elicitorsinduce major but
distinctly different shifts in isoflavonoid metabolism in proximal and distal
soybean cell populations. Mol. Plant Microbe Interact. 460-68.
Graham, T. L., and Graham, M. Y. (1991~).Cellular coordination of molecular
responses inplant defense. Mol. Plant Microbe Interact.4:415-421.
Graham, T.L., Kim, J. E., and Graham, M. Y.(1990). Role of constitutiveisofla-
vone conjugates in the accumulation of glyceollin in soybean infected with
Phytophthora megasperma.Mol. Plant Microbe Interact.3:157-166.
Habereder, H., Schroder, G., and Ebel, J. (1989). Rapid induction of phenylalanine
Phenylpropanoid
ammonia-lyase and chalcone synthase mRNAsduring fungus infection of soy-

bean (Glycinemay L.) roots or elicitor treatment of soybeancell cultures at the
onset of phytoalexin synthesis.PIanta 17258-65.
Hadwiger, L. A., Jafri, A., von Broembsen,S., and Eddy, R. (1974). Plant Physiol.
53~52-55.
Hahlbrock, K., and Scheel, D. (1989). Physiology and Molecular Biology of Phenyl-
propanoid Metabolism. Ann. Rev. Plant Physiol.40347-369.
Hahn, M. G., Bonhoff, A., and Grisebach, H. (1985). Quantitative localization of
the phytoalexin glyceollin I in relation to fungal hyphae in soybean roots in-
fected with Phytophthora megasperma f. sp. glycinea. Plant Physiol. 72591-
601.
Ingham, J. L., Keen, N. T., Mulheirn, L. J., and Lyne, R. L. (1981). Inducibly-
formed isoflavonoids from leaves of soybean.Phytochemistry 20795-798.
Keen, N. T., and Yoshikawa, M. (1982). Physiology of disease and the nature of
resistance to Phytophthora. Phytophthora: Its Biology, Taxonomy, Ecology
and Pathology (D. C. Erwin, S. Bartnicki-Garcia, and P. H. Tsao, eds.),
American Phytopathological Society, St. Paul, MN, pp. 279-298.
Keen, N. T., and Yoshikawa, M.(1983). 8-1,3-Endoglucanase from soybean re-
leases elicitor active fragments from fungus cell walls. Plant Physiol. 71:460-
465.
of carbohydrates from Phytophthora megasperma f. sp. glycinea and other
sources. Plant Physiol.71:466-471.
Lambert, M. R., and Graham, T. L. (1987). Alternative induced resistance path-
ways in soybeanand their regulation. Phytopathology 721739.
Lazarovits, G., Stoessel, R., and Ward, E. W. B. (1981). Age-related changes in
specificity and glyceollin production in the hypocotyl reactions of soybean to
Phytophthora megasperma var. sojae. Phytopathology71:94-97.
Leube, J., and Grisebach, H. (1983). Further studies on induction of enzymes of
phytoalexin synthesis in soybean and cultured soybean cells. Z. Naturforsch.
38~730-735.
Lundry, D. R., Bass, J., Castanho, B., and Graham, T. L. (1981). Protection of
soybean plants against diseaseby phytoalexin elicitors.Plant Physiol.6275.
Lyne, R. L., and Mulheirn, L. J. (1978). Minor pterocarpinoids of soybean. Tetra-
hedron Lett. 34:3127-3128.
Lyne, R. L., Mulheirn, L. J., and Leworthy, D. P. (1976).New pterocarpinoid
phytoalexins of soybean.J. Chem. Soc. Chem.Commun.:497-498.
Mauch, F., and Staehelin, L. A. (1989). Functional implications of the subcellular
localization of ethylene-induced chitinase
and beta-l ,3-glucanase in bean leaves.
Plant Cell. 1~447-457.
Mauch, F., Hadwiger, L. A., and Boller, T. (1984). Ethylene: symptom, not signal
for the induction of chitinaseand beta-l,3-glucanase in peapods by pathogens
and elicitors. Plant Physiol.76507-61 1.
Miller, C. 0. (1969). Control of deoxyisoflavone synthesis in soybean tissue. PIanta
8226-35.
Moesta, P., and Grisebach, H. (1981). Investigation ofthe mechanism of glyceollin
114 Graham
accumulation in soybean infectedby Phytophthora megqpermaf. sp. glycinea.

Arch. Biochem. Biophys. 212:462-467.
Morris, P. F., and Ward, E. W. B. (1992). Chemoattraction of zoospores of the
soybean pkhogen Phytophthora sojae, by isoflavones. Physiol. Mol. Plant
Pathol. 40: 17-22.
Moms, P. F., Savard, M. E., and Ward, E. W. B. (1991). Identification and
accumulation of isoflavonoidsand isoflavone glucosides in soybean leaves and
hypocotyls in resistance responsesto Phytophthora megasperma f. sp. glycinea.
Physiol. Mol. Plant Pathol. 39229-244.
Nothnagel, E. A., McNeil, M., Albersheim,P., and Dell, A. (1983). Host-pathogen
interactions. XXII. A galacturonic acid oligosaccharide from plant cell walls
elicits phytoalexins.Plant Physiol. 71:916-926.
Olah, A. F., and Sherwood, R. T. (1973). Glycosidase activity and flavonoid accu-
mulation in alfalfa infected by Ascochyta imperfecta.Phytopathology 63:739-
742.
Paradies, I., Konze, J. R., Elstner, E. F., and Paxton, J. (1980). Plant Physiol. 66
1106.
Parker, J. E., Hahlbrock, K., and Scheel, D.(1988). Different cell wall components
from Phytophthora megasperma f. sp. glycinea elicit phytoalexin production in
soybean and parsley. Planta 17675-82.
Partridge, J. E., and Keen, N. T. (1977). Soybean phytoalexins: rates of synthesis
are not regulated by activation of initial enzymes in flavonoid biosynthesis.
Phytopathology 6750-55.
Paxton, J. D., and Chamberlain, D. W. (1969). Phytoalexin productionand disease
resistance in soybeansas affected by age. Phytopathology59:775-777.
Pierce, M., and Essenberg, M. (1987). Localization of phytoalexins in fluorescent
mesophyll cells isolated from bacterial blight-infected cotton cotyledons and
separated from other cells by fluorescence-activated cell sorting. Physiol. Mol.
Plant Pathol. 31:273-290.
Pierson, P. E., and Graham,T. L. (1987). A cellular systemfor biochemical studies
on soybean-Phytophthorainteractions. Phytopathology 7Z1756.
Reinhardt, D., Wiemken, A., and Boller, T. (1991). Induction of ethylene biosyn-
thesis in compatible and incompatible interactions of soybeanroots with Phy-
tophthora megasperma f. sp. glycineaand its relationto phytoalexin accumula-
tion. J. Plant Physiol. 138394-399.
Rivera-Vargas, L. I., Schmitthenner, A. F., and Graham, T. L. (1993). Flavonoid
effects on andmetabolism byPhytophthorasojae. Phytochemistry 32:851-857.
Roby, D., Toppan, A., and Esquerre-Tugaye, M. T.(1986). Cell surfaces in plant-
micro-organism interactions. Plant Physiol. 81:228.
Rowlan,A.R., Hall, J. A.,Barfield-Schneider, T., and Essenberg,M. (1991).
Protection of cotton leaf palisade cellsfrom light-activated toxicity of a phyto-
alexin by red epidermal cells.Phytopathology81:1139.
Rumeau, D., Mazau, D., Panabieres,F., Delseny, M., and Esquerre-Tugaye, M. T.
(1988). Accumulation of hydroxyproline-rich glycoprotein mRNAs in infected
or ethylene treated melon plants. Physiol.Mol. Plant Pathol. 33A19-428.
Phenylpropanoid
Responses of Soybean to P. soiae 115
Ryan, C. A.(1981). Proteinase inhibitors. TheBiochemistry of Plants, Vol.6

(p.K. Stumpf and E. E. Conn, eds.), Academic Press, New York, pp. 351-
370.
Schmelzer, E., Borner, H., Grisebach, H., Ebel, J., and Hahlbrock, K. (1984).
Phytoalexin synthesisin soybean (Glycinem a ) . Similar time courses of mRNA
induction in hypocotyls infected with a fungal pathogen and in cell cultures
treated with fungal elicitor. FEBS Let. 17259-63.
Schmelzer, E., Kruger-Lebus, S., and Hahlbrock, K. (1989). Temporal and spatial
patterns of gene expressionaround sites of attempted fungal infection in pars-
ley leaves.Plant Cell1:993-1001.
Schmitthenner, A. F. (1985). Problems and progress in control of Phytophthora
root rot of soybean.Plant Dis. 69362-368.
Sharp, J. K., McNeil, M., and Albersheim, P. (1984). The primary structures of one
elicitor-active and seven elicitor-inactive hexa@-D-glucopyranosyl)-D-glucitols
isolated from the mycelial walls of Phytophthora megasperma f. sp. glycinea.
J. Biol. Chem. 25911321-11336.
Sims, J. J., Keen,N. T., and Honwad, V. K. (1972). Hydroxyphaseolin, and
induced antifungal compound from soybeans. Phytochemistry 11:827-828.
Sinclair, J. B. (1982). Compendium of Soybean Diseases,American Phytopatholog-
ical Society, St. Paul, MN.
Snyder, B. A., and Nicholson, R. L. (1990). Synthesis of phytoalexins in sorghum
as a site-specific response to fungal ingress. Science 2481637-1639.
Stermer, B. A., Schmid, J., Lamb, C. J., and Dixon, R. A. (1990). Infection and
stress activation of bean chalcone synthase promoters in transgenic tobacco.
Mol. Plant MicrobeZnteract.3:381-388.
Takeuchi, Y., Yoshikawa, M., and Horino, 0. (199Oa). Immunological evidence
that /3-1,3-endoglucanase isthe major elicitor-releasingfactor in soybean. Ann.
Phytopathol. Soc. Jpn.56523-53 1.
Takeuchi, Y., Yoshikawa, M., Takeba, G., Tanaka, K., Shibata, D., and Horino,.
0. (1990b). Molecular cloning and ethylene induction of mRNA encoding a
phytoalexin elicitor-releasing factor, /3-1,3-endoglucanase, in soybean. Plant
Physiol. 93:673-682.
Teasdale, J. R., and Hadwiger, L. A. (1977). Effect of pisatin-inducing fungi on
pea polyamines.Phytochemistry 16:681-683.
Vera, P., and Conejero, V. (1989). The induction and accumulation of the patho-
genesis-related P69proteinase in tomato during citrus exocortis viroid infection
and in response to chemical treatments. Physiol. Mol. Plant Pathol. 34323-
334.
Wang, M. C., and Bartnicki-Garcia, S. (1973). Novel phosphoglucans from the
cytoplasm ofPhytophthorapalmivora and their selective occurrencein certain
life cycle stages. J. Biol. Chem.2484112-41 18.
Wang, M.C., and Bartnicki-Garcia, S. (1974). Mycolaminarans: storage 0-1,3-
glucans from the cytoplasm of the fungus Phytophthora palmivora. Carbohy-
drate Res.32331-338.
Ward, E. W. B., and Buzzell,R. I. (1983). Influence of light, temperature and
116 Graham
wounding on the expression of soybean genes for resistance to Phytophthora

megasperma f. sp. glycinea. Physiol.Mol. Plant Pathol.23:401-409.
Ward, E. W. B., and Lazarovits, G. (1982). Temperature induced changesin speci-
ficity in the interaction of soybeans with Phytophthora megasperma f. sp.
glycinea. Phytopathology72:826-830.
Yoshikawa, M. (1988). Molecular mechanisms for induction of host defenses in
fungal diseases. Molecular Strategiesfor Pathogenicity and Host Defense in
Viral, Bacterial and Fungal Diseases(R.Heitefuss and S. Ouchi, eds.), Satellite
Meeting of the 5th International Congress for Plant Protection, August 22-23,
Kyoto, pp. 3-7.
Yoshikawa, M., Yamamuchi, K., and Masago, H. (1978). Glyceollin: its role in
tophthora megasperma var. sojae. Physiol. Plant Pathol. 12:73-82.
Zahringer, U.,Ebel, J., and Grisebach, H.(1978). Induction of phytoalexin synthe-
sis in soybean: elicitor-induced increasein enzyme activities of flavonoid bio-
synthesis and incorporation of mevalonate into glyceollin. Arch. Biochem.
Biophys. 188:450-455.
6
Culture Darkening, Cell AggregateSize,
and Phytoalexin Accumulation in
Soybean Cell Suspensions Challenged
with Biotic Agents
Robert M. Zacharius
R. M. Zacharius and Associates, Science Consultants, Highland, Maryland
William F. Fett
Eastern Regional Research Center, Agricultural Research Service,
US.Department of Agriculture, Philadelphia, Pennsylvania
Prakash C. Kadkade
Phyton Inc., Zthaca, New York
1. INTRODUCTION
Fett and Zacharius (1982, 1983) demonstrated that bacteria as well as fun-
gal cell wall elicitors can induce the accumulation of the phytoalexin gly-
ceollin in soybean (Glycine max L. Merr.) cell suspension cultures. More-
over, the typical hypersensitive response(HR) as defined by rapid host cell
death displayed by the intact plant was not a prerequisite for phytoalexin
induction in soybean cell cultures. The concentrationof glyceollin produced
declined with successive culture transfers. Some interesting responses by
separate cell lines ofthe same cultivarto challenge with a biotic agentwere
observed, suggestingfurther study.
A subsequent study by Zacharius and Kalan (1990) revealed that soy-
bean cell suspension cultures producing glyceollin when challenged with
Pseudomonas syringae pv. gZycinea (Psg) or fungal cell wall elicitor did not
undergo an HR but rather darkened with a gradual decline in culture viabil-
ity. A cell line, Sb-l (CV. Mandarin), which failed to darken on challenge
with Psg, also didnot accumulate glyceollin withinthe cells or media. This
Sb-l culture was found to have rather low levels ofconstituent isoflavonoid
which exhibited a small declineon exposure to biotic agentsand produced
only trace amountsofglyceollin.The other celllinesof CV. Mandarin
having high levels of constituent isoflavonoids, exhibited a dramatic decline
117
118 Zacharius et al.
in their isoflavonoids along with accumulation of glyceollin following expo-

sure to biotic agents.
High levelsof constitutive isoflavonoids, particularly daidzein,seemed
indicative of culture potentialfor glyceollin production concomitant witha
decrease in daidzein, genistein, and coumestrol when either fungal wall
elicitors or live bacteria was the stressingagent.However,cellcultures
containing very high levels of these isoflavonoids didnot accumulate higher
levels of the phytoalexin than those with lesser levels. Augmentation theof
weaklyresponsive Sb-l culturewithexogenouslysuppliedisoflavonoids
followed by fungal elicitor challenge had little measurable effecton glyceol-
lin production, but resulted in a metabolic breakdown of the exogenously
supplied daidzeinand genistein.
Thus, on the basis of this study, there appeared to be a link between
endogenous levels of cell isoflavonoids, ability for culture darkening, and
glyceollin production. From other observations withCV. Mandarin cell sus-
pensions, we noted the larger cell aggregates tended to darken more readily
with exposureto fungal wall elicitor than the smaller aggregates. Therefore,
our earlier studies were extended to include both smaller and larger cell
suspension aggregations of soybean CV. Mandarin and CV. Clark challenged
by live bacteria and fungal elicitor.We speculated that large cell aggregates
may have a closer relationship to in vivo tissues of an organized structure
than single cellsor small cell aggregates.
II. METHODOLOGY
A. Suspension Cell Cultures
Calluses of soybean CV. Mandarin cell line Sb-4a were initiated from epi-
cotyl tissue of 7- to 10-day old plants by the procedure described by Fett
and Zacharius (1982). Calluses of soybean CV. Clark were also initiated
from epicotyl explants on B5 media (Gamborg, 1975) containing 10 mg/
liter 2,4-D and 0.21 mg/liter kinetin. A suspension cell culture was devel-
oped from friable calluses of each cultivar and grown in 1B5 media on a
*
reciprocating shakerat 150 rpmat 26-27OC under 17.5 4.5 pmol/m2/sec
photosynthetic photon flux (PPF).
B. Bacterial Cultures
Xanthomonas campestrispv. glycines strains XP175and S-9-8 were grown
overnight on nutrient agar at 28OC, suspended in sterile distilled water,
washed three times, and resuspended in sterile distilled waterto 1.0 OD at
600 nm. Strain XP175 is virulentand causes bacterial pustule disease on CV.
Mandarin and CV. Clark while strain S-9-8 isavirulent (Fett, 1984).
Culture Darkening 119
C. Fungal Elicitor
A cell-free mycelial elicitor
was prepared from Phytophthora infestans race
0 by the procedure of Alves et al. (1979)with an added final filtration
step through a 0.45-pm Millipore filter before autoclaving. The elicitor
preparation contained 6.4mg dry wt per milliliter distilled water.
D. Interaction of Suspension Cells with Bacteria or Fungal Elicitors

Suspension cultures of each cultivar of soybean were and grown
maintained
in 60-m1 volumes of 1B5 medium in 250-m1 Delong flasks and were used 5
days after the last transfer. The contents of several flasks of Sb-4A or
Sb-Clark were each aseptically sievedon 100-mesh wire screen to separate
large, > 150-pm aggregatesand small, c 150-pm aggregates. Aggregates of
one size and culture were pooled and 10 m1 of loosely packed aggregates
was redistributed into flasks with40 ml of fresh 1B5 media. Fungal elicitor
was applied to each flask at 0.75 d / 5 0 ml of aggregated cell culture and
each bacterialstrain was added to give an initial concentrationof approxi-
mately 1 X lo7CFU/ml. Each aggregate size/cultivar/biotic-elicitor inter-
action and controls were carried out with three replicates. All flasks were
shaken at 150 rpm at 26-27OC under 17.5 f 4.5 pmol/m2/sec PPF;
Cell viability was followed by a dye exclusion test with 0.4% trypan
blue (Phillips,1973).
E. lsoflavonoid Extraction, Identification, and Quantitation

Following 68 hr of interaction, cells plus media were extracted by vortex
mixing four times, each with an equal volume of chloroform (Zacharius
and Kalan, 1990). The combined extracts were dried at ambient tempera-
ture under,a stream of nitrogen and taken up in 1.0 m1 methanol 0.5
g" dry weight. Components of the extract were separated qualitatively by
thin-layer chromatography (TLC)on Analtech silica gel G plates (250 pm)
irrigated with cyclohexane-ethyl acetate(1:l) (Zacharius and Kalan, 1984).
Quantitation was performed by high-performance liquid chromatography
(HPLC) using a 250-mm CI8reverse phase column (Whatman Partisil 10
ODS-325) attached to a Waters Model 6000A solvent delivery system with a
Reodyne (70-10) loop injector valve. The column effluent was monitored
with a Perkin-Elmer Model LC-55B spectrophotometric detectorat 262 nm
(genistein and daidzein), 343nm (coumestrol), and 290 nm (glyceollins).
Integrations were made with the Hewlett-Packard 3390A integrator and
compared with thoseof reference compounds (Zachariusand Kalan, 1990).
The column eluantwas 40% aqueous acetonitrile containingtrifluoro-
acetic acid (0.4 ml/liter) at a flow rate of 1.5 ml/min. Retention times
(min)wereasfollows:daidzein, 2.2; genistein, 3.3; formononetin, 4.5;

coumestrol, 6.0; glyceollin isomers, 7.8, 8.0; biochanin A, 9.5. The pres-
ence of glyceollin, daidzein, genistein, and coumestrol in selected samples
was further confirmed by comparison with authentic compounds by mass
spectrometry usinga Finnegan MAT311A mass spectrometer.
F. PeroxidaseAssay
The peroxidase assay was carried
out according to the method of Worthing-
ton (1972).
111. RESULTS
The large cell aggregates of both Sb-Clark and Sb4A dramatically dark-
ened when challenged with either virulent strain XP175 or avirulent strain
S-9-8 or the fungal elicitor. On the other hand, these challenges with the
smallaggregateculturesofeithersoybeancultivarproducednovisual
change in coloror color intensity (Fig.1).
Figure 1 Culture darkening 68 hr after inoculation of small or large cell aggre-

gates of Sb-Clark withXanthomonas campestrispv. glycines. Mixed aggregate cul-
ture control, A; virulent X . c. pv. glycines strain XP175 in small (B) or large (B')
aggregate cultures; avirulentX.c. pv. glycines strain S-9-8 in small (C)or large (C')
aggregate cultures; fungal elicitor in small(D) or large (D') aggregate cultures.
Culture 121
Both size cell aggregates of Sb-Clark responded to each of the stressing

agents producing similar levels of glyceollin with a marked but unequal
decrease inthe constitutive level of daidzeinand genistein. Avirulentstrain
S-9-8 effected a virtual disappearance of these two isoflavonoidsfrom the
cell culture (Table 1).
Both large and small -aggregates ofSb4A when exposed to avirulent
strain S-9-8 produced similarlevels of glyceollin, whereas the virulent strain
XP175 failed to do so with either size cell aggregates. Althoughthe isofla-
vonoids declined in both size aggregates exposed to strain XP175, S-9-8
caused an almost total loss of daidzein and genistein (Table1) as inthe case
of Sb-Clark.
Interaction of the large cell aggregatesof either soybean cultivar with
the fungal elicitor produced marked darkening, glyceollin accumulation,
and a concurrent large decrease in daidzein and genistein. While the latter
two phenomena occurred during the interaction with the small aggregates
of both cultivars, neither small aggregate cellculture darkened (Fig.1).
Both aggregate sizes of Sb4A and the large aggregate culture of Sb-
Clark became very viscousand gel-like when exposed to strain S-9-8. This
was presumably due to copious productionof bacterial exopolysaccharides
accompanied by a concurrent increase in CFU per milliliter to 10". This
was not observed in the other interactions. Where the challenged cultures
became gel-like, a sharp decline in soybean cell viability occurred 68 by hr.
Table 1 Changes in the Isoflavonoids of LargeandSmallCellAggregates of

Sb-Clark and Sb-4A(CV.Mandarin) upon Biotic Stressing &g/g dry wt)
Glyceollin Genistein Daidzein

Treatment
Sb-Clark
Sb-Clark
Sb-4A
Sb-Clark
Sb-4A
Sb-4A
Small aggregates
2560 Control 0 0
576 XP175'Strain 0
S-9-8b Strain 50 73 38 908 1395
P. infestans elicitor 1201 975
880
870
1180
236
Large aggregates
3240 Control 0 0
812 Strain
3585 xP175"
515 1837 742 0
S-9-8b Strain 867 40
107 77 1140
P. infmamelicitor 1441 385 1 1 01100
9 912 980
'Strain m175 = X.campestdspv. glycines XP175.
bStrainS-9-8 = X . campestrispv. glycines S-9-8.
There was little change inthe soybean cell viability inthe other interactions
(Table 2).
Following exposureto the stressing agent strain XP175,the larger cell
aggregates ofboth soybean cultivarswere found to have higher peroxidase
levels than the small aggregates. The changes from the prestressed levels
were of similar magnitude for both cell cultivars (Table3).
Attachment of the bacterial strainsto soybean cells of either aggregate
size of Sb-Clark or Sb-4A cultures was not observed using phase contrast
light microscopy and bacterial-induced clumping of cells of any of the
cultures did not occur.
W. DISCUSSION
In our earlier report (Zacharius and Kalan, 1990), data werepresented
which appeared to relate soybean suspension culture darkening with gly-
ceollin induction during bacterial or fungal elicitor stress. A high level of
cellular daidzein also seemed prerequisiteto both phenomena. Concurrent
with glyceollin induction, a large decline of the cellular daidzein and gen-
istein usually occurred. In the present study,the small aggregates were able
Table 2 Cell Viability of Sb-Clark and Sb-4A (CV.Mandarin) Cultures and

Bacterial Growth after68 hr Interaction
Cell 070 Viable

suspension sue Treatment CFWml cellssoybean
mall Clark 0 89
Control Large 0 92
Small XP175"
2.20 x 10' 89
Large XP175 1.12 x 109 85
Small 2.23
S-9-8" X 10' 83
Large 1.13
S-9-8 x 10'' 0
elicitorFungal Small 0 88
Fungal Large elicitor 0 84
Sb4A
Control Small 0 85
Control Large 0 88
Small XP175 8.60 x lo7 85
Large XP175 6.47
87 X 10'
Small 1.17
S-9-8 X I
O' 45
Large S-9-8 1.40 x 10" 15
elicitorFungal Small 0 85
Large elicitor
, Fungal 0 83
'Xanthomonas campestris pv. glycines.
Culture 123
Table 3 Changes in Peroxidase Levels in Cell Aggregates of Sb-Clark and Sb4A

(CV. Mandarin) on Challenge with Xanthomonascampestris pv. gIycines strain
XP175
Absorbance change at
460 nm/mg protein
after Suspension Aggregate
Soybean cultivar size (pm) Suspension m175
Sb-Clark < 150 0.60 0.85

> 150 0.70 2.70
Whole unsieved 0.68 2.50
culture
Sb4A < 150 0.65 0.85
> 150 0.75 2.70
Whole unsieved 0.70 2.50
culture
to accumulate glyceollin without observable darkening. When glyceollin

accumulated in either size aggregates, the levels were virtually the same
irrespective of whether the aggregates were darkened or not. In addition,
in the case of Sb-4A exposed to virulent strain XP175, neither the small
undarkened cell aggregates northe large darkened cell aggregates accumu-
lated glyceollin althoughthe constituent isoflavonoidsof the culture under-
went a decline. This would suggest that metabolic consumption of these
constitutive isoflavonoidsneed not be linked to glyceollin induction.
The propensity of the soybean culture to darken with biotic stressing
was reflected in the size of the cell aggregates of the culture. Large aggre-
gates darkened on exposure to any of the three stressing agents while the
small ones did not. Earlier Zacharius and Kalan (1990)had reported that
an unsieved CV. Mandarin culture of mixed size aggregates did not show
observable darkeningon stressing withfungal elicitor but the cultures hada
low level ofconstitutive daidzeinand genistein.
Cell darkening of suspension cultures with or without biotic stressing
would appear to reflect the level of phenolic compounds present the in cells
and medium. Interestingly, small cell aggregates used in this study had
lower constitutive isoflavonoid levelsthan the larger cell aggregates.Siege1
and Enns (1979)were able to prevent the discoloration and aggregation of
soybean suspension cultures grown in B5 medium with 2,4-D by adsorbing
the excess cellular polyphenols with either polyvinvlpyrrolidine or bovine
serum albumin inthe medium. Additionally, Singh et al. (1982)also associ-
ated higher levels of polyphenols in cowpea callus (Vigna unguiculata L.
Walp. subsp. unguiculata) with higher activities of peroxidase and poly-

phenol oxidase. Moreover, polyphenol content increased with 2,4D con-
centration above 1 pg/ml while supplements of casein hydrolysateand co-
conut water produced the lowest polyphenol accumulation. In our study,
soybean cultureswere grown with 1 pg/ml of 2,4-D without casein hydroly-
sate and coconut water. Although we did not compare polyphenol levels,
increased levels of peroxidase were associated with those cell aggregates
which darkened markedly withaddition of strain XP175.
Peroxidase levels werefound by Verma and Van Huystee (1970) to be
2.5-fold greater in large peanut cell aggregates (2-4 mm) than in small ones
(150 pm). It was also found that a cell mass less than 0.5 mm in diameter
consists of undifferentiated uniform cells while cellular differentiationap-
pears in large cell aggregates. Working with tobacco suspension cultures,
Kuboi and Yamada (1976, 1978) found stimulation of the lignin biosyn-
thetic pathway, during tracheid differentiation, occurs in large aggregates
but low activities of shikimate dehydrogenase, cinnamic acid-4-hydroxy-
lase, 5-hydroxyferulicacid-0-methyltransferase, and caffeic acid-0-methyl-
transferase inhibit differentiation in small aggregates. Both phenylalanine
ammonia-lyase (PAL) and peroxidase exhibited activities in the small aggre-
gates. Hahlbrock et al. (1974) found that the highest specific activity of
PAL is associated with single cellsand small aggregates, while the specific
activities in large aggregates were considerably lower. This would explain
finding greater secondary compound production in cultures of small aggre-
gates and viable single cells. Kinnersley and Dougall (1980) succeeded in
increasing the anthocyanin yieldin Daucus carota L.cell cultures by
screening for small cell aggregates and subculturing; a consequence of this
selection was small aggregates with a lower level of endogeneous cytokinin
(whichreduceanthocyaninyield).Wattsetal.(1984)foundevidence
to suggest that the presenceofgreen,aggregatedcells or low-tempera-
ture stress contributesto the ability of celery cell suspensionsto synthesize
secondary compounds. Nevertheless, the level of glyceollin accumulated
in the large and small aggregates of soybean described here were quite
similar.
Perhaps Apostol et al.(1989) provide inpart the most rational explana-
tion of our observations regardingthe relationship of aggregate size,dark-
ening, and glyceollin production. Theyfound that elicitor-stimulated plant
cell cultures responded with the rapid productionH202 of which was subse-
quently used by extracellular peroxidases. Exogenous H202 alone stimu-
lated phytoalexin production in the soybean cell suspension culture, and
inhibition of elicitor stimulated glyceollin production was observed upon
addition of catalase or other inhibitorsof the oxidative burst.For inhibition
to occur, the presence of catalase was necessary during elicitor addition.
Culture 125
Montillet and Degous6e (1991) havereportedmuchgreaterglyceollin-

eliciting activity in soybean seedlings by two organic hydroperoxidesthan
H202. H202eliciting efficiencywas comparable to the two organic hydro-
peroxides when tissue catalase activitywas suppressed. Of further interest,
Graham and Graham (1991) found that deposition of phenolic polymers in
soybean cotyledon cell walls is an early and major response to treatment
with fungal cell wall glucan. This is accompanied by a rapid and massive
increase in activity of a specific group of anionic wall-bound peroxidases.
Analysis of our observations in light of the above suggeststhat the large
aggregates of Sb-Clark and Sb-4A responded to biotic stress by severely
darkening and at the same time a burst of H20, induced (asindicated by
fourfold increase in peroxidase level)the glyceollin production. The small
aggregates ofboth cultivars didnot discolor because of a limited H20zburst
indicated by only a slight increase in peroxidase level, but these aggregates
presumably contained high activities of the early phytoalexin pathway en-
zymes, allowing for glyceollin production. The sharp decline in isoflavo-
noids in soybean suspension cultures treated with biotic elicitors (Zacharius
and Kalan, 1990) almost certainly can be explained by the oxidative burst
occurring during elicitation.Culture darkening and the level of destruction
of the measured isoflavonoids following elicitation probably reflects firstly
on the intensity of the oxidative burst and secondly on the induced levelof
the peroxidases present inthe culture.
Strain S-9-8 is avirulent and strain XP175 is virulent on both CV. Clark
and CV. Mandarin leaves. Fett (1984), however, found moderate glyceollin
accumulation in leaves of CV. Clark inoculated with virulent strain XP175
and a weak hypersensitive response with no glyceollin accumulation follow-
ing inoculation with avirulent strain S-9-8. Growth of strain S-9-8 was
restricted in leaves CV.Clark 24 hr after inoculation when compared to
growth of strain XP175. The leaf observations withCV. Clark were inconsis-
tent with those described here for the cell cultures. Leaves of Sb-4A (CV.
Mandarin) were not treated with eitherstrain.
The level of glyceollin which accumulated in the inoculated large and
small aggregate cultures did not appear to determinethe bacterial cell popu-
lations attained (Table 2). Strain S-9-8 grew to much higher levels than
strain XP175 in both the large and small aggregate Sb-4A cultures, while
strain S-9-8 but not strain XP175 induced accumulation of appreciable
levels of glyceollin. Both aggregate size cultures of Sb-Clark stressed with
either strain of X. campestris yielded similar amounts of glyceollin (Table
l ) , yet strain S-9-8 grew 200 times greater in the large cell aggregations
than
in the small aggregates. Strain S-9-8 attained higher populations than did
strain W 1 7 5 under all cell culture conditions except for the small aggregate
sized Sb-Clark. Evidence offered here indicates littleor no control by gly-
ceollin on the growth of X.campestris in the cell suspension cultures.The

percentage of viable soybean cells remainingafter 68 hr exposureto the X.
campestris agrees well withthe bacterial populationsattained.
Observations with phase microscopy didnot indicate any attachment
of strains S-9-8 or XP175 to Sb-Clark suspension cells of either aggregate
in of the findings of Jonesand Fett (1984) that
type. This is of interest view
strain S-9-8 is immobilized by electron-dense material in leaf intercellular
spaces of CV. Clark while strain XP175 is not.
V. EPILOG
The use of plant cell suspension cultures to study the interaction of plants
with biotic elicitors such as bacteria may lead to results that do not accu-
rately reflect responses of the intact plant. This is evidenced by the result
for glyceollininduction,bacterialgrowth, and bacterialimmobilization
presented here and in our earlier studiesof the interaction of soybean cells
with plant pathogenic bacteria (Fett,1984; Fett and Zacharius, 1982, 1983).
Careful considerations need to be given to the age of the cell suspension
culture, the culture medium, and cell aggregate size distribution. The fact
that plant cells in suspension cultures are continually bathed in large vol-
umes of liquid whereas leaf intercellular spaces (where leaf-spotting bacte-
ria reside and grow) are initially deficient in free water may preclude certain
important interactions such as prolonged bacterial cell-plant cell contact
from taking place. However, response of cell suspension culture to bacterial
inoculationcanaccuratelyreflectcertainin planta phenomenaduring
plant-bacterial interactionsas demonstrated by recent studies of Orlandi et
al. (1992).
Future research should be directedat determining optimal cell suspen-
sion culture conditionsand cell aggregate sizesto be used in order to more
closely mimic inplanta phenomena.
REFERENCES
Alves, L. M., Heisler, E. G., Kissinger, J. C., Patterson, J. M., and Kalan, E. B.
(1979). Effectsofcontrolledatmosphere on production of sesquiterpenoid
stress metabolites by white potato tuber. Possibleinvolvement of cyanide-
resistant respiration.Plant Physiol.63:359-362.
Apostol, I., Heinstein, P. F., and Low, P. S. (1989). Rapidstimulation of an
oxidative burst during elicitation of cultured plant cells. Plant Physiol.90:109-
116.
Fett, W. F. (1984). Accumulation of isoflavonoids and isoflavone glucosides after
inoculation of soybean leaves with Xanthomonas campestrispv. glycines and
Culture 127
24
pv. campestris and a study of their role in resistance. Physiol. Plant Pathol.
303-320.
Fett, W. F., and Zacharius, R. M. (1982). Bacterially induced glyceollin production
in soybeancell suspension cultures.Plant Sci.Lett. 24:303-309.
Fett, W. F., and Zacharius, R. M. (1983). Bacterial growth and phytoalexin elicita-
tion in soybeancell suspension culturesinoculated with Pseudomonas syringae
pathovars. Physiol. Plant Pathol.22151-172.
Gamborg, 0 . L. (1975). Callus and cell culture. Plant Tissue Culture Methods(0.
L. Gamborg and L. R. Wetter, eds.), National Research Council of Canada,
Saskatoon.
Graham, M. Y., and Graham, T. L. (1991). Rapid accumulation of anionic peroxi-
dases and phenolic polymersin soybean cotyledon tissues followingtreatment
with Phytophthora megasperma f. sp. glycinea wall glucan. Plant Physiol.97:
1445-1455.
Hahlbrock, K., Ebel, J., and Oaks, A. (1974). Determination of specific growth
stages of plant cell suspension cultures bymonitoring conductivity changes in
the medium. Planta 11875-84.
Jones, S. B., and Fett, W. F. (1985). Fate of Xanthomonas campestrisinfiltrated
into soybean leaves: an ultrastructural study. Phytopathology 45:733-741.
Kinnersley, A. M., and Dougall, D. K. (1980). Increase in anthocyanin yield from
wild carrot cell cultures by a selection system based on cell aggregate size.
Planta 149200-204.
Kuboi, T., and Yamada, Y. (1976). Caffeic acid-0-methyl transferase in a suspen-
sion of cell aggregates of tobacco.Phytochemistry 15:397-400.
Kuboi, T., and Yamada, Y. (1978). Regulation of the enzyme activities relatedto
lignin synthesis in cell aggregates of tobacco cell culture. Biochim. Biophy.
Acta542:181-190.
Montillet, J. L.,and Degouske, N. (1991). Hydroperoxydes induce glyceollin accu-
mulation in soybean.Plant Physiol. Biochem. 29689-694.
Orlandi, E. W., Hutcheson, S. W., and Baker, C. J. (1992). Early physiological
responses associated with race-specific recognition in soybean leaf tissue and
cell suspension treated with Pseudomonas syringaepv. glycinea. Physiol.Mol.
Plant Pathol.40173-180.
Phillips, J. H. (1973). Dye exclusion testsfor cell viability. Tissue Culture Methods
andApplications (F. F. Kruse, Jr., and M. K. Patterson, Jr., eds.), Academic
Siegel, N. R., and Enns, R.K. (1979). Soluble polyvinvlpyrrolidine and bovine
serum albumin adsorb polyphenols from soybean suspension cultures. Plant
Physiol. 63:206-208.
Singh, B. D., Rao, G. S. R. L., and Singh, R. P. (1982). Polyphenol accumulation
in callus culturesof cowpea (Vigna sinensis). Indian J. Exp. Biol.20387-389.
Verma, D.P. S., and van Huystee,R. B. (1970). Cellular differentiation and peroxi-
dase isozymes in cellculture of peanut cotyledons. Can. J. Bot. 48429-43 1.
Watts, M. J., Galpin, I. J., and Collin, H. A. (1984). The effect of growth regula-
tors, light and temperature on flavour production in celery tissue cultures. N.
Phytol. 98583-591.
WorthingtonBiochemicalCorporation(1972). EnzymesandEnzymeReagents,
Freehold, NJ, pp. 43-44.
Zacharius, R. M., and Kalan, E. B. (1984). Biotransformation of the potato stress
metabolite, solavetivone, by cell suspension cultures of two solanaceous and
three non-solanaceous species.Plant Cell Rep. 3~189-192.
Zacharius, R. M., and Kalan, E. B. (1990). Isoflavonoid changes in soybean cell
suspensions when challenged with intact bacteria or fungal elicitors. J. Plant
Physiol. 135~732-736.
l
Phytoalexins as a Factor in
Wilt Resistance of Cotton
M. H. Avazkhodjaev, S.S. Zeltser, and H. V. Nuritdinova
Institute of ExperimentalBiolofl of Plants, Academyof Sciences
of Uzbekistan, Tashkent, Uzbekistan
Raviprakash C. Dani
for Cotton Research, Nagpur, Maharashtra, India
Central Institute
1. INTRODUCTION
The Verticillium wilt of cotton (Gossypiurn spp.) always remaineda serious
problem in the cultivation of this crop. According to contemporary scien-
tific thinking, the plant cell, likethe animal cell, possessesan immunologi-
cal control system, the function ofwhichinvolves not only to defend
against pathogenic microorganisms,but also to support the structural and
functional integrity of the body. Disease resistance is significant in this
context, for it is the result of interaction betweenthe host plant genotype,
the pathogen, and the surrounding environmental factors.
Among the principal causesof increase in susceptibilityto wilt disease
of cotton, the foremost are crop monoculture and planting of identical
cultivars. Second comesthe public’s compulsive use of pesticides, different
chemicals, and fertilizers. Third, there is the annual carryover of infected
plant partsin the soil.
It is well knownthat in wild pathosystems a more or less rigid selection
operates, thus upholding the balance of nature, whereas under cultural
conditions two genotypesoperate, i.e., that of the host plant and that of
the parasite, which interact through metabolic changes within these two
bodies. In contrast to animals, the plant body system isnot endowed with
highly specific antigen-antibody reactions. Immunity in plants is brought
about by a whole series of less specific defense reactions. In the past 10-15
years, thanks to developments in molecular biology, particularly molecular
genetics, it has been demonstrated that immunity involvesnot only a bodily
reaction but a complex of several reactions, directed support
in of its func-
tional integrity, i.e., homeostasis. Considerable biochemicaland molecular
129
130 Avazkhodjaev et al.
studies are being directed to disease resistance in cotton because of the

unique tools of biotechnologythat are now available for plant genetic im-
provement, involvingthe aspects of biochemical regulation of phytoalexin
expression (Stewart, 1991) and selection at the cellular level for resistance
to physiological stress(Dani, 1992), and so forth.
In the study of the physiological and biochemical processes of the
host-pathogen interaction in the Verticillium wilt, two distinct phases are
involved. The first phase consists of determination in which some very
exclusive reactions occur, with transmission of signals, directly to'the ge-
netic apparatus. The second phase is the expression, characterized by the
entire range of defense reactions.
It has been established through our long-term research that one of
the fundamental wilt defense reactions in cotton consists of the so-called
hypersensitivity reaction, which in itself embodiesan entire array of param-
eters, starting with some nonspecific reactions (e.g., strengthening of oxi-
dation recovery enzymes, formation of their isozymes, oxidation of poly-
phenols, formation of quinones) and culminatingin the formation of
phytoalexins.
Muller and Borger (1940) postulated a theory onthe existence of phyto-
alexins, which has met with experimental confirmation over the past two to
three decades. A wide range of phytoalexins have now been identified in
beans, potatoes, orchids, beetroot, and other crops includingcotton (Tomi-
yama, 1970; Cruickshank and Perrin, 1971; Keen et al., 1971; Metlitski et
al., 1976; Fraile et al., 1982; KuC and Rush, 1985; Bailey and Mansfield,
1985; Sun et al., 1989; Essenberg et al., 1990).
II. RESISTANCE REACTIONS O F COTTON

A. Phytoalexins
Studies such as those mentioned above provided impetus to further the
genetic and biochemical investigations of phytoalexins.The series of inves-
tigations by Bell (1967, 1969, 1981) made a significant contribution to the
understanding of phytoalexins in cotton. Many of his studies were con-
cerned with flowering and boll initiation phases. During infection, in cer-
tain plants fungitoxicity of xylem vessels and the leaves supporting bolls
often registered such an increase at flowering stage as to completely sup-
press the germination of fungal spores. It was shown that the fungitoxicity
of the resulting products was due to certain ether-soluble phenolic com-
pounds. Gossypol was primarily recognized. Besides gossypol,at least four
other gossypol-related compoundswere also isolated fromxylem vessels of
Wilt Resistance 131
infected cottons. The complex of gossypol-like compoundsakin to phyto-

alexins was termed “gossypol equivalents.”
In the courseof our own investigationson phytoalexins, attention was
focused primarilyon the method involving“drop diffusates,” the fungitox-
icity ofwhichwasused as an index for the activity of phytoalexins in
cotton tissues. This method permits the recovery of phytoalexins without
destruction of tissues (Metlitski et al., 1971; Mwamedova and Turakulov,
1974). Resultsof our researchconfirm that wiltresistancein cotton is
closely relatedto the ability of the plant to produce phytoalexins in response
to infection. A s can beseen from the data presentedinTable 1, high
phytoalexin activity is characteristically found in the wilt-resistant wild rela-
tive of cotton, namely, subsp.mexicanum, while the lowest activity is noted
in the wilt-susceptible G.hirsutum cvs. S 4727 and 1306 DB.
Thus a major reason for the high wilt resistance of the G.hirsutum cvs.
Express and the chemo-mutants L3 and L4 would be their comparatively
higher abilityto produce phytoalexins in response to infection by the fungus
V. dahliae Kleb., the causal organism of wilt. It was noted that the fungi-
toxicity of the diffusates increased with increasing density of suspended
fungal conidia,attaining a maximum value in the concentration oflo4-lo6
conidia/ml. However, with higher concentrations of the inoculum of the
pathogen, the phytoalexin activity in the diffusates tended to register a
sharp decline.
A s indicated above, phytoalexin activityof the leaf-derived diffusates
forms a clear criterionof wilt resistance of different cultivars cotton.
of To
understand the role of phytoalexins inthe incompatibility reactionsof cot-
Table 1 Fungitoxicity of Leaf Diffusates from Various Cotton

Cultivars
070 Suppression
Cultivar/form Hyphal Conidial
of cotton growthgermination
Tashkent-l
Tashkent-2
66.4 29.0
Tashkent-3
77.5 24.2
Chemomutant L3 75.9 29.7
L4 23.0 50.0
60.9
L-4727 24.0
1306 7.4 4.0
Express 71.8
132al. et Avazkhodjaev
ton and the fungus V. dahliae, the structure of the substances that deter-
mine the fungitoxicity of the diffusates was resolved by means of IR, UV,
nuclear magnetic resonance, and mass spectrometry (Sadykov et al.,1974;
Karimdjanov et al., 1976). In the light of these investigations, and those
carried out at the National Cotton Pathology Research Laboratory in the
United States, the chemical structure of the cotton phytoalexins has been
clearly established. Thesefall into the category of sesquiterpenoid andtri-
terpenoid aldehydes, akinto gossypol (Fig. 1).
B. Location and Role of Phytoalexins

Our subsequent efforts were directedto investigations on the function-and
role of phytoalexins in disease-resistant cotton, and the difference in infec-
tion response reactionsof the resistant and susceptible varieties, considering
the localization of the processes inthe infected tissuesand the dynamicsof
their development.It was noted that a more complete picture the of defense
reaction mechanism reflects at the initial determination phase of the wilt
disease whenthe pathogen comes incontact with the host plant during the
latent, symptomless period ofthe disease (fromthe moment of plant infec-
tion until the localizationand cessation of artificially induced infection in
the event of incompatibilityor until the actual appearance of symptoms in
the event of compatibility).
To determine the quantity of isohemigossypol in the xylem vessels of
cotton varieties differing in their degree of wilt resistance, the method of
Avazkhodjaev and Zeltser (1980) was adopted. The results revealed that
production of phytoalexins was characteristic of both susceptible and resis-
tant varieties. The difference lay in the speed offormation of isohemigossy-
pol in responseto infection. In wilt-resistant varieties, formation/accumu-
lation of phytoalexins progresses at a faster rate than in susceptible ones.
Corresponding with this,quantitative differences also manifest inthe con-
tent of isohemigossypol during the early days of incubation period. In
resistant plants, duringthe initial 5 days following infection, a more or less
faster accumulation of phytoalexins was observed in the xylem vessels. In
the resistant subsp. mexicanum and in varieties belonging to the class of
Tashkent, as also inthe case ofthe chemomutant L3, tracesof isohemigos-
sypol appearedon the chromatogram within10-12 hr of infection. During
this particular phase, phytoalexins werenot detected in the tissues of the
susceptible variety S-4727, while traces of isohemigossypol appeared only
after 24-30 hr of incubation. Through such analysesof dynamics of phyto-
alexin formation, it is possible to judge the level of susceptibility or resis-
tance of cotton to the Verticillium wilt. This property is also displayed by
an array of new, promising, wilt-resistant G. himturn varieties recently
-a:
Wilt Resistance 133
Hoa
OH HOC HOC OH
Ho \ CH /
HEMIGOSSYPOL 3 ISOHEMIGOSSYPOL
4 CH, H,C
CH, H,C
OH HOC CH0 OH
A
GOSSYVERTIN CH
6-METOXYHEMIGOSSYPOL
CH0 OH
CH
CH 6-D IOX YHEM
IGOSSYPOL 4 HEMIGOSSYPOL
CH, H,C
CH, H,C
Figure 1 Chemical structure of phytoalexins of cotton.
released bythe IEBP, including Uzbekistan1 and 2, Express-2, Tashkent-6,

AN-402, and several new lines in selection. All the experiments involving
the tissues of resistant plants were marked by a much faster rate of forma-
tion and accumulation of isohemigossypol. It may be noted from the data
represented in Fig. 2that in the tissues of the susceptible CV. S-4727, in the
Avazkhodjaev et al.
A
Tashkent 6
sbusp. mexlcanum
c cheao-mutant L-3
2 5 10
DAYS AFTER INOCULATION
Figure 2 Dynamics of phytoalexin formation in xylem tissues of different cotton

cultivars.
dynamics of the incubation period, a similar trend of quantitative increase

in hemigossypol is registered, although the quantity of phytoalexins per se
was lessthan that of the resistant ones, untilthe tenth day of the postinfec-
tion period.
At different stages of wilt, depending on environment and features
of pathogenesis, the content of phytoalexins in the tissues of susceptible
cottons may even be higher. However, this isnot connected wit the incom-
patible reactions between the plant and the fungus V. dahliae but on the
contrary characterises a progressiveintensification of the wiltdisease.
At the beginning of the incubation period (2-5 days), wilt-resistant plants
characteristically show fewer necrotized cells and consequently relatively
higher doses of phytoalexins get accumulated in the infectionspots, which
Wilt Resistance 135
may prove lethalto the pathogen. In tissues of susceptible plants, progres-

sion ofthe disease is faster since during these very days latent
of infection a
lesser quantity of phytoalexins is formed, although the number of necro-
tized cells is significantly higher. In other words, the rate of growth of the
pathogen is faster than that of production of isohemigossypol and hence
the fungitoxic compounds cannot completely block the development of
infection. Thisis one of the characteristic properties of phytoalexins, deter-
mining their role in the wilt resistance of the cotton plant. A decisive mo-
ment in the event of infectionis the speed of production of postinfectional
inhibitors in tissues of host plant in response to infection and their accumu-
lation in lethal doses thein sites of infection.
A s was revealed from our subsequent studies, phytoalexinsare formed
in the cells surrounding the xylem vessels,and thus in comparisonto other
plant species the functioning of this defense reactioncotton in is unique. In
those spots where infective structures of pathogen are present, primary
cell walls of visceral parenchyma are destroyed inside the vessels, with the
formation of tangentially isolated sectors (Fig. 3). A kind of “isolation” of
the diseased sectorsfrom the healthy ones occurs, which can be clearly seen
under methyl bluestain. At times, theseare in such large numbers that they
completely cover parts of vessel segments, as in the natural formation of
tyloses in several wood species (Yatsenko-Xmelevski, 1954). Results of our
experimentsalsoconfirm that phytoalexinsaccumulatein the spherical
tyloses of xylem vessels of infested plants and are clearly visible when their
preparations are stained with antimony trichloride. Thus, in the tissues of
cotton plant during infection with V. dahliae, certain mechanical barriers
are formed, by way of blocking of vessel segments or their portions, with
simultaneous localization of fungitoxic phytoalexins in them, which may
completely suspend the development of the disease. In the susceptible CV.
S-4727, within 48 hr of infection a larger number of necrotized cells were
obtained than those in the CV. Tashkent-6. However, the concentration of
phytoalexins was obviously not adequate for a complete suppression of
infection. Consequently, despitethe fact that phytoalexins are produced in
response to infectioninthesusceptible plant, the pathogenappeared to ’
continue to settle in xylem vessels, leaving behind a certain number of
dead parenchyma cells. Onthe other hand, in the resistant cultivar, during
incubation period, cells are necrotized in lesser numbers in response to
infection, and higher doses of phytoalexinsare accumulated in the zone of
infection, inhibiting the subsequent spread of the pathogen in the plant
body. Thus, the necrotized cellsof the cotton plant may serve as barriers in .
the process of ramification of the pathogenic fungus, only if they contain
lethal doses of phytoalexins.
Transverse section
VP
Longitudinal section
Figure 3 Pathological changesin xylem of cotton infected with Verticillium wilt.
C. Search for Inducers from Fungal Cell Wall

Albersheim and Anderson-Prouty (1975) stipulated a theory, according to
which the mechanism of phytoalexin formation relates to the interaction of
surfacemoleculesoftheinteracting partners: thereceptors,products
of genes of the resistant host plant, and the pathogen molecules, products
of the avirulence genes.
Wilt Resistance 137
It has been establishedthat not only the pathogen but certain cultural
fluids containing inducers, extracts of fungalmycelia, and severalhigh
molecular weight metabolites ofthe pathogen, after their contact with the
plant can selectively act as inducers, resulting in the formation of phyto-
alexins inthe plant. Analogous behavioris displayed byan array of chemi-
cal substances, such as saltsof heavy metals, antibiotics, inhibitors of fer-
ments, fungicides, etc.
On the basis ofreports in the literature on the nature of plant defense
reactions (Metlitski et al., 1976; Dyakov, 1983; Albersheim and Valent,
1978; Zaki et al., 1972), we initiated a search for the inducer metabolites
among various fractions isolated from mycelia and cultural fluids of the
pathogen. Initially, a certain protein-lipopolysaccharide complex (PLPC)
and its componentswere studied (Malyshevaand Zeltser, 1968). Our experi-
ments established the phytoalexin activity of phytotoxic metabolites of V.
dahliae, isolated by the scientists of the Institute of Microbiology of the
UzbekAcademyofSciences(Borodin, 1978). The metabolites included
di-2-ethylhexylphthalate, transaminic acid, an oligosaccharide, a polypep-
tide, a pigment, a common protein, and lipids from fungal mycelia and
other sources.
Our subsequentefforts were directedto the study of pathogen cell wall
derivatives. A s per contemporary thinking, first in order of importance,
considering the interaction of higher plants and their specific pathogens in
the initial stage of infection, come host membrane and membranolytic
agents of the inducer pathogen.
Subsequent progress of the disease would be mainly decided by the
status of the plant membrane sensitivity towardthe specific pathogen me-
tabolites that interact with them. This interaction presupposes the presence
of receptor areason the host membranes as also the presencethe of specific
factors complementary to them on the pathogen envelopes. This theory is
reflected in the series of reports concerning certain fungal, bacterial, and
viral diseases (Hadwigerand Schwochan, 1969; Dyakov, 1976,1983; Metlit-
ski, 1976; Metlitski and Ozeretskovskaya, 1985). In this context, investiga-
tions by Albersheim and his group (Anderson-Prouty and Albersheim,
1975; Albersheim and Valent, 1978) are of considerable interest. Utilizing
the methods of isolation and fractionation of cell walls with some modifica-
tions, wewere able to isolate and identify the nature of the most active
phytoalexin-inducing components of cell walls, differing in the degree of
virulence of the races of V. dahliae (race 2 and the avirulent radiomutant
R-177). Results of our experiments showedthat comparatively high phyto-
alexin-inducing activity is exhibited by alkaline and lipid fractions of the
cell walls of the fungus V. dahliae.
Spectral data characterized the activephytoalexin-inducingcompo-
nents of the cell walls as substances with typical polysaccharidestructure,

with characteristic functional group and bonds (see Avazkhodjaev et al.,
1984a,b; Avazkhodjaev,1985).
Analyses of the elicitor isolated from cell walls have indicated that the
elicitor, complex in structure, is principally composed of carbohydrates
(over 50%) and proteins (up to ~ T O )with
, traces of K and Ca cations and
about 1% phosphorus. In the hydrolysates of the elicitors, the carbohydrate
components are predominantly galactose, glucose,and mannose, whilethe
amino acids include aspartic and glutamic acids and alanine. IR spectro-
scopic data confirmed the complexity of the elicitors and their glycoprotein
nature (Figure 4).
D. Induction of Phytoalexins by Elicitors

The induction of phytoalexins by elicitors was subsequently studied, using
the relatively resistantCV. Tashkent-6 and the susceptibleCV. S4727 (Table
2). Dynamics of quantitative contents of phytoalexins inxylem tissues was
monitored. The results of thin-layer chromatography in plates of silufol-
254 (Czechoslovakia) (Avazkhodjaev et al., 1984a,b) revealed that, with the
introduction of elicitor, a small quantity of phytoalexin appeared as early
as on the second day in stems of both cultivars. Within 2-5 days, larger
quantities of phytoalexins were found accumulated in the relatively resis-
tant CV. Taskhent-B, especially againstthe elicitor from the avirulent race.
However, at a much later stage, the reverse was found to be the case: the
content of'phytoalexins was higher in the xylem of the susceptible variety,
especially as against the inducer from the virulent race.
In some of our recent studies, further observations were made on the
nature of changes in xylems in different cotton varieties varying in the
degree of their wilt resistance, under the influence of the elicitor from V.
Figure 4 IR spectrum of the elicitor of the fungus Verticilliumdahliae.

VI
3
m
m 8
0
5
g
Y
*
si
dahfiae (see Avazkhodjaev et al., 199Oa). The pathological changes that

occurred were the formation of tyloses and the choking of vessels, with
subsequent formation of phytoalexins (Fig. 5). During the first 10 days
after the introduction of the elicitor in the stems, there was a gradual
development of these processes, from the lower portion of the stem to the
upper regions. In the susceptible CV. S-4727, under the influence of the
elicitor from the virulent races, such reactions proceeded more slowly
than
those inthe relatively resistantCV. Tashkent-6. The typeof response of the
Figure 5 Pathological changes in xylem of cotton CV.S-4727 (1) and Tashkentd

(2) in the spotof isolation of the elicitorof the fungus (after48 hr).
wilt Resistance 141
given plant variety was determined by the speed of induction of defense

reaction, which is also influenced by its metabolite, the inducer of phyto-
alexin formation in the plant, in doses lethalto the pathogen.
It is evident from the results that in the very initial phase following
introduction of elicitor inxylem of resistant plant, higher amounts of phy-
toalexins are accumulated, as compared to those inthe susceptible varieties.
This leadsto the confinementof infection to the lower portions of the stem,
i.e., in the primary infection spotand obstructs the further growth of the
fungus in the plant.
Observations on the inducible characterof the defense reactionof cot-
ton have been used as the basis for a laboratory-devised method of defense
against wilt, on the lines of artificial intensification of phytoalexin forma-
tion in plant tissues by various means, includingthe use of chemical prep-
arations that influence the natural mechanismswhichlimit or block
infection. The function of such preparations relates to the induction of
phytoalexin production in tissues. High ability to activate the defense reac-
tion was shown by ionophore cyclopolyether derivatives, i.e., Dibenzo-18
and Crown-6, effecting two- to threefold lowering of mortality in field
experiments (Avazkhodjaev et al., 1984a,b).
Amoreeffectiveantiwiltimmunizer was the preparation Biosol-2,
which was synthesizedin collaboration with chemists at the Bashkir branch
of theAcademy of Sciences of the erstwhile USSR. Preparations of the type
Bisol have been introduced in cotton cultivation, since incidence of wilt
under their influence is lowered by50-70%, while the fiber yield increases
by up to three to four quintals per hectare. Purpose-oriented screening and
subsequent testing of the preparations is in progress.
During the recent years, researchers of phytoimmunity, having resolved
the nonspecific characterof indiction, have turnedto compositional analy-
sis of the metabolites from pathogens, which condition the specificity re-
sponse ofthe host plant. Metlitski al. et (1986), considering phytoimmunity
from a biological viewpoint wherein incompatibility is a natural rule and
compatibility is almost exception, feltthat specificity ofthe response reac-
tion of the host plant hasto be associated with various productsthat condi-
tion compatibility of the participants. Such a compatibility is the conse-
quence of the lengthy process of natural evolution of the host and the
pathogen, in which new genes for virulence appear and then overcome the
genes that condition resistance in the host plant. Specificity thus may be
controlled by productsof the susceptibility gene of the plant and that of its
complimentary parasite gene.
Several hypothetical models propose that in addition to nonspecific
inducers (elicitors) conditioningthe induction of phytoalexins in association
with the receptor and the resistant plant, the parasites produce species-
specific and race- and cultivar-specific suppressors, which obstruct the ac-
tion of the elicitors (Terekhova and Dyakov, 1980; Heath, 1981; Bushnell
and Rowell, 1981). Specificity of suppression is conditioned through a se-
ries of corresponding areas on suppressor and receptor, which must be
complementary to each other, and thus effectively ensure the suppression
of the effects of the elicitor by rendering cellthe incapable of switching on
the defense reaction.
Our experience points to the fact that the high varietal breakdown in
cotton is largely due to the strength of the virulent race 2 of V. dahliae t o
inhibit the synthesis and accumulation of lethal doses of phytoalexins in the
infected tissues.
Subsequent studies have underlined the significant role of suppressor
metabolites in races of V. dahliae, blocking the defense reaction of cotton
in the initial stage of Verticillium wilt. Suppression of phytoalexin forma-
is displayed under influenceof the avirulent
tion is race-specific, since it not
mutant R-177, which apparently is one of the means to overcome the culti-
var resistance of cotton (Table 3; Fig. 6; see also Avazkhodjaev et al.,
1984a,b). As can be seen from the Fig. 6, in the experiment involving the
introduction of metabolite (analogous to the suppressor of the virulent
race 2) from the avirulent mutant R-177, no suppression of induction of
phytoalexin formation was noticeable: the xylem vessels of the host plant
showed the presence of lethal doses of the phytoalexin isohemigossypol.
Suppression of the defense reaction -the formationof phytoalexin -by the
suppressor of virulent race 2 was noticed in the host plant only in the
initial determinative stages of the disease, since the process of phytoalexin
formation was restored in the infected tissues on the fifth day of the experi
ment.
Table 3 Induction and Blocking of Phytoalexin Formation in Cotton Tissues

Isohemigossypol content
after 48 hr (pg/g dry
Experimental variant wt of tissue)
1. Control (intact tissues) Absent

2. Injection of avirulent mutant P-l77 race 38.4 f 1.3
3. Injection of virulent race 2 (2.5 m1 spores/ml) 21.8 f 0.9
4. Influence of metabolite (avirulent suppressor
of race 2) 49.7 f 1.2
5. Influence of suppressor of race 2 + after a
day, inductor of phytoalexins Absent
6. Introduction of phytoalexin inductor 46.2 f 2.1
Wilt Resistance 143
m - Inductor
- Supressor o f race 2
inductor
t
--Metabolite o f race p-l77

(isolated from analogous
supressor o f race 2 ) +
inductor
1 3 5
exposure of experiment (days)
Figure 6 Induction and blockage of phytoalexin formation in xylem vessels of

the cotton CV. Tashkent-6.
The results confirm that the high incidenceof breakdown in wilt resis-
tance of cotton is related to the ability ofthe race to suppress the formation
and accumulation of lethal dosesof phytoalexins. Since the suppression is
race-specific, it is presumably one of the ways to surmount the cultivar
resistance. On the basis of the collected information, our laboratory has
proposed a method for selectionof the wilt-resistantformsof cotton
(Anonymous, 1986), which is used by plant breeders for obtaining an in-
ducer of the Verticillium wilt of the selection material, whichis resistant to
the race prevalentat the given time.
Results of chemical analysis of the suppressor metabolite isolated from
virulent race 2 of V. dahliae, as well as the phytoalexin-inducing prepara-
tions from the cell walls of the pathogen, indicate that the suppressor me-
tabolite from virulent race 2 of Verticillium dahliae is glycoproteinaceous
in nature and contains 34.4% carbohydrates and 56.3% common proteins.
It was established with the help 'of IR spectroscopy, that the suppressor
metabolite relates to lipoglycoproteins, with their characteristic complex

bonds of functional groups.
In the hydrolysate of the suppressor, the following amino acids were
identified: aspartic acid, glutamic acid, asparagine, aminobutyric acid, ala-
nine, serine, lysine, threonine, histidine, leucine, isoleucine, arginine,tyro-
sine, cysteine and oystine, proline, tryptophan, phenylalanine, valine, and
methionine. The sugars detected were raffinose, lactose, glucosamine, ga-
lactose, glucose, mannose, and rhamnose. It was also shown that the sup-
pressor molecule had a lower molecular weight (of the order 8OOO-10,OOO)
than the phytoalexin inducers.
111. RECEPTOR SUBSTANCES

In the light of the fact that the process of recognition of pathogen inducers
is accompanied bythe participation of the cell wallcomponents of the host
plant receptors, we were able to isolate certain lectin-like substancesfrom
the cell walls of the host plants, possessing the specificities of receptors.
According to earlier reports (Sequeira, 1978; Lyutsik et al., 1981), lectins
are proteins, possessingthe ability to associate ina complementary manner
with pathogen cell wall inducers of polysaccharide character.
It has been knownthat lectins participate in the formation of glycopro-
tein and polysaccharide complexesand in the process of intercellular recog-
nition, which take place during host plant and pathogen interaction (Lyut-
siketal., 1981; MarkovandXavkin, 1983; Faye and Crispeels, 1987).
Results of our research on their physicochemical constitution indicatethat
lectins are mainly proteins forming a complex withtrace amounts of carbo-
hydrates (Nuritdinova,1988).
In the courseof our investigations, two fractions of lectins were traced
in the helium columns, marked with quantitative differences of carbohy-
drate content. The amount of carbohydrate inthe first fraction (1 1.7mg/g
dry wt) was about half the quantity of the second (22.8 mg/g dry wt). In
the hydrolysates of the two fractions, the sugars galactose, glucose, man-
nose, and ramnose were present in a ratio of 4:3:1:2. Lectins of the first
fraction contained about 31.4% proteins whereas thoseof the second con-
tained 78.8%. Some quantitative variation in the intensity of absorption
bands in the IR absorption spectra of the two fractions of lectins was
observed, although the spectral data characterized the lectins as substances
that are mainly proteinaceous in character, withfunctionalgroups and
bonds inherent to proteins. Also, the IR spectrum displayed insignificant
absorption bands characteristicof carbohydrates, which permits the lectins
to beseen as a complex union of proteins with polysaccharides with a
preponderance of protein structures, and thus confirms the data obtained
Wilt Resistance 145
Table 4 Analysis of Lectin Fractions for Content(To) of Cations of Various

Elements
Lectin C H S No Cu Ca AI Fe Si Mg Mn
Fr. I 21.97
4.33
1.66
3.27
0.009
0.0031
0.054
0.16
0.26 t t
Fr.
I1 27.22
5.11
1.17
4.42
0.003
0.0261
0.018
0.11
0.11 t t
in earlier analysis (Avazkhodjaev et al., 1987). The lectin fractions con-

tained small amounts of cations such as Cu, Ca, AI, Fe, Si, Mg, and Mn.
(Table 4 shows the percentage content of the various elements.) The two
fractions of lectins differed not only in terms of their proteins, carbohy-
drates, amino acids, and sugars but also in their individual physiological
actions.
A. Lectins, Biological Properties, Function, and Immobilization

Data on the influence of the two fractions of lectins in the formation of
phytoalexins are presented in Table 5 and Fig. 7. It is seen that the two
fractions are different in their induction of phytoalexins (Fig.7). The first
fraction suppressedor inhibited the induction byabout 53.1070, whereas the
second strengthenedit by about 131.8%.
Detailedstudieswereconducted on the biological properties of the
lectins, i.e., their abilityto react to the infectious structures and metabolites
of the pathogen,formingdeterminantcomplexesthroughspeciallyde-
signed in vitro experiments. The lectin and its two fractions can react with
V. dahliae, influencing germination of conidia and hyphal growth (Table
6). Inhibition of growth of pathogenis most evident duringthe reaction of
lectins with an avirulent race of the fungus. By producing incubation mix-
tures of the complete lectin with the inductor isolated from V. dahliae, it
has been establishedthat lectins are concerned withthe fungal metabolite-
the elicitor of phytoalexins in the plant also.
Table 5 Influence of Lectin on the Formation of Phytoalexins in Hypocotylsof

Cotton Plants(pg/g W)
Lectin Gossypol Sum of v0 of
Isohemigossypol
fraction
equivalent
phytoalexins
control
Control 36.0 f 1.2 35.6 f 0.9 71.6 f 4.2 100.0
Fr. I 17.0 f 0.9 21.0 f 0.5 38.0 f 2.1 53.1
Fr. I1 50.4 + 3.2 44.0 f 2.3 5.4
94.4 131.8
Avazkhodjaev et al.
,D K
1
-
-
control
1st fraction of lectin
2 - 2nd fraction
I HG
IHG - lsohemigossypol
Figure 7 Chromatogram of chloroform extracts from hypocotyls of cotton plant

treated with lectinefractions.
During subsequent research on the sites of contact of lectin with the

components of the elicitor of pathogen (which immobilize the inducement
of the disease) itwas found that therewas a discretionary contact of lectin
-
with the sugars glucose, galactose, and mannosebasic components of the
hydrolysate of the carbohydrate portion of the elicitor of V. dahliae. Such
components were termed as heptanes (Avazkhodjaev al., et 1989a,b).
The ability of the lectin fractions isolated from the hypocotyls of cot
plant t o agglutinate the conidiaof the two pathogenic racesof Verticillium
differing in virulence (race2 and the avirulent mutantR-177)as well as the
influence of the heptane sugars on this process were investigated. The re-
sults showed that addition of heptanes to the incubation mixture inhibited
Table 6 Phytopathological Characteristics of the Experiment on the Influence

of Lectins of Cotton Hypocotyls and Its Fractions on thePathogen of
Verticillium Wilt
Sum of
Control Lectin I Lectin I1 Lectins
Conidial Hyphal v0 Suppression
Fungal
growth germination
race (a) (b) (a) (b) (a) (b) (a) (b)
R-l77 100 100 85.0 52.0 71.2 33.6 81.1 41.0

Race 2 100 100 80.0 38.3 28.1 11.7 58.8 29.4
Wilt Resistance 147
the agglutinization of conidia of both races by lectins and their fractions

(Table 7). Thus, the introduction of heptane sugars individually to the
incubation mixture lowersthe intensity of agglutinationof conidia of viru-
lent race 2 by almost four times and that of avirulent raceby five times as
compared to control (without the addition of sugars), and three and four
times, respectively, as compared to the introduction of xyloses and ara-
binosesin the mixture, taken as control. The aggregateofheptanesin
the same concentration as that of the individual heptanes inhibited the
agglutination of conidia of both the races caused bytotal lectin, even more
significantly (8-10 times) (Avazkhodjaev et al., 1989a,b).
The difference in the agglutination responses shown by the races 2 and
R-l77 was noticed also when the first and second fractions of lectins were
introduced into the incubation mixture. The aggregate of lectins, like its
individual fractions, agglutinized the conidia of the avirulent race R-l77
more intensivelythan the conidia of virulent race2. Also, the first fraction
of lectins displayed a higher abilityto agglutinize the conidia of the aviru-
lent race. It was concluded that the presence of the sugar heptanes in the
incubation mixture inhibitedthe agglutination in all variantsof the experi-
ment significantly while the addition of xylose and arabinose in theform of
control was obviously ineffective. The aggregate of sugar heptanes more
strongly inhibited agglutination than each sugar taken individually. This
Table 7 Agglutinization of Conidia of V. dahliae with

Lectin of Cotton and Its Fractions (in Relative
Units,
Average Aggregatesof Agglutinized Conidiain 0.01 m1 of
Incubation Mixture)
R-l77
raceFungal
2 R-l77 2
Lectin fraction
Aggregate
Variant of lectins I I1 I I1
Sugar heptanes
Glucose 9.1 8.7 14.9 11.0 7.7 11.1
10.7
Galactose
11.1 14.2 17.0 10.8
11.7
Mannose 9.8 9.9 17.4 15.3 9.7 7.3
Aggregate 6.0 6.3 6.1 7.4 0.8 5.0
Control
Xylose 33.0 41.6 32.9 36.8 74.3 57.3
Arabinose 31.2 42.6 32.3 36.4 74.2 57.8
Buffer 45.0 51.1 40.9 46.6 77.1 58.8
148 al. Avazkhodjaev et
was most evident in the variant with the first fraction of lectin and the
conidia of R-177.
The experimental resultsthus pointed to the fact that agglutination is a
specific indicator of the interaction of lectin with components of sugars
from the surfaces of conidial cell walls of the fungus V. dahliae. Specific
inhibition of agglutinationof the conidia by lectins and by its fractions-
sugar heptanes (glucose, galactose, mannose)-indicates that lectins form
glycoconjugates with these sugars, i.e., they are componentsof the fungal
cell walls, reacting withthe lectins ofthe cotton plant.
Thus the property of the lectins to participate in the twin processes
of recognition and the interaction of cells in the cotton plant fungus V.
dahliae, whichemphasizestheirdefensiverole and ability to immobi-
lize the pathogen, has been ably established from the resultsour of investi-
gations. The degree of agglutination of the fungal conidia may serve as
a test for the resistance of the plants and the virulence of the initiator of
wilt.
B. Receptor Functions of Plasmalemmal and Microsomal Fractions
Since surface metabolites localized the in membranes ofboth the host plant
and the pathogen are known to interact, it was expected that the lectins
participating inthe defense mechanisms would be concentrated in the plas-
malemma and in membranes of the microsomal fraction of plants. We
investigated the receptor function of the plasmalemma and microsomal
fractions of the cotton plant, with the intention to gain an insight into the
relation of its components withthe membrane-active fungal metabolites-
the elicitor of phytoalexin formation. The chemical composition of the
hydrolysates of plasmalemma of the unaffected cells is knownto be akinto
that of lectin. Model studies on the composition of incubation mixtures
of plasmalemma and microsomal fractions with elicitor showed that the
incubating components possessed the ability to form conjugates. In the
presence of microsomal fractionsand plasmalemma, we observed a strong
agglutination of conidia, especially in the avirulent race of V. dahliae (see
Avazkhodjaev et al., 199ob). We have thus established that a property of
metabolites of plant and pathogen (elicitors, immunodepressors, lectins) is
that when present on the cell membrane surface they participate in the
host-pathogen interaction, resulting in realizationof either the markers of
resistance of the host plant or the markers of virulence of the pathogen.
With the help of the elicitor of phytoalexin formation, isolated from the
fungal cell wall, specific sites of reaction have been identified
the receptor
in
sections of microsomal fractions cotton
of cells.
Wilt Resistance 149
IV. EFFECT OF INOCULUM LOAD ON PHYTOALEXIN FORMATION
It has been notedthat the ability of microsomal fractionsto form glycocon-

jugates with elicitor (and, consequently, phytoalexins in plant tissues in
response to induction) may be affected by certain unfavorable factors of
the surrounding environment. One such important factor is the load of
pathogen, i.e., quantity of the inoculum per unit surface of the plant. Data
obtained on this factor (Table 8) revealthat the infection load significantly
influences the rate of accumulation of lethal doses of phytoalexins. Against
the background of a relatively small load of infection (infection with a
weaker intensity),the plants displayed high concentrations of isohemigossy-
pol in the lower portions of the stem, ranging between42.7 and 45.2 pg/g
of the tissue, as early ason the fifth day following inoculation.An increase
in the dose of artificially induced infection by a factor of 5 yielded a lesser
quantity of phytoalexin during the same initial 5-day period (between29.2
and 31.4 g), wherein isohemigossypol was also detected in the central por-
tion of the infected stem.Thus, in this particular variantof the experiment,
the infection zonewas larger, and the content of the phytoalexins per unit
infected tissue would be less than lethal to pathogen. This equally applies to
phytoalexin accumulation patterns in the infected stems of cotton during
the 10-day period of incubation: the high infection load caused a faster
growth of the disease and the zone of infectionwas larger, i.e., the rate of
proliferation of infection is higher than the rate of accumulation of lethal
doses of phytoalexins.
Besides known external factors that influence the phytoalexin forma-
tion (temperature, humidity,pH of the medium, etc.),the race composition
of the fungus V. dahliae present in the soil is also a significant influential
factor. Results of experimentson the dynamics of phytoalexinformation in
cotton, under the influence of races differing in virulence, are shown in Fig.
8. These indicate that the virulent race 2 of V. dahliae has the ability to
delay the accumulation of high doses of phytoalexins in the tissues of resis-
tant plants, which get strongly affectedby the disease causedby this patho-
gen. Introduction of the avirulent mutant R-l77 into the xylemvessels
of cotton plant, on the contrary, induces a much faster formation and
accumulation of phytoalexins inthe sites of infection.
Virulent forms of pathogens are endowed withthe ability to circumvent
the defense mechanism of the host plant, rendering it inactive (Hadwinger
and Schwochan, 1969; Metlitski and Ozeretksovskaya, 1973, 1985). This is
one of the means of adaptation of the pathogen to the disease-resistant
host plant, in which the defense mechanism is not destroyed but remains
unstimulated. In this context, we conducted some experiments concerning
the so-called cross-protection phenomenon.
2
."$
U
U
%
E,
<
c)
."
B
C
."0
c)
td
P
2
."
c
C
0
5
n
VI
Wilt Resistance 151
A
-Virulent race 2
B
Avirulent radio-mutant
cl? R - 177.
m
J
E 50 A
1
"4
4J
y.
0
40 I.
c, 30 B
3
m 20
0
l.4
* 10
00
'M60
-50
B
0
a A
VI 40
0 II
2 30
E
0
$i
VI
20 B
I
2 5 IO
Days after inoculation
Figure 8 Dynamics of phytoalexin formation under the influence of races of Ver-

ticillium dahliae differing in virulence in the
cotton cultivars " 4 2 7 (I) and Uzbekis-
tan (11).
We made some interesting observations on cotton plants in the tissues

of which spore suspensionsof avirulent mutant R-l77 were first introduced
and after 1-3 days the plants were reinoculated with a virulent race of
the wilt fungus. Tissues of xylem vessels of the plant of the resistant cvs.
Uzbekistan and S-4727 showed a remarkable production of isohemigossy-
pol within 48 hr (35-40 pg/g) in response to infection by the avirulent
mutant R-177. It can be assumedthat such fast formation of high doses of
phytoalexins delayedthe development of artificial infection bythe virulent

race 2. This was confirmed from the growth patterns of wiltin xylem vessels
as wellas from observations made during 60 daysof the postinfection
period (Table 9). In the control variant involving infection withthe aviru-
lent mutant R-177, the plant was apparently healthy for about 50 days,
without any signs of wilt disease, while in the control variant involving
infection with race2 all the inoculated plants displayedstrong symptoms of
the disease and perished within18-20 days.
Stimulation of phytoalexin formation through injection of avirulent
race increased the resistance to the strongly virulent race 2 of V. dahliae.
Plants of the CV. S4727 suffered higherwilt infection than those of the CV.
Uzbekistan. The differences were evident the in speed of growth ofthe wilt
as well as inthe degree of infection ofthe experimental plants. However,a
complete death of anyplant, as was the case with thecontrol variants, was
not noticed. On the contrary, when reinoculated after 2-3 days, the degree
of wilt infection was found reduced in both kinds of plants. In the CV.
Uzbekistan, the plants remained practically healthy, especially3 days after
reinoculation with race 2.
Table 9 Infection of Cotton with Wilt inthe Experiment with“Cross-Protection”

(Infection Load 2.5 million Conidia/ml)
070 of infectedplants
With
high
With
low 070 Healthy
Experimental variant plants intensity
intensity
CV. S4727
Control (injection of race 2) a a -
Avirulent mutant + reinoculation
with race2 after 1 day50 50 -
with race2 after 2 days 45 55 -
with race2 after 3 days 30 70 -
CV. Uzbekistan-l
Control (injection of race2) a a

with race2 after 1 day 10 80 10
with race2 after 2 days - 40 60
with race2 after 3 days - 10 90
“All plants died, i.e.. 100% mortality.
Wilt Resistance 153
It was further observed that if the period between inoculation and

reinoculation is increased up to 10 days, mass infection results, since the
initial induction of phytoalexin by the avirulent race lasts onlyfor a short
duration and fades in time. Thusthe fast rate of formation and accumula-
tion of lethal doses of phytoalexinsin response to avirulent mutant R-177
checks further development of the pathogen and subsequently the postin-
fectional inhibitors were not produced. Virulent racesof V. dahliae cotton
plant tissues; hence they settle rapidly in the xylem vessels, striking new
sectors. Molecular mechanisms behind such responses have not been clearly
understood, but their foreplay, with the resultant synthesis of inhibitors of
the phytoalexin, is fully known.
V. EXPERIMENTS INVOLVING INTERACTION O F DISEASES
Occurrence of cross-protectionwas noticed when variousother pathogenic

microorganisms, such as those causing gummosis and root rot, were al-
lowed to infect a cotton plant prior to its infection by Verticillium. The
effect was either congenialor adverse dependingon the reaction of variety
and species to the particular pathogen and its level of virulence. Thus the
cross-contact of the plant with the inducers of root rot, gummosis, and
Verticillium wilt may lead eitherto the stimulation of its defense reactions
or to the decline of its immunosystems and death. In the first case, if the
variety turns out to be resistant to the inducer of black rot and gummosis,
there may be nonspecific synthesis of phytoalexins in the plant tissues,
which, with subsequent infection withVerticillium, would suppress the de-
velopment of the fungus. In the susceptible cultivar, with escalation of
disease the plant either perishes or is unable to withstand the subsequent
wilt infection consequentto increased sensitivityto the pathogen.
It hasbeenconfirmedin our studies that the inducersof root rot
(Rhizoctonia solani) and gummosis (Xanthomonas malvacearum) bring
about the formation of phytoalexins in the tissues of seedlings of cotton,
depending on the level of varietal resistance against these microorganisms
(Table 10). Plants in the early phase of development were artificially in-
fected by the inducers of gummosis and root rot. Those plants which sur-
vived were infected with Verticillium during the boll initiation stage. In all
variants of the experiment with the preliminary infection of gummosis and
root rot, a similar tendency for increase in the induction of phytoalexin
formation was observed. In CV. Uzbekistan, following infection withVerti-
cillium, the content of isohemigossypol was 56.8 pg/g fresh wt, while in
those showing primary infection withthe bacteria and root rot, the corre-
sponding values were found to be 68.0 and 104.6 pg/g, respectively. In CV.
154 et Avazkhodjaev al.
Table 10 Quantitative Analysis of Phytoalexins of Cotton in Hypocotyl

Tissues Infectedwith the Bacteria Xanthomonas malvaceammand the
Fungus Rhizoctonia solani(pg/g)
Cultivar
Uzbekistan White-gold 108 F

Fungus Bacteria
Fungus
Bacteria Fungus Bacteria
IHG
GE
IHG
GE
IHG
GE
IHG
GE
IHG
GE
IHG
GE
10.0
39.8
19.2
7.3
15.4
32.4
13.0
21.0
20.2
80.5
54.0
7.4
IHG, isohemigossypol; GE, gossypol equivalent.
108 F, the corresponding values were 36.0, 50.0, and 72.0 pg/g fresh wt,
respectively.
VI. XENOBIOTICS IN THE IMMUNOSYSTEM OF COTTON

Considerableinfluence on phytoalexin formation isexertedbyvarious
chemical preparations, such as fungicides, pesticides, and defoliants. Our
data indicate that certain preparations under test (e.g., the defoliants Buti-
fos and Butylcaptaks) substantially increased the disease tolerance of the
cotton plant. When applied in low concentrations, the intensity of wilt
disease was also low in a majority of experimental plants. Inoculation of
the plant with V. dahliae following the introduction of the testpreparations
in xylem strengthened the formation of phytoalexins in the unaffected tis-
sues of the plant, depending on the concentration of the defoliant. It was
further noted that it is possibleto increase the resistance of the plant to wilt
by using the preparations for enhancing the overall immunityof the plant,
but only with a precise control of timing, with a short duration of applica-
tion and proper concentrationfor their action inthe plant, and also taking
care to minimize their harmful influence, if any, on the surrounding envi-
ronment.
Pesticides at high concentrations may cause elevated phytotoxicity rela-
tive to the Vericillium wilt inducer and the plant, suppressing the hypersen-
sitive reactionand phytoalexin formation in the tissues of the cotton plant,
due to the consequent short period of latent infection and the high wilt
incidence.
Certain unfavorable environmentalfactors may reduce the disease re-
sistance of the plant. Presently, thanks to developments in molecular biol-
ogy and molecular genetics, it has been established that immunity is not
Wilt Resistance 155
only the resistance of the body but also a complex set of reactions directed
in support of its functional integrity, i.e., homeostasis.
One ofthe modes ofcontrol of the immunosystem incotton, subject to
the compounding influence of various anthropogenic factors (xenobiotics
such as pesticides, herbicides, fertilizers, as well as ontogenic xenobiotics,
such as pathogen toxins, phytoalexins, quinones, etc., that are induced in
the plant tissues during disease incidence), isthe elucidation of activity of
the enzyme system, metabolizing or expelling the foreign elements fromthe
plant body. Endogenic xenobiotics, such as phytoalexins, accumulating in
the tissues in high concentrations, display toxic effects not only toward the
pathogen, but toward the plant itself. It has been noted byus that one such
system, very effective in detoxification and utilization of both exogenic and
endogenicxenobiotics,
the
issystem
of g1utathione-glutathione-S-
transferase(Avazkhodjaevetal., 1991). It hasbeenestablished that
under the influence of this enzyme the phytoalexin isohemigossypol isme-
tabolized to 60-70%. Varietal differences have been detected the in activity
of this system in various cotton cultivars and species, which leads to the
recommendation for its use asa physiological test for the desired selection
of varieties with overall resistanceto diseases and pesticides.
VII. EPILOG
Results of our research have elucidatedthe phenomenon of phytoalexins in
considerable details. Oneof the characteristics of phytoalexin formation in
the cotton plant differing from other host-pathogen systems is its produc-
tion in the cells surrounding infected tissues into which the fungus does not
penetrate. Defense reactions, relating to the incompatibility of the plant
and the wilt fungus, are developed inthe parenchyma, cell walls of which
subsequently collapse from within,and the gaps are sealed due to infected
segments of the xylem vessels. Mechanicaland chemical barriersare simul-
taneously created in the path of infection by the cells since the latter are
saturated with fungitoxic phytoalexins.It is demonstrated that the process
of phytoalexin formation, in response to races of V. dahliae differing in
virulence, is a feature of all varieties of cotton regardless of the status of
their wilt resistance. A major factor limiting the disease resistance is the
rate of formation and accumulation of lethal (to pathogen) dosesof phyto-
alexins. Induction of phytoalexin formation in the cotton plant, similar to
that in many other crops, is polygenic in character. Various metabolites of
V. dahliae which induce phytoalexin formation (toxins, elicitors, proteins,
their different fractions, etc.), as well as metabolites that block the induc-
tion of phytoalexin formation in cotton and suppressors that display high
race specificity, have been identified. Suppression of phytoalexin formation
and partial metabolization are two of the means employed by the virulent
races of the wilt fungusto overcome cotton plant resistance (Avazkhodjaev
et al., 1991).
The characteristics of receptor substances of the surface structures of
the cells ofthe cotton plant, which displaythe ability to bind complementa-
rily with the pathogen inductors such as lectins, have been enlisted. We
have identified two fractions of the cotton plant lectins, differing with
respect to chemical compositionas well as in their physiological actions: the
first fraction blocks phytoalexin formation, whereas the second inducesit.
Also elucidated is the ability of lectins to immobilize infectious structures
of the fungus V. dahliae, hyphal growth, and the association of inductor
with phytoalexinformation.
It may be noted in conclusion that for resolving important problems
related with wilt disease of cotton it is imperative that systems of an all-
inclusive and objective assessment of realization of immunological control
of the plant be exploited. Ascertainingthe realistic criteria of immunologi-
cal control in plant permits not only the stimulation of activity of the
protector metabolites but alsothe realization of simultaneous selectionfor
resistance to diseases, vermins, pesticides, etc.
In confirmation of the same,
the data obtained byus on the activityof glutathione-glutathione-4
transferase system in thecotton plant and its role inthe biotransformation
of xenobiotics should prove useful. In the course of research on wild and
cultivated new forms of cotton, some differences were detected, both the in
phytoalexin-inducing ability and in the glutathione-S-transferase activity.
Several of these would be useful in selection as initial materials
for constitu-
tion of complex-resistant varieties of cotton.
ACKNOWLEDGMENTS
The authors acknowledge several friends and colleagues for tireless help
and active interest. Thanks are especially due to our numerous colleagues
in the Institute of Microbiology and the Bashkir branch of the erstwhile
USSR, and to Dr. R. W. Jayashankar of IEBP, for valuable assistance.We
are also grateful to Dr. RamDas Akella,SeniorLecturerinRussian,
Nagpur University, for a reviewof the English version of the text and
valuable advice on translation. The fourth author (R. G. D.) thanks the
Ministryof Human ResourceDevelopment,Governmentof India, and
the Indian Council of Agricultural Research, for award of a postdoctoral
fellowship inthe erstwhile USSR. He also acknowledges the help by way of
useful personal discussions with Dr. Margaret Essenberg and her colleagues
at the Oklahoma State University, and with Dr. AloisBell of the National
Wilt Resistance 157
Cotton Pathology Research Laboratory, Texas A&M University, College

Station, Texas.
REFERENCES
surfaces and the biochemistry of pathogenesis. Ann. Rev. Plant Physiol. 2 6
31-52.
Albersheim, P.,and Valent, B. (1978). Host-pathogen interactions in plants. J. Cell
Biol. 6(3):627-643.
Anderson-Prouty, A. J., and Albershiem, P. (1975). Isolation of a pathogen-
synthesized fraction rich in glycon,that elicits a defence response in the patho-
gen host. Plant Physiol. 56(2):286-291.
Anonymous (1986). Patent No. 1507264, 19-12-1986. Institute of Experimental Bi-
ology of Plants, Tashkent (in Russian).
Avazkhodjaev, M. X. (1985). Physiological basis of wilt-resistance ofcotton plant
and the paths of its induction. D. Sc. dissertation, Academy of Sciences, USSR,
Baku. (in Russian).
Avazkhodjaev, M. X., and Zeltser, S. S. (1980). Physiological Factors of Wilt-
Resistant Cotton Plant,FAN, Academy of Sciences, Uzbekistan (in Russian).
Avazkhodjaev, M. X., Jumaniyazov, I., Tashmuxamedov, B. A., and Gagelgans,
A. (1984). Phytoalexin inducing properties of cyclopolyethers. Proc. of Second
All-Union Conf. Macrocycles, 1984 (in Russian).
Avazkhodjaev, M. X., Zeltser, S. S., and Adilova, A. N. (1984a). Characteristics
of a metabolite of the fungus Verticillium dahliae, inducer of phytoalexins in
the cotton plant. Uzbekskii BiologickeskiiZ . 1:3-5 (in Russian).
Avazkhodjaev, M. X., Zeltser, S. S., and Adilova, A. N. (1984b). Induction and
blockage of phytoalexin formation in the cotton plant. Doklady Academii
Nauk. Uzbek. SSR 8:260-266 (in Russian).
Avazkhodjaev, M. X., Nuritdinova, X. V., Zeltser, S.S., and Adilova, A.N.
(1987). Some properties of lectines of cotton plant in relation to the Verticillium
wilt. Doklady Academii Nauk. Uzbek. SSR 156-58 (in Russian).
Avazkhodjaev, M. X., Nuritdinova, X. V., Zeltser, S. S., and Madaminova, L. M.
(1989a). Formation of determinate complexes of lectine of the cotton plant,
with components of the elicitor of the Verticillium wilt pathogen. Uzbekskii
Biologickeskii Z . 3:6-8 (in Russian).
Avazkhodjaev, M. X., Nuritdinova, X. V., Zeltser, S. S., and Madaminova, L. M.
(1989b). Receptor function ofplasmalemmain the wiltdiseaseof cotton.
Doklady Academii Nauk.Uzbek. SSR 1154-56 (in Russian).
Avazkhodjaev, M. X., Nuritdinova, X. V., and Zeltser, S. S. (1990a). Dynamics of
pathological changes in xylem of the cotton plant under the influence of the
fungus Verticillium dahliae. Micologia i Fitopatologia 24(3):240-244 (in Rus-
sian).
Avazkhodjaev, M. X., Zeltser, S.S., Nuritdinova, X. V., and Musaev, X. A.
(1990b). Receptor function of plasmalemma in the wilt disease of the cotton
plant. Abstr. All-Union Convention“Physiological and biochemical aspects of

plant immunityto diseases and injury.”Ufa, Bashkiria 1990,pp. 5 (in Russian).
Avazkhodjaev, M. X.,Musaev,X.A.,Zeltser, S. S., and Nuritdinova, X. V.
(1991). Role of glutathione glutathione-S-transferases in the detoxication of
xenobiotics. Uzbekskii BiologickeskiiZurnal459-60 (in Russian).
Bailey, J. A., and Mansfield, J. B. (1985). Phytoalexins. Naukova Dumka Publ.
Kiev. Ukr. SSR. 320 p. (in Russian).
Bell, A. A. (1967).Formation of gossypol in infectedor chemically irritated tissues
of Gossypium Species. Phytopathology 57(7):759-764.
Bell, A. A. (1969). Phytoalexin production and verticillium wilt resistance in cot-
ton. Phytopathology59(8):1119-1127.
Bell, A. A. (1981). Biochemical mechanisms of disease resistance. Annual Review
of Plant Physiology32:1221-1281.
Borodin, G. I. (1978).Physiological-biochemical basisof pathogenesity of the fun-
gus Verticillium dahliae, theinducer of Verticillium wilt of cotton. D.Sc.
Dissertation, Academy of Sciences. Uzbek SSR,Tashkent 1978.(in Russian).
Bushnell, W. R., and Rowell, J. B. (1981). Suppressors of defence reactions: a
model for roles in specificity.Phytopathology, 71:1012-1014.
Cruickshank, J. A., and Perrin, D.R. (1971). Studies of phytoalexins. XI. The
induction antimicrobial spectrum and chemical assay of phaseollin. Phytopa-
thofogy,70(3):209-229.
Dani, R. G. (1992). Biotechnological research of cotton- two decades in Soviet
retrospection. Advances in Plant Sciences(India),5(2): (In press).
Dyakov, Y. T. (1976).Membrane aspects of phytopathology. Obzor. MSX,WZZSE
ZSX, M.2. (in Russian).
Dyakov, Y. T. (1983).Physiological-biochemical mechanisms of resistance ofplants
to fungal diseases. Itogi Nauka i Texniki WZZTI, Zashita Rasyenyii, 1983,pp.
5-90. (in Russian).
Essenberg, M., Grover, P.M., and Cover, E.C. (1990).Accumulation of antibacte-
rial sesquiterpenoids in bacterially inoculated Gossypium leaves and cotyle-
dons. Phytochemistry29(10)3107-3113.
Faye, L., and Crispeels, M. (1987). Transport and processing of the glycosylated
precursor of concanavalin. A.in Jack-bean. PIanta 170(2):217-224.
Fraile, A., Garsia-Arenal., J., Garsia-Serrano, J. J., Sagastra, L. (1982). Toxicity
of phaseollin, phaseolliniso-flavan and kievitone to Botrytis cinerra. Phyto-
fogia Journal 102(2):161-169.
Hadwiger, L., and Schwochan, M. (1969).Host-resistance responses- an induction
hypothesis. Phytopathology 59(2):223-227.
Heath, M. C. (1981). A generalized concept of host-parasite specificity. Phytopa-
thology, 71~1121-1123.
Karimdjanov, A. K., Ismailov, A. I., Abdullaev, Z. S., Islambekov, S. Y.,Kamaev,
F. T., and Sadikov, A. S. (1976). Structure of gossyvertin-a new phytoalexin
of cotton. Ximiya Prip. Socd. 2:238-242.
Keen,N.T.,Sims, J. J., Erwin, D. C.,Rice, E., and Patridge, J. E. (1971).
6-a-Hydorxyphaseollin: an antifungal chemical induced in soyabean hypocot-
yls by Phytophtora megasperma var. soyae. Phytopathology,61(9): 1084-1089.
Wilt 159
KuC, J., and Rush, Y. S. (1985). Phytoalexins. Archieves Biochemistry et Biophysics

236(2):455-472.
Lyutsik, M. H.,Panasyuk, E. N., and Lyutsik, A. D. (1981). Lectines. Lyvov Pub.,
Moscow, p. 153.
Malysheva, K. M., and Zeltser, S. S. (1968). Protein-lipid-polysaccharide complex
cultural fluid and the mycelii of the fungus Verticillium dahliae, inducer of
Verticilliumwilt of cotton. Doklady AkademiNauk, SSSR, 179(1):231-234 (in
Russian).
Markov, E. Y., and Xavkin, E. V. (1983). Plant lectines: proposed functions.
Phisiologia Rastyenii,30(5):852-865. (in Russian).
Metlitski, L. V. (1976). Phytoimmunity. Molecular mechanisms. Nauka Publishers,
Moscow. (in Russian). 50 pp.
Metlitski, L. V., and Ozeretskovskaya, 0 . L. (1973). Phytoalexins.cNauka Publish-
ers, Moscow. (in Russian). 176 pp.
Metlitski, L. V., and Ozeretskovskaya, 0 . L. (1985). Howplants defend themselves
against diseases.Nauka Publishers, Moscow.P 114. (in Russian).
Metlitski, L. V., Lyubimova, N. V., and Muxamedova, R. A. (1971). Phytoalexin
activity and wilt tolerance of cotton. Izvestii Akademy Nauk, SSSR, Seria
Biologia 2:72-75.
Metlitski, L. V., Ozeretskovskaya, 0. L., Yurganova, A. L.,Savelyeva, 0. H.,
Chalova, L. I., and Dyakov, Y. T. (1976). Phytoalexin induction in potato
tubers with a metabolite of the fungus Phytopthra infeSans (Mont) de Bary.
Doklady Akadmii Nauk, SSSR, 226(5):1217-1220. (in Russian).
Muller, K. O., and Borger, H. (1940). Experimentalle untersuchungen uber die
phytophora-Resistenz der kartoffel. Arb. Biol. Reichsanst. Land Forstwirt.,
Berlin Dehlem23(2):189-231. (in German).
Muxamedova, R. A., and Turakulov, Y. X. (1974). Phytoalexin activity of cotton
and its wilt tolerance. Uzbeksi BilogicheskiZhurnal65-7 (in Russian).
Nuritdinova, X. V. (1988). Functionalpropertiesof exoosomose of amino acidsand
lectines during the expression of wilt disease of cotton. Ph.D. Dissertation,
Academy of Sciences. Uzbek SSR,Tashkent. (in Russian).
Sadykov, A. S., Metlitsky, L. V., Karimdjanov, A. K., Ismailov, A. I., Muxamed-
ova, P. A., Avazkhodjaev, M. X., and Kamaev, F. G. (1974). Isohemigossy-
pol: phytoalexin of cotton. Doklady Akademy Nauk, SSSR 281(6):1274-1276
(in Russian).
Sequeira, L. (1978). Lectins and their role in host-pathogen specificity. Ann. Rev.
Phytopathol. 16453-481.
Stewart, J. McD. (1991). Biotechnology of Cotton ICAC Review articles on cotton
production, No. 3. CAB International, Wallingford.
Sun, J. T., Essenberg, M., and Melcher, U. (1989). Phytoactivated DNA nicking,
enzyme inactivation, and bacterial inhibition by sesquiterpenoid phytoalexins
from cotton. Mol. Plant Microbe Interact.2(3):139-147.
Terekhova, V. A., and Dyakov, Y. T. (1980). Isolation of metabolites of Phytop-
thora infestans participating in the reaction of the parasite with potato under
the influence of osmotic shock. Bull. Moskovskova Obshyestva isp. Prirody,
Otdel Biologii85(5):68-77 (in Russian).
Tomiyama, K. (1970). Phytoalexins. Chem. Regul. Plants5(2):105-115.

Yatsenko-Xmelevski. A. A. (1954). Aspects and Methods of Anatomical Research
of Wood. Academy of Sciences, USSR (in Russian).
Zaki, 0. J., Keen, N. T., and Erwin, D. C. (1972). Implication of vergosin and
hemigossypol in the resistance of cotton to Verticillium albo-atrum. Phytopa-
thology 62(12):1406-1407.
8
Cotton (Gossypium hirsutum) Strategies
of Defense Expression
Accumulation of Unique Aspergillus flaws
Control Substances
H. J.Zeringue, Jr.
Southern Regional Research Center, Agricultural Research Service,
U.S. Department of Agriculture, New Orleans, Louisiana
1. INTRODUCTION
Toxigenic strains of the fungi Aspergillusflaws Link ex Fries and Aspergil-

lusparasiticus Speare produce secondary toxic metabolites called aflatoxins
(Detroy et al., 1971). These toxins, particularly aflatoxinBI, have the dis-
tinction of being the most carcinogenic of allnatural products (Groopman
et al., 1981). Much research effort has been focused on aflatoxins since
elevated levels ofthe toxins are found in pre-and postharvest crops. Inthe
United States a guideline of 20 ppb is the minimum aflatoxin allowed by
the Food and Drug Administration (FDA)for interstate shipment of foods
and feeds.
Currently, good farming practices have produced some positive results
in control of aflatoxin contamination of cottonseed in the field. These
include 1.) irrigation during periodsof drought (the fungusseems to thrive
during drought conditions),2.) use of insecticidesat marked stages of cot-
ton boll development to prevent fungal entry through insect wounds, and
3.) the use of crop varieties that exhibit some resistanceto A . f l a w infec-
s
tion or aflatoxin accumulation.
The basic planfor elimination or control of aflatoxin in cottonseed has
involved research on natural products in the cotton plant that inhibit A .
f l a w s or toxin production. Identification of natural inhibitors and their
mechanisms of production can be used to enhance aflatoxin inhibitory
traits in cotton plants through either classical plant breedingor contempo-
rary molecular engineering techniques.
161
162 Zeringue
II. METHODOLOGY
A. Fungi and Elicitor Preparation
In most cases, A. Javus SRRC I O 0 0 was maintained on potato dextrose
agar platesat 29OC. Fungal sporeswere inoculated on a defined medium of
Adye and Mateles (1984) contained in 2.8 liter Fernback culture flasks.
Inculated mediawere keptat 29OC without shakingfor 10-14 days. Mycelia
were washed with distilled H20,homogenized with 0.1M phosphate buffer,
filtered, defatted with CHCL,-MeOH (1:l v/v), washed with MqCO, and
air-dried (Zeringue et al., 1982). Mycelia were resuspended in HzO, auto-
claved for 3 hr at 12I0C, filteredonfilter paper, and the filtrate was
concentrated by vacuumdistillation,dialyzedagainst H20 at 2OC, and
lyophilized. The hot-water-soluble mycelial extract elicitor was prepared
from a IO-day-old culture and contained 44.6% protein and 37.5% carbo-
hydrate.
B. Plants and Treatments of Leaf Disks and Cotton Boll
Disks and TLC Quantitation
Acala SJ-2 cotton plants were grown under greenhouse conditions and were
at least 2 months old at the time of leaf treatment. Both leaves and cotton
boll surfaces were slightly abraded with 6- to 8-mm diameter wounds pro-
duced by carborundum disks or by a scalpel blade. Two areas of fifth or
sixth true leaves per plant were scratched in areas midway between the
outer margin of the leaf to the medium. Similar wounds were produced on
developing cotton boll carpel surfaces in areas between the suture lines of
the boll. Usually 10 pg of the mycelial extractin 10 p1 sterile, distilled H 2 0
was applied to the wounded areas, and 2 days after treatment wounded
areas wereexcisedwith14- to 18-mm-diametercorkborers.Leafdisks
(20 disks at a time) were placed in 40 m1 50% aqueous EtOH and were
vacuum-infiltrated; the flaskscontaining the disksweresubsequently
placed on a reciprocating shaker for 12 hr at room temperature. Cotton
boll disks (24 disks) were vacuum-infiltrated with 40 m1 methyl alcohol-
acetone (4:lv/v). The disks were removedfrom the alcohol extractand the
extract was concentrated to dryness byrotary evaporation under vacuumat
5OOC. The residue was dissolved in MeOHto produce a 5% (w/v) solution
whichwas spotted on silica gel thin-layer chromatography (TLC) plates
and developed in CH3CN-CHC1, :4 (1v/v). Plates were spotted with treated
leaf or boll extracts and standards and scanned densitometrically at 365
nm with a Schoeffelspectrodensitometer(Model SD 3000) to obtain a
semiquantitative analysis of the fluorescent spots. The Rf’ values of the
compounds in plant extracts weredeterminedin three different solvent
systems (Table1).
Cotton Strategies of Defense Expression 163
Table 1 TLC of Plant Extracts and Standards in Three

Solvent Systems
R' values
Band color
(365 nm) System In System I I ~ System I I I ~
Y (0.26)d 0.26 0.22 (0.22)d 0.28 (0.28)d

B 0.38 (0.39)' 0.35 (0.34)e 0.10 (0.11)c
Y 0.48 0.46 0.42
Y (0.58)' 0.58 0.55 (0.56)' 0.45 (0.46)'
Y-G 0.64 0.65 0.55
'CHCI,-MQCO-HCOOH (80 19: 1).
b ~ (M).
~ ~ - ~ ~ ~ ~ ,
CHexane-EtOAc-MeOH (60:40:1).
dLacinilene C standard.
'Scopoletin standard.
'Lacinilene C 7-methyl ether standard.
Y. yellow; B, blue; Y-G,yellow-green.
Source: Data from Zeringue (1984).
C. Collection and Identification ofMajor Volatiles Emitted

from Cotton Leaves
Volatiles were trapped on glassTenaxcolumns(0.1-gTenax GC 60-80
mesh) packed between glass wool in tubes (3/8 x 3 in.). Air streams were
passed over the plant material and volatiles were collectedfor 30 min. The
Tenax tube was loaded into an external inlet apparatus (Scientific Instru-
ment Service, River Ridge, LA) interfaced with a Finnegan MAT GUMS
400Series instrument(Legendreetal.1979).Thevolatiles were heat-
desorbed onto a 50-m SE-54 column.The temperature program usedwas 3
min from -3OOC to 3OoC (to trap the volatiles at the head of the column)
followed by 30-150°C at 2.5O/min and then 150-250°C at 10°/min. The
peaks on the reconstructed ion chromatogram were tentatively identifiedby
comparisonwith the Finnegan NBS library and quantitated using the
ARICQ program.
111. PHYTOALEXINS
A. Induction of Phytoalexins
Early research at the Southern Regional Research Center (SRRC) showed
that five cotton autofluorescent (365 nm illumination) phytoalexins could
be induced in the cotton leaf or the developing cotton boll by a cell-free,
hot-water-soluble, mycelial extractof A. flaws (Zeringue, 1984), including
164 Zeringue
1.)the sesquiterpenoid phytoalexins (Fig.l), 2,7-dihydroxycadalene (DHC)

and 2-hydroxy-7-methoxycadalene(DHMC); 2.) their oxidation products
lacinilene C (LAC) and lacinilene C 7-methyl ether (LACME);and 3.) the
coumarinphytoalexinscopoletin(SCOP).Thedenotedsubstanceswere
induced when cell-free mycelial extracts were applied to artificially wounded
areas on the cotton leaf or on the surfaces of the developing cotton boll
(Zeringue, 1984, 1987, 1988). Fungal extracts appeared to elicit the produc-
tion of the phytoalexins by the redirection of the terpenoid biosynthetic
pathways inthe cotton plant. A similar modificationof terpenoid metabo-
lism was demonstrated in potato tubers (Shih et al., 1973; Shih and KuC,
1973).
A 10-day, time-course studywas conducted on the induction of DHC,
DHMC, LAC, and LACME in AcalaSJ-2 cotton leaves that were slightly
abraded by a 6-mm-diameterfine, carborundum sandpaperdisk and
treated with 10 pg of a cell-free hot-water-soluble extract of a toxigenic
strain of A. flavus (Table 2) (Zeringue, 1987a). Initial accumulation ofthe
compounds occurred 2 days after treatment with elevated levels on day 6
posttreatment. A gradual decline ofthe compounds occurredafter the peak
periods. Control wounded leaves also showed increasesof the compounds
on day 6 with a gradual decline toward the end of the test period. To
characterize the distribution of induced phytoalexins in reference to the
6-mm-diameter wounded, treated areas, cork borers of various diameters
were used to excise rings encircling the wounded areas (Table 3). Afterthe
second day oftreatment, the greatest concentrationof induced compounds
Lacinilene C (l),2,7-Dihydroxycadalene (21,

Lacinilene C 7-methyl ether (3).
and 2-Hydroxy-7-methoxycadalene (4)
Figure 1 Lacinilene C (l), 2,7-dihydroxycadalene (2), lacinileneC 7-methyl ether
(3),
and 2-hydroxy-7-methoxycadalene(4).
166 Zeringue
Table 3 Percentage Distribution After 2 and 5 Days of Induced Components in

&mm-diameter Wounded/Fwngal-Treated Areas and in Areas Encircling the 6-mm
Areas in the Cotton Leaf
Outer diameter (mm)
Components 9 6 11 14 16 21
2,7-Dihydroxycadalene
29.1" 58.3 12.6 0 0 0
86.2b 13.8 0 0 0 0
Lacinilene C 18.9" 77.0 4.1 0 0 0
92.6b 7.4 0 0 0 0
2-Hydroxy-7-methoxycadalene 34.5" 58.0 4.6 2.1 0.9 0
81.5b 18.5 0 0 0 0
Lacinilene C 7-methyl ether 42.7' 53.1 4.2 0 0 0
94. lb 5.9 0 0 0 0
'Two days after treatment.
bFive days after treatment.
Source:Data from Zeringue (1987a).
was localized in a 3-mm area around the 6-mm wounded/fungal-treated

area (Table 3). The 6-mm-diameter area was brownish in color and the
adjacent 3-mm area had a light yellow-green appearance. Five days after
treatment the four induced compounds were found predominately in the
6-mm-diameterwoundedkreated area; this area plus the adjacent encircling
3-mm area were both brownish in color. The resultsare in agreement with
Hargreaves and Baily (1978) whodemonstrated that the phytoalexin phase-
ollin from the french bean (Phaseolusvulgaris) and other isoflavonoid
phytoalexins are probably synthesized in tissues around necrotic cells and
are absorbed and accumulate in the dead tissue.
B. Phytotoxic Effects
To understand the cyclic production of the induced phytoalexins incotton
leaves, a sensitive bioassay was developed that utilized the aquatic macro-
phyte Lemna minor L. (Einhelling et al., 1985; Zeringue, 1987a).LAC and
LACME from excised leaf disks exhibited phytotoxic properties in the L.
minor assay (Table 3). Both lacinilenes inhibited the growth of L. minor,
but LAC produced the greatest inhibition. Both compounds produced root
discoloration and some frond bleaching occurred with LAC. The phyto-
toxic properties of the two lacinilenes may explain some of the results
described in Table4. It is possible that cell death in the cotton leaf releases
constitutive material (an endogenous elicitor) which initiates phytoalexin
biosynthesis and the accumulation of the compounds shown in Table 1
Table 4 SeparatedComponents*ofCottonLeaf Disks andControlsAssayed

After 1 Week of Incubation with L. minor
No. of
frondsafter Dry wt of
No. of
leaf 1 week of
incubation L. minor as
disks percentage of
Component control ControlTreated
extractsb
Lacinilene C 1 24' 32 77
2 41
1 l' 28
27 27 3 8d
4 1Id
28 32
5 27 Sd 30
Lacinilene C 7-
methyl ether 1 28 30 84
22' 2
3 63 2v 32
30 4 20' 62
27 5 18' 64
'Separated by preparativeTLC.
bAliquots prepared from SO-leaf disks
to represent the amountof component in 1-5 leaf disks.
aoots have a brownish discoloration.
dSome bleachingof fronds resulted.
Source: Data from Zeringue (1987a).
margraves and Bailey, 1978). This reasoning could also explainthe induc-
tion of the compounds in wounded control leaves on days5-6 after wound-
ing. The synergistic effect produced by a fungal extractand an endogenous
elicitor may explain the peak accumulations of the compounds in treated
leaf extractson days 2 and 6 (Table 1). Synergistic elicitor effects have been
observed betweenthe 6-glucan elicitor ofPhytophora megasperma and the
endogenous elicitor of soybean tissue (Darvilland Albersheim, 1984). Simi-
lar synergism has been observed between 6-glucan and fatty acid elicitorsin
potato (Block and KuC, 1983; Kuranty and Osman, 1983; Preisig and KuC,
1983).
The lacinilene and cadalene precursorsare induced in the green tissues
of the cotton plant in contrast to the terpenoid phytoalexins desoxyhemi-
gossypol and hemigossypol, together with their methyl esters, which are
formed in germinating seed, youngroots, hypocotyls, xylem vessels, cam-
bial tissues, andcotton boll endocarp (Halloin and Bell, 1979; Veech, 1978;
Hunter et al., 1978; Bell and Stipanovic, 1977). Varied levels of A. flaws
inhibition were observed in direct bioautography on TLC plates that con-
tained the five separated induced compounds (unpublished). Essenberg et
168 Zeringue
al. (1982) demonstrated that DHC and LAC are bacteriostatic and accumu-
late in leaves and cotyledons of blight-resistant lines of cotton when the
tissues weretreated with a bacterial suspension of Xanthomonas campestris
pv. malvacearum. Halloin and Greenblatt (1982) found accumulation of
lacinilenes and cadalenes in cotton boll tissues after puncture inoculation
of bolls with Diplodia gossypina; the bolls exhibited a bright, yellowish
fluorescence under long-waveUV light and also resistance to infection by
Diplodia gossypina.
C. Wound Response and Fungal Extract Treatment

Initial true leaves of Acala SJ-2 cotton plants wounded by gentleabrasion
resulted in accumulations of ferulic acid (4-hydroxy-3-methoxyci~amic
acid) detected on the third and fourth days after treatment. When similar
wounds were treated with cell-free extracts ofA. flavus, elevated levels of
scopoletin (6-methoxy-7-hydroxycoumarin)were produced after 1 day of
treatment and continuing for 7 days (Zeringue et al.,1985). Elevated levels
of scopoletin were not detected in leaves that were only wounded, and
enhanced quantities of ferulic acid were not observed in leaf tissue after
wounding and exposure to the fungal extract (Table 5).
Increased ferulic acid in wounded cotton leaves and scopoletin in the
Table 5 Ferulic Acid and Scopoletin Concentrations' in Leaf Disks from the First True
Leaf of Acala SJ-2 Cotton Plants That Were Wounded and Wounded-Plus-Treated with
Cell-Free A. flavus Extract
Wounded
Wounded plus treated
Ferulic acid extract
with fungal
Day after
wounding Cis trans Scopoletin Scopoletin
Ferulic acid
1 tb t 2.4 f 0.12 t 179.6 f 2.64
2 t t 2.1 f 0.20 t 61.8 f 3.28
3 262 f 5.4' 1268 f: 6.2 1.3 f 0.30 t 53.5 f 2.73
4 47 f 2.3 472 f 6.7 1.0 f 0.03 t 36.2 f 1.66
5 t t t t 27.4 f 4.84
6 1 1 t t 18.1 f 2.13
7 1 1 t t 14.3 f 2.55
'Data reported, in &g fresh wt, are from analyses of extracts of 80 leaf disks for each day (20
wounded leaves and20 wounded-plus-treated leaves). Each leaf disk was15 mm in diameter, includ-
ing a wounded area6 mm in diameter. These data represent the meanof three test runs.
%ace amounts, less than 1 pg/g fresh wt.
'Mean * SE.
Source: Data from Zeringue et al.
(1985).
Strategies
Cotton of Defense Expression 169
fungal extract-treated, wounded leaves indicates either an increased synthe-

sis of the compounds or an active transport of the substances from cells
surrounding the wounded area. Apparently ferulic acid in the wounded
leaves is derivedfrom cinnamic acid andother intermediates of the phenyl-
propanoid pathway leadingto the formation of lignin. It is possiblethat the
rapid accumulation of scopoletin after a day of wounding and fungal ex-
tract treatment results from a redirection of cinnamic acid in the phenylpro-
panoid metabolic pathway via ortho-hydroxyl derivatives and lactonization
to the production of the coumarin scopoletin (Rhodes and Wooltorton,
1978). Accumulation ofscopoletin and itsglyconescopolinhavebeen
shown in potatoes and tobacco infected with fungi, bacteria, and viruses
(May et al., 1963; Tanguy and Martin, 1972; Sequeria, 1969; Huges and
Swain, 1960). Scopoletin acts as a stimulator of IAA oxidase (Imbert and
Wilson, 1970), and peroxidase activities (Schafer et al.,1971). Accumulated
scopoletin also acts as an inhibitor of catalase activity in tobacco infected
by Peronospora tabacina (Hochberg and Cohen, 1977). The fact that sco-
poletin can alter the host enzyme systems may be of significance in the
plant’s defense system.
D. Carpel Wall Phytoalexinsin the Developing Cotton Boll

Developing cotton bolls of Deltapine 61 that had been grown under con-
trolled conditionswere artificially woundedon the carpel surfacesand sub-
sequently treated with the hot water-soluble mycelial extract of A. flaws at
weekly intervals for 8 weeks postanthesis. Two days after treatment, the
bolls were harvested and disks containingthe treated surfaces were excised
and extracted to determine induction of DHC, DHMC, LAC, LACME,
and SCOP (Zeringue, 1988). Concentrations of the lacinilenes and cada-
lenes increased over the 8-week testing period and peaked either at the
seventh or eighthweek postanthesis boll age (Table 5). However, scopoletin
concentrations were just the opposite with higher concentrations in extracts
of younger bollsthat peaked at 3 weeks postanthesis.
The highest concentrations of lacinileneand cadalene phytoalexins oc-
curred during the period of boll opening (41-49 days postanthesis; Table
6). Lee (1988) inoculated cotton boll sutures with A. flaws spores at the
initiation of boll opening, harvested the bollsafter 2 or 4 weeks, and found
no aflatoxin contamination in the seeds of the treated bolls. It is possible
that induced phytoalexins couldbe translocated to deeper tissues withinthe
boll either passively or actively. The lacinilene and cadalene phytoalexins
are slightly water-soluble (Stipanovicand Wakelyn, 1974), and a penetrat-
ing carpel wall wound could allow dew or rain to transfer the phytoalexins
into deeper tissues of the bollto effect a more widespread defense.
U
C
Cotton Strategiesof Defense Expression 171
E. Environmental Stress Effectson Phytoalexin Production

The concentrations of DHC, DHMC, LAC,LACME, and SCOP were
determined in cotton leaves treated with the hot water-soluble mycelial
extract of A. flaws under a light regime of 14 hr light110 hr dark and
temperature cyclesof25/20°,30/25O,35/30°, and 40/35OC (Zeringue,
1990). The highest concentrations of the sesquiterpenoid and coumarin
phytoalexins were produced at temperaturecycles of 30125 O and 2 daysafter
treatment withthe fungal elicitor (Table 7). Temperature regimes lower than
30/25OC resulted ina 35% reduction in induced phytoalexins while the high-
est temperaturecycle tested (40/35 “C) yielded a 43% reduction.
Concentrations of the individual phytoalexins were measuredcotton in
leaves treated with the fungal elicitorand maintained at different levels of
plant moisture stress (PMS). PMS values less than - 11.7 bars resulted in
decreased amounts of the five induced compounds (Jable 8). Even slight
leaf wilt (- 15.5 bars) at the time of fungal elicitor treatment after 2 days
(-24.3 bars) resulted in45-58% reduction of the five phytoalexins.
From these findingsit is apparent that an “ideal” set of environmental
conditions of temperature and moisture is necessary for the maximum in-
duction of the cotton leaf phytoalexins. Cotton plants grown in hot, dry,
desert regions may not favor maximum production of the phytoalexins.
Inability of host plants to produce defensive chemicals quickly when at-
tacked by A. flavus may be related to the elevated incidence of aflatoxin
contamination of cottonseed grown in desert regions.
IV. VOLATILE ELICITORS A N D GASEOUS PHYTOALEXINS

Baldwin and Schultz (1983) damaged leaves of potted poplar(Populous X
euroarnericanu) ramet and sugar maple ( h e r saccharum) seedlings and
demonstrated increased concentrations of phenolic compounds inthe dam-
aged leaves as well as in the leaves of undamaged plants sharing the same
enclosure. Rhodes (1982) found that Stika willow (Salix) trees attackedby
tent caterpillars (Mulacosoma California pluviale) and nearby unattacked
control treesexhibitedalteredleafquality.Fallwebworms (Hyphantria
cunea) fed on the attacked leaves grew more slowlythan those fed on the
leaves from unattacked willows. These studies suggested that an airborne
signal from damaged tree tissues stimulated biochemical changes in neigh-
bouring undamaged trees and influenced the feeding and growth of phy-
tophagous insects. Initial SRRC tests with volatiles involved interactions
between A. Jaws and host cotton plants and an assessment of a cotton
plant’s defense system relative to airborne signals. Zeringue (1987b) ex-
posed cotton leaves for 7 days to volatile chemicals originating from1.) A.
flaws-infected cotton leaves, 2.) A. fravus cultures, or 3.) mechanically
172 Zeringue
174 Zeringue
damaged cotton leaves. It was observed that volatiles from A. flavus-in-

fected leaves caused significant increases(52% and 34%) in phloroglucinol-
reactive compounds, expressed as “gossypol equivalents,’’ in wounded and
undamaged leaves, respectively (Table 9). Bell (1967) induced accumulation
of phloroglucinol-reactive compounds in tissues of cotton by inoculating
with conidia of Verticillium alboatrum or with sporangiosporesof Rhizo-
pus nigricans. He described the accumulation of the phloroglucinol-reactive
phenolic compounds in terms of gossypol equivalents and discussed their
accumulation relativeto that of phytoalexin production. Subsequent stud-
ies at SRRC demonstrated that heliocide Hz (Czs terpenoidaldehyde, a
natural cotton insecticide) was one of the predominant products formed in
the volatile recipient (wounded)and nonwounded cotton leaves. Heliocide
Hz is formed naturally by a Diels-Alder addition of hemigossypolone and
myrcene in the pigment glands of the cotton plant, Stipanovic and Bell
(1977). ZeringueandMcCormick (1989) found that myrcenerepresents
17.7% of the total volatiles in nonwounded Acala SJ-2 leaves and 24.8% of
the total volatiles in wounded Acala SJ-2 leaves. Leaf damage caused by
penetrating hyphae of A. fravus might have released myrcene from the
pigment glands, and its subsequent reaction with hemigossypolone could
have increasedthe heliocide Hz level inthe volatile recipient leaves.
To elucidate the effects of cotton leaf volatiles on A. fravus cultures,
microbial-freecompressed air waspassed continuously (2 and 7 days)
Table 9 Effects of Volatile Chemicals Emitted by Infected or Wounded Cotton

Leaves or by A. flavus Cultures on the Gossypol Equivalent Content of Receptor
Cotton Leaves after 7 Days Exposure
Treatment condition means” innm/mol gossypol

equivalentdg dry leaf tissue
Source
Receptor
Control
leaves’
leaves
leaves
receptor
Volatile
source
Volatile
337
Normal
leaves
leaves
Cut 225 a‘ b 197 b
A. flavus inoculatedleaves
Woundedleaves 1003 c 702 c 337 d
A. flovus inoculatedleaves
Normal
leaves
481
e 339
225f g
A. flavus
295leaves
Normal
cultures h h 229
i 317A. flavus
i 401
Wounded
cultures
leaves
’There were at least three replicatesof each treatment condition, and the data represent the
means of the analysesof three subsamples in each treatment.
bControl leaves treated the same as receptor leaves, except they received only filtered com-
pressed air.
’Means in separatecolumns in each treatment followed by the same letter are not significantly
different atP = 0.05 accordingto Duncan’s Multiple Range Test.
Source: Data from Zeringue (1987b).
Table 10 Effects of 2- and 7-DayIncubation of A.

flavus in Contact with Cotton Leaf Volatiles
Drywtasa
percent of
cultivar
Cotton Time control
2 days
8160 NW" 90.7 9.1C
8160 Wb 32.4 f 8.6
SJ-2 NW 115.3 f 6.7
SJ-2 W 32.4 f 3.2
7 days
8160 NW 100.6 f 3.3
8160 W 96.6 f 4.0
S5-2 NW 179.3 f 5.6
SJ-2 W 178.3 i 1.9
'Nonwounded.
wounded.
'Standard error of mean of three separate experiments.
Source: Data from Zeringue and McCormick (1989).
through enclosed systems containing wounded or nonwounded leaves of

glandular or glandless cotton; the emitted volatiles were bubbled through
liquid culturesof A. flaws (Zeringue and McCormick, 1989). After 2 days
of incubation (Tablelo), volatiles from wounded, glandular,and glandless
cotton leaves retarded the growth of A. flavus. After 7-day incubations
(Table 9) fungal growthwas stimulated by volatiles from woundedor non-
wounded glanded cotton leaves, but not from either type of glandless cot-
ton leaves. The results showed that wounded cotton leaves release antimi-
crobialvolatiles from both wounded SJ-2 and 8160 after 2 days. In
addition, SJ-2 wounded and SJ-2 nonwounded volatiles stimulated fungal
growth after 7 days; the substances apparentlywere produced from lysigen-
ous pigment glands in the leaves since the stimulatory effect was not ob-
served in cultures receiving volatiles derived
from glandless 8160 leaves.
To characterize the active componentsfrom wounded leaves, volatiles
were trapped on Tenax collection tubes and volatile profiles were deter-
mined by direct injectiodgas chromatography/mass spectral analysisat 2-
and 7-day periods (Zeringue and McCormick, 1989). Over 90 compounds
were identified by this method. Purified compounds of selected identified
cotton leaf-derived volatiles were assayedin culture to determine their bio-
activity. Of the individual volatile components tested,C&, alkenals, espe-
ciallytrans-2-hexena1,exhibited the maximuminhibitoryeffect on the
growth of fungus (Table 11). Stimulatory effectsof the individual volatile
176 Zeringue
0 0 0 0 0 0 m 0 0 0 m m
iiiiiiiiiiiiiiii-tlii~ii
o o o o o o u o o o ~ d
d
n e
"
a
178 Zeringue
components tested were not as pronounced as the inhibitory effects. Un-

branched cS-Cl2 alkanals, 2- and 3-CSalkanones, and a- and &pinene
stimulated the growth ofA.flaws.
The bioactive alkenals are breakdown products'of the unsaturated fatty
acids linoleic and linolenic acids with their concentrations being greatly
increased by mechanical damage ofplant tissue (Lyr and Banasiak, 1983).
The key enzymein alkenal biosynthesis isa membrane-bound lipoxygenase
which is believedto be locatedboth in chloroplastsand mitochondria (Sek-
iya et al., 1979; MacLeod and Pikk, 1979; Hatanaka et al., 1978). Because
the alkenals are highly inhibitoryto A.flaws growth, they maybe consid-
ered as gaseous phytoalexins because of their mode of production and
antifungal effects.
Since C&o alkenals are released from damaged cotton leaves, it seems
logical to ask if newly released volatiles couldeffect a defense expression in
the same plant or in neighboring plants. In order to address the question
pertinent experiments were conducted. Microbial-free, compressedair was
designed to carry individual c 6 9 C7, CS, C9, and Cloalkenals and alkanals
into enclosedsystemscontainingartificiallywoundedandnonwounded
Acala SJ-2 developing cotton bolls (Zeringue, 1991). Two days after treat-
ment, diskswere excised from the treated cotton boll surfacesand extracted
to determine the induction of DHC, DHMC, LAC, LACME, and SCOP.
Results showed that all the alkenals produced elevated levels of the five
induced phytoalexins in artificially wounded, developing cotton bolls. All
tested volatile alkanals produced slightly elevated levels of SCOP in the
artificiallywounded cotton bollwhencompared to woundedcontrols.
Wounding or tissue damage appears necessaryfor c6-Clo alkenals to func-
tion as active "volatile elicitors''
for phytoalexin production.
Results of SRRC studies demonstrate that C&o alkenals function
both as volatile elicitors and as gaseous phytoalexins in the cotton plant.
Airborne signals consisting of c69 C7,CS, C9, andCloalkenals emittedfrom
wounded cotton bolls or wounded cotton leaves can inducethe production
of the lacinilene, cadalene, and scopoletin phytoalexins on the carpel sur-
faces of the cotton bolls on the same plant or in bolls of nearby plants
(Zeringue,1991). It also has been demonstrated that damage of cotton
leaves or cotton bolls can initiate production of the c6-cIo alkenals, the
fungitoxic gaseous phytoalexins. Volatile compounds appear to represent a
specialized niche in basic cotton plant defense.
V. EPILOG
Research at SRRC will continueto focus on plant-derived metabolites that

inhibit the growth of toxigenicstrains of Aspergillus sp. or inhibit aflatoxin
Cotton Strategies
Expression
of Defense 179
biosynthesis. This brief chapter has described several strategies that the
cotton plant may utilize in defense against Aspergiffussp. It is apparent
that induction of a combination of defense mechanisms in cotton plants
will protect cottonseed from aflatoxin contamination. Elucidation of the
defense systems will provide a basis for development of commercialcotton
cultivars with enhanced resistance to A. fravus and aflatoxin.
REFERENCES
Adye, J., and Mateles, R. I. (1964). Incorporation of labelled compounds into
aflatoxins. Biochimica and Biophysica Acta 86418420.
Baldwin, I. T., and Schultz, J. C. (1983). Rapidchangesin tree leafchemistry
induced by damage: Evidencefor communication between plants.Science 221:
277-279.
Bell, A. A. (1967). Formation of gossypol in infected or chemically imtated tissues
of Gossypium Species.Phytopathology 57:756-764.
Bell, A. A., and Stipanovic, R. D. (1977). Biochemistry of disease and pest resis-
tance in cotton. Mycopathologia 65:91-106.
Block, G.B., and KuC, J. (1983). Elicitation of phytoalexins by arachidonic and
eicosapentaenoic acids: a host survey. 75th Annual Meeting AmericanPhyto-
pathological Society, Ames,IA, Paper No.A486.
Darvill, A. G., and Albersheim, P. (1984). Phytoalexins and their elicitors-a de-
fense against microbial infection in plants. Ann. Rev. Plant Physiol. 35243-
275.
Detroy, R. W., Lillehoj, E. B., and Ciegler, A. (1971). Aflatoxin and related com-
pounds. Microbial Toxins, Vol. 6. (A.Ciegler, S. Kadis, S. J. Ajl, eds.),
Einhelling, F. A., Leather, G. R., and Hobbs, L. L. (1985). Use of Lemma minor
L. as a bioassay in allelopathy.J. Chem. Ecol.11:65-72.
Essenberg, M., Doherty, M. &A., Hamilton, B. K., Henning, V. T., Cover, E.
C., McFaul, S. J., and Johnson, W. M. (1982). Identification and effects on
Xanthornonas campestris pv. malvacearum, of two phytoalexins from leaves
and cotyledons of resistant cotton. Phytopathology 72: 1349-1356.
Groopman, J. D., Groy, R. G., and Wogan, G . N. (1981). In vitro reactions of
aflatoxin B,-aducted DNA. Proc. Natl. Acad. Sci. USA 78:5445.
Halloin, J. M., and Bell,A.A. (1979). Production ofnonglandularterpenoid
aldehydes within diseased seeds and cotyledons of Gossypium hirsutum. J.
Agric. Food Chem.221407-1409.
Halloin, J. M., and Greenblatt, G. A. (1982). Abstract, Beltwide Cotton Production
Research Conference,January 3-7, 1982, Las Vegas, p. 46.
Hargreaves, J. A., and Bailey, J. A. (1978). Phytoalexin production by hypocotyls
of Phaseolus vulgaris in response to constitutive metabolites release by dam-
aged bean cells.Physiol. Plant Pathol.13:89-100.
Hatanaka, A., Sebiya, J., andKayiwara, T. (1978). Distribution of an enzyme
system producing cis-3-hexenal and n-hexanal from linolenic and linoleic acid
in some plants.Phytochemistry 12869-872.
180 Zeringue
Hochberg, M., and Cohen, Y. (1977). Scopoletin-inducedcatalaseinhibitionin

tobacco leaves infected by Peronospora tabacinaAdam. Israel Journalof Bot-
any 2648-52.
Hughes, J. C., and Swain, T. (1960). Scopolin production inpotato tubers infected
with Phytophthora infestans. Phytopathology 50398-402.
Hunter, R. E., Halloin, J. M., Veech, J. A., and Carter, W. W. (1978). Terpenoid
accumulation in hypocotyls of cotton seedlings during aging and after infection
by Rhizoctonia solani. Phytopathology 68:347-350.
Imbert, M. P., and Wilson,L.A. (1970). Stimulatory and inhibitory effects of
scopoletin in IAA oxidase preparations from sweet potato. Phytopathology 9
1787-1794.
Kuranty, M. J., and Osman, S. F. (1983). Class distribution, fatty acid composition
and elicitor activity of Phytophthora infestans mycelial lipids. Physiological
Plant Pathology22:363-370.
Lee, L. S. (1988). Aflatoxin in Arizona cottonseed: lack of toxin formation follow-
ing Aspergillus flaws inoculation at sutures. Journal of the American Oil
Chemists’society 653127-128.
Legendre, M. G.,Fisher, G. S., Schuller, W. H., Dupuy, H. P., and Rayner, E. T.
(1979). Novel techniquefor the analysis of volatiles in aqueous and nonaqueous
systems. Journal of the American Oil Chemists Society 56552-555.
Lyr, H., and Banasiak, L. (1983). Alkenals, volatile defense substances in plants,
their properties and activities. Acta Phytopathology Academy Science Hun-
gary. 183-12.
MacLeod, A. I., and Pikk, H. E. (1979). Formation of (E)-hex-2-enal and (Z)-hex-
3-en-1-01 byfresh leaves ofBrassica oleracea. Journal of Agriculture and Food
Chemistry 22469-475.
Mayr, H.H., Diskus, A., and Beck,W. (1963). Accumulation of scopoletin in
tissue ofNicotiana tabacumL. infected by Peronospora tabacinaAdams. Phy-
topathologie Zeitschuft4295-97.
Preisig, C., and KuC, J. (1983). Regulations of the sesquiterpenoid eliciting activity
of C-20 fatty acids in potato by carbohydrate isolated from Phytophthora
infestans. 75th. Annual Meeting American Phytopathological Society,Ames,
IA., USA. Paper No. ASII.
Rhodes, J. M., and Wooltorton, L. S. C. (1978). The biosynthesisofphenolic
compounds in wounded plant storage tissues. In Biochemistry of wounded
plant tissues(Gunter Kohl ed.) Walterde Gruyter, Berlin andNew York.
Schafer, P., Wender, S. H., and Smith,E.C. (1971). Effectofscopoletin on
two anodic isoperoxidases isolated from tobacco tissue culture W 38. Plant
physiology 48232-233.
Sekiya, J., Kajiwara, J. T., and Hatanaka, A. (1979). Volatile C,-aldehyde forma-
tion via hydroperoxide from C,,-unsaturated fatty acids in etiolatedalfalfa and
cucumber seedings.Agriculture Biological Chemistry43:969-980.
Sequeira, L. (1969). Synthesis of scopolinand scopoletin in tobacco plants infected
by Pseudomonas solanacearum. Phytopathology 63:826-829.
Shih, M., Kric, J., and Williams, E.B. (1973). Suppression of steroid glycoalkaloid
accumulation as related to rishitin accumulation inpotato tubers. Phytopathol-
ogy 63~821-826.
Strategies
Cotton Expression
of Defense 181
Shih, M., and Kut, J. (1973). Incorporation of I4C from acetate and mevalonate
into rishitin and steroid glycoalkaloids by potato tuber slices inoculated with
Phytophtora infestans. Phytopathology 63:826-829.
Stipanovic, R. D.,Bell, A. A., O’Brien, D. H., and Lukefar, M. J. (1977). Heliocide
H,: an insecticidal sesquiterpenoidfrom cotton (Gossypium). Tetrahedron Let-
ters 6567-570.
Stipanovic, R. D.,and Wakelyn, P. J. (1974). Proceedings ACGIH Cotton Dust
Symposium. Atlanta, GA, November 1974,225.
Tanguy, J., and Martin, C. (1972). Phenolic compounds and the hypersensitivity
reaction in nicotiana tabacum L. infected with tobacco mosaic virus. Phyto-
chemistry I1 :19-28.
Veech, J. A. (1978). An apparent relationship between methoxy-substituted terpe-
noid aldehydesand the resistance ofcotton to Meloidogyne incognita. Nemato-
logica24:81-87.
Zeringue, H. J., Jr., Neucere, J. N., and Parrish, F. W. (1982). Hemagglutinins of
some Aspergilli. Biochemical Systematicsand Ecology 10217-220.
Zeringue, H. J., Jr. (1984). The accumulation of five fluorescentcompounds in the
cotton leaf induced by cell-free extracts ofAspergillusflavus. Phytochemistry
23( 11):2501-2503.
Zeringue, H. J., Jr., Conkerton, E. J., and Chapital, D. C. (1985). Canadian
Journal of Botany 632470-2472.
Zeringue, H. J., Jr. (1987a). A possible relationship between phytoalexin produc-
tion in the cotton leaf and a phytoxic response.Phytochemistry26:975-978.
Zeringue, H. J., Jr. (1987b). Changes in cotton leaf chemistry induced by volatile
elicitors. Phytochemistry 261357-1360.
Zeringue, H. J., Jr. (1988). Production of carpel wall phytoalexins inthe developing
cotton boll. Phytochemistry2Z3429-3431.
Zeringue, H. J., Jr., and McConnick, S. P. (1989). Relationships between cotton
leaf-derived volatiles and growth of A . flavus. Journal of the American Oil
Chemists Society66581-585.
Zeringue, H. J., Jr. (1990). Stress effects on cotton leaf phytoalexins elicited by
cell-free-myceliaextracts of Aspergillusflavus. Phytochemistry291789-1791.
Zeringue, H. J., Jr. (1992). Effects of Cs-Clo alkenals and alkanals on eliciting a
defense response in the developing cotton boll. Phytochemistry 31:2305-2308.
9
Sesquiterpenoid Phytoalexins
Synthesized in Cotton Leaves and
Cotyledons During the Hypersensitive
Response to Xanthomonas campestris
pv. malvacearum
Margaret Essenberg andMargaret L. Pierce
Oklahoma State University, Stillwater, Oklahoma
1. INTRODUCTION
Our work has focused on evaluation of the role of phytoalexins in heritable
resistance of cotton to bacterial blight. The most thorough studies of this
type by other workers have been with fungal diseases, and mostly with
isoflavonoid phytoalexins (Mansfield, 1982; Hahn et al., 1985). This study
evaluates the role of sesquiterpenoid phytoalexins in a bacterial leafspot
disease. Cotton (Gossypiurnspp.) andXanthomonas campestrispv. malva-
ceamrn are a genetically interesting system because of the race/cultivar
specificity due to a number of major host genes for resistance, each of
which confers resistanceto a different setof races (Brinkerhoff,1970). The
pathogen'sracephenotypeisdeterminedbyavirulencegenes(DeFeyter
and Gabriel, 1991), each of which contains several 102-nucleotide-base-pair
repeats of unknown function (D. W. Gabriel, personal communication).
Upland cotton (Gossypiurn hirsutumL.) lines possessing several resistance
genes have been developed whose resistance'to bacterial blight is so high
that no macroscopic symptomsor only pinpoint lesions are produced after
inoculation with most races of the pathogen (Brinkerhoff et al.,1984).
Objectives of our work have been to determine whether phytoalexins
contribute significantly to the high resistance of those cotton lines and to
learn how the phytoalexins are biosynthesized. In evaluating the role of
phytoalexins, we have taken an analytical approach by asking whether the
pathogen is exposed to inhibitory doses of the phytoalexins at the time it
becomes inhibited during incompatible interactions. Results of the study
183
184 Essenberg and Pierce
yielded information on whether phytoalexin accumulation is sufficient to

account for resistance. The difficulttask in such a study is determining
the
local phytoalexin concentration in contact with the pathogen. Our discov-
ery that the cellular localizationof sesquiterpenoid phytoalexins incotton
leaves and cotyledons could be directly observed by fluorescence micros-
copy of fresh, unstained tissue offered the opportunity to determine such
local concentrations.
II. MATERIALS AND METHODS
The various plant lines and bacterial strains are described in the articles
cited in the text. Cotton plants (Gossypium hirsutum L.) were grown in a
growth chamber with a 14-hr light/lO-hrdark cycle with 3OoC maximum/
19OC minimum temperatures (Pierce and Essenberg, 1987). Xanthomonas
campestris pv. malvacearum Smith (Dye) was cultured, inocula were pre-
pared as bacterialsuspensionsinsterile saturated CaCO, solution, and
cotyledons were inoculated by infiltration from a syringe (no needle) as
described by Pierce and Essenberg (1987). Entire leaves were infiltrated
with a pressure-drive spray of inoculum through stomata of their abaxial
surfaces. Bacterial growth trends planta
in were determined by diluting and
plating tissue homogenates (Essenbergal., et 1979a).
Sesquiterpenoids were extracted from frozen foliar tissues, worked up,
resolved by reverse-phase high-performance liquid chromatography (HPLC),
and quantitated by several methods (Essenberg et al., 1982, 1990; Pierce
and Essenberg, 1987). Bioassays of their antibacterial activities were pre-
formed on logarithmically growing cultures of X. campestris pv. malva-
cearum in defined medium in the dark (Essenberg et al., 1982) or with 300-
700nm radiation at 70 pEinsteins/m’/sec (Steidl, 1988). Photoactivated
nicking of DNA plasmids and inactivation of DNAseI and malate dehydro-
genase by 2,7-dihydroxycadalene (DHC) or lacinilene C (LC) were per-
formed with 300-700 nm radiation at 630 pEinsteins/m’/sec and detected
as described by Sun et al. (1989). Inactivation of cauliflower mosaic virus
was performed similarly, was detected by inoculation ofturnip leaves, and
was characterized by polyacrylamide gel electrophoresis (PAGE) of viral
polypeptides after denaturation with sodiumdodecylsulfate and 2-mer-
captoethanol and by alkaline agarosegel electrophoresis of viral DNA (Sun
et al., 1988).
The yellow-green fluorescent cells of inoculated foliar cotton tissues
were observed by fluorescence microscopy with filter sets designed for de-
tection of fluorescein (450-490 nm excitation, 520-560 nm emission) (Es-
senberg et al., 1992b). Brightly fluorescent cells were separated from weakly
fluorescent cells with a fluorescence-activated cellsorter, using filters with
PA Synthesis and X. campestris pv. malvacearum 185
the same excitation and emission ranges as employed for fluorescence mi-
croscopy (Pierce and Essenberg, 1987). Necrotic cells were stained in 0.25-
cm2 leaf discs withan aqueous solution of Evan’s blue, Exhaustive extrac-
tion of phytoalexins was performed by vacuum infiltration with ethanol/
water (70:30) and sonication for 30-45 min at 20-30°C. The relative fluo-
rescence intensities of cells before and after tissue extraction were viewed
with a video camerafitted to the fluorescence microscope, amplified, digi-
tized, and analyzed by computer (Essenberg et al., 1992b). Alkaline hydro-
lysis of materials remainingafter extraction was performed with 1N NaOH
at room temperature for 24 hr. Detailsofthemethod for determining
average phytoalexin concentration in the yellow-green fluorescent cells of
foliar tissue were described by Essenberg et al. (1992a). Important in that
method was determination of the fraction of the leaf tissue waterthat was
in the hypersensitively responding (i.e., yellow-green fluorescent and/or
brown) cells.The fraction of leaf tissue waterthat was in all mesophyll cells
was determined by stereological analysis of cross-sections. The fraction of
mesophyll cells that were hypersensitively responding was determined in
fresh, unsectioned leaf samples by counting cells that were fluorescent and/
or Evan’s blue-stainable, aswell as total mesophyll cells.
Incorporation of [1 ,2-13C]acetateinto DHC was by infiltration of inoc-
ulated OK1.2 cotyledons with a20 mM solution at 52 hr postinoculation,
followedbyharvest at 70 hr (Essenberg et al., 1985). Incorporation of
[5-3H]mevalonolactone into DHC and2-hydroxy-7-methoxycadalene(HMC)
was by infiltration of inoculated cotyledons with 30 a pCi/ml solutionat 26
hr postinoculation, followedby harvestat 43 hr. Degradation wasby oxida-
tion with a ruthenium-containing catalyst, followed by esterification of the
resulting labeled isobutyric acid and HPLC of the esterto constant specific
radioactivity (Davis etal., 1991). 6-Cadinene wasidentified by comparison
of its proton nuclear magnetic resonanceand mass spectra with spectra of
authentic 6-cadinene isolatedfrom cade oiland by cochromatography with
authentic 6-cadinene (Davis, 1993). Cell-freepreparations of the sesquiter-
pene cyclase were obtained by homogenizing inoculated, glandless cotyle-
dons in cold buffer containing enzyme-stabilizing additives and then pre-
paring from the homogenate a 27,OOOg supernatant (Munck and Croteau,
1990). Cyclase activity was assayed by incubating an aliquot of the superna-
tant (10 pg protein per assay) with0.3 pCi [l-3H]FPPfor 30 min at 3OoC,
followed by hexane extractionof the reaction mixture, passage ofthe hex-
ane through silica to removepolarcompounds, and liquid scintillation
counting of the eluate. Cochromatography of the eluate with authentic
8-cadinene by normal phase HPLC verified the identity ofthe predominant
product as 6-cadinene (Davis, 1993).
111. RESULTS AND DISCUSSION

A. Association of Phytoalexins with Resistance
to Bacterial Blightof Cotton
A search for organic solvent-extractable substances produced during the
hypersensitive response of leavesand cotyledons ledto isolation and identi-
fication of the sesquiterpene phenols DHC and LC (Essenberget al., 1982).
Methyl ethers ofboth compounds were also found: HMCand lacinilene C
7-methyl ether (LCME) (Essenberg et al., 1990).
DHC LC HMC LCME
In three highly resistant cotton lines, phytoalexin accumulation oc-

curred duringthe hypersensitive, resistant response. In inoculated leaves or
cotyledons, the four sesquiterpenoidsaccumulated to levelsof 100-500
nmol/g dry wt by the time bacterial multiplicationwas inhibited (3-4 days
after inoculation) (Essenberget al., 1982, 1990). Only very small amounts
(1-30 nmol/g dry wt) were found in mock-inoculated resistant plantsor in
inoculated susceptible plants at this time. In the highly resistant lines, the
four compounds continued to accumulate for several days after bacterial
inhibition, each reaching levels of 1000-5000 nmol/g dry wt before de-
clining.
Each highly resistantcotton line possesses twoor three major genes for
bacterial blight resistance plusa complex of “polygenes” (Essenberg et al.,
1982; Pierce and Essenberg, 1987). Also of interest were a collection of
lines of related genetic background, each of which possesses a different,
race-specific resistance gene(Hunter and Brinkerhoff, 1961). Three of these
lines, as wellas the related, fully susceptible line Ac44, were tested for
phytoalexin production after inoculation with each of six strains of X.
campestris pv. malvaceamm (of three different race phenotypes). In all
14 incompatible (resistant) interactions, higher amountsof all four of the
sesquiterpenoids shown above were found at the timeof bacterial inhibition
than in all 10 of the compatible (susceptible) interactions (Shevell,1985).
Different resistance genes evidentlycontrol accumulation of the same phy-
toalexins. Furthermore, A&, which possesses no known genes for resis-
tance to bacterial blight, accumulated high amounts of the four compounds
PA Synthesis and X. campesfris pv. malvacearum 187
during its nonhost response to X. campestris pv. campestris, a pathogen

of crucifers (Essenberg et al., 1990). Thus in the many compatible and
incompatible race/cultivar interactionsand in the one heterologouspatho-
var/cotton interaction that we have analyzed, sesquiterpenoid phytoalexin
accumulation correlated invariably with effective resistance to bacterial dis-
ease. Correlation of phytoalexin production with incompatibility has also
been observed within differential race/cultivar setsof Pseudomonas syrin-
gae pv. pisi and pea (Hadwigerand Webster, 1984) and of P . syringae pv.
glycinea and soybean (Long et al.,1985).
The abnormal behavior of inoculatedcotton plants held in continuous
darkness after inoculation gave us another opportunity to observe phyto-
alexin levels in relationto resistance (Morgham etal., 1988). In continuous
darkness, inoculated leaves of both susceptible and highly resistant lines
underwent confluent necrosis. However, the sesquiterpenoid phytoalexins
did not accumulate in either line. In both lines, the bacteria multiplied to
high population densities typicallyfound in susceptible plants maintained
in a normal light/dark cycle. Thus light was required for phytoalexin pro-
duction incotton leaves, and in its absence rapid tissue necrosis was insuffi-
cient to inhibit bacterial multiplication. These observationsare consistent
with the hypothesis that phytoalexin accumulationis responsiblefor bacte-
rial inhibition under normal light/dark conditions, although it is possible
that other light-dependent functions were instead responsible.
B. Antibacterial Activities of the Sesquiterpenoids
Bioassays of the sesquiterpenoids’ effects on growth of X. campesfris pv.
matvacearurn in vitro were at first conducted in darkness to avoid photooxi-
dation, to whichall four compounds are sensitive. DHC was the most
potently antibacterial (Table 1) (Essenberg et al., 1982, 1990). HMC, its
methyl ether, was not detectably active, perhaps becauseis almost
it insolu-
ble in water. The S enantiomer of LC was more active than the R enanti-
omer; however, the two enantiomers of LCME had comparable activities.
Although the water solubilities of the phytoalexins in planta may differ
somewhat from those reported in Table1, it appears likelythat the phyto-
alexins are soluble enough to diffuse through the apoplast in concentrations
that in combination are bacteriostatic.
It was reported in 1983 that phaseollin and several other isoflavonoid
phytoalexins were activated byirradiation with ultraviolet lightto produce
free radicals and to cause inactivation of the enzyme glucose-6-phosphate
dehydrogenase inan in vitro assay system (Bakker et al., 1983). This report
stimulated us to test the effects of irradiation on the biological activitiesof
the cotton phytoalexins. In vitro studies showedthat when irradiated with
Pierce
188 and Essenberg
Table 1 Antibacterial Concentrations, Water Solubilities, and Average Cellular

Concentrations In Planta" of Cotton Leaf Phytoalexins
Antibacterial
concentrations Local concentration
(mWb
Phytoalexin ED, ED,, Solubility (mM)' Day4
Day3
DHC 0.35 0.5 0.6 6.2 6.3
LC -1.5 -1.8
4.2 1.8 4.9
LCME 1.7 -
0.6 -
1.1 1.2
No. HR necrotic cells
per bacterial colony7 day3: 0.61 day 4: 4.5
Avg. no. HR cells per
cluster: day 3: 2.4 day 4: 4.8
Ratio of HR cell clusters
to bacterial colonies: day 3: 0.25 day4: 0.93
'Leaves of the highly resistant line OK1.2 were spray-infiltrated with 3 x lo6 cells/ml of X.
carnpmtrapv. rnalvacearum.
%D, and EDw were the phytoalexin concentrations which permitted 50% and 10% as many
bacterial generations, respectively. as in an uninhibited control culturegrowing logarithmically
in defined liquid medium at 30°C in the dark (Essenberg et al., 1982. 1990). The LC and
LCME used for these bioassays were isolated from inoculated Im 216 cotyledons and were
predominantly the R enantiomers. LCME did not exhibit an ED, within its solubility range.
'Solubility in water at 3OoC (mM) (Essenberg et al., 1990).
dConcentrations inhypersensitively responding (HR) cells, determined as described in the text
with the minor modification that Evan's blue stain was used to facilitate the detection of
responding cells that fluoresced little. Phytoalexin concentrations are means of determinations
from two plants (Essenberg, Pierce, Cover, Richardson and Scholes, unpublished) and cor-
rected for recovery rates of phytoalexin standards in the extraction and separation procedures:
33%, 24%. and 36% for DHC, LC, and LCME, respectively.
Tluorescent and/or Evan's blue-stainable cell numbers per cm' of leaf corrected for cells
wounded by the inoculation procedure,determined from mock-inoculated leaves. The number
of bacterial colonies per cm2in inoculated leaves wasassumed to be equal to theinitial number
of bacteria per cm2(Essenberg et al., 1979a,b).
white light (300-700 nm), DHC and LC can induce single-strand breaks in
DNA, inactivate enzymes,and destroy the infectivity of cauliflower mosaic
virus, apparently by crosslinkingthe viral DNA to coat protein (Sun et al.,
1988, 1989). During exposure of DHC to light and oxygen, free radicals
were detected by spin trapping with phenyl-tert-butylnitrone(Steidl, 1988).
DHC, LC,and LCME were more inhibitoryto liquid cultures of X.cam-
pesfris pv. malvacearum when they were exposed to white light (300-700
m)than in the dark (Steidl, 1988; Sun et al., 1989; Samad and Essenberg,
unpublished). In the light, 0.10 mM DHC inhibited bacterial growth com-
pletely. HMC remained inactiveand therefore does not qualify asa phyto-
alexin. The broadly destructive effects observedfrom photoactivated DHC
suggest that during the hypersensitive response in sunlit
cotton plants, these
phytoalexins probably damage many components of the infecting bacterial
cells. The only other report of phototoxic phytoalexins of which we are
aware is on thiophenes elicited in Tagetes by Fusarium oxysporum (Kour-
any et al.,1988). However,the growing numbers ofplant species discovered
to contain phototoxins (Downum etal., 1991) suggestthat other phytoalex-
ins may have unrecognized phototoxicity.
C. localization of the Phytoalexinsin and Around All

Hypersensitively Necrotic Cells
When foliar tissues of resistantcotton lines are inoculated withX.campes-

tris pv. malvacearum by infiltration with a suspension of fewer than lo7
bacteria/ml, the tissue does not undergo confluent collapse and necrosis,
but instead develops clusters of dark brown, necrotic cells. Each cluster is
the site of a bacterial colony which has grown from a single bacterial cell
deposited during inoculation in the adjacent intercellular space. Our early
work demonstrated that in resistant plants, growthof each colony is inhib-
ited by a local resistant response (Essenberg et al., 1979a) manifested as
a small bacteriostatic zone around the colony (Essenberg et al., 1979b).
Fluorescence microscopy revealedthat all of the brown, necrotic cells ex-
hibit yellow-green fluorescence that is spectrally similarto that of the phy-
toalexins LC and LCME (Pierce and Essenberg, 1987; Essenberg et al.,
1992b). Fluorescent materials may also be present in theapoplast, as sug-
gested by apparent fluorescence of the cell walls of the necrotic cells and
often also of neighboring cells.
To determine whetherthe most potent phytoalexin,DHC, is also local-
ized in the fluorescent cells, we undertook the physical separation of yel-
low-green fluorescent cells from symptomless cells so that they could be
analyzed directly for their phytoalexin contents. Macerating enzymes were
used to obtain suspensions of palisade and spongy mesophyll cells from
inoculated resistant cotyledons, and the cellswerequicklysubjected to
fluorescence-activated cell sorting. From the phytoalexin contents deter-
mined in the brightly fluorescent and less fluorescent fractions and from
the numbers of brightly fluorescentand symptomless, less fluorescent cells
in each fraction, we calculated the average phytoalexin contentof each of
these two cell types (Pierceand Essenberg, 1987).A brightly fluorescent cell
contained 10-25 times as much LC and 40 times as much DHC as a less
fluorescent cell. When the relative numbers of brightly fluorescent and
symptomless, less fluorescent cells in the cotyledonary tissue were taken
into account, it was calculated that more than 90% of the DHC and more
than 75% of the LC and LCME were associated with the fluorescent cells
at infection centers (Table 2).
Since the bacterial colonies are in the intercellular spaces rather than
inside the necrotic cells, itwas important to know whetherthe phytoalexins
can penetrate the host plasma membrane to contact the bacteria. Evan's
blue, a dye that stains onlycells with damaged plasma membranes (Taylor
and West, 1980), was usedto assess membrane integrity.Of the fluorescent
cells observed in leaf samples 2, 3, and 4 days after inoculation, 98 2% *
were stained by Evan's blueand/or were dark brown (Pierceand Essenberg,
unpublished). We know that the dark brown cells have dysfunctional mem-
branesbecausethey are collapsed(Essenbergetal.,1979a, 199213).We
conclude that all of the fluorescent cells have damaged membranes perme-
able to small molecules.
An efflux experiment demonstratedthat the phytoalexins were able to
diffuse from the fluorescent cells. The abaxial epidermis was peeled from
strips of inoculated resistant cotyledons,and the strips were gently stirred
in a buffered 0.4 M mannitol solution (Pierce et al., unpublished). The
phytoalexins LC, LCME, and DHC diffused into the medium with first-
order kineticsand half-times of3 hr, 6 hr, and 16 hr, respectively.
However, not all of the yellow-green fluorescent material in hypersen-
Table 2 Distribution of Phytoalexins Between Brightly Fluorescent Cells and

Unit Area of Cotyledon'
Symptomless, Less Fluorescent Cells per
Phytoalexins Contribution to
Fraction of total
mesophyll @mol/mm2)
tissue W O ) DHC LC
LCME DHC LC
LCME
Experiment 1
68.0
Brightly fluorescent 23 93
86 62 100 88
5.3
Less fluorescent 77 11 0 7 12 0
Experiment 2
110.0
Brightly fluorescent 23 140 71 79
92 75
9.6
Less fluorescent 77 8 47 18 25 21
'Cells were isolated from resistant OK1.2 cotyledons three days after infiltration with 5 x
10' cells/ml of X. campestris pv. malvacearum and subjected to fluorescence-activated cell
sorting.
was assumed that the phytoalexin amounts lost during isolation and sorting had the same
distribution between brightly fluorescent and less fluorescent cells in the intact tissue as the
amounts recovered in the sorted cells.
Source: Reproduced with permissionfrom Pierce and Essenberg, 1987. Copyright 1987,Aca-
demic Press, Ltd.
sitively responding cotton foliar tissue was extractable. After exhaustive

extraction with ethanol/water (70:30), cotyledons still contained fluores-
cent cells. They were the same cells that fluoresced in fresh tissue: of 507
fluorescent cells in resistant cotyledons harvested3-4 days after inoculation
and examined before and after extraction, 503 (99Vo) had visually detect-
ableresidualfluorescence(Essenbergetal.,1992a).Alkalinehydrolysis
partially removed this residual fluorescence, indicating its probable origin
in both wall-bound phenolic monomers and phenolic polymers.
This findingthat essentially all ofthe fluorescent cells possessed nonex-
tractable fluorescence raised the question of whether all or only some of
them also containedthe fluorescent phytoalexins.To answer this question,
fluorescence intensity (520-560 nm) of individual yellow-green fluorescent
palisade cells was measured with a videodensitometer before and after tis-
sue extraction. Extraction removedat least 35% of the fluorescence from
allcells, and the mostfrequentloss was85-90%. Thus all fluorescent
cells of hypersensitively responding tissue possessed extractable as well as
nonextractable fluorescence. Whenan extract of hypersensitively respond-
ingtissue was fractionated, LC and LCMEaccounted for 75%ofthe
yellow-green fluorescence in the set of fractions (Essenberg et al., 1992b).
These results strongly suggestthat in inoculated resistantcotton foliar tis-
sue, all hypersensitively responding cells accumulate both free phytoalexins
and covalently bound fluorescent phenolics.
Autofluorescence or absorbance that is spectrally similar to that of
phytoalexins has also been used to localize phytoalexins in leaves of broad-
bean (Mansfield et al., 1974)and grapevine (Langcakeand Pryce, 1976) in
response to Botrytis infection, in resistantoat infected with Pyricularia or
Puccinia (Mayama and Tani, 1982), and in juvenile sorghum leaves infected
with Colletotrichum (Snyderet al., 1992). In broadbean and grapevine,
healthy cells adjacent to necrotic cells fluoresced most intensely, whereas in
oat and sorghum, as incotton, the collapsed cells werethe sites of fluores-
cence or pigmentation. Histochemistry has also been usedto locate phyto-
alexins at or near the invading pathogen during expression of resistance in
cotton stems, bean leaves and hypocotyls, flax leaves, and tomato roots
and stems (reviewed in Pierce and Essenberg, 1987).
D. Phytoalexin Concentrations in the

Hypersensitively Necrotic Cells
The findings described above, that the phytoalexins are predominantly lo-
calized in cells which can be identified
by their autofluorescence, enabled us
to determine the average phytoalexin concentrations in those cells. Tissue
water content was determinedas the differencebetweenfresh and dry
weights. Other samples of the same leaves were examined by fluorescence

microscopy so that the fraction of mesophyll cells exhibiting yellow-green
fluorescence and/or brown pigmentation could be determined by counting.
The fraction of the leaf tissue water that wasin those hypersensitively
responding cellswas computed. The phytoalexin content of a given amount
of leaf tissue was then divided by the computed volumeof its hypersensi-
tively necrotic cells to give average millimolar concentration in those cells
(Essenberg etal., 1992a).
Resultsofapplying this method to highlyresistantline OK1.2 are
shown in Table 1. Each of the phytoalexins accumulatedthe in hypersensi-
tively necrotic cellsto a concentration higher than its water solubility (Es-
senberg et al., 1990).At both 3 and 4 days after inoculation, DHCand LC
concentrations were greater than their ED, levels in the dark. However, at
3 days there was only about one cluster of detectably fluorescent and/or
Evan’s blue-stainable (necrotic) cells for every four bacterial colonies. By
day 4, numbers of necrotic cells had increased so that there were 4.5 ne-
crotic cells per bacterial colony. Since the average cluster consisted of 4.8
cells, there was nearly one (0.93) cluster per bacterial colony. This corre-
spondence agreeswell withthe observation that at4 days bacterial multipli-
cation in these leaveswas inhibited.
Cellular phytoalexin concentrations high enoughto account for bacte-
riostasis on the day it was observed have also been determined for the
highly resistant cotton lines Im 216 and WbM(O.0) (Pierce et al., unpub-
lished results). On the same day, in the genetically related susceptible line
WbM(4.0)’ phytoalexin levels were too low to be inhibitory, which is consis-
tent with the fact that bacteria were still multiplying.
Cellular phytoalexin concentrations were also determined in several of
the Ac44 lines possessing single resistance genes (Shevell et al., unpublished
results). At the time of bacterial inhibition, inhibitory phytoalexin concen-
trations were found in leaves infected with incompatible bacterial strains,
but in leaves infected with compatiblestrains, cellular phytoalexin concen-
trations were too low to be inhibitory.
In summary, the phytoalexins are localized in the right place to be
effective-in the leaky, dead, or dying mesophyll cells closest to the inter-
cellular bacterial colonies.In the various cotton lines studied, which have
different bacterial blight resistancegenes and different levels of resistance,
the phytoalexins have been found at concentrations higher than in vitro
bioassays indicate that they are needed to fully inhibit multiplication of
the bacterial pathogen. These phytoalexin-loaded cells become numerous
enough to inhibit essentially all bacterial colonies
on the day when bacterio-
stasis was observed. We conclude that our determinations of cellular phyto-
alexin concentrations provide strong evidence that phytoalexin accumula-
tion is sufficient to account for resistance ofcotton foliar tissueto bacterial

blight.
Studies by other investigators have yielded similar evidence for a few
plant/fungal pathogen interactions. One approach has been to cut thin
tissue slices and to subject alternate or replicate slices either to staining
for presence of the pathogen or to quantitative analysis for phytoalexins.
Phytoalexin concentration is computed by dividing the slice’s phytoalexin
content by its water content. This approach was applied to the soybean
hypocotyl/Phytophthora megasperma var. sojae system by Yoshikawa et
al. (1978), using 0.25-mm-thick slices,and at high resolutionby Hahn et al.
(1985), using 15-pm-thick slices. Results support the hypothesis that the
phytoalexin glyceollinI has an important early role, but may not be solely
responsible for resistance. Metlitskii et al. (1985) employed the tissue slice
approach in studyingpotato tuber inoculated on a cut surface with compat-
ible and incompatible racesof Phytophthora infestans and found antifun-
gal concentrations of rishitin at the arrested front of incompatible fungal
growth. During the resistant response of juvenile sorghum to Colletotri-
chum graminicola, epidermal cells respond to fungal appressoria by accu-
mulating orange-red deoxyanthocyanidin phytoalexins. Snyder al. et (1991)
obtained an estimate of 150 mM phytoalexin in subcellular vesicle-like in-
clusions by microspectrophotometry. Although dilution occurswhen these
vesicles release the phytoalexin within the cell, it seems likelythat the con-
centration is still higherthan the 9 pM that inhibits fungal growth in vitro.
Thus in the plant/fungal pathogen systemsin which local phytoalexin con-
centrations at sites of pathogen growth have been experimentally estimated,
as well as in the plant/bacterial pathogen system that we study, results
indicate that phytoalexins can contribute significantly to observed resis-
tance.
E. BiosyntheticPathways
The structures of phytoalexins produced in cotton leaves and cotyledons
suggest that they are terpenoid in origin, and isotope incorporation into
DHC and HMC from labeled acetate and mevalonolactone has confirmed
this (Essenberg et al., 1985; Davis et al., 1991). Several things have been
learned about how trans, tram-farnesylpyrophosphate(FPP), a common
intermediate of terpene biosynthesis, enters the specialized pathway to the
cotton leaf phytoalexins. When acetate containing I3C at 90% enrichment
in both positions was fed to cotyledons during phytoalexin biosynthesis, the
positions of adjacent carbon atoms labeledfrom the same acetate molecule
were deduced from the I3C nuclear magnetic resonance spectrum of DHC
(Essenberg et al., 1985). Thisinformation showed which of thethree possi-
Pierce
194 and Essenberg
ble folding patterns of the farnesyl precursor occurs during formation of

the bicyclic ringstructure of DHC. The same cyclizationpattern has subse-
quently been demonstrated in gossypol,a bisesquiterpene from cotton that
functions in resistanceto herbivores (Masciadri etal., 1985).
During the biosynthesis of DHC in planta, a tritium atom from [5-
3H]mevalonolactonewas transferred to the methine carbon of the isopropyl
side chain (Davis etal., 1991). It seems likely that this transfer occurs as a
1,3-hydride shift from C-l of the farnesyl precursor following its cycliza-
tion to a 10-membered ring cation. The retentiontritium of on the isopropyl
group throughout the pathway to DHC suggestedfurther that if [1-3H]FPP
were used as the substrate with cell-free preparations from phytoalexin-
synthesizing tissues, intermediates the
of pathway would be tritium-labeled.
To simplify the mixture of labeled compounds produced, a cotton line
which lacks pigment glands was used. Its leaves and cotyledons have only
very low amounts of gossypol and the other terpenoids stored in pigment
glands of other cotton lines (Elzen et al., 1985), but it produces high levels
of DHC, HMC, LC,and LCME after bacterial infection. Cell-free prepara-
tions from its inoculated cotyledonscyclized [1-3H]FPP or unlabeledFPP
to 6-cadinene, an unsaturated, bicyclic hydrocarbon with carbon skeleton
of DHC (Davis, 1993).
A
6-Cadinene
This enzymic activity is strongly infection-induced. Thus it appears that
biosynthesis of the phytoalexins from FPP begins with cyclization and is
followed by hydroxylationand aromatization. Inthis, it follows the pattern
that has been found for biosynthesis of other cyclic sesquiterpenes (Cane,
1990).
IV. EPILOG
We willcontinue our exploration ofthe foliar tissue’s responseto individual
bacterial colonies. The observation that cells become Evan’s blue-stainable
(i.e., necrotic) as they become visibly fluorescent raises the question of
where biosynthesis of these fluorescent phytoalexins occurs. Is ita inshort
burst just before death? Or in living, neighboring cells followed by trans-
port to the dead or dying cells? Or inboth places? Our biosynthetic studies
are providing tools for answering these questions. We are currently per-
forming pulse label experimentsto gain information about time and place
of biosynthesis relativeto host cell death. We are also purifyingthe sesqui-
terpene cyclase described above. itIfis provento catalyze a reaction on the
biosynthetic pathway to the phytoalexins, then an immunocytochemical
stain for it or an in situ hybridization probe for its messenger RNA can
reveal the cells in which phytoalexin biosynthesis is induced.
We are also working to elucidate the rest of the biosynthetic pathway.
Several putative intermediates have been identified. Isotopic incorporation
and dilution experiments will be conducted to show whether these com-
pounds are biosynthetic intermediates betweenFPP and the phytoalexins.
Assays for later enzymes of the pathway may then be developed, so that
these enzymes may be isolated.
Preparation and identification of cDNA clones for biosynthetic en-
zymes will give us the opportunity to block these steps in plants by trans-
forming them with antisense constructsof the genes. This strategy is now
being applied by other investigatorsto legumes to test the role of isoflavo-
noid phytoalexins in disease resistance. Such manipulation of genes affect-
ing phytoalexin concentration or activity offers promise of powerful tests
of whether phytoalexins are necessary for resistance. The analyses of phyto-
alexin concentrations inplanta that we have described inthis chapter have,
by comparison, provided evidence concerning whether phytoalexin accu-
mulation is sufficient to account for observed disease resistance.
The cDNA clonesfor biosynthetic enzymes may be used as probes for
identification of genomic clones. The cadalene and lacinilene phytoalexins
that we study are only four of a large group of terpenoids in cotton that
have the same carbon skeleton and function variously as constitutive de-
fenses against herbivory, as attractants to pollinating insects, or as infec-
tion-induced defenses against fungal and bacterial pathogens (Bell, 1986).
Isolation of genes for the biosynthetic enzymes, along with their variously
regulated promoters, will open up possibilities of crop improvement by
enhancing or repressing synthesis of these compounds inparticular organs
of the plant or by introducing genes for foreign defense proteins under the
regulation of cotton’s own stress-responsive promoters.
ACKNOWLEDGMENTS
We thank our colleagues Ellen C. Cover, Paul E. Richardson, Vernon E.
Scholes, Roushan A. Samad, and Judith L. Shevell for permission to de-
scribe their unpublished work, and Oral Roberts University for use of the
microspectrophotometer and the cell sorter. The work described here was
supported by a grant from the Herman FraschFoundation, by the Cooper-
ative State Research Service,U.S. Department of Agriculture, under Agree-

ment No. 5901-0410-9-0236-0, by the National Science Foundation under
Grants Nos. PCM-8117015, PCM-8316759, and DMB-8616650, and by the
Oklahoma Agricultural ExperimentStation, of which this is journal article
5-6312.
REFERENCES
Bakker, J., Gommers, F. J., Smits, L., Fuchs, A., and De Vries, F. W. (1983).
Photoactivation of isoflavonoid phytoalexins: involvement of free radicals.
Photochem. Photobiol.38(3): 323-329.
Bell, A. A. (1986). Physiology of secondary products. Cotton Physiology (J. R.
Mauney and J. McD. Stewart, eds.), Cotton Foundation, Memphis, pp. 597-
621.
Brinkerhoff, L. A. (1970).Variation inXanthomonas malvacearum and its relation
to control. Annu. Rev. Phytopathol. 8 85-1 10.
Brinkerhoff, L. A., Verhalen, L. M., Johnson, W. M., Essenberg, M., and Rich-
ardson, P. E. (1984). Development of immunity to bacterial blight of cotton
lications for other diseases. Plant Dis. 68(2): 168-173.
Cane,and T
D.itsE. ( 990). Enzymatic formation of sesquiterpenes. Chem. Rev. 90(7):
1089-1 103.
Davis, G. D. (1993). 6-Cadinene synthase from hypersensitively responding cotton
cotyledons: identification of substrate and product and partial purification of
the enzyme. PhD thesis, Oklahoma State University, Stillwater.
Davis, G. D., Eisenbraun, E. J., and Essenberg, M. (1991).Tritium transfer during
biosynthesis of cadalene stress compounds in cotton. Phytochemistry, 30(1):
197-199.
De Feyter, R., and Gabriel, D. W. (1991).At least six avirulence genesare clustered
on a 90-kilobase plasmid in Xanthomonas campestrispv. malvacearum. Mol.
Plant-Microbe Interact.4(5): 423-432.
Downum, K. R., Swain, L. A., and Faleiro, L. J. (1991). Influence of light on plant
allelochemicals: a synergistic defense in higher plants. Arch. Insect Biochem.
Physiol. I7(4): 201-212.
Elzen, G. W., Williams, H. J., Bell, A. A., Stipanovic, R. D., and Vinson, S. B.
(1985).Quantification of volatile terpenes of glanded and glandless Gossypium
hirsutum L. cultivars and lines by gas chromatography. J. Agric. Food Chem.
33(6): 1079-1082.
Essenberg, M., and Pierce, M. L. (1994). Role of phytoalexins in resistance of
cotton to bacterial blight.Bacterial Pathogenesis and Disease Resistance(S.-D.
Kung and D. D. Bills, eds.), World Scientific, Singapore, pp.303-304.
Essenberg, M., Cason, E. T. Jr., Hamilton, B., Brinkerhoff, L. A., Gholson,
R. K., and Richardson, P. E. (1979a). Single cell colonies of Xanthomonas
malvacearum in susceptible and immune cotton leaves and the local resistant
response to colonies in immune leaves. Physiol. Plant Pathol.15(1): 53-68.
Essenberg, M., Hamilton, B., Cason, E. T. Jr., Brinkerhoff, L. A., Gholson, R.

K., and Richardson, P. E. (1979b). Localized bacteriostasis indicatedby water
dispersal of coloniesof Xanthomonas malvacearum within immune cotton
leaves. Physiol. Plant Pathol.15(1): 69-78.
Essenberg, M., Doherty, M.d'A., Hamilton, B. K., Henning, V. T., Cover, E.
C., McFaul, S. J., and Johnson, W. M. (1982). Identification and effects on
Xanthomonas campestris pv. malvacearum of two phytoalexins from leaves
and cotyledons of resistant cotton. Phytopathology 72(10): 1349-1356.
Essenberg, M., Stoessl, A., and Stothers, J. B. (1985). The biosynthesis of 2,7-
dihydroxycadalene in infected cotton cotyledons: the folding pattern of the
farnesolprecursor and possibleimplications for gossypolbiosynthesis. J.
Chem. Soc. Chem. Commun.9 556-557.
Essenberg, M., Grover, P. B. Jr., and Cover, E. C. (1990). Accumulation of anti-
bacterial sesquiterpenoids in bacterially inoculated Gossypium leaves and coty-
ledons. Phytochemistry29(10): 3107-3113.
Essenberg, M., Pierce, M. L., Cover, E. C., Hamilton, B., Richardson, P. E., and
Scholes, V. E. (1992a). A method for determining phytoalexin concentrations
in fluorescent, hypersensitively necrotic cells in cotton leaves. Physiol. Mol.
Plant Pathol.41(2): 101-109.
Essenberg, M., Pierce, M.L., Hamilton, B., Cover, E. C., Scholes, V. E., and
Richardson, P. E. (1992b). Development of fluorescent, hypersensitively ne-
crotic cells containingphytoalexins adjacent to coloniesof Xanthomonas
campestris pv. malvacearum in cotton leaves. Physiol. Mol. PlantPathol.
41(2): 85-99.
Hadwiger, L. A., and Webster, D. M. (1984). Phytoalexin production in five culti-
vars of peas differentially resistantto three races of Pseudomonas syringaepv.
pisi. Phytopathology 74(11): 1312-1314.
Hahn, M. G., Bonhoff, A., and Grisebach, H. (1985). Quantitativelocalization of
fected with Phytophthora megasperma f. sp. glycinea. Plant Physiol. 77(3):
591-601.
Hunter, R. E., and Brinkerhoff, L. A. (1961). Report of the Bacterial Blight Com-
mittee, Oklahoma, Proc. Cotton Dis. Council 21:7-8.
Kourany, E.,Amason, J. T., and Schneider, E. (1988). Accumulation of phototoxic
thiophenes in Tageteserecta (Asteraceae) elicited by Fusarium oxysporum.
Physiol. Mol. Plant Pathol.33(2): 287-297.
Langcake,P., and Pryce, R. J. (1976). The production of resveratrol by Vitis
viniferaand other members ofthe Vitaceae as a response to infection or injury.
Physiol. Plant Pathol. 1):9( 77-86.
Long, M., Barton-Willis, P., Staskawicz, B. J., Dahlbeck, D., and Keen, N. T.
(1985). Further studies on the relationship between glyceollin accumulation
and the resistance of soybean leaves to Pseudomonas syringae pv. glycinea.
Phytopathology, 75(2): 235-239.
Mansfield, J. W. (1982). The role of phytoalexins in disease resistance.Phytoalexins
(J. A. Bailey and J. W. Mansfield, eds.), Blackie,London, pp. 253-288.
Mansfield, J. W., Hargreaves, J. A., and Boyle, F. C. (1974). Phytoalexin produc-

tion by live cells in broad bean leaves infected with Botrytis cinerea. Nature
252(5481): 316-317.
Masciadri, R., Angst, W., and Arigoni, D. (1985).A revised schemefor thebiosyn-
thesis of gossypol.J. Chem. Soc. Chem. Commun. 22: 1573-1574.
Mayama, S., and Tani, T.(1982).Microspectrophotometric analysisof the location
of avenalumin accumulation in oat leaves in responseto fungal infection.Phys-
iol. Plant Pathol.242): 141-149.
Metlitskii, L. V., Ozeretskovskaya, 0. L., Chalenko, G. I., Karavayeve, K. A.,
and Vasiukova, N. I. (1985). Methodological approaches to studying potato
phytoalexins. Mikologia i Fitopathologia19(5):429-435.
"orgham,A.T., Richardson, P. E., Essenberg, M., and Cover, E.C. (1988).
Effects ofcontinuous dark upon ultrastructure, bacterial populations and accu-
mulation of phytoalexins during interactions between Xanthomonas campestris
pv. malvacearum and bacterial blight susceptibleand resistant cotton. Physiol.
Mol. Plant Pathol.32(1): 141-162.
Munck, S. L., and Croteau, R. (1990). Purification and characterization of the
sesquiterpene cyclasepatchoulol synthase from Pogostemon cablin. Arch. Bio-
chem. Biophys.282(1): 58-64.
Pierce, M., and Essenberg, M. (1987).Localization of phytoalexins in fluorescent
separated from other cells by fluorescence-activatedcell sorting. Physiol. Mol.
Plant Pathol.31(2): 273-290.
Shevell, J. L. (1985). Phytoalexin production in congenic cotton lines challenged
with races of Xanthomonas campestrispv. malvacearum, MSc. thesis, Okla-
homa State University, Stillwater.
Snyder, B. A., Leite, B., Hipskind, J., Butler, L. G., and Nicholson, R. L. (1991).
Accumulation of sorghum phytoalexins induced by Colletotrichum graminicola
at the infection site.Physiol. Mol. Plant Pathol.39(6): 463-470.
Steidl, J. R. (1988). Synthesis of 2,7-dihydroxycadalene,a cotton phytoalexin;
photo-activated antibacterial activity of phytoalexins from cotton; effect of
reactive oxygen scavengers and quenchers on biological activity and on two
distinct degradation reactions of DHCi PhD thesis, Oklahoma State Univer-
sity, Stillwater.
Sun, T. J., Essenberg, M., and Melcher, U. (1989). Photoactivated DNA nicking,
enzyme inactivation, and bacterial inhibition by sesquiterpenoid phytoalexins
from cotton. Mol. Plant-Microbe Interact.2(3): 139-147.
Sun, T. J., Melcher, U., and Essenberg, M. (1988). Inactivation of cauliflower
mosaic virus byphotoactivatable cotton phytoalexin. Physiol. Mol. Plant Pa-
thol. 33(1): 115-126.
Taylor, J. A., and West, D. W. (1980). The use of Evan's blue stain to test the
survival ofplant cells after exposure to high salt and high osmotic pressure. J.
EXP.Bot. 31(121): 571-576.
Yoshikawa,M.,Yamauchi, K., and Masago, H. (1978). Glyceollin: its rale in
tophthora megasperma var. sojae. Physiol. Plant Pathol. 12(1): 73-82.
10
Chemistry, Biology, and Roleof
Groundnut Phytoalexinsin
Resistance to Fungal Attack
P. V. Subba Rao*
Centre de CoopPration Internationale en Recherche Agronomique
pour le
DPveloppement (CIRAD), Montpellier,France
Richard N. Strange
University College London, London, England
1. INTRODUCTION
A. The Groundnut Crop
Groundnuts (Arachis hypogaea L.), or “peanuts” as theyare known in the
United States, are an important oilseed crop belonging to the Fabaceae
(Leguminosae). They are cultivated in most of the tropical regions of the
world, where rainfall is adequate, lying between the latitudes 40°N and
4OoS (McDonald, 1984). The total area devoted to groundnut cultivation is
approximately 20 million ha (FAO, 1990; Wynneetal., 1991) and the
average annual world production is estimated at 23.1 million t of pods
(FAO, 1990).
The growth habit of the plant is unusual in that the pods are borne
underground. After fertilization, the gynophore (or peg) elongates toward
the ground and, on penetrating the soil, begins to form a pod. Once the
kernels have differentiatedthe pod tissue becomes woodyand dies, giving
rise to the familiar pittedstructure of the mature groundnut.
B. ProductionConstraints
Only a third of the world production of groundnuts comes from large
commercial plantations, the remainder being produced by small holdings in
developing countries where yieldsare extremely low, averaging 780 kg/ha.
*Current offiation: University College London, London, England

199
Strange200 and Rao Subba
Low yields are caused by variety of constraints, among which pests and
diseases are the most significant (Gibbons,1980). In contrast, yields are far
higher in developed countries, averaging 3000 kg/ha, and some cultivars
are capable of yields as high as5000 kg/ha under ideal growing conditions
(McDonald, 1984).
Groundnuts are host to approximately 50 genera of fungi, 1bacterium,
15 viruses, 16 nematodes, and 2 phanerogamic parasites, aswell as a num-
ber of insect pests; theyare also susceptibleto physiological disorders such
as deficiency diseasesand injuries caused by climaticfactors (Jackson and
Bell, 1969; Feakin, 1973; COPR, 1981; Porter et al., 1984; Subrahmanyam
and Ravindranath, 1988). Many of these agents are major constraints to
groundnut production worldwideand reduce yields as wellas quality of the
crop substantially.
Two of the more serious disorders ofgroundnuts are leaf spot diseases
and infection of kernels by fungi of the Aspergillusflavus group. The main
causal agents of leaf spot diseases are Cercospora arachidicola (early leaf
spot), Phaeoisariopsis personata (late leaf spot), and rust, Puccinia ara-
chidis, which can cause losses ofup to 50% (Subrahmanyam and McDon-
ald, 1983, 1987). In addition, web blotch, causedby Phorna arachidicola, is
important in some African countries (Cole,1982). The A. flavus group of
fungi consists of A . flavus itself and the closely related A . parasiticus.
They produce powerful mycotoxins which can have serious consequences if
consumed (see below).
C. Disease Control Strategies
Effective chemicalcontrol measures are available for leaf spot diseasesbut
not for infection of kernels by the A . flavus group of fungi, which often
occurs in the soil. Use of chemical methods to control leaf spot diseases is
limited in developing countriesfor several reasons including lack of money
to pay the high costs of the chemicals, lack of local availability of the
chemicals themselvesand the means of applying them, and lack of knowl-
edge about their use. Hence, control measures have been directedto select-
ing and breeding disease-resistant cultivars.
Although disease control by breeding resistant cultivars can be very
effective and is particularly appropriate for developing countries, the pro-
cess is painfully long. Moreover, resistance genes for each disease and pest
have to be introduced separately into agronomically acceptable cultivars.
One way in which this process might be accelerated is to enhance the plant’s
defense mechanisms.
In this chapter we review one defense mechanism, the phytoalexin re-
sponse, and suggest ways in which it might be exploited to increase the
resistance ofgroundnuts to its more important parasites.
Groundnut Phytoalexins and Disease Resistance 201
II. PHYTOALEXINS
A. Definition and Role in Defense
Phytoalexins are low molecular weight antimicrobial compounds that are
synthesized byand accumulated in plants in response to microbial challenge
(Paxton, 1981). Although the role of these compounds in resistance is not
included inthis definition, there isnow ample evidence to suggest that rapid
accumulation of sufficient concentration of phytoalexin in the vicinity of
the challenging parasiteis inhibitory for its growthand is crucial for defense
(reviewed by Mansfield, 1982; Strange, 1992). Such evidence comes from
two types of experiments: biochemical and genetic.
Biochemicalevidencedemonstrates that phytoalexinsaccumulate to
inhibitory concentrations at the time that the parasite ceases to grow. For
example, Hahn etal. (1985) showed,using a radioimmuneassay, that
within 8 hr of inoculation the concentration of the soybean phytoalexin,
glyceollin I, exceeded the amount required to inhibit the in vitro growthof
Phytophthora megaspermaf. sp. glycinea in roots of a resistant cultivarbut
not in those ofa susceptible cultivar. Similar results were also obtained for
interactions of soybean roots and the soybean cyst nematode, Heterodera
glycines; again usinga radioimmune assay,Huang and Barker (1991) found
that concentrations ofglyceollin I in the head region of the nematode
reached 0.3 pmol/ml by 24 hr after inoculationin a resistantcultivar
whereas nonewas found ina susceptible cultivar.
Genetic evidence for the role of phytoalexins in resistance has been
reviewedby VanEtten et al. (1989) with particular reference to the pea
parasite, Nectria haematococca. In brief, only strains of the fungus that
were tolerant of the pea phytoalexin pisatinand were able to degrade it by
demethylation were virulent. When these were crossed with strains that
were avirulent and sensitive to pisatin, the progeny generally fell into the
two parental phenotypes, i.e. 1.) those that were tolerant of pisatin, were
able to degrade it, and were virulent (Pda'), and 2.) those that were sensi-
tive to pisatin, couldnot degrade it, and were avirulent (Pda-).
The importanceof pisatin demethylase,the enzyme responsiblefor the
degradation of pisatin, has been corroborated by experiments in which a
Pda- isolate was transformed with a fragment of DNA containing a gene
which conferred the ability to demethylate pisatin (Ciufetti et al., 1988).
Three of the transformants were more tolerant of pisatin and two were
significantly more pathogenic to pea.
Recently, VanEtten's team (Miao et al., 1991) mapped one Pda gene
(Pdad) to a small, meiotically unstable chromosome which was dispensable
for normal growth. They point out that these properties are characteristic
of B chromosomes and suggest that such genetic elements might be an
202 Strange Subba Rao and
important meansbywhichvariationcouldbegeneratedinpathogenic
fungi. This would be of particular significance
if such genetic elements were
laterally transmissibleand could confer phytoalexin tolerance
to other fungi
in a manner analogousto R plasmids in bacteria (Miao al.,
et 1991).
B. Chemistry of Groundnut Phytoalexins

When a phytoalexin is first suspected to be involved in the defense of a
plant against parasiticattack it is normally necessaryto be ableto extract it
and to detect it in the extract. Unless the compound is already known,
detection inevitably involves bioassay with a suitable test microorganism.
The next step is to isolate the active compound and determine its chemical
characteristics. Oncethis has been achieved, physicochemical techniques or
radioimmune assays maybe used for quantifying it in plant extracts.
Extraction and Partial Purification

Both groundnut kernels (cotyledons) and groundnut leaves present prob-
lems for the extraction of phytoalexins. Cotyledons are rich in oil and this
can interfere with the subsequent chromatography if it is not removed.
Chlorophyll can also interfere with chromatographic procedures if too
much is extracted when making preparations from leaves.
Arora and Strange (1991) compared three methods for extracting coty-
ledons. The cotyledons were either1.) vacuum-infiltrated with acetonitrile
and incubated for 48 hr at 25OC, 2.) homogenized in 80% acetone, or 3.)
homogenized in95% ethanol.
Leaves were extracted by vacuuminfiltration of either 50% methanol
(Subba Rao et al., 1988a, 1990, 1991) or 60% ethanolvdwards and Strange,
1991). Higher concentrations of the alcohols resulted in the extraction of
too much chlorophyll.
The phytoalexins were partially purified by either solid phase extraction
or solvent partitioning. In solid phase extraction, the acetonitrile extracts
of cotyledons were diluted to 25% acetonitrile with water and applied to
a cartridge containing ODS silica. After washing with 25% acetonitrile,
phytoalexins were eluted from the cartridge in 100% acetonitrile. Ethanol
extracts of leaves wxe treated similarly but with this material the original
60% ethanol concentration of the extractant was reduced to 25% by film
evaporation before application to the cartridge.
When phytoalexins were partially purified by solvent partitioning,the
organic solvent was first removed from extracts by film evaporation. The
resulting aqueous preparations were partitioned three times against hexane
to remove oils and then three times against ethyl acetate (Subba Rao et al.,
Phytoalexins
Groundnut
and Disease Resistance 203
1988a, 1990, 1991; Edwardsand Strange, 1991; Arora and Strange, 1991).
Ethyl acetate extracts were usually dried over anhydrous sodium sulfate
before reductionof the volume by film evaporation. (See Fig. 1 for a sum-
mary of these techniques.)
Bioassay
Phytoalexins are often detected initially by their antifungal activity. The
most widely used technique is the thin layer chromatography (TLC) bioas-
say (Homan and Fuchs, 1970). Solutions of phytoalexins extracted from
the plant are spotted on a TLC plateand chromatographed. After removal
of the solvent, the plate is sprayed with a suspension of spores ofa darkly
pigmented fungus in a nutrient solution and the plate incubated at high
humidity and suitable temperaturesfor growth of the fungus for 48-72 hr.
The phytoalexins appear as white spots devoid of fungal growth whereas
the remainder of the plate is covered by a dark mat of fungal mycelium
(Fig. 2). Suitable test fungi are Cladosporium cucumerinum (Aguamah et
al., 1981; Edwards and Strange, 1991), C. herbarum (Ingham, 1976), and
C.cladosporioides (Subba Rao et al., 1988a, 1990, 1991).
Chemical and Physical Detection

Chromogenic Reagents. Manyphytoalexins are phenolic and there-
fore form colored reaction products with reagents such as diazotized p-
nitroaniline (DPN), diazotized sulfanilic acid (DSA), and Gibbs reagent.
These color reactions can be an aid to identification, particularly when
combined withRf data (Ingham, 1982). Other reagents such as iodine vapor
and antimony chloride (SbCl,), which detect lipoidal phytoalexins, have
also been used (Subba Rao et al., 1988a).
W Properties. Some phytoalexins appear as green or light or dark
blue spots on chromatograms when illuminated with long-wavelength UV
light but many do not. The W absorption properties of solutions of pure
compoundscanprovideuseful information for identification(Ingham,
1982).
Identification
Once pure preparations of phytoalexins have beenobtained, mass and nu-
clear magnetic resonance spectrometryare powerful meansfor their identi-
fication and generally provide enough data to propose a structure if the
compound is new. For example, these techniques were usedto deduce the.
structure of the arachidins fromgroundnut cotyledons (Keen and Ingham,
1976; Aguamah et al., 1981; Cooksey etal., 1988).
204 Subba Rao and Strange
AGUAMAH ET AL.. 1981

Groundnut kernels soaked H
in
,O
1
Cut into1-2 mm thick slices
Incubated 48 h in dark at 25OC
1
Homogenized in 95% EtOH
1
Filtered and evaporated at 40°C
1
Partitioned three times with
petroleum ether b.p. 40-60°C
l
l
Aqueous frkction Petroleum fraction
1 (discarded)
Partitioned three times
with EtOAc
EtOAccombinedfractionsAqueousfraction
1 (discarded)
Preparative HPLCon Si column
(mobile phase, Petroleum ether
b.p. 40-60°C:EtOAc -
7:3 V/V)
1
Active fractions dissolved in
acetonitrile
1
Analytical HPLC ODS column
(mobile phase MeCN-H,O 1:l V/V)
Figure 1 Methods for the extraction,separation,andquantitativeanalysis of

groundnut phytoalexins.
Groundnut Phytoalexins and Disease Resistance 205
SUBBA RA0 ET AL. 1988a. 1990. 1993.
Infected leaves ground in 50% MeOH

1
Incubated at room temperature in
dark for 48h
1
Filtered and reduced to
volume
# in vacuo at 40°C
1
Centrifuged 10 min (15 000s)
1
l 1
Pellet
(discarded) 1
Partitioned three times
with hexane
1 1
queous
ion
fraction
Hexane
1
Partition three times
with EtOAc
1 1
mbined
ctions
EtOAc
phase
Aqueous
(discarded) 1
Open column chromatography on Si gel (mobile
phase CHC1,-MeOH
50:l and 50:lO or
Hexane:EtOAc:MeOH -
50:50:5 ( V / V ) )
1
Analytical HPLCon a Si column (mobile phase
Hx-EtOAc 65:35 V/V or a methanol gradient in
Hexane)
1
Analytical HPLCon a C,. column (mobile phase
a gradient of methanol in 1% acetic acid
from 50-99% in 20 min).
Infected leaves
+c
Method 1
4
Method 2
Air-dried samples vacuum Samples vacuum infiltrated

infiltrated with60% EtOH with 60% EtOH
1 1
EtOH removed in vacuo EtOH diluted to 25%
at 4OoC
1 1
Aqueous solution partitioned Solid phase cartridge (500 mg
three times withEtOAc. Aqueous Techoprep C,.: phytoalexins
phase discarded eluted in acetonitrile
1
Combined EtOAc fractions
evaporated to dryness and
dissolved in CHC1,
1
Flash chromatography ona
silica gel column: phytoalexins
eluted in a stepwise gradient
of EtOAc in hexane
1
Semi-preparative HPLC on a C,.
column: phytoalexins eluted in
acetonitrile-H,O 1 :1 (V/V)
Analytical HPLCon a C,.

column: phytoalexins eluted in
a gradient of acetonitrile in
1% acetic acid
Figure 1 Continued.
Resistance
Disease
Groundnut
Phytoalexins
and 207
Figure 2 Detection of phytoalexins using theCladosporiumTLC test. White areas

indicate thelocation of phytoalexins where fungal growth was inhibited.
Quantitative Determination
For many phytoalexins, high-performance liquid chromatography (HPLC)
is the methodof choice (Strange,1987; Subba Rao et al., 1988a, 1990,1991;
Edwards and Strange, 1991; Arora and Strange, 1991). The compounds are
usually separated by reverse phase chromatography using an ODS silica
column and quantitated by referenceto internal standard compounds and/
or external standards of the authentic phytoalexins.
Although no work has yet been achieved with a radioimmunoassay
of groundnut cotyledons, this sensitive technique will undoubtedly be of
importance for establishing whether phytoalexins accumulate at the right
time and in the right place to explain the cessation of growth of invading
microorganisms (Hahn et al., 1985; Huang and Barker, 1991).
Structures and Chemical Characteristics

The compounds so far identified as phytoalexinsfrom groundnut fall into
three main classes: stilbenes, flavonoids, and compounds related to long-
chain fatty acids (Table 1 and Fig. 3). Chromatographic data and their
reactions to chromogenic reagentsare given in Table2. Spectral characteris-
tics are found in Table 3.
C. Biology of Groundnut Phytoalexins

Elicitation
Wounding of sterile tissue caused elicitation of stilbene phytoalexinsin ground-
nut cotyledons (Aguamah et al., 1981; Narayanaswamy and Mahadevan,
1983; Arora and Strange, 1991). Subba Rao (1987) also found that the
wounding caused by detaching leaflets from plants elicited phytoalexins.
Although saltsof heavy metalsand UV light have commonly been used
to elicit phytoalexins in other plants, Edwards and Strange (1991) found
that these techniques only resulted in low levels ofaccumulation inground-
nut leaflets. Accordingly, these and other workers have used fungal chal-
lenge asa means of elicitation.For example, Ingham (1976) elicited phyto-
alexins in groundnut leaflets by inoculation with the nonpathogen Helmin-
thosporiumcarbonum, and similarly,Narayanaswamy and Mahadevan
(1983) used the nonpathogens Curvularia spicata and Helminthosporium
oryzae. Variation in phj.toalexin yield occurred according to the challenging
organism. For example, C. spicata induced 33 pg of trans-resveratroVm1
of diffusate, whereas H . otyzae induced 667 pg/ml (Narayanaswamy and
Mahadevan, 1983).
Substantial yields of phytoalexins have been obtained from plants that
were naturally infected with major fungal pathogens such as Cercospora
arachidicola (Edwards and Strange, 1991), Phoma arachidicola(Strange et
H o v o H - OH 0
OH OH
OH
OH
OH
OH
R2
6.7
OH 10.11
I
O"(CH2)n-0
12
13.14.15
Figure 3 Structures of groundnutphytoalexins.1, Resveratrol. 2, Arachidin I

(4-(3-methyl-but-l-eny1)-3,5,3'4'-tetrahydroxystilbene). Arachidin
3, I1 (4-(3-
methyl-but-2-enyl)-3,5,4'-trihydroxystilbene). 4, Arachidin I11 (4-(3-methyl-but-l-
eny1)-3,5,4'-trihydroxystilbene. 5, Arachidin IV (3-isopentadienyl4,3',5'-trihydroxy-
stilbene). 6, R, = OH, R, = OCH, medicarpin. 7, R, = R, = OH demethylmedi-
carpin. 8, R, = R, = OH daidzein. 9, R, = OH, R, = OCH, formononetin. 10,
R, = R, = OCH,,R, = OH 7,4"dimethoxy-2'-hydroxyisoflavanone.11,R, = R,
= OH, R, = OCH, 7,2'-dihydroxy-4'-methoxyisoflavanone. 12,Nonylphenol.
13-15, Alkyl bk phenyls (n = 11, 12, or 13). 16, Methyl linolenate. 17, 13-Hydroxy-
octadecadiene-9,l l-methyloate. 18,9-Hydroxyoctadecadiene-10,12-methyloate.19,
R, = C,H,, R, = COCH, 1,2,3(2-acetyloxy)tricarboxylic propanic acid.
212
2 5 2
0
z?!
a00
0
E
n
>
\
>
..
rl
W
3 3
I I
0
4
2
I
I l l
f
0 0 0 . 0 0 0 0 0
Groundnut Phytoalexins andDisease Resistance 215
Table 3 Absorption Spectraof Groundnut Phytoalexins

Absorption
Compound (%lax, nm)
Stilbenes
1. 33.4'-Trihydroxystilbene (sh),281,297 306,
(resveratrol)
2. Arachidin I (4-(3-methylbut-l-enyl)- 220,245
(sh),
310
3,5,3',4'-tetrahydroxystilbene) (sh), 346 (sh)
3. Arachidin I1 (4-(3-methylbut-2-enyl)- 220,295
(sh), 307,
3,5,4'-trihydroxystilbene) (sh) 324,340
4. Arachidin I11 (4-(3-methylbut-l-enyl)- 219,241
(sh), 327
3,5,4'-trihydroxystilbene) 331,346 (sh),
(sh), 364 (SW
5. Arachidin IV (3-isopentadienyl-
4,3 ',5 "trihydroxystilbene)
296
Flavonoids
carpin 6. 287
Demethylmedicarpin
7. 282 (sh), 287
212,2388. Daidzein
262 (sh), 305
mononetin
8 9. 249,
262 (sh), 305
10. 7,4'-Dimethoxy-2'-hydroxyisoflavanone 277,311
11. 7,2'-Dihydroxy4'-rnethoxyisoflavanone 275,311
Fatty acid-related compounds
12. phenol -
sphenyls 13-15. Alkyl -
16.
17. 13-Hydroxyoctadecadiene-9,ll-methyloate -
18. 9-Hydroxyoctadecadiene-10,12-methyloate -
19. 1,2,3(2-Acetyloxy)tricarboxylic propanic acid -
al., 1985), Puccinia urachidis (Subba Rao et al., 1988a, 1990, 1991), and
Rhizoctonia bataticola(Narayanaswamy and Mahadevan, 1983).
Two other factors that profoundly affect phytoalexin elicitation and
accumulation are the genotype of the plant and environmental conditions.
These topics will be discussed later along with the role of phytoalexins in
defense.
Biosynthesis and Degradation
Elicitor treatment causesthe derepression ofthe biosynthetic pathway lead-
ing to phytoalexin synthesis. Once synthesizedthe phytoalexin may be sub-
ject to degradation by either host or parasite (Yoshikawa,1983; VanEtten

et al., 1989).
The phytoalexins of groundnuts so far described are stilbenes, flavo-
noids, and compounds related to long-chain fatty acids (Table 1 and Fig.
3). A considerable amount of information about the biosynthetic pathways
for flavonoid and stilbene phytoalexins is available and the reader is re-
ferred to the reviewsof Hahlbrock and Griesbach (1975,1979), Stoessl
(19821, Dewick (1982), Dixon et al. (1983), Ebel and Hahlbrock (1983),
Kind1 (1985), Ebel (1986), Smith and Banks(1986), Keen (1986), and Van-
Etten et al.(1989).
Inbrief,isoflavonoids and stilbenes are formed by the shikimic-
polymalonate acid route. The first step is the synthesis of cinnamic acid
from phenylalanine catalyzed by phenyl alanine ammonia-lyase. This is
then converted to 4-coumaroyl-CoA by the action of cinnamate hydroxy-
lase and 4-coumarate:CoA ligase. Reaction of 4-coumaroyl-CoA with three
units of malonyl-CoA givesan intermediate which can cyclize in differ-two
ent ways leading to two series of compounds. Isoflavonoids are found in
one of these series and stilbenes in the other. Medicarpin, an important
isoflavonoid phytoalexin of groundnut leaves, is probably synthesized via
daidzein and formononetin, compounds that have also beenfound in leaves
infected by Cercospora arachidicola (Edwards and Strange, 1991). The al-
ternative cyclization leads to the biosynthesis of stilbene phytoalexins of
groundnut via decarboxylationand the appropriate hydroxylation and pre-
nylation of the 4-coumaroyl-CoA-malonyl-CoAintermediate.
At presentthere is little information concerning the biosynthesis ofthe
other groundnut phytoalexins such as methyl linolenate, the dienols, and
alkyl bis phenyl ethers. Presumably, methyl linoleateis formed according
to the normalroute of fatty acid synthesis, i.e., by the progressive conden-
sation of two-carbon units derived from malonyl-CoA. RavisC (personal
communication) suggested that the dienols are also formed via this route
and that the alkyl bis phenyl ethers are derivedfrom acetyl-coA.
Mechanisms for the degradation of groundnut phytoalexins are virtu-
ally unknown, although Edwards and Strange (1991) suggested that the
occurrence of demethylmedicarpin in leaves infected withCercospora ara-
chidicola, by analogy withthe demethylation of pisatin by Nectria hamato-
cocca (VanEtten et al., 1989), was a degradation product. Demethylmedi-
carpin was not detected ingroundnut leaves infected by rust.
BiologicaI Activity
The spectrum of biological activity of groundnut phytoalexins has hardly
begun to be explored and so far has been confined to antifungal activity,
especially against parasites ofgroundnut (Table 4). The lowest EDsovalue
Groundnut Phytoalexins and Disease Resistance
(2.2 pg/ml) was that of the two isomeric dienolsfor the inhibition of germ
tube growth of uredospores ofPuccinia arachidis, and the highest (140 pg/
ml) was that of demethylmedicarpinfor inhibition of spore germination of
CIadosporium spp. (Table4).
Role in Defense
Attempts to assess the role of phytoalexins in resistance have been made
mainly with leaf spot fungi and fungi of the A . fravus group. In order to
provide a context for evaluating thesedata, it is worth reviewing the criteria
that have been proposed to establish whether a phytoalexin can be consid-
ered to play a role in defense. These
are as follows:
1. The compound must accumulate in response to infection.
2. The compound mustbe inhibitory to the invading organism.
3. The compound must accumulate to inhibitory concentrations in the
vicinity of the parasite at the time the parasite ceases growing.
4. Varying the rate of accumulation of the phytoalexin causes a corre-
sponding variation in the resistance ofthe plant.
5. Varying the sensitivity of the invading organism causesa corresponding
variation in its virulence (Strange, 1992).
Of these criteria, the first three are mandatory and the remaining two
provide corroborativeevidence.
Resistance to Toxigenic Aspergihs Species. Groundnuts are frequent-
ly infected by the mycotoxigenic fungiA . flavus and A. parasiticus. Strains
of A. fravus normally synthesize aflatoxinB, and GI whereas strains of A .
parasiticus usually produceaflatoxin B,, B1,G,, and G2.In addition to their
toxigenicity, these compoundsare reported to be carcinogenic, mutagenic,
and teratogenic (Hayes, 1980). Aflatoxin B1 is the most toxic of the four
compounds and is alsoa powerful liver carcinogen (Heathcote and Hibbert,
1978). Infection of groundnuts usually takes place before harvest from soil
infested by the fungi but may also occur during harvest and storage (Wil-
liams and McDonald, 1983; Strange, 1991).
In attempts to find a simple method for screening genotypesfor resis-
tance to mycotoxigenic AspergiIIus spp. an in vitro technique has been
established. Groundnut cotyledons are hydrated to 20% moisture and
rolled ina spore suspension ofthe fungi before incubation. Results of such
tests have shownthat cultivars which are resistant inthe in vitro screenare
also resistant inthe field (Mehan et al.,1981; Zambettakis, 1983).
Susceptibility to Aspergillus spp. isinfluenced by the environment
(Strange, 1991). In particular, drought stress and high temperatures increase
susceptibility (Hill et al., 1983; Coleet al., 1985; WottonandStrange,
1985,1987).
Resistance
Disease
Phytoalexins
Groundnut
and 219
The evidencethat phytoalexins play a role in resistanceto Aspergillus

spp. will be reviewed according to the five criteria setout in the preceding
section.
1. Stilbene phytoalexins accumulated in groundnut cotyledons in re-
sponse to challenge by A. flavus using the in vitro technique described
above (Wotton and Strange,1987; Arora and Strange, unpublished).After
a lag of2 days phytoalexins accumulated rapidly, reaching > 140 pg/g fresh
wt of kernels (Wotton and Strange, 1987).
2. Wotton and Strange (1985, 1987) found that arachidins I, 11, and
I11 (Fig. 3) inhibited both spore germination and hyphal extension of A.
flavus in vitro with ED,, values ranging from 8.9 to 12.8 pg/ml and from
4.9 to 9.7 pg/ml, respectively (Table4).
3. Wotton and Strange (1987) found that the fungus essentially ceased
to grow after 2 days in the in vitro seed testing technique. This coincided
with the time that phytoalexins began to accumulate rapidly, going from
- 12 pg/g fresh wt to > 50 pg/g fresh wt. This latter figure represents a
concentration that is about four times the ED,, value.
4. Wotton and Strange (1987) and Arora andStrange (1991, and
unpublished results) found both genotypic and environmentally induced
variation in phytoalexin accumulation. When inoculatedA.with flavus, the
more susceptible cultivar, TMV2, accumulated resveratrol and arachidin
IV, whereas a more resistant cultivar, J l l , accumulated both these com-
pounds and additionally arachidin 111. Moreover, total phytoalexin in the
more resistant cultivarwas > 300 pg/g fresh wt 120 hr after inoculation by
the in vitro technique compared with 90 pg/g fresh wt for the susceptible
cultivar. These differenceswere reflected inthe accumulation of aflatoxins
-
which reached 150 ppb in the more resistant cultivarbut > 12,000 ppb in
the susceptible cultivar.
Wotton and Strange (1987) confirmed that drought stress enhancedthe
susceptibility of groundnut kernels to A. flavus when tested by the in vitro
technique. Reduction in water supply also inhibited the phytoalexin re-
sponse of such kernels.
Since phytoalexinsare elicited ingroundnut kernels by wounding in the
absence of pathogenic challenge was it possible to evaluate the phytoalexin
responsewithout the complicationsconsequent on the presenceof the
pathogen in infection experiments. In time course studies of three cultivars,
phytoalexins accumulated within 24 hr of wounding kernels and reached
maximain 96-120 hr, after whichtheybegan to decline(Wotton and
Strange, 1985). Cultivars differed in their speed of response, the susceptible
cultivar producingthe least amount of phytoalexinat 24 hr but outstripping
the other cultivars at 96 hr. In other experiments, phytoalexin accumula-
tion, measured in a set of 10 groundnut cultivars at 24 hr after wounding,
ranged from 28 to 935 pg/g fresh wt and was negatively correlated with
susceptibility in the in vitro screen (see above). No such correlation was
obtained with phytoalexin accumulation at 96 hr after wounding (Wotton
and Strange, 1985). It seems, therefore, that resistance to A. flavus may be
related to early accumulationof phytoalexin rather than the final amounts
found after prolonged incubation. Similar conclusions have been reached
by workers investigating other systems (Mansfield, 1982).
5. At present thereare no data reporting variation in the sensitivity of
A. flavus to groundnut phytoalexins.
Resistance to Leaf Spot Fungi. The possibility that phytoalexins play
a role in limiting infection bythe leaf spot fungi, Cercospora arachidicola
(early leafspot), Phaeoisariopsispersonata (late leaf spot), Phoma arachid-
icola (web blotch), and Puccinia arachidis (rust) will be discussed in the
context of the five criteria previously stated.
1. Cole (1982) reported that early colonization of groundnut leaflets
by C. arachidicola prevented their subsequent parasitism by P. arachidi-
cola. Juice from the C. arachidicola-infected leaflets, but not healthy leaf-
lets, contained a compound that was inhibitory to P. arachidicola. The
compound was isolated by Strange et al. (1985) and identified as medicar-
pin. These results suggested that the accumulation of medicarpin prevented
the colonization of early leaf spot-infected leaves with web blotch.
In similar experiments Subba Rao (1987) found that the in vitro toxicity
of crude extracts from resistant genotypes infected with rust were always
more toxic toward uredospore germination of the fungus than those of
susceptibleones.Moreover, the highertoxicitylevelsweremaintained
throughout the latent period (Subba Rao, 1987).
In further experiments Subba Rao et al. (1988b) isolated 12 antifungal
compounds from rust-infected leaves and determined the structures of eight
ofthem.Methyllinolenate,twoisomericdienols,namely13-hydroxyocta-
decadien-9,l l-methyloate and9-hydroxyoctadecadien-l0,12-methyloate, a nonyl
phenol, 1,2,3(2-acetyloxy)tricarboxylic propanic acid, and three alkyl bis
phenyl ethers with aliphatic chains correspondingto 11, 12, and 13 carbon
atoms were reported as phytoalexins for the first time. Medicarpin and
arachidin I1 were also isolatedbut in very low concentrations (SubbaRao,
1987).
In contrast, Edwards and Strange (1991) found that concentrations of
both formononetin and medicarpin exceeded 120 pg/g fresh wt in leaves
infected by Cercospora arachidicola. These workers, in addition, found
demethylmedicarpin (>50 pg/g fresh wt) a possible degradation product
of medicarpin. Cole et al. (unpublished) also found medicarpin and low
concentrations of demethylmedicarpin inthe healthy upper leavesof plants
infected naturally inthe field withC. arachidicola or P. arachidicola. These
Resistance
Disease
Phytoalexins
Groundnut
and 221
leaves did not develop symptoms of disease when detachedand incubated

under conditions that were conducive to fungal growth. It is not known
whether the phytoalexins were translocated from the lower infected leaves
or whether they were synthesized in situ in response to a signal from the
infected part of the plant.
2. An important component of the phytoalexin response to leaf spot
diseases was generally medicarpin although this was not true for all culti-
vars. TheED,, for medicarpin when testedfor inhibition of spore germina-
tion of C.arachidicola was 5 pg m1 and for demethylmedicarpin, a probable
degradation product, 92 pg m1 (Table 4;Edwards, 1992). It was difficult to
obtain an ED,, value for medicarpin against hyphal extension since the
fungus degraded itto demethylmedicarpin (Edwards,1992). Spore germina-
tion of C.arachidicola was also inhibited by 7,4’-dimethoxy-2’-hydroxyiso-
flavanone, a prominent component of the phytoalexin response of some
cultivars to infection by this fungus (Table4;Edwards, 1992).
The inhibitory activity of methyl linolenate, a mixture of the dienol
isomers and a mixture of the alkyl bis phenyl ethers,was estimated against
uredospores of P. arachidis in vitro. TheED,, values for uredospore germi-
nation were 37.5, 3.8,and 40 pg/ml for methyl linolenate,the dienols, and
alkyl bis phenyl ethers, respectively. Similarly, the ED,, values for germ
tube growth were 2.2 and 35 pg/ml for dienols and alkyl bis phenyl ether,
respectively (Table4).
3, 4, and 5 . No adequate evaluation of the role of phytoalexins in
resistance to leaf spot diseases has been carried out according to these
criteria.
111. STRATEGIES FOR INCREASING PLANT RESISTANCEBY

EXPLOITING THE PHYTOALEXIN RESPONSE
The evidence reviewed above for phytoalexin involvement inthe resistance
of groundnut to fungal diseases is circumstantial. A major requirement is
to show that phytoalexin accumulationslows the progress of infection. As
previously discussed, such evidence may be biochemical or genetic. In the
biochemical approach, the concentrations of the different compounds
should be assessed at infection sites in order to ascertain the most impor-
tant. These could then be assayed routinely by HPLC. Alternatively, if
HPLC is not sufficiently sensitiveto detect inhibitory concentrations of the
phytoalexins in the vicinity of the parasite, radioimmune assays could be
developed.
One geneticapproach to establishing phytoalexin involvement in resis-
tance hinges on the discovery, or the development by mutation, of strains
of pathogens that differ markedly in their sensitivity. A correlation of
virulence with insensitivity to phytoalexin is evidence for a role of the

phytoalexin in resistance (VanEtten et al.,1989). If the mechanism of toler-
ance is phytoalexin degradation, as seems likely inthe case of Cercospora
arachidicola, which appears to demethylate medicarpin, then transforma-
tion of nondegrading isolates with the gene for medicarpin demethylase
should enhance the virulence of the recipients (cf. VanEtten et al., 1989).
Another approach is to prevent the production of phytoalexins by trans-
forming plants with an antisense gene to one of the key biosynthetic en-
zymes. If the phytoalexin isimportant to disease resistance, such transgenic
plants will be more susceptiblethan control plants.
Once phytoalexins have been demonstrated unequivocally to be crucial
in defense against infection, then it will beimportant to exploit the response
in orderto combat disease. This may be done in three ways:
1. Plants may be selected for a rapid phytoalexin response to chal-
lenge byparasites resulting in the accumulation of inhibitory concentrations
of the compounds before the parasite is able to grow awayfrom the infected
area. It will be important that this response occurs under conditions that
are likely to obtain in the field. Critically, with regard to infection of
groundnut cotyledons by Aspergillus spp., the response should occureven
if the plant is drought-stressed (see above; Wotton and Strange, 1987).
2. Plants may develop systemic acquired resistance to infection. For
example,asreportedabove,leaves from the tops of groundnut plants
whose lower leaves were infected by leaf spot fungi remained healthy and
contained medicarpin. One possibility is that the compound was synthe-
sued in situ in response to a signal. Recently, salicylic acid has been shown
to function as such a signal for a number of defense reactions in tobacco
and cucumbers (Malamy et al., 1990; Mktraux et al., 1990). It would be
interesting to know if salicylic acid canact in thisway in groundnuts and if
the response isthe production of medicarpin. Sucha finding would prompt
experiments in determining whetherthe application of salicylic acid might
promote resistance without seriously compromising the yield of the crop. If
yieldswere not significantly depressed, the application ofsalicylicacid
through a sprinkler system might be worth investigating asa possible con-
trol measure.
3. Isoflavone reductaseis a key enzyme in the biosynthesis of medic-
arpin. In groundnuts it leads to the formation of (+)-medicarpin (Strange
et al., 1985) whereas inalfalfa the enantiomer (-)-medicarpin is formed. It
is unlikely that parasites that rely on the degradation of one enantiomerof
medicarpin for virulence would be ableto degrade the other. This could be
tested with groundnut pathogens. If (-)-medicarpin is not degraded by,
for example, C.arachidicola or P. arachidicola then it would beimportant
to genetically engineer plants to produce this form of the phytoalexin by
Resistance
Disease
Phytoalexins
Groundnut
and 223
transforming them withthe isoflavone reductasefrom alfalfa. Such plants

could then be tested
for their resistanceto the leaf spot fungi.
IV. EPILOG
We now have some knowledge of the chemistry of the phytoalexins pro-
duced by groundnuts. However, it is likely that this is still incomplete.
Much work remains to be done in establishing the variation in the phyto-
alexin response among Arachis spp. as well as the mechanisms by which
they are elicited. A further area of research is the effect of environmental
conditions on the phytoalexin response, particularlydrought stress.
Research aimed at assessing the role of phytoalexins in resistance is
urgently needed.The evidence so far is circumstantial. Critical biochemical
and genetic experiments along the lines outlined inthis chapter now needto
be done.
If, as seems likely, the phytoalexin response is found to play a crucial
role inthe resistance ofgroundnuts to infectious disease,then this response
should be exploited by conventional plant breeding and, perhaps, by genetic
engineering.
ACKNOWLEDGMENTS
The authors thank Dr. A. RavisC (Director of Research Retd, ORSTOM,
Paris) for his comments.
REFERENCES
Aguamah, G. E., Langcake, P., Leworthy, D. P., Page, J. A., Pryce, R. J., and
Strange, R. N.(1981). Two novel stilbene phytoalexins from Arachis hypogaea.
Phytochemistry20 1381-1383.
Arora, M. K., and Strange, R. N. (1991). Phytoalexin accumulation ingroundnuts
in response to wounding. Plant Sci. 78 157-163.
Ciufetti, L. M., Welltring, K."., Turgeon, B. G., Yoder, 0. C., and VanEtten, H.
D. (1988). Transformation of Nectria haematococca with a gene for pisatin
demethylating activity, and the role of pisatin detoxification in virulence. J.
Cell Biochem.12C: 278 (Abstr).
Cole, D. L. (1982). Interactions between Cercospora arachidicolaand Phoma ara-
chidicola, and their effectson defoliation and kernel yield ofgroundnut. Plant
Pathol. 31: 355-362.
Cole, R. J., Sanders, T. H., Hill, R. A., and Blankenship, P. D. (1985). Geocar-
posphere temperatures that induce preharvest aflatoxin contamination of pea-
nuts under drought stress. Mycopathologia 91:41-46.
Cooksey, C. J., Garratt, P. J., Richards, S. E., and Strange, R. N.(1988). A dienyl
stilbene phytoalexin fromArachis hypogaea. Phytochemistry27: 1015-1016.
COPR (Centre for Overseas Pest Research) (1981).Pest Control in Tropical Grain
Legumes, Overseas DevelopmentAdministration, London.
Dewick, P. M. (1982). Isoflavonoids, The Flavonoids: Advancesin Research (J. B.
Harborne and T. J. Mabry, eds.), Chapman and Hall, London, pp. 535-640.
Dixon, R. A., Dey, P. M., and Lamb, C. J. (1983). Phytoalexins: enzymologyand
molecular biology.Adv. Enqmol. Rel. AreasMol. Biol. 55: 1-136.
Ebel, J. (1986). Phytoalexin synthesis: the biochemical analysis of the induction
process. Annu. Rev. Phytopathol. 24:235-264.
Ebel, J., and Hahlbrock, K. (1982).Biosynthesis. The Flavonoids: Advances in
Research (J. B. Harborne and T. J. Mabry, eds.), Chapman and Hall, London,
pp. 641-679.
Edwards, C. (1992). Investigation of the Phytoalexin Response of Arachis hypogaea
L. and its possible involvementin resistance to pathogens, PhD thesis, Univer-
sity ofLondon.
Edwards, C., and Strange, R. N. (1991). Separation and identification of phytoalex-
ins from leaves ofgroundnut (Arachis hypogaea) and development of a method
for their determination by reversed-phase high-performance liquidchromatog-
raphy. J. Chromatogr. 547: 185-193.
FAO. (1990). Production YearBook, 44: 109.
Feakin, S. D. (ed.) (1973). Pest control in groundnuts, PANS Manual No. 2, Over-
seas DevelopmentNatural Resources ResearchInstitute, ODA, London.
Gibbons, R.W. (1980). Groundnut improvement research technologyfor the semi-
arid tropics, Proceedings of the International Symposium on Development and
Tramfer of Technologyfor Rainfed Agriculture and the SAT Farmer, August
28-September 1, 1979, International Crops Research Institute for the Semi-
Arid Tropics, Patancheru, India, pp. 27-37.
Hahlbrock, K., and Griesbach, H. (1975). Biosynthesis of flavonoids. The Flavo-
noids: Advancesin Research (J. B. Harborne and T. J. Mabry, eds.), Chapman
and Hall, London, pp. 866-915.
Hahlbrock, K., and Griesbach, H. (1979). Enzymiccontrols in biosynthesis of lignin
and flavonoids. Annu. Rev. Plant Physiol.3 0 105-130.
Hahn, M. G., Bonhoff, A., and Griesbach, H. (1985). Quantitative localization of
fected with Phytophthora megasperma f. sp. glycinea. Plant Physiol. 77: 591-
601.
Hayes, A. W. (1980). Mycotoxins: a review of biological effects and their role in
human diseases. Clin. Toxicol. 17: 45-83.
Heathcote and Hibbert (1978). Production of aflatoxins. Developments in Food
Science. 1. Aflatoxins: Chemicaland BiologicalAspects, Elsevier, Amsterdam,
pp. 26-29.
Hill, R. A., Blankenship, P. D., Cole, R. J., and Sanders, T. H. (1983). Effects of
soil moistureand temperature on preharvest invasion ofpeanuts by the Asper-
gillusflavus group and subsequent aflatoxin development. Appl. Environ. Mi-
crobiol. 45:628-633.
Homans, A. L., and Fuchs, A. (1970). Direct bioautography on thin layer chroma-
Resistance
Disease
Phytoalexins
Groundnut
and 225
tograms as a method for detecting fungitoxic substances. J. Chromatogr. 51:

327-329.
Huang, J. S., and Barker, K. R. (1991). Glyceollin-Iinsoybean-cystnematode
interactions: spatial and temporal distribution in roots of resistantand suscepti-
ble soybeans.Plant Physiol.96 1302-1307.
Ingham, J. L. (1976). 3,5,4’-Trihydroxystilbeneas a phytoalexin from groundnut
(Arachis hypogaea). Phytochemistry 15: 1791-1793.
Ingham, J. L. (1982). Phytoalexins from the Leguminosae. Phytoalexins (J. A.
Bailey and J. W. Mansfield, eds.), Blackie, Glasgow, pp. 21-80.
Jackson, C.R., and Bell,D. K. (1969). Diseases of peanut (groundnut) caused
by fungi. ResearchBulletin, 56, Univ.ofGeorgia,CollegeofAgriculture,
Experiment Station.
Keen, N. T. (1986). Phytoalexins and their involvement in plant disease resistance.
Iowa State J. Res. 6 0 477-499.
Keen, N. T., and Ingham. J. L. (1976). New stilbene phytoalexins from American
cultivars ofArachis hypogaea. Phytochemistry 15: 1794-1795.
Kindl, H.(1985).Biosynthesisof stilbenes.BiosynthesisandBiodegradationof Wood
Components(T. Highchi, ed.), Academic Press, New York, pp.345-377.
Malamy, J., Carr, J. P., Klessig, D. F., and Raskin, I. (1990). Salicylic acid:a likely
endogenous signal in the resistance response of tobacco to viral infection.
Science 250 1002-1004.
McDonald, D. (1984). The ICRISAT Groundnut Program, Proceedings of the Re-
gional Groundnut Workshopfor Southern flrica, March 26-29, 1984, Lilon-
gwe, Malawi, International Crops Research Institute for the Semi-Arid Trop-
ics, Patancheru, India.
(J. A. Bailey and J. W. Mansfield, eds.), Blackie, Glasgow, pp. 133-180.
Mehan, V. K., McDonald, D., Nigam, S. N., and Lalitha, B. (1981). Groundnut
cultivars withseed resistance to invasion by Aspergillusflavus. OIPagineux36
501-505.
MCtraux, J. P., Signer,H.,Ryals, J., Ward, E., Wyssbenz,M., Gaudin, J.,
Raschdorf, K., Schmid, E., Blum, W., and Inverardi, B. (1990). Increase in
salicylic acidat the onset of systemic acquired resistance in cucumber.Science
250 1004- 1006.
Miao, V. P., Covert, S. F., and VanEtten, H. D. (1991). A fungal gene for antibi-
otic resistance on a dispensable (B) chromosome. Science 254: 1773-1776.
Narayanaswamy, P., and Mahadevan, A. (1983). Isolation of phytoalexins from
germinating seeds ofgroundnut (A. hypogaea).Acta Phytopathologica Acade-
miae Scientiarum Hungaricae1 8 33-36.
Paxton, J. (1981). Phytoalexins: a working redefinition. Phytopathol. Z. 101: 106-
109.
Porter, D. M., Smith, D. H., and Rodriquez-Kabana, R. (eds.) (1984). Compen-
dium ofpeanut diseases. Am. Phytopathol. Soc.
Smith, D.A., and Banks, S. W. (1986). Biosynthesis,elicitation and biological
activity of isoflavonoid phytoalexins. Phytochemistry 25: 979-995.
Stoessl, P. (1982). Biosynthesis of phytoalexins. Phytoalexins (J. A. Bailey and J.

W. Mansfield, eds.), Blackie, Glasgow, pp.133-180.
Strange, R. N. (1987). HPLC of phytoalexins.Modern Methodsof Plant Analysis,
New Series5 (H. F. Linskens and J. F. Jackson, eds.), Springer-Verlag, Berlin,
pp. 121-148.
Strange, R. N.(1991). Natural occurrence of mycotoxins groundnuts,
in cottonseed,
soya and cassava. Mycotoxins and Animal Feed Stuffs; Natural Occurrence,
Toxicity and Control (J. E. Smith and R. S. Henderson, eds.), CRC Press,
Boca Raton, pp. 341-362.
Strange, R. N. (1992). Resistance: the role ofthe hypersensitive responseand phyto-
alexins. Plant Response to Foliar Pathogens (P. G. Ayres, ed.), Biosis Scien-
tific, Oxford.
Strange, R. N., Ingham, J. L., Cole, D. L., Cavill, M. E., Edwards, C., Cooksey,
C . J., and Garatt, P. J. (1985). Isolation of the phytoalexin, mediwpin from
leaflets of Arachis hypogaeaand related species ofthe tribe Aeschynomeneae.
2.Naturforschung. 40c:313-316.
Subba Rao, P. V. (1987). La rouille de l'arachide: Ctude de quelques mkcanismes de
dCfense de l'hdte, These de Doctorat, UniversitC de Paris XI, Orsay, France.
Subba Rao, P. V., Geiger, J. P., Einhorn, J., Malosse, C., Rio, B., Nicole, M.,
Savary, S., and RavisC, A. (1988a). Isolation of methyl linolenate, a new anti-
fungal compound from Arachis hypogaea L. leaves infected with Puccinia
arachidis Speg. OMagineux 43: 173-177.
Subba Rao, P. V., Geiger, J. P., Einhorn, J., Rio, B., Malosse, C., Nicole, M.,
Savary, S., and RavisC, A. (1988b). Host defence mechanisms against ground-
nut rust. Znt. Arachis Newslett.(IAN)4: 16-18.
Subba Rao, P. V., Einhorn, J., Geiger, J. P., Malosse, C., Rio, B., and RavisC, A.
(1990). New dienol phytoalexinsfrom Arachis hypogaeaL. infected withPucci-
nia arachidisSpeg. OlpaSineux 45: 225-227.
Subba Rao, P. V., Einhorn, J., Geiger, J. P., Malosse, C., Rio, B., and RavisC, A.
(1991). Alkyl bis-phenyl ethers, new phytoalexins produced by Arachis hypo-
gaea L. infected withPuccinia arachidisSpeg. OI&agineux 46:501-507.
Subrahmanyam, P., and McDonald, D. (1983). Rust disease of groundnut. ICRI-
SAT Information Bulletin No. 13, International Crops Research Institute for
the Semi-Arid Tropics, Patancheru, India.
Subrahmanyam, P., and McDonald, D. (1987). Groundnut rust disease: epidemiol-
ogy and control, Groundnut Rust Disease.Proceedings of a Discussion Group
Meeting Held at ICRISAT Center, Patancheru, India. Sept. 24-28, 1984, pp.
23-39.
Subrahmanyam, P., and Ravindranadh, V. (1988). Diseases of groundnut caused
by fungi and nematodes. Groundnut (P. S. Reddy, ed.), ICAR,New Delhi, pp.
453-507.
VanEtten, H. D., Matthews, D. E.,and Matthews, P. S. (1989). Phytoalexin detoxi-
fication: importance for pathogenicity and practical implications. Annu. Rev.
Phytopathol. 27: 143-164.
Williams, R. J., and McDonald, D. (1983). Grain molds in the tropics; problems
and importance. Annu. Rev. Phytopathol. 21: 153-178.
Resistance
Disease
Groundnut
Phytoalexins
and 227
Wotton, H. R., and Strange, R. N. (1985). Circumstantial evidence for phytoalexin

involvement in the resistance of peanuts to Aspergillusflavus. J. Gen. Micro-
biol. 131: 481-494.
Wotton, H.R., and Strange, R. N. (1981). Increased susceptibility and reduced
phytoalexin accumulation in drought stressed peanut kernels challenged with
Aspergillusflavus. Appl. Environ. Microbiol.53: 210-213.
Wynne, J. C., Beute, M. K., and Nigam, S. N. (1991). Breeding for disease resis-
tance inpeanut (Arachis hypogaeaL.). Annu. Rev. Phytopathol.2 9 279-303.
Yoshikawa (1983). Macromolecules, recognition and the triggering of resistance.
Biochemical Plant Pathology (J. A. Callow, ed.), Wiley, New York, pp. 267-
298.
Zambettakis, C. (1983). Results of research on selected groundnut hybrids to limit
Aspergillus flavusinfection. Comptes Rendu deSt?ances de I'Academie d'Agri-
culture deFrance 69: 4-50.
l1
Phytoalexins from the Crucifers
Thierry Rouxel, Albert Kollman, andMarie-Hekne Balesdent
Institut Nationalde la RechercheAgronomique, Versailles, France
1. INTRODUCTION
A. GeneralInformation
Crucifers are distributed all around the world and are present in almost
every available kind of environment from the Arctic Circle to the tropics
(Hedge, 1976). However, the greatest number of genera are found in tem-
perate regionsof the Northern Hemisphere.
Cruciferous crops are grown as 1.) oilseed crop for seed-based condi-
ments and oils, including oils of industrial use, 2.) forage and fodder crops
for animal feeds, and 3.) vegetables for human consumption. Manyof the
cruciferous crops, especially within the genus Brassica, have been cultivated
in the neolithic age and have been described by the ancient Greeks, Ro-
mans, Indians, and Chinese (Prakash and Hinata, 1980). Age-old breeding
has selected a wide range of crop types within each of the species, and
the extensive variation within species has caused considerable taxonomic
confusion(Crisp, 1976; Prakash and Hinata, 1980; Williams and Hill,
1986). Each of the main cultivatedBrassica species thus contains numerous
varieties which are grown for oil, vegetables, or animal feeds. However,
varieties of different species, i.e., B. napus, B. juncea,and B. rapa, grown
for oil are often referred to by the samecommonname,oilseedrape.
Rapeseed oil isthe fourth commonly traded oil inthe world. Mostnorthern
European countries produce rapeseed, i.e.,B. napus var. oleifera, as their
main edible oil crop (Williams and Hill, 1986). Canada, India, and the
European Economic Community(EEC)are currently the main oilseed rape
producers.
The six major cultivatedBrassica species are genetically closely related.
Three diploid species, B. nigra (bb), B. rapa syn. campestris (aa), and B.
oleracea (cc), are the natural genitors of the three allotetraploid species B.
napus (aacc), B. juncea (aabb), and B. carinata (bbcc) (Fig. 1) (U,1935).
229
230 Rouxel et al.
B.y-uncea B. carinata
(aebb) (bbcc)
/
B. rapa
(ea)
n = 18
8. napus
(eecc)
n = 17
\ 8. olerucea
(cc)
n = 10 n = 19 n=9
Figure 1 Genetic relationships of the cultivated Brassica species. (From U,1935.)
B. Brassica napus and Blackleg Disease of Crucifers

Diseases of B. napus are mainly due to fungal pathogens, most of them
having only a low incidence on yield. One of the major diseases of oilseed
rape is blackleg of crucifers caused by Leptosphaeria rnaculans (Desm.)
Ces. et de Not. [imperfect stagePhoma lingam (Tode ex Fr.) Desm.] (for a
review, see Gabrielson, 1983). The disease was the limiting factor for the
growth of most Brassica crops worldwide in the 1970s and the early 1980s.
The breeding of B. napus cultivars displaying adult stage resistance, i.e.,
field resistance, has provided fairly good control of the disease since then.
The life cycle of the pathogen (Gabrielson,1983; Regnault et al., 1987)
'is characterized by a very long symptomless development inside the host
plant. The ascospores, discharged from the previous year's stubbles,are the
main source of primary inoculum (Brunin and Lacoste,1970). The patho-
gen penetrates the leaves or cotyledons of the plantlets through stomata
or wounds. These primary infections of leaves have been shown to be a
prerequisite for stem infection (Hammond et al., 1985). After entering the
plant, the pathogen grows profusely between the epidermis and palisade
layer, forming a sheet of hyphae. Branches behind the hyphal front grow
downward into the spongy mesophyll. The pathogen then causes the typical
primary symptom of the disease on leaves, stems, or cotyledons, a greyish
green tissue collapse on which pycnidiae differentiate. Pycniospores can
cause secondary infection of the diseased or surrounding plants. During the
appearance of the necrotic symptom, hyphae ahead of the necrotic zone
eventually reach minor veins and, using the intercellular spaces of
the cortex
beneath the veins, initiate a systemic growth through the petiole into the
main stem (Hammond et al., 1985). During this systemic growth, the fun-
rom Phytoalexins the Crucifers 231
gus is biotrophic, and no macroscopic symptom can be observed. Eventu-

ally, upon reaching the basis of the stem, the pathogen turns necrotrophic
and causes the stem canker responsible for the lodging of the plants. The
sexual stage of the fungus grows as a saprophyte on Brassica stubbles,
and it can thus survive in the soil for 4-5 years, still remaining infectious
(Alabouvette and Brunin, 1970).
Brassica with the B genome, i.e., B. juncea, B. nigra, and B. carinata
(Fig. l ) , display a complete resistanceto L. maculans, expressed asa hyper-
sensitive response, at all growth stages ofthe plants. In contrast, B. napus
displays a field resistance only, which is incompleteand only expressed at
the adult stage.This latter resistancecould be due to the development
of morphological barriers restricting the systemic infection of the plants
(Hammond and Lewis, 1986).
According to Roy (1978), genetic determinantsof adult stage resistance
could be localized inthe A genome ofB. napus and be polygenic innature.
In contrast, the hypersensitive resistance ofthe mustards could be located
in the B genome and be oligo- or monogenic (Roy, 1978). Recent findings
by Keri et al. (1990) confirmed that the hypersensitive responseof B. juncea
to two L. maculans isolates was controlled by two nuclear genes with domi-
nant recessive epistatic action.
Adult stage resistance allows a reduction in yield losses. The plants
remain infected, but a small percentage of them develop a stem canker
causing the lodging at the moment of the harvest. In France, the develop-
ment of a “new” disease,the premature ripening of oilseed rape, in which
L. maculans maybeinvolved @run and Jacques, 1990), along with an
increased aggressiveness of L. maculans to the double-low B. napus culti-
vars inEurope, Canada, and Australia, questions the adequacy of the adult
stage resistanceto control the disease.
As a consequence, one ofthe main challenges ofthe breeders is now to
introduce the complete resistance of Brassica with the B genome into B.
napus. Until now, diverse approaches have been tested, using interspecific
hybridprogeny B. napus x B. juncea (Roy, 1978,1984; Sacristan and
Gerdemann, 1986; Rouxel et al., 1990a), B. napus x B. nigra interspecific
hybrids(Sjodin and Glimelius, 1989), B. napus-B. nigra addition lines
(Jahier et al., 1987; Rouxel et al., 199Oc), or B. napus x B. insularis inter-
specific hybrids (Mithen and Lewis, 1988). Most of this work is still in
progress today.
C. Objectives
The aim of our study of Brassica defense responses was, first, to have a
better knowledge ofthe biochemical events associated with resistance
to L.
232 Rouxel et al.
maculans and, second, to know whetherit was possible to screen for disease
resistance genesat an early stage usinga qualitative and quantitative analy-
sis of plant phytoalexin response. Breeding programsto transfer resistance
genes from the B genome ina B. napus background were mainly interested
in this work. Third, we wished to test whether this host-pathogen system
could bea suitable modelfor further research of the molecular and genetic
determinants of plant-pathogen interaction, and for a study of the earliest
events leadingto resistance or susceptibility.

A.MicroorganismCultures
The virulent isolates(1-144, 2-3, and single-ascospore isolateIIal) and the
nonvirulentisolate SV1(Badawy and Hoppe, 1989;Kochet al., 1989;
Rouxel et al., 1989, 199Ob; Hassan et al., 1991) of L. maculans were used.
They were maintained on malt-agar or V8-agar medium in the conditions
previously described (Rouxel et al., 1989).
Erwinia chrysanthemi pv. dahliae, E. carotovora spp. atroseptica, and
Xanthomonas campestris pv. campestris were grown on YDA agar me-
dium, inthe dark, at 3OOC.
B. Plant Material and Inoculations

The interspecific hybrids progeny were issued from a cross between B.
napus and B. juncea. These plants were self-pollinated twiceand screened
for resistance to L. maculans at SO and S1 generations, according to a
cotyledon inoculation test. Cultivars or breeding lines of B. carinata, B.
nigra, B. juncea, B. rapa syn. campestris, B. napus, B. oleracea, and prog-
eny issuingfrom an interspecific crossB. napus x B. juncea were grown in
the greenhouse conditions previously described (Rouxelal., et 1989).
The results of the interactions between plants and L. maculans were
evaluatedaccording to a cotyledon inoculation test (Williams and Del-
wiche, 1979). Cotyledons of plantlets at growth stage l (Harper and Berk-
enkamp, 1975)werewoundedwith a needle and a'lO-pl droplet of L.
maculans pycniosporesuspension (lo6 sporedml) wasplacedover the
wound. Incubation took place in a growth chamber at 18OC. Ten to l 4
days after the inoculation, symptomswere assessed usingthe Williams and
Delwiche rating system (Williamsand Delwiche, 1979). This scale is divided
in 10 classes, with 1 representative of the hypersensitive resistance and 9
representative of a highly susceptible interaction.
Phytoalexins from the Crucifers 233
C. Abiotic Elicitation of Indole Phytoalexins

Except for time-course studies whereby whole plants were elicitedand incu-
bated under greenhouse conditions, abiotic elicitations were performed on
detached leaves or leaf discsfloating on water. Plant samples ranged from
200-400 mg fresh wt (one leaf) to 2 g. Leaves were elicited by spraying 10
mM CuCl, or AgN03 in 5 mg/liter Tween 80. Incubation took place in the
growth chamber, at 18OC, under continuous white fluorescent light, for
48 hr.
UV elicitation was obtained by placingthe detached leaves25 cm under
a 25-W germicidal lamp, and irradiating them for 2 hror more. Incubation
lasted 48 hr as described previously.
D. Effect of Synthesis Inhibitors and Ethylene

on Phytoalexin Accumulation
B. juncea leaf discs were floated for 3 hr on solutions of actinomycin D,
cycloheximide, or sirodesmin PL. Leaf discs were either maintainedon the
inhibitor solution or rinsed and floated on water. Elicitation with copper
chloride was performed as described above and incubation took place in the
growth chamber,at HOC, under continuous illumination,for 12 hr.
Low concentrations of ethylene were applied for 12 hr on B. napus or
B. juncea leafdiscsmaintainedonwater,inclosed transparent plastic
boxes.CuC1, elicitation ofethylene-treatedleaves and incubation took
place as described above.
E. Biotic Elicitation of Indole Phytoalexins

L. maculans was used asa IO6 per m1 pycniospore suspension in water.
Twenty-four-hour-old bacteria were suspended in water and the con-
centration adjusted to 5 x IO’ cells per ml, by measuring the absorbance
of the suspension at 600 nm.
Mycelial wall extracts or culture filtrates ofL. maculans were obtained
from 14-day-and 20-day-old cultures in modified Fries medium (FCrCzou et
al., 1977).
Mycelial walls were washed, homogenized, and further purified accord-
ing to Ayersetal.(1976).Acid or base hydrolysis of the purified cell
walls was performed accordingto Ricci (1986). Purified mycelial wallsand
extracts were freeze-dried and maintained at -2OOC. For the elicitation
they were diluted in sterile twice-distilled water (500 mgdry wt per ml).
In all cases of biotic elicitation, leaves were surface-sterilized, rinsed
with sterile water, and wounded using a multineedle device. Elicitors were
234 Rouxel et al.
applied by sprayingthe solutions or suspensions on leaves.Incubation took

place in the growth chamber conditions described above, under a 16-hr
photoperiod.
F. HPLCApparatus
A Waters multisolvent delivery system equipped with a Waters U6K injec-
tor was used for high-performance liquid chromatography(HPLC). Chro-
matographic and spectral data from the eluate were acquired witha Waters
990 photodiode array detector. Data were computed by the Waters 990+
software loadedon NEC APC 111.
G. Extraction and Purification of Brassilexin and

Cyclobrassinin Sulfoxide
Detached B. juncea leaves (10-12 g fresh weight) were elicited using copper
chloride as described above.
Phytoalexins were extracted eitherby macerating the elicited tissues in
ethanol (95% v/v) with a Waring blender for 1-2 min, at room tempera-
ture, or by immersing the samples in8OoC ethanol (95% v/v) for 15 min.
The ethanolic extract was filtrated and evaporated to dryness at 4OoC
under reduced pressure. The dried extract was partitioned between twice-
distilled water and diethyl ether (100 m1150 ml, three times) and the com-
bined diethyl ether phases were evaporated to dryness. The residue was
taken up in 1 m1 ethanol and the sample was applied to a column of silica
gel 40 (35-70 mesh) and eluted with ethyl acetate-hexane (1:l). The thin
layer chromatography (TLC) Cladosporium bioassay (Rouxel et al., 1989)
permitted the detection of fungitoxic fractionswhich were collected, com-
bined, dried (4OoC, reduced pressure), and dissolved in a small volume of
ethanol. The phytoalexins werefurther purified by HPLC. Fifty-milliliter
samples were injected on a Brownlee C18 column (4.6 X 220 mm, 5-mm
particle size) eluted with methanol-water(1: 1 v/v, 1.5 ml/min). The fungi-
toxic fractions were combined, dried, recovered in a small volume of di-
chloromethane, and purified on a Brownlee Spheri-5 silica column (4.6 x
220 mm, 5-mm particle size) eluted with dichloromethane (1.5 ml/min).
Further purification and characterization of the phytoalexins are described
in Devys etal. (1988, 1990).
H. Determination of Indole Phytoalexins by HPLC

Elicited leaves or leaf discs were extracted and the hexane samples were
as
cleaned up using Spe-ed octadecyl cartridges (Applied Separations, Inc.)
described in SectionI11 (Results).
The colorless eluates obtainedafter clean-up were dried under reduced

pressure (4OOC) and recovered in dichloromethane. The solvent was evapo-
rated at 5OoC under a stream of nitrogen. The residuewas finally takenup
in 100 p1 95% (v/v) ethanol and S- to 20-p1 samples were chromatographed
through a Brownlee S-mm particle size C18 column (4.6 x 220 mm). The
solvent delivery consisted of a linear gradient from 50Vo methanol v/v in
water to 100% methanol in 5 min which was then maintained constant for
10 min. The flow rate was constant at 1.5 ml/min. Pure samples of the
phytoalexins were used as standards.
111. RESULTS
A. First Reports
The fiist evidence of the accumulation of antifungal compounds, synthe-
sized de novo byBrassica plants, was obtained in parallelby our team and
that of Dr. Takasugi, in Japan (Takasugi et al., 1986; Sarniguet, 1986;
Rouxel et al., 1987). Using the standard Cladosporium TLC bioassay, we
showed that elicited B. juncea extracts displayed a large fungitoxic spot,
which was not detected in control plants, when plants were elicited using
the abiotic elicitor silver nitrate (Rouxel et al., 1989). The fungitoxic spot
was localized at Rf 0.55 when silica gel TLC plates were developed using
ethyl acetate.B. napus plants challenged by the abiotic elicitor displayeda
much weaker fungitoxic spot at the same RP Such a weak fungitoxic spot
was also observed whenB. junceaplants were inoculated withL. maculans.
B. Chemical Structures
The phytoalexin was extracted from elicitedB. juncea leaves byeither mac-
erating the tissues in ethanol with a Waring blender or immersion of the
samples in hot ethanol.It was then purified using column chromatography
and HPLC.
The chemical structure of the compound was established on the basis
of physicochemical data, i.e., UV, IR, high-resolution mass spectroscopy
(HR MS), 13Cand 'H NMR, and nuclear Overhauser enhancement (nOe)
difference experiments (Devys et al.,1988). These data confirmed that the
new compound, brassilexin, was a sulfurcontaining indole sharing struc-
tural similarities with brassinin, cyclobrassinin, or methoxybrassinin, pre-
viously characterized from other Brassica species (Takasugi et al., 1988)
(Fig. 2).
Using similar elicitation, purification, and analysis procedures,an ad-
ditional phytoalexin, cyclobrassinin sulfoxide,was obtained fromB. juncea
(Fig. 2) (Devyset al., 1990).
236 Rouxel et al.
1
S
dks- QJ$L H
W" 3 12
10 11
14: R=H
H
15: R=OCHa
e-
\ F4
Figure 2 Chemical structures of phytoalexins from the crucifers (1-15) and some
indole glucosinolates (16-18). (1) Brassilexin (Devys et al., 1988), (2) brassinin, (3)
methoxybrassinin (Takasugi et al., 1988), (4) 4-methoxybrassinin (Monde et al.,
1990a), (5) cyclobrassinin (Takasugi et al., 1988), (6)cyclobrassinin sulfoxide(Devys
et al., 1990),(7) spirobrassinin (Takasugi et al., 1987),(8) brassitin (Monde and
Takasugi, submitted), (9) methoxybrassitin (Takasugi et al., 1988), (10) methoxy-
brassenin B (Monde et al., 1991c), (11) brassicanal A (Monde et al., 1990b), (12)
brassicanal B (Monde et al., 1990b), (13) brassicanal C (Monde et al., 1991b), (14)
camalexin(Browne et al., 1991; Tsuji et al., 1992), and (15)methoxycamalexin
(Browne et al., 1991); (16) glucobrassicin, (17) neoglucobrassicin, (18) 4-methoxy-
glucobrassicin.
1 2
OH
PN OH
3 4
Figure 3 Synthesis of brassilexin starting from 3-indolecarbddehyde. (From

Devys and Barbier, 199Oa.)
Apart from these two compounds, 15 other phytoalexins have been

characterized from Brassica species, i.e., B. rapa and B. oleracea (Takasugi
et al., 1988; Monde et al., 199Oa, 199Ob, 1991b), or other cruciferous plants,
i.e., Raphanus sativus (Takasugi et al., 1987), Camelina sativa (Browne et
al., 1991) and Arabidopsis thaliana (Tsuji etal., 1992) (Fig. 2). Up to now,
the phytoalexins from cruciferous plants havebeen reported to possess an
indole or oxindole nucleus withthe appendage containing one or two sulfur
atoms. No departure from this scheme has yet been reported.
C. Synthesis
Brassilexin has been synthesized with an overall 11% yield, starting from
3-indolecarbaldehyde(Fig.3)(Devys and Barbier, 199Oa). In addition,
brassilexin was obtained in30% yield by periodate-induced degradation of
cyclobrassinin (Devysand Barbier, 199Ob).
Three other phytoalexins have been successfully synthesized:
1. Brassinin was obtained from 3-(aminomethy1)indole treated with car-
bon disulfide in the presence of pyridine
and triethylamine. The dithio-
carbamate saltthus obtained was methylated with methyl iodine to give
brassinin in66% yield (Takasugiet al., 1988).
238 Rouxel et al.
2. Cyclobrassinin was obtained, in 35% yield, by bromination of bras-

sinin with pyridinium bromide perbromide, followed by dehydrobromi-
nation (Takasugi etal., 1988).
3. A simple synthetic method was developed for methoxybrassinin start-
ing from indole-3-carboxaldehyde (Somei et al., 1992). The compound
was obtained in seven steps, in 12% overall yield.
D. Toxicology Studies
Brassilexin inhibitedboth the germination of L. maculans pycniospores in
water and its hyphal growth. The germination was completely suppressed
at 12.5 mg/liter (75 PM), and a significant reduction in hyphal growth was
obtained for concentrations aslow as 1.5-3.12 mg/liter (9-18 pM) (Rouxel
et al., 1989). A similar effect on Alternaria brassicae conidia or mycelium
was observed (Rouxel, 1988).
These data can be compared with those reported for other phytoalexins
from various plant families, i.e., toxicity values ranging from 10 to 100 p M
(Smith, 1982).
Since itwas observed that the compound may have a fungistatic effect
rather than a lethal one (Rouxel etal., 1989), the possible metabolization of
brassilexin by L. maculans has been assessed in liquid growth medium (Fig.
4). Pycniospores of the virulent isolate IIal and the nonvirulent one SVl
were grown in liquid Fries medium, theinpresence of increasing concentra-
tions of the phytoalexin. After21 days of growth,the mycelium was freeze-
dried and weighed, and the phytoalexin was extracted from the medium
and determined. Under these conditions, the growth of both isolates was
totally suppressed for concentrations above12 mg/liter (70 PM). However,
the virulent isolate IIal displayed a highly heterogeneous sensitivityto the
phytoalexin. In a few cases it was even able to grow in the presence of 16
mg/liter brassilexin, which was not the case for the nonvirulent isolateSV1.
Except for lethal concentrations, only low a rate of the initial concentration
of brassilexin could be extractedfrom the medium thus suggesting, in both
the virulent and nonvirulent isolate, an ability to metabolize the compound
(Fig. 4).
Similar experiments, performed with brassinin, showed that this phyto-
alexin was metabolized into methyl(3-indolylmethy1)dithiocarbamate S-
oxide, which is metabolized twoto three times fasterthan brassinin (Soled-
ade et al., 1991). The carboxylic acid that eventually resulted from this
degradation was at least 10 times less toxic to the fungus than brassinin.
.Moreover, recent findings by Taylor et al. (1991b) showed that virulent
isolates were able to specifically and rapidly metabolize brassinin, as com-
pared to nonvirulent isolates.
T
12
l4 I loo
0 6 10 12 16 14
Phytoalexin concentration (mg.T')
Figure 4 Toxicity of brassilexinto L. maculans and degradation of the compound

by the fungus. The toxicity wasassessed on the virulent isolate, IIal, and the
nonvirulent one, SV1, by weighing their freeze-dried mycelium after 21 days of
growth inthe presence of the phytoalexin. Brassilexin degradationby isolate SVl is
expressed as a percentage of the concentration of phytoalexin initially addedto the
medium. Each value represents the mean of five replications. The experiment was
repeated twice.
Up to now, only a few other data are available on the toxicology of

phytoalexins from Brassica. As it is the case for most of the phytoalexins
isolated today, they would seem to act as general toxicants (Smith, 1982).
Most of these compounds have been reported to be toxic to numerous
fungal species including L. maculans (Takasugi et al., 1988; Dahiya and
Rimmer, 1988a; Monde etal., 1990a, 1991b; Browne etal., 1991).
The toxicity of brassilexin to human KB cancer cells and human normal
cells has been compared to that of cyclobrassinin and two other synthetic
sulfur-containing indoles. Under these conditions, the LDSo was found to
be 47 pM (8 mg/liter) for brassilexin and 94 pM (22 mg/liter) for cyclobras-
sinin. The toxic concentrations weresimilar for KB culture or normalcells.
In addition, thetoxicity of brassilexin to plants has also beeninvestigated.
It suppressed the germination of watercress seeds at 300 p M (50 mg/liter)
(TempCte et al., 1991).
240 Rouxel et al.
E. Analytical Conditions
Most of the quantitative analyses of phytoalexin production by cruciferous

plants were performed usingHPLC procedures (Dahiyaand Rimmer, 1989;
Rouxel et al., 1989; Monde and Takasugi, 1992). However, using chloro-
phyllous parts of plants, these analyses were usually complicated by the
presence of pigments, waxes, and other interfering compounds. Moreover,
using aqueous gradients, e.g., water-methanol, these compounds were also
highly damaging to HPLC equipment and columns. The authors usually
bypassed these problems using nonchlorophyllous parts of the plants, or
callus tissues (Dahiya and Rimmer, 1988b; Monde et al., 1991a). The ex-
tracts could also be passed through silica and C18 cartridges priorto analy-
sis, to remove highly polaror nonpolar compounds (Monde et al.,1991a).
As an alternative, diffusates from elicited plant tissuesand etiolated stems
or leaves were analyzed (Dahiyaand Rimmer, 1988a, 1989).
Since our aim was an investigation of the early interaction betweenthe
plants and a leaf pathogen, we chose to study the response of nonetiolated
elicited leaves.The use of an unusual clean-up procedure, priorto analysis,
allowed interfering compoundsto be removed with retention ofthe phyto-
alexins (Kollmann et al., 1989; Rouxel et al., 1991) (Fig. 5). Elicited tissues
were homogenized in 95% (v/v) ethanol, evaporated to dryness under re-
duced pressure, and the residues takenup in hexane. Hexane samples were
passed through dried, unsolvated C18 cartridges and the sorbent washed
with hexane until elution of a colorless solution occurred. The cartridges
were then eluted with methanol-water (50:50) and the eluate, colorless or
slightly colored, contained the phytoalexins. The yield of the method was
very high for brassilexin, brassinin,and methoxybrassinin, since morethan
80% of a known amount of phytoalexin was recovered after extraction,
clean-up, and HPLC analysis. It was lower for cyclobrassinin (61-73%),
but elution of the cartridges with methanol-water (65:35) increased the
yield ofthis compound.
In addition, due to the characteristicUV spectra of these phytoalexins
(Rouxel, 1988),the use of a photodiode detector (PAD Waters 990) permit-
ted quantitative analyses.The three-dimensional display ofthe UV spectra
of the compounds duringthe HPLC, along with a possible further compu-
tation of the spectra and chromatograms, allowed an unequivocal recogni-
tion of each of the phytoalexins, even when weak responses to the elicita-
tion were obtained or when small plant samples (< 1 g fresh wt) were
analyzed.
A major improvement of reverse phase HPLC analysis of indole phyto-
alexins has recently been proposed by Monde and Takasugi (1992; see also
Monde et al., 1991a). The authors used a complex acetonitrile-methanol-
Phytoalexinsfrom the Crucifers 241
P LOADING
THE SAMPLE p ELUTION
WITH p WASHING
WITH
T
AND HeOH/WATER ETHANOL
WASHING (SO/SO)
WITH
HEXANE
STRONG
RETENTION
OF
PIGMENTS
YELLOW T O COLORLESS DARK

GREEN ELUATE ELUATE GREEN ELUATE
CONTAINING
THE PHYTOALEXINS
0
Figure 5 Sampleclean-up protocol using reverse phasecartridges.(From Koll-
mann et al., 1989.)
water gradient system (Fig.6) which allowed themto obtain a well-resolved

chromatogram of 13 phytoalexins and three relatedindolemetabolites.
They finally suggestedthat a combinationof Kollmann’s protocol to clean
up colored plant extracts with their HPLC analysis method could be the
“standard analytical procedureof cruciferous phytoalexins.”
F. Induction of Phytoalexin Accumulation

The efficiency of various elicitors was testedon B. juncea. Abiotic elicitors
like silvernitrate, copper chloride,or UV light allowedthe accumulation of
higher amounts of phytoalexins than L. maculans (Table 1). The fungus,
extracts of its cell walls, or most of the chromatographic fractions of its
culture filtrate were also slow and low inducers of phytoalexin accumula-
tion. The only fraction that allowed both the development of visible reac-
242 Rouxel et al.
0 4 17 14 32 37 42 47
REI€NTIONTlME(min)
Figure 6 Complex gradient system used for an optimal resolution of cruciferous

phytoalexins during reverse phase HPLC analysis. (From Monde and Takasugi,
1992.)
tions and phytoalexin accumulation,fraction 4, was also the one that con-
tained the main toxic activity. Even though sirodesmin PL, the main toxin
produced in vitro, did not induce a phytoalexin response inBrassica plants
(see below), we did not succeed in separating the toxic from the eliciting
activity using liquid chromatography or HPLC procedures. It is thus un-
clear whether uncharacterized elicitorsor toxic compounds originated the
response. So far, the best biotic inducers we tested were various nonhost
bacteria which consistently induced a visible hypersensitive responseand a
rapid and seemingly specific accumulation of phytoalexins (Table 1).
Similar resultswere obtained by Dahiya and Rimmer (1989) when com-
paring the efficiency of chemicalsto that of L. maculans to induce cyclo-
brassinin and methoxybrassinin accumulation in Brassica plants. AgNO,
and, to a lesser extent, CuCI, and HgClz were found to induce the highest
accumulation of both methoxybrassinin and cyclobrassinin. The authors
also observed that “the effectiveness of a chemical agent as a phytoalexin
elicitor appearsto be specificfor a plant species as
well asfor plant organs.”
Dahiya and Rimmer (1989) pointed out the importance of the age of the
plants or temperature conditions for the accumulation of phytoalexins.
They showed that plant tissues from 35-day-old plants accumulated more
oggoo 0 0 0 0 0 0 0
x ss
m
0 0
8
0
243
244 Rouxel et al.
12
10
0
0 4h 6h 15h 18h Id 2d 3d 7d 10d 14d
Tinw
Figure 7 Time-course study of brassilexin accumulation in B. juncea CV.Aurea

and B. napus CV.Brutor. Plants in the greenhouse weresprayedwith a 10 mM
aqueous copper chloride solution. Incubation took place under greenhouse condi-
tions. Each value is the mean of three experiments with five replications per experi-
ment. (From Rouxel etal., 1989.)
phytoalexin than tissues from20-, 2 5 , and 40-day-old plants.Plants inocu-

lated byL. maculans accumulated an optimal amount of phytoalexins at 20
and 25OC. Higher incubation temperatures (i.e., 30, 35, or 4OOC) resulted
in a drastic decrease of phytoalexin synthesis.
C. Time Course Studies on Phytoalexin Accumulation

Due to their greater efficiency, we used abiotic elicitors to compare the
kinetics of brassilexin accumulation ina plant susceptible to L. rnaculans,
B. napus CV.Brutor, and a hypersensitive one,B. juncea CV. Aurea (Fig.7).
Under these conditions, the first traces of brassilexin were detected 6 hr
after the elicitation inB. juncea, and 12 hr later inB. napus. Moreover, B.
juncea always accumulated more brassilexin than B. napus.
A moreelaboratestudy was recentlyperformed by Mondeetal.
(1991a). Using turnip roots, the authors studied the time-course accumula-
tion of four indole phytoalexins (brassinin, methoxybrassinin, cyclobras-
sinin, and spirobrassinin) and that of nine glucosinolates following UV

irradiation (Fig. 8). Under these conditions, the first traces of brassinin,
methoxybrassinin, and cyclobrassinin were observed 8 hr after the elicita-
tion. In contrast, the first traces of spirobrassinin were only observed 2
days after elicitation, and it then became the main part of the phytoalexin
response. The increase in spirobrassinin accumulationwas paralleled by a
decrease in cyclobrassinin.
H. Effect of Synthesis Inhibitors on theAccumulation of Brassilexin

To assess whether brassilexin
was indeed synthesized de novo,
the effects of
cycloheximide as a protein synthesis inhibitorand that of actinomycinD as
an inhibitor of transcription were evaluated by floating B. juncea leaf discs
for 3 hr on a solution of the inhibitor, prior to abiotic elicitation. In addi-
tion, the effect of sirodesmin PL, a toxin produced by the fungus in high
amounts in culture medium (Fbrbzou etal., 1977; Rowel et al., 1988), was
assessed underthe same conditions. Both actinomycin D and cycloheximide
were potent inhibitorsof brassilexin accumulation, since concentrations of
16 pM for actinomycin D, and 35 p M for cycloheximide nearly totally
suppressed phytoalexin accumulation (Table2). These data thus suggested
de novo synthesis. In contrast, sirodesmin PL only moderately inhibited
Methoxybrassinin 0 Brasslnln [B Cyclobrasdnin Splrobrassinln
f
3 120
Figure 8 Time-course study of indole phytoalexin accumulation in B. rapa root

tissue, following UV irradiation. (From Mondeet al., 1991a.)
246 Rouxel et al.
Table 2 Effect of Synthesis Inhibitors and SirodesminPL on

Brassilexin Accumulation byB. juncea'
Brassilexin accumulation
&g/g fresh wt)
Elicited
Unelicited
Actinomycin D Control 0 4.97 f 1.85

2 PM 0 2.56 i 0.62
8 PM 0 1.56 f 0.18
16 pM 0 0.33 f 0.12
Control
Cycloheximide 0 4.08 f 1.66
18 pM 0 0.66 f 0.03
35 pM 0 0.22 i 0.13
35 pM/rinsed 0 0
Control
Sirodesmin
PL 0 5.65 f 1.05
10 pM 0 2.46 f 1.46
20 pM 0 3.36 f 0.32
40 pM 0 3.28 f 0.34
20 pM/rinsed 0 2.29 f 1.46
'Leaf discs were floatedfor 3 hr on a solutionof the inhibitor prior to the abiotic
elicitation. They were either rinsed and floatedon water, or maintained on the
inhibitor solution during the elicitation. Each data is the mean SD of three
experiments with five replications per experiment. 0, not detected.
brassilexin accumulation(40-59% of the control plant), whatever the tested

concentration (Table 2). In addition, under our experimental conditions,
sirodesmin PL did not by itself elicit the plant defense response. However,
the weak efficiency of sirodesminPL on brassilexin accumulation brought
into questionitspossibleinvolvementasasuppressorofplantdefense
responses (Rouxelet al., 1988).
1. Effect of Ethylene on the Accumulation of Phytoalexins

Ethylene was produced in high amounts in response to an abiotic elicita-
tion, either by B. juncea or B. napus (Bouchenak etal., 1990). Exogenously
applied ethylene never induced phytoalexin accumulation. However, low
concentrations of exogenously applied ethylene (0.1-1 pl/liter) highly en-
hanced brassilexin, cyclobrassinin, or methoxybrassinin accumulation, in
B. napus or B. juncea, whenleafdiscswere treated for 12 hr with the
growth regulator prior to abiotic elicitation (Fig.9). From these preliminary
results, it was suggestedthat ethylene may act as a secondary messenger to
enhance defense responses.
16 -
14 --
12 "
10 "
8 "
6 --
4 "
2 "
o ! I I I I I I
0 0.05 0.1 0.5 1 2
Ethylene treatment W1.1")
Figure 9 Effect of exogenously applied ethylene on brassilexin accumulation by

*
B. juncea. Each datum represents the mean SD of three replications. The experi-
ment was repeated three times.
J. Species-specific Accumulation of Phytoalexins

To compare phytoalexin accumulation inBrassica species, we chose to use
a nonspecific abiotic elicitation and a short time of incubation, i.e., 48 hr,
which has been shown to allow an optimal accumulation of brassilexin in
B. juncea (Fig. 7). Using copper chloride,we elicited 42 cultivars or breed-
ing lines of Brassica belonging to the U triangle plus populationsor culti-
vars of related species, B. adpressa, Raphanus sativus, and Sinapis alba
(Rouxel et al., 1991). Five phytoalexins, brassilexin, brassinin, methoxy-
brassinin, cyclobrassinin, and cyclobrassinin sulfoxide were analyzed. Un-
der these conditions,three of the phytoalexins, i.e., brassilexin, cyclobras-
sinin, and cyclobrassinin sulfoxide,were found within at least some lines of
all species, whereas brassinin was only detected inB. oleracea and B. napus
and methoxybrassinin within these two species and B. rapaand B. carinata.
Brassinin, however, was alwaysa minor part of the response. None of the
five indole phytoalexins could be found in Raphanus sativus or Sinapis
alba, although compounds displaying related UV spectra at different reten-
tion times were observed during the HPLC analysis ofthe extracts.
248 Rouxel et al.
The profiles obtainedfor representative lines withina species (Fig. 10)

showed 1.) an overall low accumulation of these phytoalexins within B.
oleracea; 2.) the important part of brassilexin plus cyclobrassinin sulfoxide
accumulation inthe total response of plant with theB genome (see Fig. l),
as compared to species lacking it; 3.) the importance of methoxybrassinin
in B. oleracea and B. napus.
B. oleracea linesusuallyaccumulated the fivephytoalexinsin low
amounts, even though a few lines displayed an important response, due to
their high ability to rapidly accumulate methoxybrassinin. Only traces of
brassilexin, brassinin, and cyclobrassinin sulfoxide were detected.
Low amounts of brassilexinand cyclobrassinin sulfoxidewere also pro-
duced by B. rapa and B. napus. In the former case, however, one line
displayed a quite unusualpattern of phytoalexin accumulation (see below).
The production of cyclobrassinin was highly variable from one line to the
other. However, both B. napus and B. rapa produced high amounts of
cyclobrassinin. B. napus also consistently accumulated high levels of meth-
oxybrassinin.
Figure 10 Phytoalexin profiles of representative Brassica lines following a CuCl,

elicitation. (From Rouxel et al., 1991.)
from Phytoalexins 249
B. juncea and B. nigra accumulated no brassinin or methoxybrassinin

and high levels of brassilexin plus cyclobrassinin sulfoxide. One B. nigra
line, however, or individuals withinother lines only producedlow levels of
brassilexin. B. nigra usually accumulated low amounts of cyclobrassinin,
whereas the production of this compoundby B. juncea was either very low
or as high asthat of B. napus, depending on the line.
Finally, the one cultivar of B. carinata that we tested mainly accumu-
lated brassilexin and, to a lesser extent, cyclobrassinin and cyclobrassinin
sulfoxide. Only traces of methoxybrassinin were found within that species.
When considering a pool of all of the cultivars within a species, and
calculating the accumulation of a given phytoalexinas a percentage of the
total accumulation of each species (Table3) it can be noticed that each of
the diploid species produceddifferent major phytoalexins and the profiles
of the allotetraploid species were intermediate between their two diploid
genitors (Table 3). Hence methoxybrassinin and, to a lesser extent, cyclo-
brassinin are mainly accumulated by plants withthe C genome, cyclobras-
sinin by plants withthe A genome, and brassilexin and cyclobrassinin sulf-
oxide by plants with the B genome. The A and C genomes are supposed to
be phylogenetically more closely related to each other than either are to the
B genome (Prakash and Hinata, 1980), and the phytoalexin accumulations
of these two genomes are more similarto each other than to the B genome.
Most ofthe cultivars withina species couldnot be clearly discriminated
according to their phytoalexin profiles. Some of them differed from the
average pattern in the production of one given phytoalexin. A few cultivars,
or lines, nonetheless displayed unusual profiles departing from the profile
of the species.
K. Phytoalexin Accumulation and Hypersensitive Resistance

to L. maculans
There was a strong correlation between resistanceto L. maculans, accord-

ing to a cotyledon inoculation test, and the accumulation of brassilexin
(Fig. 11). No susceptible plant accumulated amounts of brassilexin compa-
rable to that of the majority of lines possessing the B genome when chal-
lenged bythe nonspecific elicitor,and no line witha high abilityto accumu-
late brassilexin was susceptible to the disease (Fig. 11). A lesser correlation
was observed to occur for cyclobrassinin sulfoxide, and low or no correla-
tion for the three remaining phytoalexins (Rouxel et al.,1991). Finally, the
total accumulation of the five indole phytoalexins couldnot be correlated
with hypersensitive resistance to the pathogen either.
While there isa high correlation coefficient between the ability to accu-
mulate brassilexin following an abiotic elicitation and disease resistance,
250
from Phytoalexins the Crucifers 251
"O T m Classes 1-2 0 classes 7-9
0-2 4-62-4 6-8 8-10 10-12

12-14 14-16 16-18
18-20
20-22
22-24
Brassllexin accumulation classes (Irg.g" hv)
Figure 11 Distribution of Brassica plants, for a givensetoflesionclasswhen

inoculated by L. maculans, as a function of their belonging to aclass of brassilexin
accumulation. Lesion classes 1-2 are representative of high-level resistance. Classes
7-9 are representative of susceptibility. Intermediate disease ratings were not consid-
ered for the analysis. Forty-two accessions of Brassica species and three lines of
interspecific hybrid progeny B. napus x B. juncea are depicted onthe graph. (Mod-
ified from Rouxel etal., 1991.)
and between the presence of these two variables and the B genome, there
are some exceptionsto this relationship, which means that the data should
be interpreted with care. 1.) One B. nigra line, obtained from wild plants
collected in Turkey, and individuals within otherlines with the B genome,
which consistently gavea hypersensitive reactionto the pathogen, accumu-
lated only low amounts of brassilexin. 2.) Two B. rapa lines, 76-1 and 75-1,
which were obtained from wild plants collected in Sicily and Algeria and
lack theB genome, gave a hypersensitive reactiont o inoculation. However,
only 75-1 accumulated amounts of brassilexin similar to that of plants
possessing theB genome. Recent findingsthat 76-1 segregates for resistance
suggest that the genetic determinants of hypersensitive resistance could be
different from that of species withthe B genome (Mithen, personal commu-
nication). 3.) The accessions of S. alba and R. sativus, which are very
closely related t o Brassica, developed a superficially similar hypersensitive
reaction to the pathogen but did not accumulate any of these five indole
phytoalexins. Closely related compounds have nonetheless been observed
to be elicited in R . sativus (Takasugi et al., 1987), S. alba, Camelina sativa,
252 Rouxel et al.
and Capsella bursa-pastoris(Conn et al., 1988; Browne et al., 1991; Tewari,

personal communication) (seeFig. 2) and couldbeinvolvedindefense
responses.
L. Use of Phytoalexin Quantitative Analysis in Screening

for Disease Resistance Genes
Using copper chloride as a nonspecific elicitor, we studied the brassilexin
production of interspecific hybrid progeny between B. napus and B. juncea
(Roy, 1978, 1984) and compared it to the response of both parent plants
(Table 4). In parallel,the plants were assessedfor resistance to L. maculans
according to a cotyledoninoculationtest.Threedifferentinterspecific
progeny, termed 85-2, 85-3,and 85-4, were analyzed. 85-4 displayed a high
level of resistance. It also accumulateda high level of brassilexin similarto
that of species withthe B genome. 85-3 was very susceptible and its response
to the inoculation did not differ significantly from that of B. napus. Fi-
nally, 85-2showed very heterogeneous responses to the pathogen. Both
85-2 and 85-3 accumulated an amount of brassilexin which was comparable
with the one of B. napus (Table 4). The most heterogeneous interspecific
Table 4 Disease Severity and Brassilexin Accumulation in VariousBrassica

Species and Interspecific Hybrid Progenya
Brassilexin
Mean disease accumulation
severity class oLg/g)
Brassica B. juncea CV.Aurea
1.3 f 0.5 a 5.9 f 1.9 b
B. nigra CV.Junius
1.1 f 0.4 a 4.3 f 0.5 b
B. carinata CV.Awassa 67 1.0 + 0.4 a 4.6 f 2.0 b
B. napus CV. Brutor 7.8 f 0.5 1.7c f 0.3 a
CV.Primor 7.9 f 0.6 1.8c f 0.7 a
CV.Jet Neuf 7.2 f 0.7 c 2.0 f 0.8 a
CV. Bienvenu 7.5 f 0.7 c 2.5 f 0.7 a
Interspecific 85-2 4.9 f 2.7 bc *
2.4 0.6 a
hybrid 85-3 7.7 f 0.8 c 1.9 f 0.7 a
progeny 85-4 2.5 f 0.7 b 5.3 f 1.7
b
'Disease severity was assessed according to a cotyledon inoculation test (Williams and Del-
wiche, 1979). The scale is divided into 10 classes, with the higher ratings being representative
of susceptibility. Brassilexin accumulation was quantitated in individual plants following an
abiotic elicitation. Values are mean f SD. In each column, values followed by different letters
are significantly different(P = 0.01).
Source: From Rouxel et a1 (1990a).
hybrid progeny, 85-2, was further separated into five classes according to
the result of the cotyledon inoculation test and its ability to accumulate
brassilexin. Each of the plants were self-pollinated twice again
and individ-
uals belonging to each of the classes were assessed for resistance to L.
rnaculans. All of the families displayed moderate to high susceptibility to
the pathogen, and noneaccumulatedmorebrassilexin than the parent
plants. The lowest susceptibility of line 85-2-4 was not associated with a
higher accumulationof brassilexin. However, plants of line 85-2-4 accumu-
lated higher amountsof cyclobrassinin than the other lines. This difference
was of low significance due to the highly variable accumulation of this
compound from one plant to the other (data not shown).
To assess the usefulness of this methodology, we are currently under-
taking a similar work with B. napus-B. nigra addition lines (Jahier et al.,
1987; Rouxel et al.,1990~).
IV. DISCUSSION
The data presented here demonstrate that crucifers, like most plant fami-
lies, accumulate fungitoxic compounds when challenged by microorganisms
or abiotic stress. Accordingto Paxton’s definition(1981), these compounds
are phytoalexins: 1.) they are toxicto Brassica pathogens aswell as to other
microorganisms; 2.) theyare low molecular weight compounds; 3.) they are
absent, or nondetected, in healthy plants and are synthesized de novo; 4.)
they accumulate duringa plant-microorganism interaction.
As compared to other plant families, e.g., the Fabaceae (Legumino-
seae) (Ingham,1982), in which more than 100 phytoalexins have been char-
acterized, only limited numbers of phytoalexins from cruciferous plants
have been described. However, the first characterizations of these com-
pounds are quite recent and up to now only a few cruciferous plants have
been assessedfor phytoalexin production.
As has been described for other plant families, all the phytoalexins
from the crucifers displaystructural similarities. As comparedto the usual
phytoalexins from other plant species, these compounds are characterized
by their structural originality, i.e., the presence of an indole or oxindole
system linked withone or more sulfur atoms. Cruciferous plants all contain
a large group of sulfur-containingglycosides, the glucosinolates.These
compounds may be hydrolyzed, e.g., following tissue damage, by the en-
dogenous enzyme myrosinase. Amongthe glucosinolates some, like gluco-
brassicin (Fig. 2),contain an indole moiety,thus suggesting close biochemi-
cal relationships between these compounds and indole phytoalexins. Indole
phytoalexins have been shown not to be degradation products of indole
glucosinolates (Takasugi etal., 1988), but the hydrolysis products of indole
254 Rouxel et al.
glucosinolates are nonetheless toxic to L. rnaculans (Mithen et al., 1986).

Time-course studies of phytoalexins and indole glucosinolate accumulation,
following UV irradiation of plant tissues, showed that a decrease in the
phytoalexin cyclobrassinin paralleled an increase in indole glucosinolate
accumulation. Finally,the close biochemical relationship of both classes of
products is also supported by deuterium-labeled tryptophan incorporation
in both brassinin and cyclobrassinin (Monde and Takasugi, 1991). How-
ever, the control of the synthesis of both classes of products is different
since indole glucosinolate accumulation is strongly induced on wounding of
plant tissues, which is not the case for phytoalexins. Further studies are thus
still neededto “elucidate the exact nature of the biosynthetic relationship,if
any, between these two classes of biologically active compounds’’ (Monde
et al., 1991a).
Using an abiotic elicitor, a species-specific accumulation of phytoalexin
was observed. Devys and Barbier (1992) recently demonstrated the biosyn-
thetic sequence cyclobrassininto brassilexin through cyclobrassinin sulfox-
ide. In this respect, the high correlation coefficient observed between the
ability to accumulate brassilexinand, to a lesser extent cyclobrassinin sulf-
oxide, and blackleg resistance,and between the presence of these two vari-
ables and the B genome suggests that the presence, or efficiency, of a few
enzymes could be responsiblefor the observed differences. However, these
data only represent part of the plant response, since the complete phyto-
alexin response of cruciferous plants,or most of the biochemical relation-
ships that may exist betweenthe phytoalexins, is still unknown. Moreover,
even the high-resolution HPLC analyses do not permit the quantitative
analysis of all the phytoalexins in an extract (Mondeand Takasugi, 1992).
It also remains difficult to compare our results with that obtained by the
Takasugi team or Dahiya and Rimmer. Actually, each of the teams used
different Brassica species or cultivars, analyzed different organs of the
plants (and even different phytoalexins), and used different inducers of
phytoalexin accumulationand quite different incubation conditions. Asan
illustration, the overall amount of phytoalexin quantified by Dahiya and
Rimmer(1989)wasmuchhigher than obtained byus or the Takasugi
group, whatever the tested elicitor. This result is not surprising, however,
since theseauthors only analyzedthe reacting parts of the plant and usually
incubated the tissues for longer times than we did. To suppress such dis-
crepancies, a critical assessment ofthe differential responsiveness of plant
tissues as a function of the inducer, and an assessment of the effect of
physical factors, e.g., light conditions,on the response of the plants remain
to be done. The mastering of these variables and a better knowledge of
indole phytoalexin biosynthetic pathways will then allow, using sophisti-
cated HPLC analysis procedures, minimization of plant-to-plant variations
from Phytoalexins the Crucifers 255
and better differentiation of species-specific or cultivar-specific accumula-

tion.
The quantification of one phytoalexin, brassilexin, permittedus to dif-
ferentiate interspecific hybrid progeny resistant to L. rnaculans from sus-
ceptible ones. The techniquewas assessed using small-plant samples(200-
400 mg fresh wt) so that a single leaffrom a single plant could be examined
and the plant still be used for further breeding. Its objective was to screen
for an intrinsic abilitywhich may bea property of the part of the B genome
responsible for complete resistanceto the pathogen, without using the para-
site. Along with molecular biology characterization, phytoalexin “finger-
prints” ofsingleplants,usingsophisticated HPLC methodologies, will
allow a better differentiation between individualsand may be incorporated
into breeding programs.
A strong correlation was observed between resistance to L. rnaculans
and the ability to accumulate high levels of brassilexin, together with the
presence of the B genome. However, recent findings cast doubts on the
significance ofthe macroscopic reactions usually associated with resistance
(Gugel et al., 1990). Moreover, correlations do not imply that the com-
pounds really playa role in resistance.A comprehensive study ofthe vari-
ability of the pathogen, along with the development of methods allowing
an unequivocal quantification of the result of the interaction between the
plant and the pathogen, are now needed to confirm these correlations.
Genetic studies of phytoalexin detoxification, or specificity of interaction
isolate/cultivar, are now at hand and will hopefully unequivocally deter-
mine the involvement of phytoalexins inthe resistance responses of crucif-
erous plants.
V. EPILOG
According to Hill and Williams (1988) and to the data presented here, L.
rnaculans in combination with Brassica has considerable potential as a
model host-pathogen system for the study of the genetic and molecular
bases of plant-microorganism interactionsand plant disease resistance.
The main interests of this systemare 1.)the ease ofculture of both host
and pathogen; 2.) the development of a methodology allowinga reproduc-
tive induction of the sexual state of the pathogen in culture along with
the ability to develop tetrad analyses (Mengistu et al., 1990); 3.) the high
pathogenic variability of the pathogen in natural populations along with
the finding that cultivar-isolate type-specific interactions occur in this
sys-
tem; 4.) the broad genetic basis of Brassica crops, like B. napus, and the
ease with which the interaction can be studied due to 5.) the rapid and
intense production of pycniospores which can easily bequantified, 6.) the
256 Rouxel et al.
rapid evaluation ofthe result of the interaction using a seedling test which
allows a differentiation of specific resistance vs. susceptibility (Williams
and Delwiche, 1979; Rouxel et al., 1990a), 7.) the possible use of rapid-
cycling Brassica to undertake genetic analyses. Moreover, 8.) the recent
development of powerful biochemicaland molecular tools for the study of
microorganisms now allows an unequivocal characterization of L. macu-
lansisolates (Koch et al., 1991; Taylor et al., 1991a; Balesdent et al., 1992).
9.) Both the host and the pathogen are amenable to cellular and molecular
biology techniques. In this respect, it is important to keep in mind that
Arabidopsis thaliana, one of the most studied model plants, is a crucifer.
Finally, 10.) the data presented here, along with similar studies performed
by other teams, give a new insight into Brassica defense reactions, thus
providing new tools to develop a quantification of the interaction and fa-
voring a questfor defense genes or disease resistance genes.
ACKNOWLEDGMENTS
This work was supported by a grant from the Institut National de la Re-
chercheAgronomique (AIP, 1986). The authors wish to thank H. H.
Hoppe (GesamthochsuleKassel,FachbereichLandwirtschaft,Witzen-
hausen, F.R.G.) for single-ascospore lines of L. maculans; M. Takasugi
(Department of Chemistry, Faculty of Science,Hokkaido University, Sap-
poro, Japan), M. Devys and M. Barbier (Chimie des Substances Naturelles,
CNRS, Gif-sur-Yvette)for authentic samples of indole phytoalexins;N. N.
Roy (Department of Agriculture,Perth, Australia) for interspecific hybrid
progeny; A.-M. Chkre, M. Renard (AmClioration des Plantes, INRA, Le
Rheu), L. Boulidard (AmClioration desPlantes, INRA, Versailles), and R.
Mithen (John Innes Centrefor Plant Science Research, Colney Lane, Nor-
wich NR4 7UH, UK) for crucifer seeds and breeding lines; J. F. Bousquet
(Pathologie VCgCtale, INRA, Versailles) who permitted this researchto take
place in his lab and then allowed the development of a fruitful collabora-
tion; N. T. Keen and D. Cooksey (University of California, Plant Pathol-
ogy, Riverside,CA92521,USA) for reviewof the manuscript. Special
thanks are due to A. Sarniguet (PathologieVCgCtale, INRA, Le Rheu) who
initiated this work.We are very grateful to C. Sutre, A. Delaunay, and F.
Bouchenak, whose help to perform part of this work was greatly appreci-
ated.
REFERENCES
Alabouvette, C., and Brunin, B. (1970). Recherches sur la maladie du colza due A
Leptosphaeria maculans (Desm.) Ces. et De Not. I. R61e des restes de culture
dans la conservation etla disskmination du parasite. Ann. Phytophathol. 2(3):

463-475.
Ayers, A. R., Ebel, J., Valent, B., and Albersheim, P. (1976). Host-pathogen
interactions. X. Fractionation and biologicalactivityof an elicitor isolated
from the mycelial walls of Phytophthora megasperma var. sojae. Plant Physiol.
57:760-765.
Badawy, H. M. A., and Hoppe, H.H. (1989). Production of phytotoxic sirodes-
mins by aggressivestrains of Leptosphaeria maculans differing in interactions
with oilseedrape genotypes. J. Phytopathol. 127: 146-157.
Balesdent, M. H.,Gall, C., Robin, P., and Rowel, T. (1992). Intraspecific varia-
tion in soluble mycelial protein and esterase patterns of Leptosphaeria macu-
fans French isolates. Mycol. Res. 9 6 677-686.
Bouchenak, F., Rouxel,T., Kollmann, A., Touraud, G., and Bousquet J.-F. (1990).
Ethylene: phytopathologicalsignificance inthe hypersensitive response of Bras-
sica juncea leaves to Leptosphaeria maculans. Plant Physiol. 93( 1): 53 (abst).
Browne, L. M., Conn, K. L., Ayer, W. A., and Tewari, J. P. (1991).The camalex-
ins: new phytoalexins produced in the leaves of Camelina sativa (Cruciferae).
Tetrahedron47(24): 3909-3914.
Brun, H.,and Jacques, M. A. (1990). Le desstchement prkmaturk des pieds de
colza. Quelques sympti3mes et agents pathogtnes associks. La D& Vdg. 262:
7-12.
Brunin, B., and Lacoste, L. (1970). Recherches sur la maladie du colzadue A
Leptosphaeria maculans (Desm.) Ces. et De Not. 11. Pouvoir pathogtne des
ascospores. Ann. Phytopathol. 2(3): 477-488.
Conn, K. L., Tewari, J. P., and Dahiya, J. S. (1988). Resistance to Alternaria
brassicae and phytoalexin elicitation in rapeseed and other crucifers. Plant Sci.
5 6 21-25.
Crisp, P. (1976). Trends in the breeding and cultivation of cruciferous crops, The
Biology and Chemistryof the Cruciferae(J. G. Vaughan, A. J. Macleod, and
B. M. G . Jones, eds.), Academic Press, London, pp. 69-118.
Dahiya, J. S., and Rimmer, S. R. (1988a).Phytoalexin accumulation in tissues of
Brassica napus inoculated with Leptosphaeria maculans. Phytochemistry 27:
3105-3107.
Dahiya, J. S., and Rimmer, S. R. (1988b). High-performance liquid chromatogra-
phy of phytoalexins in stem callus tissues of rapeseed. J. Chromatogr. 448
448-453.
Dahiya, J. S., and Rimmer, S. R. (1989).Phytoalexin accumulationin plant tissues
of Brassica spp. in response to abiotic elicitors and infection with Leptos-
phaeria maculans.Bot. Bull. AcademiaSinica 3 0 107-115.
Devys, M., Barbier, M., Loiselet, I., Rouxel, T., Sarniguet, A., Kollmann, A., and
Bousquet, J. F. (1988). Brassilexin,anovel sulphur-containing phytoalexin
from Brassica juncea L., (Cruciferae). TetrahedronLett. 2 9 6447-6448.
Devys, M., Barbier, M., Kollmann, A., Rouxel, T., and Bousquet, J. F. (1990).
Cyclobrassininsulphoxide,a sulphur-containing phytoalexin from Brassica
juncea. Phytochemistry2 9 1087-1088.
Devys, M., and Barbier, M. (199Oa). A convenient synthesis of isothiazolo[5,4b]in-
258 Rouxel et al.
dole (brassilexin) via apolyphosphoric acid initiated ring closure. Synthesis 3:

214-215.
Devys, M., and Barbier, M. (199Ob). Oxidative ringcontraction of the phytoalexin
cyclobrassinin: a way to brassilexin. J. Chem. Soc. Perkin Trans.2856-2857.
Devys, M., and Barbier, M. (1992). In vitro demonstration of a biosynthetic se-
quence for theCruciferae phytoalexins.2.Naturforschung. C 47c: 318-319.
Fkrkzou, J. P., Riche, C., Quesneau-Thierry, A., Pascard-Billy, C., Barbier, M.,
Bousquet, J. F., and Boudart, G. (1977). Structures de deux toxines isolkes
des cultures du champignon Phoma lingam Tode: la sirodesmine PL et la
desacetylsirodesminePL. Nouv. J. Chim. I : 327-334.
Gabrielson, R. L. (1983). Blackleg disease of crucifers caused by Leptosphaeria
maculans (Phoma lingam)and its control. Seed Sci. Technol. 11: 749-780.
Gugel, R. K., Seguin-Swartz, G., and Petrie, G. A. (1990). Pathogenicity of three
isolates of Leptosphaeria maculans on Brassica species and other crucifers.
Can. J. Plant Pathol. 12: 75-82.
Hammond, K. E., Lewis, B. G., and Musa, T. M. (1985). A systemic pathway in
the infection of oilseed rape plants by Leptosphaeria maculans. Plant Pathol.
34: 557-565.
Hammond, K. E., and Lewis, B. G. (1986). Ultrastructural studies of the limitation
of lesions caused by Leptosphaeria maculans in stems of Brassica napus var.
oleifera. Physiol.Mol. Plant Pathol.28: 251-265.
Harper, F. R., and Berkenkamp, B. (1975). Revised growth-stage key for Brassica
campestris and B. napus. Can.J. Plant Sci. 55: 657-658.
Hassan, A. K.. Schulz, C., Sacristan, M. D., and Wostemeyer, J. (1991). Biochemi-
cal and molecular tools for the differenciation of aggressiveand non-aggressive
isolates of the oilseed rape pathogen, Phoma lingam.J. Phytopathol. 131: 120-
136.
Hedge, I. C. (1976). A systematic and geographic survey of the old world cruciferae.
The Biology and Chemistry of the Cruciferae(J. G. Vaughan, A. J. Macleod,
and B. M. G. Jones, eds.), AcademicPress, London, pp. 1-45.
Hill, C. B., and Williams, P. H. (1988). Leptosphaeria maculans,cause of blackleg
of crucifers. Advances in Plant Pathology, Vol. 6, Genetics of Plant Patho-
genicFungi (D. S. Ingram and P. H. Williams, eds.), AcademicPress, London,
pp. 169-174.
Ingham, J. L. (1982). Phytoalexins from the Leguminosae. Phytoalexins (J. A.
Bailey and J. W. Mansfield, eds.), Blackie, Glasgow, pp. 21-80.
Jahier, J., Tanguy, A.”., Chevre, A.-M, Tanguy, X., .and Renard, M. (1987).
Extraction of disomic addition lines B. napus-B. nigra and introduction of
B. nigra type Phoma lingam resistance to rapeseed. Proceedings of the 7th
International Rapeseed Congress, Poznan, Poland, pp. 445-450.
Keri, M., Rimmer, S. R., and van den Berg, C. G . J. (1990). The inheritance of
resistance ofBrassica juncea to Leptosphaeria maculans. Can. J. Plant Pathol.
12(3): 335 (abst).
Koch, E., Badawy,H. M. A., and Hoppe, H.H. (1989). Differencesbetween
aggressive and non-aggressive single spore lines of Leptosphaeria maculans
in cultural characteristics and phytotoxin production. J. Phytopathol. 124:
52-62.
Koch,E.,Song, K., Osborn, T. C., and Williams, P. H. (1991). Relationship

between pathogenicity and phylogenybased on restriction fragment length
polymorphism in Leptosphaeriarnaculans. Mol. Plant-MicrobeInterac. 4:
341-349.
Kollmann, A., Rouxel, T., and Bousquet, J. F. (1989). Efficient clean up of non-
aqueous plant extracts usingreversed-phasecartridges.Applications to the
determination of phytoalexins from Brassica spp. by high-performance liquid
chromatography. J. Chromatogr. 473: 293-300.
Mengistu, A., Rimmer, S. R., and Williams, P. H. (1990). In vitro induction of
pseudothecia, ascospore release and variation in fertility among geographic
isolates of Leptosphaeriamaculans. Phytopathology 8 0 1007 (abst).
Mithen, R. F., Lewis, B. G., and Fenwick, G . R. (1986). In vitro activity of glucosi-
nolates and their products against Leptosphaeria maculans. Trans. Br. Mycol.
Soc. 87(3): 433-440.
Mithen, R. F., and Lewis, B. G . (1988). Resistance to Leptosphaeria maculans in
hybrids of Brassica oleracea and Brassica insularis. J. Phytopathol. 123: 253-
258.
Monde, K.,Sasaki, K., Shirata, A., and Takasugi, M. (199Oa). CMethoxybrassinin,
a sulphur-containingphytoalexin from the Chinese cabbage Brassica campestris
L. ssp. pekinensis. Chem.Lett. 2 209-210.
Monde, K., Katsui, N., Shirata, A., and Takasugi, M. (199Ob). Brassicanal-A and
brassicanal-B, novel sulphur-containing phytoalexins from Brassica oleracea.
Phytochemistry 29(5): 1499-1500.
Monde, K., Takasugi, M., Lewis, J. A., and Fenwick, G. R. (1991a). Time-course
studies of phytoalexins and glucosinolates in UV-irradiated turnip tissue. Z.
Naturforschung C46c: 189-193.
Monde, K., Sasaki, K., Shirata, A., and Takasugi, M. (1991b). Brassicanal C and
two dioxindoles from cabbage. Phytochemistry 30(9): 2915-2917.
Monde, K., Sasaki, K., Shirata, A., and Takasugi, M. (1991~).Methoxybrassenins
A and B, sulphur-containing stress metabolites from Brassica oleracea var.
capitata. Phytochemistry3q12): 3921-3922.
Monde, K., and Takasugi, M. (1991). Biosynthesis of cruciferous phytoalexins: the
involvement of a molecular rearrangement in the biosynthesis of brassinin. J.
Chem. Soc. Chem. Commun. 1582-1583.
Monde, K., and Takasugi, M. (1992). High-performance liquid chromatographic
analysis of cruciferous phytoalexins using a ternary mobile phase gradient. J.
Chromatogr. 598 147-152.
Paxton, J. D.(1981). Phytoalexins-a working redefinition. Phytopathol. 2. 101:
106-109.
Prakash, S., and Hinata, K. (1980). Taxonomy, cytogenetics and origin of crop
brassicas, areview. Opera Botanica55: 1-57.
Regnault, Y., Laville, J., and Penaud, A. (1987). Cahiers techniques:Les maladies
du colza d‘hiver, CETIOM, Paris, France.
Ricci, P. (1986). Etude des relations h6te-parasite dans l’interaction compatible
entre Dianthus caryophyllus L. et Phytophthora parasitca Dastur. Nature et
intervention d’unmecanisme de dCfenseClicitable. These de 1’UniversitC de
Paris XI, Orsay, France.
260 Rouxel et al.
Rowel, T.(1988). Pouvoir pathoghe de Leptosphaeriamaculans et reaction hyper-

sensible de Brassica spp. Intervention d’une pathotoxine, la sirodesmine PL, et
d’une phytoalexine, la brassilexine. Thbe de l’Universit6 de Paris XI, Orsay,
France.
Rouxel, T., Sarniguet, A., Kollmann, A., and Bousquet, J. F. (1987). Identification
d’une phytoalexine dans Brassica juncea (moutarde brune) et Brassica napus
(colza) en relation avec la resistance ZL Phoma lingam. In ler Congrh de la
Soci6t6Francaisede Phytopathologie, Nov. 19-20, 1987, SFP, Rennes, France
(abst).
Rouxel, T., Chupeau, Y.,Fritz, R., Kollmann, A., and Bousquet, J. F.(1988).
Biological effects of sirodesmin PL, a phytotoxin produced by Leptosphaeria
maculans. PlantSci. 6 8 77-86.
Rouxel, T., Sarniguet, A., Kollmann, A.,and Bousquet, J. F. (1989). Accumulation
of a phytoalexin in Brassica spp. in relation to hypersensitive reaction to Lep-
tosphaeria maculans. Physiol.Mol. Plant Pathol.34: 507-517.
Rouxel, T., Renard, M., Kollmann, A., and Bousquet, J. F.(199Oa). Brassilexin
accumulation and resistance to Leptosphaeria maculans in Brassica spp. and
progeny of an interspecific cross B. juncea x B. napus. Euphytica 46 175-
181.
Rouxel, T., Kollmann, A., and Bousquet, J. F. (1990b). Zinc suppresses sirodesmin
PL toxicity and protects Brassicanapus plants against the blackleg disease
caused byLeptosphaeria maculans. Plant Sci. 68: 77-86.
Rouxel, T., Kollmann, A., Sutre, C., Chkre, A.”., Renard, M., and Boulidard,
L. (199Oc). Induction abiotique de l’accumulation de phytoalexines soufrees
chez les Brassica et resistance hypersensible ZL Leptosphaeria maculans. 26me
CongrPs de la Soci6t6 Francaise de Phytopathologie, Nov. 28-30, 1990, SFP.
Montpellier, France (abst.)
Rouxel, T., Kollmann, A., Boulidard, L., and Mithen, R. (1991). Abiotic elicitation
of indole phytoalexins and resistance to Leptosphaeria maculans within Brassi-
ceae. Planta 184 271-278.
Roy, N. N. (1978). A study on disease variation in the populations of an interspe-
cific cross ofBrassica juncea L. x Brassica napus L. Euphytica 27:145-149.
Roy, N. N.(1984). Interspecific transfer of Brassica juncea-type highblackleg
resistance to Brassica napus. Euphytica33: 295-303.
Sacristan, M. D., and Gerdemann, M. (1986). Different behavior of Brassicajuncea
and B. carinata as sources ofPhoma lingam resistance in experiments of inter-
specific transfer to B. napus. Plant Breed.97: 304-314.
Sarniguet, A.(1986). Accumulation d’une phytoalexine chez Brassica juncea en
relation avec la resistance hypersensible ZL Phoma lingam. Dipldme d’Etude
Approfondie en Phytopathologie, Universite de Paris-Sud-Orsay, Paris,
France.
Sjodin, C., and Glimelius, K. (1989). “Brassica naponigra” a somatic hybrid resis-
tant to Phoma lingam. Theor.Appl. Genet. 77: 651-656.
Smith, D.A. (1982). Toxicity of phytoalexins.Phytoalexins (J. A. Bailey and J. W.
Mansfield, eds.), Blackie, Glasgow, pp. 218-252.
Soledade, M., Pedras, C., and Taylor, J. L. (1991). Metabolic transformation of
the phytoalexin brassinin by the “blackleg” fungus. J. Org. Chem. 56: 2619-
2621.
Somei, M., Kobayashi, K., Shimizu, K., and Kawasaki, T. (1992). A simple synthe-
sis of a phytoalexin, methoxybrassinin. Heterocycles 33:77-80.
Takasugi, M., Katsui, N., and Shirata, A. (1986). Isolation of three novel sulphur-
containing phytoalexins from the Chinese cabbage Brassica campestris L. ssp.
pekinensis (Cruciferae). J. Chem. Soc. Chem. Commun.1631-1632.
Takasugi, M., Monde, K., Katsui. N., and Shirata, A. (1987). Spirobrassinin, a
novel sulphur-containing phytoalexin from the daikon Raphanus sativus L.
ssp. hortensis (Cruciferae). Chem. Lett. 1631-1632.
Takasugi, M., Monde, K., Katsui,N., and Shirata, A. (1988). Novel sulphur-
containing phytoalexins from the Chinese cabbage Brassica campestris L. ssp.
pekinensis (Cruciferae). Bull. Chem. Soc. Jpn.61: 285-289.
Taylor, J. L., Borgmann, I., and S6guin-Swartz, G.(1991a). Electrophoretic karyo-
typing of Leptosphaeria maculans differenciates highly virulent from weakly
virulent isolates. Curr. Genet.1 9 273-277.
Taylor, J. L., Soledade, M., Pedras, C., and Morales, V. M. (1991b). Molecular
genetic and chemical characterization of the rapeseed “blackleg” causalorgan-
ism. 3rd Congress of the International Societyfor Plant Molecular Biology:
Molecular Biology of Plant Growth and Development (R. B. Hallick, ed.),
Tucson, Arizona, October, 6-11, 1991 (abst).
TempCte,C.,Devys,M., and Barbier, M. (1991). Growth inhibition of human
cancer cell cultures with the indole sulphur-containing phytoalexins and their
analogues. Z. Naturforschung. C 46c:706-707.
Tsuji, J., Jackson, E. P., Gage, D. A., Hammerschmidt, R., and Somerville, S. C.
(1992). Phytoalexin accumulation inArabidopsis thaliana during the hypersen-
sitive reactionto Pseudomonas syringaepv. syringae. Plant Physiol.98: 1304-
1309.
U.N. (1935). Genome-analysis inBrassica with special reference to the experimental
formation of B. napus and peculiar mode of fertilization. Jap. J. Bot. 7: 389-
452.
Williams, P.H., and Delwiche, P. A. (1979). Screening for resistance to blackleg of
crucifers in the seedling stage, Proceedings of a Eucarpia Conference on the
Breeding of CruciferousCrops, Wageningen, Netherlands,pp. 164-170.
Williams, P. H., and Hill, C. B. (1986). Rapid-cycling populations of Brassica.
Science232: 1385-1389.
12
Scoparone (6,7-Dimethoxycoumarin),
a Citrus Phytoalexin Involvedin
Resistance to Pathogens
Uzi Afek
The Volcani Center, Bet Dagan, Israel
Abraham Sztejnberg
The Hebrew Universityof Jerusalem, Rehovot, Israel
1. INTRODUCTION
Not much work has been done on induced resistance of Citrus to patho-
gens. Studies with Phytophthora citrophthora (Smith and Smith) Leonian
indicated that morphological exclusionof the fungus inmature cells cannot
be responsible for Citrus resistance to this pathogen, and biochemical or
physiological factors may also be involved in resistantand susceptible reac-
tions. Mature cells of resistant Citrus species may possess inhibitors, or
hypersensitivity reaction may be induced by substances produced by the
plant cells after infection (Broadbent,1969).
Two fungitoxic compounds were found by Hartmann and Nienhaus
(1974a, b) in bark of Citrus limon (L.)Burm. infected by P. citrophthora
and Hendersonula toruloideaNatrass. Oneof these compoundswas identi-
fied as xanthoxylin(2-hydroxy-4,6-dimethoxyacetophenone).No xanthoxy-
lin could be demonstrated in healthy bark or following mechanical wound-
ing or chemical treatments. Musumeci and Olivera (1975, 1976) reported
about total phenol increase in sweet orange [Citrus sinensis (L.)Osbeck]
and sour orange (C.aurantium L.) following inoculation withP. citroph-
thora. However, the increaseintheresistantspecies(sourorange) was
higher than in the susceptible species (sweet orange). These investigators
pointed to compound 1 as a phytoalexin accumulated inCitrus tissue after
inoculation withP . citrophthora.
Xanthyletin (6,7-dimethylpyranocoumarin)(Kahn et al., 1985) and ses-
elin (7,8-dimethylpyranocoumarin)(Vernenghi et al., 1987) were isolated
from Citrus tissue infected with P . citrophthora and P . parasitica Dastur.
Both xanthyletin and seselin showed an inhibitory activity against these
263
264 Afek and Sztejnberg
pathogens in vitro. Ismail et al. (1978) reported that the synthesis of umbel-
liferone (7-hydroxycoumarin) was greatly enhanced during healing of in-
jured grapefruit. Ben-Yehoshua et al. (1987, 1988) isolated several antifun-
gal substances from pomelo fruit, some of whichare coumarin derivatives.
Several studies reported that scoparone is involved in defense mecha-
nisms ofCitrus against pathogens such as P. citrophthora (Afek and Sztejn-
berg,1986,1988a), Guignardia citricarpa Kiely(DeLangeet al., 1976),
Penicillumdigitatum Sacc.(Kimet al., 1991;Rodovet al., 1992), and
Diaporthe citri (Faw.) (Aritmo et al., 1986). Treatments such asy irradia-
tion (Riov, 1971; Dubery and Schabort, 1987; Afek and Sztejnberg, 1993),
high temperature (Afekand Sztejnberg, 1988b, 1993; Aritmo and Homma,
1988;Kim etal.,1991), UV illumination (Rodov et al.,1992),andfo-
setyl-A1 and phosphorous acid (Afek and Sztejnberg, 1989) were found to
stimulate scoparone production in Citrus.
This chapter examinesthe role of scoparone as a phytoalexin involved
in Citrus resistance to pathogens. Different chemical and physical treat-
ments that may increase scoparone concentration in Citrus tissue and an
attempt to understand the biosynthesis of scoparone from phenylalanine
are discussed.

Fungal and Plant Material
Experiments were done with the fungus Phytophthora citrophthora(Smith
and Smith) Leonian (isolate C-16) and with 3-month-old Citrus branches
and mature fruits. Twenty-five to 30-cm-long and 7- to 10-mm-thick
branches were prunedfrom 3-year-old Citrus seedlings; Citrus sinensis (L.)
Osbeck(Shamouti); C. aurantium L.(sourorange); C. jambhiri Lush.
(rough lemon); Poncirus trifoliata Raf. (trifoliate orange); C. - reticulata
Blanco x C. sinensis (Niva); and C. macrophylla Webster (Macrophylla).
These species were chosen because one group (Macrophylla, trifoliate or-
ange, and sour orange) is resistant to P. citrophthora and the other group
(rough lemon,Shamouti, and Niva) is susceptibleto P.citrophthora.
Mature Citrus fruits of C. paradisi Macfadden (grapefruit),C. sinensis
(Valencia),Shamouti, and sour orange werepicked from a 12-year-old
orchard.
A. Inoculation
Incisions were made in the bark of branches and agar discs were cut from
an actively growing zone of mycelia of P. citrophthora on PDA medium
and placed overthe incisions, fungal side downward. The inoculated branch
Scoparone (6,7-Dimethoxycoumarin), a Cifrus Phytoalexin 265
sections were incubated ina humid chamber at 2OoC and 28OC in darkness
(Afek and Sztejnberg, 1988a). Fruits were inoculated by removing pieces of
flavedo witha 3-mm-diameter cork borerto a depth of 0.2-0.5 mm at four
to five sitesaround the equatorial plane of the fruit. A 3-mm-diameter disc,
cut from an actively growing PDA culture of P. citrophthora, was placed
on the fruit wounds and the inoculated fruits were incubated in a humid
chamber at 24OC in darkness.
B. Extraction, Purification, Identification, and

Quantification of Scoparone
Slices of inoculated, necroticbark, cut from the outer edge of the wounds,
were extracted with distilled water. The antifungal component extracted
from the inoculated bark was concentrated and was eluted with EtOAc/
petroleum ether (1 : 1, v/v) and partitioned withCHC13from which it crys-
tallized uponevaporation, as colorless needles with mp 146-147OC.
Ultraviolet (UV)spectrophotometry, infrared (IR) analysis, ‘HNMR
(nuclear magnetic resonance) and 13C NMRspectra indicatedthat the active
antifungal agentthat was extracted and purified from Citrus bark and fruit
inoculated withP. citrophthora was scoparone (Fig. 1).
C. Scoparone Quantification
Solutions of Citrus samples were analyzed spectrofluorometrically withex-
citation at 340 nm and emission reading at 430 nm. Concentration of sco-
parone in the Citrus tissue was calculated by a comparison with the stan-
dard curve (Afekand Sztejnberg, 1988a;Afek et al., 1986).
D. y Irradiation
y Ionizing irradiation was appliedwithcobalt radiant <“CO).Branches
and fruits were irradiated with 0-, loo-, 200-, 300-, and 400-krad dose.
Inoculation of the irradiated branch and fruit with P. citrophthora was
done 24 hr later.
Figure 1 Scoparone.
Sztejnberg
266 and Afek
E. Fosetyl-AI and Phosphorus Application and Bioassay

Citrus branches were immersed in aqueous @utions containing fosetyl-AI
(0-800 pm/ml) or H3P03(0-400pm/ml) for 3 hr. The branches were then
washed withtap water and inoculated withP.citrophthora. Advance ofthe
and scoparone concentrationwere measured4 days
pathogen (lesion length)
later.
F. Bioassay
EDSo values were determined by adding increasing concentrations of scopar
one or fosetyl-Al or H3P03 to cooled molten PDA immediately before it
was poured into Petri plates. Disc of P. citrophthora, taken from an ac-
tively growing colony on PDA, was placed fungal side downward in the
center of each plate. Plateswere incubated in darknessat 25OC for 8 days
(Afek and Sztejnberg, 1988a, 1989).
G. labeling and Aminooxyacetic Acid Application

Excised branches (1cm length and 0.5 cm diameter) wereimmersed in
a solution containing ['4C]phenylalanine (3.14 x los dpm) or in10mM
aminooxyacetic acid (AOA) (in sterile deionized water)
for 3 hr, inoculated
with P. citrophthora, and incubated at 24OC. The labeled scoparone, con-
centration of scoparone, and lesion length were measured
4 days later (Afek
and Sztejnberg, 1988a).
111. SCOPARONE PRODUCTION IN ClTRUS

A. Accumulation of Scoparone in Various Plant Parts
of Citrus
and in Different Varieties
Small concentrations of scoparone naturally exist in healthy Citrus barks
and fruit peel (Afek and Sztejnberg, 1988a;Tatum and Berry, 1977). How-
ever, its concentration inCitrus tissue increases following inoculation with
P. citrophthora. The production of scoparone and the lesion length in the
bark of 3-month-oldCitrus branches, resistant (macrophylla, trifoliate, and
sour orange) and susceptible (rough lemon, Shamouti, and Niva) to P.
citrophthora, were measured daily during incubation at 2OoCand 28OC for
8 days after inoculation with the pathogen. Scoparonewas induced inboth
groups of Citrus, but the concentration washigher and increased more
rapidly in the resistant species. Twenty-four hours after the inoculation,
concentration of scoparone in barkof the resistant specieswas about three
times morethan in the susceptible species. High concentration of the phyto-
alexin within the first 24-48 hr is critical in stopping the advance of the
Scoparone(6,7-Dimethoxycoumarin), a Cifrus Phytoalexin 267
pathogen in vivo. However, scoparone concentration continued to increase

and reached the maximum at 2OoC, 4days after the inoculation. The con-
centrations were 440,415, and 250 pg/g fresh wt in macrophylla,trifoliate,
and sour orange, and 42, 31, and 28 pg/g fresh .W in rough lemon, Sha-
mouti, and Niva, respectively (Fig.2).
The length of lesions caused by P. citrophthora in the bark 4 days after
inoculation was 2.5 mm in macrophylla, 3.2 mm in trifoliate, 5.0 in sour
orange, 11.0 mm in rough lemon, 15.5 mm in Shamouti, and 17.0 mm in
Niva (Fig. 2).
Scoparone showed inhibitory activity against various phytopathogenic
fungi in vitro (Table 1) and ED,, value for inhibition growth of P. ci-
trophthora was 97 pg/ml (Fig. 3).
These results indicate that scoparone is involved in resistanceand not a
resultofresistance(necrosis).Lesionlengthin the branchescaused by
P. citrophthora infection is inversely proportional to the increase in the
phytoalexin concentration.A s the lesion increasesthe concentration of sco-
parone decreases. Lesion length, 4 days after inoculation with P. citroph-
Figure 2 Accumulationof scoparone and lesionlengthin Citrus bark of the

resistant species macrophylla(H), trifoliate orange (0),sour orange (a), and the
susceptible species rough lemon (D), Shamouti (A), Niva (A),after inoculation
with Phytophthora citrophthoraat anincubation temperature of 2OOC. Significant
differences were indicated by different letters within each time period accordingto
Duncan’s multiple range test (p = 0.05). Statistical analysis of scoparone concen-
tration was done starting on the second dayand lesion lengthon the third day (Afek
and Sztejnberg, 1988a).
Table 1 Effective Dose of Scoparone for 50% Inhibition

(ED,) of Mycelial Growthof Phytophthora citrophthora
Compared with Conidial Germination Inhibitionof Six Other
Phytopathogenic Fungi In Vitro
~
ED, of scoparone
Fungal species Olg/ml)
Phytophthora citrophthora 97
Verticillium dahliae 61
Penicillium digitatum 64
Penicillium italicum 60
gloeosporioides
Colletotrichum 54
tomloideaHendersonula 90
Botryiodiplodia
(Diplodia) natalensis 85
Source: Data from Afek et al. (1986).
SCOPARONE (pg m 1 - a )
Figure 3 Dosage-response of Phytophthora citrophthora mycelial growth to log

concentration of scoparone, expressed as a linear regression (Afek and Sztejnberg,
1988a).
Scoparone(6,7-Dimethoxycoumarin), a Citrus Phytoalexin 269
thora at incubation temperaturesof 2OoC and 28OC, was negatively corre-

lated with increased accumulation of scoparone in vivo (Fig. 4). Unlike
Hartmann and Nienhaus (1974b) the authors did not find any correlation
between the degree of Citrus resistance and the accumulation of xanthoxy-
lin inthe bark after infection with P.citrophthora. Accumulation of xanth-
oxylin appearedto be a result of the necrotic reactionand seemed to play a
part in eliminatingthe pathogen after it had been inhibited by other defense
mechanisms of the plant.
Arimoto et al. (1986) reported similar results with the fungus Diaporthe
citri and suggested that scoparone was produced by the Citrus in response
to infection by this pathogen.
Musumeci and Olivera (1975, 1976) described compound 1 as a phyto-
alexin accumulated in Citrus after inoculation withP. citrophthora. Because
the final structural formula of compound 1 is still unknown, this compound
cannot be related to one of the known Citrus phytoalexins. However, ac-
cording to the description, itseems that compound 1 belongs to the couma-
rin group. Other coumarins such as xanthyletin (Khan et al., 1985) and
seselin (Vernenghi et al., 1987) may also play a part as phytoalexins in
Citrus resistance to pathogens.
8- -
- 28 C -
Y=25.70-8.93 X
L-
r2=0 - 9 5 6
-
0 I I I l I I l l
10 100 1000
SCOPARONE (pg g-x fr. wt.)
Figure 4 Relation between the concentration of scoparone and lesion length in

Citrus bark 4 days after inoculation withPhytophthora citrophthoraat an incuba-
tion temperature of 2OoC ( 0 )and 28OC (0)
(Afek and Sztejnberg, 1988b).
B. Temperature Effect
In all the tested species, scoparone concentration,after inoculation withP.
citrophthora, was higher whilethe lesion length wasshorter at 28OC (Figs.
2and 5). It was especially expressed in rough lemon, which showed resistant
reaction. The concentration of scoparone increasedup to 290 pg/g fresh wt
(as comparedto 42 pg/g fresh wt at 2OoCafter 4 days of incubation). The
lesion length inthe bark of rough lemon, at 28OC 4 days after the inocula-
tion, was 5.2 mm. However, in the other species scoparone concentrations
in the same period were 587,515, and 326 pg/g fresh in macrophylla,
trifoliate, and sour orange, and 53 and 39 pg/g fresh wt in Shamouti and
Niva, respectively. Lesion length inthe bark in these conditions was 1.3 mm
in macrophylla, 1.9 mm in trifoliate, 3.5 mm in sour orange, 10.8 mm in
Shamouti, and 11.7 mm in Niva (Fig.5).
Figure 5 Accumulation of scoparoneandlesionlengthin Citrus barkof the

resistant species macrophylla(M), trifoliate orange (0),sour orange (e), and the
susceptible species rough lemon (U), Shamouti (A), Niva (A),after inoculation
with Phytophthora citrophthoraat an incubationtemperature of 28OC. Significant
differences were indicated by different letters within each time period according
to
Duncan’s multiple range test (p = 0.05). Statistical analysis of scoparone concen-
tration was done starting on the second dayand lesion lengthon the third day (Afek
and Sztejnberg, 1988b).
In the noninoculated control the concentration of scoparone was 12-18

pg/g fresh wt (at both temperatures) and wounding had no effect on sco-
parone production.
Temperature effect on Citrus resistance was discussed by Hartmann
and Nienhaus (1974a). Lemon trees, which werefound to be susceptibleto
P . citrophthora at 2OoC, were resistant at 28OC. The optimal temperature
for mycelial growth of this pathogen in vitro is 28OC. Disease inhibition in
Citrus bark at 26-28OC seems to be caused by defense mechanisms of the
host tissue. Results of this study confirm that the elevated temperature
(28 "C)increases resistance ofCitrus to P. citrophthora by stimulating sco-
parone production (Fig.5). This may explainwhy several Citrus rootsocks
were found to be more resistantto P. citrophthora in the summer than in
the winter (Hartmann and Nienhaus, 1974a; Broadbent, 1969).
In both temperatures scoparone concentration reached the peak on the
fourth day after inoculation and then started to decline. Such a pattern of
accumulation and degradationis typical for phytoalexins in plants (Bailey
and Mansfield, 1982).
Kim et al. (1991) tested the temperature effect on lemon fruit resistant
to Penicillium digitatum. It was found that greater amounts of scoparone
accumulated in lemon peel when inoculated with P . digitatum and incu-
bated at 36OC as compared to 17OC. Heat treatment at 36OC prevented
decay development while treatmentat 17OC had no effect. However, since
the optimum temperatureis 24OC for P. digitatum in vitro itis possiblethat
both heat treatment at 36OC and scoparone accumulation have an inhibi-
tory effect on P . digitatum in vivo (Fig. 6). The effect of temperature on
scoparone production was studied by Arimoto and Homma (1988), who
found that the largest quantity of this compound was produced in Citrus
melanose spots and scars at 25OC, with progressively lower levels at lower
temperatures until no scoparone was detected at 10°C or 5OC. Similarly,
results of Afek and Sztejnberg (1988b) indicated that the high-temperature
effect was strongly expressed in rough lemon. This species is considered to
be susceptibleto P . citrophthora. Rough lemon branchthat gave a suscepti-
ble reaction at 2OoC reacted as a resistant species at 28OC. In this case
scoparone concentration in the bark was seven times higher at 28OC than at
2OoC, in parallel with the increase in the resistance of Citrus against P.
citrophthora.
C. Effect of y Irradiation and UV Illumination

Citrus branch and fruit'were inoculated with P. citrophthora 24 hr after
treatment with different y-irradiation doses. Scoparone concentration and
lesion length in the branches were measured 4 days after the inoculation
200 -
-150 -
c
3
L:
-
c
I
$100 -
Y
a
W
z
3 50-
0
0
V
cn
L\ Heat treatment
period k
ob , (4
" \
0 I 2 3 4 5 IO 40 70
Figure 6 Effect of heat treatment on accumulation of scoparone in lemon flavedo

inoculated withPenicillium digitatum (Kim et al., 1991).
whereas scoparone concentration and infectedarea of the fruits were mea-

sured after 7 days. Maximum scoparone concentration in the inoculated
branch wasachieved after treatmentwith 400 krad y irradiation and
reached up to 970 and 530 pg/g fresh wt in macrophylla and sour orange,
and 100 and 82 pg/g fresh wt in rough lemon and Shamouti, respectively
(Table 2).
In the noninoculated, irradiated (400-krad) branch, scoparone concen-
tration was much higher (about seven times) in the resistant species and
moderately (two to three times) in susceptible species as compared to the
nonirradiated control. However, y irradiation significantly increased sco-
parone concentration inboth inoculated and noninoculated branches only
when treated with a 300- or 400-krad dose. This treatment also resulted in
significant decrease of lesion length inthealltested species (Table 2).
In fruits (sour orange, Valencia, Shamouti,and grapefruit), scoparone
concentration after treatment with 400 krad and subsequent inoculation
with P. citrophthora was about two to threetimesgreater than in the
nonirradiated fruit. Maximum concentration of 94.7 pg/g fresh wt was
achieved in infected grapefruit (Table3). However, as in branches, a signifi-
cant increase in scoparone concentration occurred only after treatment with
300 and 400 krad y irradiation.
M
c , o
- E
e-
v)
U
3
E
%J
QJ
0 C
E
.e
d
8
3
c
0
E:
l
273
274
Scoparone (6,7-Dimethoxycoumarin), a Citrus Phytoalexin 275
y Irradiation affected the size of lesions only in grapefruit (Table 3).

Infected areas ofgrapefruit significantly decreasedfrom 175 mm2to 83 and
50 mm2after treatment with 300 and 400 krad, respectively.
Riov (1971)and Dubery and Schabort (1987)had reportedthat scopar-
one accumulated in Citrus fruit peel after treatment with y irradiation.
These investigators did not detect scoparone in nonirradiated tissue. Results
of this research show that a small amount of scoparone naturally exists in
branches and fruits of various species ofCitrus (Tables 2 and 3).
Furthermore, this study suggests that a correlation exists among sco-
parone concentration, lesion length,and resistance of Citrus tissue. y Irra-
diation (300and 400 krad) increased scoparone concentration inboth inoc-
ulated (P.citrophthora) and noninoculated branches and fruits (Tables 2
and 3). The levelof scoparone in the inoculated, irradiated bark, both
resistant and susceptible, and grapefruit fruit peel was sufficient to inhibit
the growth of the fungus in vivo while in sour orange, Shamouti, and
Valencia fruits scoparoneconcentration waslow and hence the fungal
growth was not inhibited.
Riov et al. (1968)reported that y irradiation increased the activity of
phenylalanine ammonia-lyase (PAL), a key enzyme in the biosynthesis of
coumarins and phenols in plants (Hanson and Havir,1981;Legrand, 1983;
Jones, 1984). Afek and Sztejnberg (1988a)also supported this finding and
related it to scoparone accumulation. They showedthat AOA, a competi-
tive inhibitor of PAL, suppressed scoparone production in Citrus and this
was accompanied with decreased resistance.
Kim et al. (1991)measured the concentrations of scoparone in lemon
fruit illuminated with UV light. The concentrations were 27, 77, and 120
pg/g fresh wt when treated with 1.5, 3.0, and 4.5 x IO4 erg/mm, respec-
tively, after 7 days of incubation. Later, scoparone concentration declined
(Fig. 7). Rodov et al. (1992)tested the effect of UV light in kumquat. The
quantity of the phytoalexin reached a peak of 530 pg/g fresh wt 1 1 days
after treatment with UV dose of 1.5 x IO3 J/m2 and then declined rapidly
to trace levels after a month (Fig. 8).
Chalutz et al. (1992)reported that grapefruit exposed to 30-60 sec of
UV illumination and subsequently inoculated with P. digitatum showed
reduction in incidence of green mold. A process of induced resistance in
grapefruit seemed to develop gradually with time after treatment. Maxi-
mum response was observed between24 and 48 hr after the treatment. The
activity of PAL in the peel of UV-treatedgrapefruit increased within24 hr
following treatment and remained high at 48 hr, compared with the low
and unchanged level of PAL activity in the peel of the nontreated control.
These results suggest a possible involvement of an induced mechanism of
resistance. This maybe due to an increase of antifungal compounds in vivo.
I20
U V - Irradiation.
/ o 1.5x10~erg/rnrn~
e 3 II
c 0 4.5 11
ov!! -
I
7
I
14
l
21
DAYS AFTER TREATMENT
Figure 7 Effect of UV dose on accumulation of scoparone in lemon fruits (Kim et

al., 1991).
600
-
c
3 500 -
c
v)
a,
400 -
-
b
2300
-
v
a,
g 200
-
0
0 IO 20 30
Days aftertreatment
Figure 8 Timecourse of scoparoneaccumulationanddepletioninUV-treated

kumquat fruit(Rodov et al., 1992).
However, the mechanism of resistance suggested by Chalutz et al. (1992)

has to be stimulated withinthe first 24-48 hr after treatment and infection
with the pathogen.
We now can assume that the concentrations of scoparone requiredfor
inhibition of growth of P. citrophthora in vitro and in vivo are similar. A
comparison between scoparone concentrations that inhibit the advance of
P. citrophthora in vivo and its growth in vitro showed that the advance
of the pathogen in marcophylla and trifoliate terminates 2 days after the
inoculation, when the concentration of the phytoalexin is about 150 pg/g
fresh wt (Fig. 2). This concentration is equivalent to EDe for P. citroph-
thora in vitro (Fig. 3). In the susceptible Citrus species when scoparone
concentrations remainlow the advance ofthe pathogen is not inhibited.
D. Effect of Fosetyl-AI and Phosphorous Acid

Fosetyl-A1 and phosphorous acidwere found to induce resistance inCitrus
against P. citrophthora by increasing scoparone accumulation in the tissue
(Afek and Sztejnberg, 1989).Studieshaveshown that fosetyl-Al (trade
name Aliette, Rhone-Poulenc Sanitaire, Lyon, France) haslittle effect on
mycelial growth of oomycetes in vitro (Sanders et al., 1983; Bompeix and
Saindrenan, 1984), but has the ability to control diseases of plants caused
by these pathogens (Farihet al., 1981a, b; Sanders etal., 1983). Coffey and
Bower(1984),Fenn and Coffey (1984,1985), and Ouimette and Coffey
(1989) suggestedthat fosetyl-Al is degradedinto phosphorous acid (H3P03)
in vivo and this compound acts directly against the pathogen asa fungicide.
Alternatively, fosetyl-A1 mayact indirectly against pathogensby activating
host defense mechanisms (Bompeix et al., 1980; Guest, 1984a, b; Vernenghi
and Ravise, 1985; Khan et al., 1986).
The accumulation of scoparone and the advance of P. citrophthora
(lesion length) in the bark of 3-month-old Citrus branches were measured
after treatmentwithfosetyl-Al or and inoculationwith P. citroph-
thora. In all the Citrus species tested except for Niva (very susceptible to
the pathogen), concentration of scoparone in inoculatedbark treated with
300 pg/ml fosetyl-Al or 125 pg/ml &PO3 was two to four times greater
than in the controls (inoculatedand nontreated branches). Treatment with
higher concentration of these compounds decreased scoparone concentra-
tion (Figs. 9 and 10).
Lesionlengthinmacrophylla, sour orange (resistant to P. citroph-
thora), and rough lemon (susceptible to P. citrophthora) decreased more
rapidly than in Niva after treatments with 0-300 pg/ml of fosetyl-AI or 0-
125 pg/ml of H3PO3. In Niva, treatments with fosetyl-Al and H3PO3 had
no effect on scoparone concentrations,and lesion length sharply decreased
1600
I ' O r l
................................ ...................................... .__.
.
h
0 aoo 400 600 eo0

FOSETYL-AL (pg ml-l)
Figure 9 Accumulation of scoparone (left)and lesion length (right) in Citrus bark

of the resistant species macrophylla ( 0 )and sour orange (0), and the susceptible
species rough lemon(A), and Niva (W), 96 hr after inoculation with Phytophthora
citrophthora at an incubation temperature of 2OOC. Branches were treated with
fosetyl-AI 3 hr before inoculation. Vertical bars are standard errors. (Afek and
Sztejnberg, 1989).
only when treated with more than 500 pg/ml fosetyl-Al or200 pg/ml H3PO3
(Figs. 9 and 10). Fosetyl-Al and H3PO3 did not induce scoparone produc-
tion in healthy tissue in any of the species tested. The regression analysis
showed that inhibition of mycelial growth area of P. citrophthora was
significantly correlated with increasing concentrations of fosetyl-Al (I? =
0.960, P < 0.01) and H3P03 (? = 0.949, P < 0.01). The EDSovalues of
fosetyl-Al andformycelialgrowth were 55 and 7 pg/ml, respec-
tively.
Probablytheargument as t o whetherfosetyl-Al andactdirectly
against pathogens in vivo (Coffey and Bower, 1984; Fenn and Coffey,
1984, 1985;Ouimette and Coffey, 1989) or act indirectly against pathogens
in vivo by increasing resistance (Bompeix et al., 1980; Guest, 1984a, b;
Khan et al., 1985, 1986; Vernenghi and Ravise, 1985) will remain open.
Results of the present study support both opinions for direct and indirect
effectsof fosetyl-A1 andon P.citrophthora in vivo. Similarly,Smil-
lie et al.
(1989) presented evidence for both direct and indirect modes of
action of phosphiteon Phytophthora spp. causing disease in plants.
le00 , I P
0 50 126 200 300 400
Figure 10 Accumulation of scoparone (left) andlesionlength (right) in Citrus

( 0 )and sour orange (0 ), and the suscepti-
bark of the resistant species macrophylla
ble species rough lemon (A), and Niva (m), 96 hr after inoculation with Phytoph-
thora citrophthora at an incubation temperature of 2OOC. Branches were treated
with H3PO3 3 hr before inoculation. Vertical bars are standard errors (Afek and
Sztejnberg, 1989).
Fosetyl-Al and H3P03 induce resistance Citrus in by increasing concen-

trations of scoparone in macrophylla, sour orange, and rough lemon. How-
ever, when fosetyl-AI penetrates the plant it is degraded t o H3PO3, which is
much more toxic to Phytophthora spp. than fosetyl-Al and inhibits the
pathogen as a fungicide (Coffey and Bower,1984; Fenn and Coffey, 1984,
1985). Results of this study show that the concentration of H3PO3 required
for the same effect as that offosetyl-A1 is 40%. This is a good indication
that fosetyl-Alisdegradedinto about 40% in vivo.
This research has led to the suggestion that fosetyl-Al and H3P03 affect
P. citrophthora in vivo in two ways. Apparently, at low-level treatments it
stimulates the host defense mechanisms, while at higher levels it acts di-
rectly as a fungicide. In the Citrus species macrophylla, sour orange, and
rough lemon, maximum production of scoparone is stimulated by 300 pg/
m1 fosetyl-A1 or 125 pg/ml H3PO3, but the concentration of scoparone in
Niva remains low (Figs. 9 and 10). This is explained by the “potential of
resistance” (the particular selection of Citrus to produce scoparone) that
these three species have and Niva has not.
Sztejnberg
280 and Afek
Scoparone, which is involved in Citrus resistance, and the fungicide

both have an effect onP.citrophthora in Citrus plants with the potential of
resistance, whereas in Niva onlythe fungicide has an effect on the fungus.
Thus, the concentrations of fosetyl-Aland H,P03 that are required to stop
the advance of the pathogen in macrophylla, sour orange, and rough lemon
are lower than those in Niva.
To support this hypothesis, we assume that in vitro and in vivo ED50
values are similar and that lesion length in vivo reasonably reflects fungal
growth in vitro. From Figs. 9 and 10 it appears that 50% lesion length in
Niva (on which fosetyl-Al and &PO3 directly affected the pathogen with-
out the involvement of scoparone) occurredafter application of fosetyl-Al
(750 pg/ml) or &P03 (350 pg/ml). In vitro, the ED50 values for fosetyl-Al
or &P03 were 55 and 7 pg/ml, respectively. It suggests that only 55 pg/ml
fosetyl-Al and 7 pg/ml are presentwhen ED50 valueisreachedin
vivo. If one assumes from this that only 2% of the applied &PO3 or 7.3%
of fosetyl-Al enter the tissue, it means 22 pg/ml fosetyl-Al and 2.5 pg/ml
H,P03, which are equivalent to only EDzoand ED25 values, respectively, in
vitro. Probably, these concentrations in macrophylla, sour orange, and
roughlemon cannot stop the advance of P. citrophthora without the
involvement of scoparone.
Furthermore, treatments of inoculated macrophylla, sour orange, and
rough lemon with 800 pg/ml fosetyl-A1 or 400 pg/ml H3PO3 did not in-
crease scoparone concentration in vivo. The explanation for this is that
treatments with high concentrations of these compounds inhibit the physio-
logical activities ofP.citrophthora. When the physiological activities ofthe
pathogen are completely inactivated, the effect of fosetyl-Al and
on scoparone production is similar to their effect on noninfected tissue.
Fosetyl-AI or H3P03 alone cannot stimulate scoparone production in healthy
tissue.
E. Probable Hypothesisfor the Biosynthetic Pathway of

Scoparone in Citrus
Cinnamic acid was reported to be a precursor of several coumarins in plants

(Sequeira, 1969; Fritig, 1972; Clarke and Baines, 1976). Other studies show
that the amino acid phenylalanine is the precursor of cinnamic acid in
plants (Amerhein and Zenk, 1977; Grisebach, 1977; Gross, 1978). The re-
sults of Afek and Sztejnberg (1988a) indicate that the precursor of scopar-
one is apparently phenylalanine. The authors tried to prove this theory in
two differentways: measurement ofthe distribution of radioactive scopar-
one in the bark 4 days after treatment with the isotope ['4C]phenylalanine
and inoculation with P. citrophthora. The results showed that scoparone
Scoparone (6,7-Dimethoxycoumarin), a C i f m Phytoalexin 281
PpL
AOA""> I
I
V
PHENYLALANINE---->CINNAMIC ACID---->P-COIJMARIC ACID-"->
"" >I-METHOXY 0-COUMARIC ACID---->7-METHOXYCOUMARIN
Figure 11 Biosynthesic pathwayof 7-methoxycoumarin from the amino acid phe-

nylalanine (Brown, 1978; Legrand, 1983).
was labeled withI4C.Total incorporation was 1 lolo and 15% in the resistant
species, sour orange, and macrophylla, respectively, as compared to 1.5%
in the susceptible species, Shamoutiand Niva.
Phenylalanineis deaminated to trans-cinnamic acid by the enzyme PAL
(Hanson and Havir, 1981; Legrand, 1983; Jones, 1984).
It is known that AOA is a competitive inhibitor of PAL (Amerhein,
1978; Hoagland and Duke, 1982). The precursor of simple coumarin and
7-methoxycoumarin, whose chemicalstructure is similar to scoparone (6,7-
dimethoxycoumarin),is cinnamic acid (Brown, 1978). We can assume that
the biosynthetic pathways of scoparone and 7-methoxycoumarin are simi-
lar. Inhibition of PAL activity by AOA may inhibit the production of
scoparone in the biosynthesis pathway (Fig. 11). Resultsof the present
research showthat resistant species responded as susceptible species follow-
ing treatment with 10 mM AOA. Scoparone concentrations inthe bark of
the resistant speciestreated with 10 m M AOA 4 days after inoculation with
P. citrophthora was 32.5-43.4 pg/g fresh wt as compared to 295-472 pg/g
fresh wt in the control (inoculated and nontreated branches).
Lesion length inthis group, 4 days after the inoculation variedfrom 9.2
to 12.0 mm as compared to 2.8 to 5.1 in the nontreated control (Fig. 12).
The effectof AOA on scoparone concentrationand on lesion length inthe
susceptible species was insignificant.
W. EPILOG
We believe that the present research establishesthe role of scoparone as a
phytoalexin conferring resistance of Citrus to P. citrophthora. This chapter
confirms the hypothesis of Broadbent (1969) that morphological exclusion
of P. citrophthora in Citrus tissue cannot be responsiblefor the resistance
to this pathogen, and that biochemical or physiological differences must
exist between resistantand susceptible species. Our study shows differences
between resistant and susceptible Citrus species following inoculation with
P. citrophthora. Scoparone accumulates inboth groups but the concentra-
and reaches 10-15 times higher
tion in the resistant species rapidly increases
1l
Afek and Sztejnberg
1
2
C
CITRUS SPECIES
Figure 12 The concentration of scoparone ([eft)and lesion length (right) in the

bark of the following Citrus species96hr after inoculation with Phytophthora
citrophthora: (1) macrophylla (resistant); (2) @foliate orange (resistant); (3) sour
orange (resistant); (4) roughlemon(susceptible); (5) Shamouti (susceptible); (6)
Niva(susceptible),treated (El)and nontreated (U) withaminooxyaceticacid
(AOA). The incubation temperature after the inoculation was 2OOC. Different let-
ters indicate significant differences accordingto Duncan’s multiple range test (p =
0.05) (Afek and Sztejnberg, 1988a).
than that in the susceptible species. The advance of the pathogen (lesion
length) at that time is 2-6 times greater in the susceptible species as com-
pared to the resistant species.
The effect of physical treatments, such as temperature, y irradiation,
and UV illumination, and of chemical treatments, such as fosetyl-Al and
phosphorous acid, onCitrus resistance is expansively discussed in this chap-
ter. Difference between the high- and low-temperature effect on Citrus can
give a good explanation for seasonal influences on Citrus susceptibility.
Furthermore, nonchemical approaches, such as high temperature and y-
irradiation treatments, applied to increase resistance against diseases may
replace, orat least reduce, the use of chemicals against pathogens.
Results of fosetyl-Al and phosphorous acid experiments give another
point of view for the argument whether these compounds act directly on
pathogens in vivo, similar to the effect of fungicides, or indirectly by in-
creasing resistance. Probably, in Citrus, these compounds act against P.
citrophthora in bothways.
An understanding of the biosynthetic pathway of scoparone in Citrus
can give a good opportunity to understand the mechanism of resistance.
Future studies should focus on the involvement of new approaches, such as

UV treatments as well as controlled and modified atmosphere treatments,
on the increased Citrus resistance to pathogens.
REFERENCES
Afek, U., and Sztejnberg, A. (1988a). Accumulation of scoparone, a phytoalexin
associated with resistance of Citrus to Phytophthora citrophthora. Phytopa-
thology 7 8 1678-1682.
Afek, U., and Sztejnberg, A. (1988b). The involvementof scoparone (6,7-
dimethoxycoumarin) in resistance of Citrus rootstocks against Phytophthoru
citrophthora. Proceedingof the Sixth International Citrus Congress,2. Mar-
graf Publishers, Weikersheim, Germany. pp. 779-785.
Afek, U., and Sztejnberg, A. (1989). Effect of fosetyl-Al and phosphorous acid on
scoparone, a phytoalexin associated with resistance of Citrus to Phytophthora
citrophthora. Phytopathology79:736-739.
Afex, U.,and Sztejnberg, A. (1993). Temperature and gamma irradiation effects
on scoparone, a citrus phytoalexin conferring resistance to Phytophthora ci-
trophthora. Phytopathology83: 753-758.
Afek, U., Sztejnberg, A., and Carmely, S. (1986). 6,7-Dimethoxycoumarin, a Cit-
rus phytoalexin conferring resistance againstPhytophthora gummosis. Phyto-
chemistry 25: 1855-1856.
Amerhein, N. (1978). Novel inhibitors of phenylpropanoid metabolism in higher
plants. Regulation of Secondary Product and Plant Hormone Metabolism (M.
Luckner and K. Schreiber, eds.), PergamonPress, Oxford. pp. 173-182.
Amerhein, N., and Zenk, M. H. (1977). Metabolism of phenylpropanoid com-
pounds. Physiologie Vegetale 15: 251-260.
Aritmo, Y., and Homma, Y. (1988). Studies on Citrus melanose and Citrus stem-
end rot by Diaporthe citri (Faw.) Wolf. 9. Effect of light and temperature on
the self-defense reaction of Citrus plants. Ann. Phytophatol. Soc. Jpn. 54:
282-289.
Aritmo, Y., Homma, Y., and Ohsawa, T. (1986). Studies on Citrus melanose and
Citrus stem-end rot by Diaporthe citri(Faw.) Wolf. 5. Identification of phyto-
alexin in melanosespot. Ann. Phytopathol. Soc. Jpn. 52: 620-625.
Bailey, J. A., and Mansfield, J. W. (1982). Phythoalexins.Wiley. New York.
Ben-Yehoshua, S., Shapiro, B., and Moran, R. (1987). Individual seal-packing
enables the use of curing at high temperatures to reduce decayand heal injury
of Citrus fruits. HortScience 22: 777-783.
Ben-Yehoshua, S., Shapiro, B., Kim, J. J., Sharoni, J., Carmeli, S.,and Kashman,
Y. (1988).Resistance of Citrus fruit to pathogens and its enhancement by
curing. Proceedings of the Sixth International Citrus Congress, 3. Margraf
Publishers, Weikersheim, Germany. pp. 1371-1379.
Bompeix, G., and Saindrenan, P. (1984). In vitro antifungal activity of fosetyl-Al
and phosphorous acid on Phytophthora species. Fruits 39 777-785.
Bompeix, G., Ravise, A., Raynal, G.; Fettouche, F., and Durand, M. C. (1980).
Method of obtaining necrotic blocking zones on detached tomato leaves using

aluminum tris. Hypotheses as to its mode of action in vivo. Annales de Phyto-
pathologie. 12: 337-351.
Broadbent, P. (1969). Observation on mode of infection of Phytophthora citroph-
thora in resistant and susceptible Citrus roots. Proceedings of the First Znterna-
tional Citrus Symposium.International Society of Citriculture. 3: 1207-1210.
Brown, S. A. (1978). Biochemistry of the coumarins. Rec. Adv. Phytochem. 12:
249-285.
Chalutz, E.,Droby, S., Wilson, C. L., and Wisniewski, M. E. (1992). UV-induced
resistance to postharvest diseases ofCitrus fruit. J. Photochem. Photobiol. 15:
367-374.
Clarke, D. D., and Baines, P. S. (1976). Host control of scopolin accumulation in
infected potato tissue. Physiol. Plant Pathol.P 199-203.
Coffey, M. D., and Bower, L. A. (1984). In vitro variability among isolates of eight
Phytophthora species in response to phosphorous acid. Phytopathology 74:
738-742.
DeLange, J. H., Vincent, A. P., Du Plessis, L. M., Van Wyk, P.J., and Ackerman,
L. G. J. (1976). Scoparone (6,7-dimethoxycoumarin) inducedin Citrus peel by
black spot, Guignardia citricarpaKiely. Phytophylactica 8 83-84.
Dubery, I. A., and Schabort, J. C. (1987). 6,7-Dimethoxycoumarin: a stress metab-
olite withantifungal activity in gamma-irradiatedCitrus peel. Suid-Afrikaanse
Tydskrif vir Wetenskap 83: 440-441.
Farih, A.,Menge, J. A., Tsao, P. H., and Ohr, H. D. (198la). Metalaxyl and
efosite aluminum for control of Phytophthora gummosis and root rot on Cit-
rus.Plant Dis. 65: 654-657.
Farih, A., Tsao, P. H., and Menge, J. A. (1981b). Fungitoxic activity of efosite
aluminum on growth, sporulation, and germination of Phytophthora citroph-
thora and P. parasitica. Phytopathology71: 934-936.
Fenn, M. E, and Coffey, M.D.(1984). Studies on the in vitro and in vivoof
fosetyl-A1 and phosphorous acid. Phytopathology 74: 606-61 1.
Fenn, M. E., and Coffey, M. D. (1985). Further evidence for the direct mode of
action of fosetyl-Aland phosphorous acid. Phytopathology 75: 1064-1068.
Fritig, B., Hirth, L., and Ourisson, G. (1972). Biosynthesis of phenolic compounds
in healthy and diseased tobacco plants and tissue cultures. Hoppe-Seyler’s 2.
Physiol. Chem. 353: 134-135.
Grisebach, H.(1977). Biochemistry of lignification. Natunvissenschaften 65: 619-
625.
Gross, G. G. (1978). Recent advances in the chemistry and biochemistry of lignin.
Rec. Adv. Phytochem. 12: 177-220.
Guest, D. I. (1984a). The influence of cultural factors on the direct anti-fungal
activities of fosetyl-Al, propamocarb, metalaxyl, SN 75196 and Dowco 444.
Phytopathol. Z.111: 155-164.
Guest, D.I. (1984b). Modification ofthe defense responsesin tobacco and capsicum
following treatment with fosetyl-A1 [Aluminum tris (0-ethyl phosphonate)].
Physiol. Plant Pathol.25: 123-124.
Hanson, K. R., and Havir, E. A. (1981). Phenylalanine ammonia-lyase. The Bio-

chemistry of Plants, Vol. 7 (P. K. Stumpf and E. E. Conn, eds.), Academic
Hartmann, G., and Nienhaus, F. (1974a). Gummosis pathogen Phytophthora ci-
trophthora (Smith and Smith) Leonian and Hendersonula toruloideaNattrass
on Citrus limon in Lebanon. 11. Seasonal influence. Z. Pfanzenkrankheit.
Pflanzenschutz 81: 433-457.
Hartmann, G., and Nienhaus, F. (1974b). The isolation of xanthoxylin from bark
of Phytophthora and Hendersonula-infected Citrus limon and its fungitoxic
effect. Phytopathol. Z. 81: 97-113.
Hoagland, R. E., and Duke, S. 0. (1982). Effect of glyphosate on metabolism of
phenolic compounds. VIII.Comparison of the effects of aminooxyacetateand
glyphosate. Plant Cell Physiol. 23:1081-1088.
Ismail, M. A., Rouself, R. L., and Brown, G. A. (1978). Wound healing in Citrus.
Isolation and identification of 7-hydroxycoumarin (umbelliferone) from grape-
fruit flavedo and its effect on Penicillium digitatumSacc. HortScience 13:358.
Jones, D. H. (1984). Phenylalanine ammonia-lyase: regulation of its induction and
its role in plant development. Phytochemistry 23:1349-1350.
Khan, A. J., Vernenghi, A., and Ravise, A. (1986). Incidence of fosetyl-Al and
elicitors on the defense reactions of Citrus attacked by Phytophthora spp.
Fruits 41:587-595.
Khan, A. J., Kunesch, G., Chuilon, S., and Ravise, A. (1985). Biological activity of
xanthyletin, a new phytoalexin ofCitrus. Fruits40 807-81 1.
Kim, J. J.,Ben-Yehoshua, S., Shapiro, B., Henis, Y., and Camely, S. (1991).
Accumulation ofscoparone in heat-treated lemon fruit inoculated withPenicil-
lium digitatum Sacc. Plant Physiol.97: 880-885.
Legrand, M. (1983). Phenylpropanoid metabolism and its regulation in disease. Bio-
chemicalPlantPathology (J. A. Callow, ed.), Wiley, New York, pp. 367-384.
Musumeci, M. R., and Olivera, A. R. (1975). Accumulation of phenolsand phyto-
alexins in citrus tissue inoculated with Phytophthora citrophthora(Smith and
Smith) Leonian.Summa Phytopathologica I : 275-282.
Musumeci, M. R., and Olivera, A. R. (1976). Accumulation of phenolsand phyto-
alexins in Citrus tissues inoculated with Phytophthora citrophthora. Summa
Phytopathologica2: 27-3 1.
Ouimette, D.G., and Coffey, M. D. (1989). Phosphonate levels in avocado (Persea
americana)seedlings and soil followingtreatment with fosetyl-Alor potassium
phosphonate. Plant Dis. 73:212-215.
Riov, J. (1971). 6,7-Dimethoxycoumarinin the peel of gammairradiated grapefruit.
Phytochemistry I0 1923.
Riov, J., Monselise, P. S., and Kahan, S. R. (1968). Effect of gamma radiation on
phenylalanine ammonia-lyase activity and accumulation of phenolic compound
in Citrus fruit peel. Environ. Exp.Bot. 8: 463-466.
Rodov, V., Ben-Yehoshua, S., Kim, J. J., Shapiro, B., and Ittach, Y.(1992).
Ultraviolet illumination inducesscoparone production in kumquat and orange
fruit and improves decay resistance.J. Am. Soc. Hort. Sci. 117: 788-792.
Sztejnberg
286 and Afek
Sanders, P. L., Houser, W. I., and Cole, H. (1983). Control of Pythium spp. and
Pythium blight ofturfgrass with fosetyl aluminum.Plant Dis. 67: 1382-1383.
Sequeira, L. (1969). Synthesis of scopolin and scopoletin intobacco plants infected
with Pseudomonassalanacearum. Phytopathology 5 9 473-478.
Smillie, R., Grant, B. R., and Guest, D. (1989). The mode of action of phosphite:
evidence for both direct and indirect modes of action on three Phytophthora
spp. in plants. Phytopathology 79 921-926.
Tatum, J. H., and Berry,R. E. (1977). 6,7-Dimethoxycoumarin in the peels of
Citrus. PhytochemistryId: 1091-1092.
Vernenghi, A.,and Ravise, A. (1985). Stimulation de la reaction dela tomate contre
le Phytophthora parasitica par lephosetyl-AI par des eliciteurs fongiques.
Fruits 40: 495-502.
Vernenghi, A., Ramiandrasoa, F., Chuilon, S., and Ravise, A. (1987). Phytoale-
xines desCitrus: seselin proprietes inhibitriceset modulation de synthese.Fruit
42: 103-1 11.
13
Stilbene Phytoalexins and
Disease Resistance in Vitis
Wilhelm Dercks
Fachhochschule Erfurt, Erfurt, Germany
L. L. Creasy and C. J. Luczka-Bayles
Cornell University, Ithaca, New
York
1. INTRODUCTION
A. General Information onStilbeneCompounds
Stilbenes have been found in a number of plant families (Ingham, 1976,
1982; Ingham and Harborne, 1976). Their antifungal nature has been impli-
cated in preventing wood decay (Hart and Shrimpton, 1979; Hart, 1981)
and in disease resistanceof plants againstdifferent pathogens (Wardet al.,
1975; Ingham, 1976; Hart, 1981; Aguamah et al., 1981). Biosynthetically,
stilbenes are derived from the shikimic-polymalonic acid pathway. Stilbene
synthase, which isthe key enzyme inthe biosynthesis of stilbenes in ground-
nuts (Arachishypogaea)and also in Vitis spp. (Fritzemeierand Kindl, 1981;
Melchior and Kindl, 1991), converts one molecule of p-coumaroyl-CoA and
three molecules of malonyl-CoA into 4,3 ',5 "trihydroxystilbene, common-
ly known as trans-resveratrol (Ingham, 1976; Rupprich and Kindl, 1978).
In UV-irradiated grapevine leaves, phenylalanine was a goodbut tyrosine a
poorprecursor (Lkgcake and Pryce, 1977b), indicating that the4'-hy-
droxylisprobablyintroduced at thecinnamicacidstage.trans-Resver-
atrol (hereafter resveratrol) appears to be the most widely distributed stil-
bene in several plant families.
B. Phytopathological Relevanceof Stilbenes in Vifis Species:

An Overview on Initial Research Work
The production of stilbenes by Vitis spp. as a response to fungal infection
was first reported by Langcake and Pryce (1976). They observed a blue
fluorescence inthe zone of apparently healthy cells adjacent to the advanc-
ing margin of lesions in leaves of Vitis vinifera infected with Botrytis cin-
287
288 Dercks et al.
erea when the leaves were examined under long-wavelength ultraviolet radi-
ation (365 nm).The compound responsiblewas identified as resveratrol.It
was not detectable in healthy leaves but accumulated locallyto high levels
at the sites of infection in diseased leaves. Resveratrol
was also foundto be
a major constituent of lignified stem tissue (Langcake and Pryce, 1976;
Pool et al., 1981). In addition to the predominant resveratrol, other com-
pounds considered oligomers of resveratrol and termed viniferins were also
synthesized. The most important ones were a- and winiferin (Langcake
and Pryce, 1977a,c; Pryce and Langcake, 1977). These three compounds
were produced sequentially inthe leaves following infection suggesting,the
occurrence of a polymerization process. A similar compound, pterostil-
bene, was also identified,but was onlyfound in very small amounts (Lang-
cake et al., 1979). The structural formulas and chemical names of these
four stilbenes are given in Fig.1.
These initial findings stimulated research projects in several different
groups around the world. Whereas each group focusedon specific aspects,
the overall goal, for a long time, has beenthe same: the elucidation of the
involvement of stilbenes in disease resistanceVitisof spp. Only recently has
attention been directed to other aspects, e.g., the significance for human
health of the contents of resveratrol in grapes.The purpose of this chapter
is to highlight the major achievements made with emphasis on the results
generated inour laboratory and to indicate some potentialfuture prospects
of stilbene research.
C. Antifungal Properties of Stilbenes

The ECSo values of various stilbenes derived from orchinol rangedfrom 10
to 50 pg/ml in a number of fungi includingPhytophthora infestans, Moni-
linia fmcticola, and Venturiainaequalis (Ward etal., 1975). Thecom-
pounds were more active against spore germinationthan mycelial growth.
With Helminthosporium carbonum,an ECSo value of 50 pg/ml for resvera-
trol against mycelial growth was obtained (Ingham, 1976). ECSo values of
resveratrol for germination of the grapevine pathogensPlasmopara viticola
and B. cinerea ranged from 71 to over 200 pg/ml. This was also the case
with mycelial.growthof the latter fungus. The ECSo valuesfor winiferin
ranged from 19 to over 200 pg/ml(Langcake and Pryce, 1976,1977a;
Langcake, 1981; Luczka, 1982; Stein, 1984; Stein and Blaich, 1985).
Compared with antibiotic agents or fungicides this activity does not
appear very potent, but there is little doubt that the quantities found in
diseasedleaftissue contribute effectively to cessationof the parasite’s
growth in the host tissue. Concentrations of 400-600 pg/ml fresh wt, de-
pendent on Vitis spp. and stilbene tested, have been measured in grapevine
Stilbenes and Disease Resistance 289
OH
Resveratrol
(trans-4, 3’,5’-trihydroxy stilbene)
(dimer
of
resveratrol) OH
’OH
a-Viniferin
(cyclic trimer of resveratrol)
0 CH3
Pterostilbene
(trans-3.5-dimethoxy-4’-hydroxystilbene)
Figure 1 Stilbene compounds identified in Vitis species.
leaves(Langcake and Pryce, 1976; Langcake, 1981). Thissuggests that

amounts at the actual infection sites may be
even higher.
Little is known about the mode of action of stilbenes in fungi, but
recent work suggestsan interference of pterostilbene withthe functionality
ofmembraneproteins(Pezet and Pont, 1988b, 1990; Pont and Pezet,
1991). Structure-activity studies with substituted trans-hydroxystilbenes de-
rived from resveratrol and pterostilbene showedthat the effects on mem-
290 Dercks et al.
brane proteins were related to the electronic character, lipophilicity, and

molecular weight ofthe substances (Pont and Pezet, 1990). An involvement
of benzylic or allylic radicals with fungicidal modeaction
of was implicated
by experiments withstructural analogs of stilbenes(Amoldi et al., 1989).
D. Objectives of the Investigations
The studies summarized herein have served to find answers to different

questions. In the following, the objectives of all investigations described in
the following sectionsare explicitly listedto clarify the flow of logic in this
chapter.
1. Determination of basic features of stilbene synthesis in leaves and ber-
ries of Vitis spp. These studieswere conducted to evaluate the methods
of stilbene induction to study patterns of synthesis, the capacity of
various Vitis spp. to synthesize stilbenes (stilbene production poten-
tial), and the role of endogenous and exogenous plant conditions for
stilbene synthesis.
2. Aspects of host-parasite interactions between Vitis spp. and selected
pathogens in leaves. B. cinerea: These experiments were carriedout to
study the rate and level of accumulation following inoculationand the
relation between stilbene production potential and disease resistance.
P. viticola: The purpose of these investigations was to determine the
stilbene production potential withrespect to resistance, the role of
inoculum density in elicitation of stilbenes,the plant’s response during
the early stages ofinfection, and the comparative relevanceof resvera-
trol and e-viniferin.
3. Occurrence of stilbenes infoods derived from grapes. The adoption of
new cultivars selectedfor disease resistance raises questions on whether
new compounds with toxicological properties will be introduced inthe
food supply. In one well-documented case, a disease-resistant cultivar
of potato (Akeley et al., 1968) was introduced only to be later with-
drawn from production when it was found that the increased disease
resistance was due to exceptionally high concentrations of mutagenic
glycoalkaloids (Zitnakand Johnston, 1970). Studies were conauctedon
both grape berriesand wines to quantify the stilbenes present in them.
4. Involvement of stilbene phytoalexinsin the modeof action of the fun-
gicide fosetyl-A1 against P. viticola in Vitis spp. These studies were
conducted to determine whether or not fosetyl-Al can induce stilbene
synthesisinleavesof Vitis spp. and a potentialfosetyl-Alinduced
accumulation of stilbenes is significant enough to be considered an
important aspect of its mode of action.

A. Plants of Vitis Species Used and Selectionof leaf and
Berry Material for Experiments
The plants used wererooted cuttings takenfrom the Vitis species collection
at the New York State Agricultural ExperimentStation, Geneva, New York
14456, USA. The vines were maintained in the greenhouse exceptfor yearly
8-week rest periods in a cold room. The second to sixth fully expanded leaf
was selectedfor experiments because very young and very old leavesdo not
synthesize high concentrations of stilbenes (Pool et al.,1981; Luczka, 1982;
Stein, 1984).
To study the phytoalexin production potential of grape berries over the
course of the season, berries were harvested from field- or greenhouse-
grown plantsat sampling times considered appropriate (Creasy and Coffee,
1988).
B. Culture of and Experiments with Pathogens

Plasmopara viticola was maintained on detached grapevine leaves incu-
bated on moist filter paper Petri
in dishes. Subsequent inoculation of leaves
or leaf discs which had been cut witha cork borer for specific experiments
were carried out with sporangia suspensions. Stilbene detoxification tests
were conducted with sporangia in aqueous suspension.
Botrytis cinerea was maintained on V-8 juice agar. Spore germination
tests for assessment of antifungal activity of stilbenes were carried out in
malt extract broth incubated in multiwell tissue culture plates. Mycelium
growth inhibition testswere conducted on solid Czapek Dox agar. Inocula-
tion experiments were .carriedout with drops of spore suspensions placed
on the lower surface of leaves or leaf discs.
C. Induction of Stilbenes
To assess the general capacityfor phytoalexin synthesis (stilbene production
potential) in Vitis spp., the lower surfaces of leaves were exposedto 8 min
UV irradiation (254 nm) from a spectroline UV lamp (0.6 mW/cm*). UV
irradiation is known to cause rapid transcription of defense genes in plants
(Chappel and Hahlbrock, 1984). After 48 hr in darkness (to avoid photo-
chemical alternations of stilbenes: Battersby and Greenock, 1961; Black-
burn and Timmons, 1969; Blaichand Bachmann, 1980) leaf discs werecut,
blotted, weighed, and stored at -2OOC until extracted.
To assess fungal elicitation of stilbenes, leaves were inoculated as de-
scribed belowand samples collectedat appropriate times following inocula-
tion (see below).
292 Dercks et al.
D. Stilbene Purification and Quantitation

Stilbenes are present in lignified stem tissueof most Vitis spp. (Langcake
and Pryce, 1977a; Pool et al., 1981). External standards for analysis were
extracted from canes in 70% methanol. Upon removal of the methanol, the
aqueous phase was extracted with ethyl acetate. Purified compounds were
obtained by high-performance liquid chromatography (HPLC). Stilbenes
were identified by their UV spectra in ethanol (Langcake and Pryce, 1976;
Ingham, 1982) as well as by mass spectrometry. Concentrations were deter-
mined spectrophotometrically.
E. Stilbene Extraction and Measurement

Stilbenes were extracted from tissue in methanol and partitioned in ethyl
acetate (Langcake and Pryce, 1976; Pool et al., 1981). Ethyl acetate was
distilled off in vacuo. All these procedures were carried out in dim lightto
avoid photochemical alterations of stilbenes (see above). Initial measure-
ments (Pool et al., 1981) were carried out with thin layer chromatography
(TLC) and gas chomatography (GC), for which the trimethylsilyl deriva-
tivesweremade. Later (Luczka, 1982; Creasy and Coffee, 1988; Dercks
and Creasy, 1989a, b), concentrations of stilbenes were measured by HPLC
on a silicic acid column with 4% methanol/96% methylene chloride as
eluent. Wine analyses required two HPLC separations, the first on silicic
acid with3% methanol in methylene chloride and the secondafter complete
isomerization to cis-resveratrol on a reverse phase column (Siemann and
Creasy, 1992).
F. Application of the Fungicide Fosetyl-AI

Leaf discs were cut and placed upper surface down on fungicide-soaked
filter paper inPetri dishes. In time-course studies on stilbene accumulation
discs were sampled as found appropriate (Dercks and Creasy, 1989b; see
also above).
111. BASIC FEATURES O F STILBENE SYNTHESIS IN

LEAVES A N D BERRIES O F VlTIS SPECIES
A. Induction and Patterns of Stilbene Production
Stilbenes, especially resveratrol,are present in lignified stem tissue of most
Vitis spp. (Langcakeand Pryce, 1977a; Pool et al., 1981),but either are not
detectable in healthy leaves and berries or are present at only very low
levels.There are, however,severalwaysofinducing theirsynthesis
(Table 1).
Stilbenes 293
Table 1 Overview of Inducing Principles of StilbeneFormation in Vitis spp.
Inducing
principles
Chemicals Phenyloxyalkyl- Blaich and Bachmann (1980)

carbonic acids;
2-desoxy-~-
glucose
Sugar solutions; Stein and Hoos (1984)
mucic acic
Fosetyl-Al Raynal etal. (1980)
Stein and Hoos (1984)
Dercks and Creasy (1989b)
UV irradiation Langcake and Pryce (1976,1977b)
Pool et al., (1981)
Luczka (1 982)
Barlass et al.(1987)
Creasy and Coffee (1988)
Dercks and Creasy (1989a)
Jeandet et al. (1991)
Mechanical injury - Langcake and Pryce (1976)
Inoculation with B. cinerea Langcake and Pryce (1976)
fungal Langcake and McCarthy (1979)
pathogens Pool et al. (1981)
Luczka (1982)
Stein and Blaich (1985)
P. viticola Langcake (1978,1981)
Langcake and Love11 (1980)
Raynal et al.(1980)
Barlass et al. (1987)
Dercks and Creasy (1989a,b)
The unspecific nature of the inducing principles led Langcake (1981) to

term the viniferins stress metabolites rather than phytoalexins. Whereas
inoculation studiesare indispensable to determine the involvements of stil-
bene, in disease resistance (see below),UV irradiation has been instrumen-
tal in the study of production patterns in Vitis spp. In all our studies we
have never detected eithera-viniferin or pterostilbene. Therefore, the focus
has been on resveratrol and E-viniferin. The synthesis of these compounds
as induced by irradiation with short-wavelength (254 nm) UV radiation is
species specific. These results of Luczka (1982) will be discussed in more
detail because they exemplify basic general patterns. Vitis spp. differed in
the quantity and quality of stilbenes synthesized. UV irradiation-induced
294 Dercks et al.
resveratrol and e-viniferin concentrations over time

were compared for four
species (Fig. 2). V. rupestris B-38produced the highest concentrations of
stilbene compounds(400 nmol/g fresh wt) while V. vinifera CV. “Chardon-
nay” producedthe lowest concentrations(90nmol/g freshwt). V. doaniana
and V. riparia B-50were two ofthe species between these two extremes.
The time course of production of resveratrol and e-viniferin found by
Langcake and Pryce (1977b)was very similarto the one shown byV. riparia
B-50. In this species, resveratrol accumulated more quickly in the tissue
than e-viniferin. A s polymerization proceeded, resveratrol concentration
declined while e-viniferin continued to build. After 3 days, levels of both
compounds began to decline. V. rupestris B-38 and V. doaniana showed
similar curves. In V. viniferaCV. “Chardonnay,” however, e-viniferin accu-
mulatedmorequickly than resveratrol and was the predominantcom-
pound. The relative proportions of the two compounds in each species were
similar. Species which produced high amounts of resveratrol also produced
high amounts of e-viniferin and vice versa. The actual ratios between the
two compounds depended on when they were measured. In addition, the
time requiredto reach maximumtotal stilbene content depended somewhat
on the amount produced. It took longer for V. rupestris B-38 to reach its
peak (3.5 days) than V. viniferaCV. “Chardonnay” (1.5 days).
B. Stilbene Production Potential of Vitis Species

From the above resultsit is obvious that each Vitis spp. may have its own
characteristic response to induction. Some species are able to synthesize
high concentrations of stilbenes, whereasothers attain medium or low lev-
els. With regard to the correlation between accumulated stilbene levels and
resistance to pathogens, attempts were made to rank Vitis spp. according
to their capacity to synthesize stilbenes (hereafter referred to as stilbene
production potential). To assess this general potential, plants of different
Vitis spp. were induced byUV irradiation and B. cinerea in 1980-81 (Table
2) and again in 1985-86 by UV light (Table 3). Resveratrol production
potential of grape berries was measured in six cultivars by UV induction in
1988 (Table 4).
A direct comparison betweenthe data of Tables 2 and 3 is not possible
(different plants, different conditions, different calculations: amounts of
stilbenes per cmz leaf area or mg fresh wt, respectively). The independent
sets of data, however, had crucial aspects in common. Whereas Vitis spp.
with “intermediate” stilbene production potential showed variable results
and were not easyto define, the order of species with highor low potential,
respectively, was never reversed.For instance, “Castor” or V. rupestris B-38
always exhibited a high potential whereas V. vinifera spp. always showed
V. doanlana
0.5 2.51.5 4;5

. 3;5 4.5 3.5 2.5 1.5 0.5
Days afterirradatii Days after irradiation
v. tiparia 8-50
Days afler irradiation Days afler irradialion
Figure 2 Production of resveratrol and e-viniferinin four Vitis spp. following

irradiation with short-wavelength(254 nm) W radiation.
296 Dercks et al.
Table 2 Total Stilbene Production (Resveratrol plus

E-Viniferin)”in Leaves of 10 Vitisspp. in 1980-81b
Stilbene
nmol stilbenes/
production
Vitis spp. potential
area leaf cm2
V. rupestris B-38 44 H
V. longii 34
V. rupestris 6544 29
V. treleasei 28
V. champini CH-3-48 27
V. riparia B-50 25
V. andersonii 20
V. cinerea 1-66 18
V. vinifera CV.
“Chardonnay” 16
V. argentifolia GBC-17
15 Low
‘Meanof three tests (one test induced by UV irradiation, two tests in-
duced by inoculations withB. cinerea).
bGreenhouse plants.
Source: Data from Luczka (1982).
a low capacity for stilbene synthesis. Another common feature was that
resveratrol was the predominant stilbene in native American Vitis spp.
whereas winiferin was morecommoninEuropean V. vinifera spp. In
summary, UV irradiation has madeit possible to select Vitis spp. with high
and low phytoalexin potentialto study specific questions like,for instance,
the relation between stilbene contents and disease resistance (see below).
In berries, no stilbenesother than resveratrol were found in significant
amounts. The resveratrol productionpotential changed during the season.
Greenhouse-grown berries had highest potentialafter set but before matu-
ration. In field samplesthe highest potentialwas reached near vkraisonbut
it decreased after late August, independentof maturity (Creasyand Coffee,
1988). Similar results were obtained by Jeandet al.et(1991). The resveratrol
production potentialwas also found to be species-specific (Table4).
C. Role of Endogenous and Exogenous Plant Conditions

for Stilbene Synthesis
Many pitfalls inthe initial work on grapevine stilbenes
were caused by the
enormous variabilityof stilbene production in grapevines. This pertainsto
variability not only between different species
but also between:
Stilbenes 297
Table 3 Total Stilbene Production (Resveratrol plus eviniferin)” in Leaves

of 17 Vitis spp. in 1985-86b
Stilbene
nmol
stilbenes/ production
Vitisspp. fresh g wt potential
Interspecific CV. “Castor” (B-7-2) 255 1 H

V. riparia B-50 200
V. rupestris B-38 176
V. doaniana 173
V. andersonii 160
Interspecific CV. “Pollux” (B-6-18 ) 139
V. acerifolia 132
V. argentifolia GBC-17 123
Interspecific CV. “Vignoles” 118
V. rupestris 6544 105
V. vinifera CV. “Riesling” 90
V. vinifera CV. “Chardonnay” 83
V. champini CH-3-48 65
V. cinerea 1-66 63
V. vinifera CV. “Bacchus” 51
V. vinifera CV. “Muller-Thurgau” 45 I
V. ireleasei 25 Low
*Mean of2 tests (induced byUV irradiation).
bGreenhouse plants.
Source:Data from Dercks and Creasy (1989a).
Table 4 Resveratrol Production Potential” in Berry Skin ofSix

Vitis spp. Near VCraison in 1988b
nmol resveratrol/
Vitisspp. cm2 berryskin
~~
V. labrusca CV. “Concord” 49

V. vinifera CV. “Cabemet Sauvignon” 49
V. vinifera CV. “Riesling” 22
Interspecific CV. “Chancellor” 22
V. labrusca CV. “Catawba” 18
Interspecific CV.White”
“Cayuga 18
“Meanof two tests (induced by
UV irridation).
bField-grown plants.
Source:Data from Creasy and Coffee (1988).
298 Dercks et al.
Table 5 Relationship Between Origin of

Leaf Material andVariability of Stilbene
Production inVitis spp.
Variability of stilbene
Origin production
Greenhouse
Intermediate
to high
Field Extreme
Source: DatafromLuczka(1982);Stern(1984);
Dercks and Creasy (unpublished).
Different clones ofthe same species

Different plants ofthe same clone
Different leaves ofthe same plant
Differences between leaf halves the
of same leaf
This set of complications is further magnified by plant vigor and leaf
age as well as by origin (greenhouseor field leaves). Tables5 and 6 summa-
rize the experience of independent researchers (Langcake and McCarthy,
1979; Pool et al., 1981; Luczka, 1982; Steinand Blaich, 1985; Barlass et al.,
1987; Bavarescoand Eibach, 1987; Dercksand Creasy, 1989a, b).
To obtain reproducible and reliable quantitative results it has become
customary to exclude field-grown leaves from specifically designed induc-
tion experiments (for instance, in SectionVI) which, although high in stil-
bene production potential, exhibit a degree of variability detrimental to
any quantitative analysis. Picking midshoot leaves from greenhouse plants
grown under the same conditions provides the best chanceto obtain homog-
enous leaf material. This, of course, does not pertain to studies where
specific aspects of stilbene production in field leaves or berries are ad-
Table 6 Relationship Between Leaf Position, DevelopmentStage,

and Stilbene Production Potential inLeaves of Vitis spp.
Stilbene production
Development
position
Leaf stage potential
Shoot base Old Low
Shoot center Mature High
Shoot tip Young Intermediate
Source: DatafromLuczka(1982);Stem(1984);DercksandCreasy(unpub-
lished).
Stilbenes 299
dressed, as in the study of production patterns during the course of the

season (Creasyand Coffee, 1988; Jeandet et al., 1991).
IV. ASPECTS OF HOST-PARASITE INTERACTIONS BETWEEN

VlTIS SPECIES A N D SELECTED PATHOGENS IN LEAVES
A. Botrytis cinerea
of Stilbene Accumulation FollowingInoculation
Rate and Level
The concentration of both resveratrol and winiferin increased 1 day after
inoculation in V. riparia B-50.This augmentation continuedfor another 2-
3 days and was very similarto the one describedfor V. riparia B-50 follow-
ing induction by UV as shown in Fig.2. These results have been described
in detail elsewhere(Pool et al., 1981). Contrary to the response induced by
UV irradiation, the concentrations did not decline until the last testingdate
(6 daysafter inoculation). This may attributed
be to the fact that in contrast
to the brief UV irradiation period, the inducing principle remained present
in the leaves. Very similar studies conducted by Blaich et al. (1982) and
Stein and Hoos (1984) showed exactlythe same pattern for nine Vitis spp.
of different stilbene production potentialand corresponding level of resis-
tance to B. cinerea. The relative ranking ofthe species (eight of which were
the same as inour studies) accordingto their stilbene production potential
was very similarto ours (Tables 2 and 3). These studies clearly demonstrate
that stilbenes accumulate at a faster rate and to a higherlevelin Vitis
spp. with high phytoalexin potential than in species with low phytoalexin
potential.
Relationship Between Stilbene Production Potential
and Disease Resistance
In the Vitis spp. used by Stein and Hoos (1984) in the study discussed
above, there was a close positive correlation between stilbene production
potential and resistance to B. cinerea. When the authors analyzed this cor-
relation in 95 Vitis spp., they found three different groups (Table7).
This is the same pattern as found in the work of our laboratory. The
results are described in detail elsewhere (Luczka, 1982).Of10 Vitis spp.
tested, the following ones fellinto the respective groups:
Group 1: V. cinerea 1-66, V. champini CH-3-48, V. argentifoliaGBC-17
Group 2: V. rupestris B-38, V. longii
Group 3: V. vinifera CV. “Chardonnay,” V. andersonii, V. riparia B-50,V.
rupestris 6544, V. treleasei. (The occurrence of V. vinifera CV. “Char-
donnay” in this group cannot be easily explained. Normallyit is very
susceptible to B. cinerea.)
300 Dercks et al.
Table 7 Groups of Relationships Between Stilbene Production Potential" and

Level of Resistance to B. cinerea in 95 Vitis spp. (according to Stein, 1984band
Stein and Hoos?
Stilbene
production
Level of resistance
Relative
frequency
potential
Group to B. cinerea of Vitis spp.
1 High Low Low

2 High
Intermediate to high High
3 Low Low High
"Inducedby inoculations withB. cinerea.
bDescriptionof results including identificationof Vitk spp.
'Description of results only.
The existence of genotypes in the third group suggests that resistance

may, in several cases, be associated with factors other than stilbene phyto-
alexins, which cannot be surprising since many of them are already known
(Fregoni, 1983; Stein, 1984; Jeandet and Bessis, 1989). Thereis a clear
message from the above results, i.e., high phytoalexin potential is never
associated with susceptibility. These findings heavily support the important
role of stilbenes in resistance of
Vitis spp. to B. cinerea. Further substantia-
tion stems from the results of Langcake and McCarthy (1979) and Lang-
cake (1981). They determinedthe amounts of resveratrol around the actual
lesions and found an inverse relationship between stilbene concentration
and susceptibility to B. cinerea. B. cinereais a more important parasite on
grape berriesthan on leaves (unfortunately, berries are much more difficult
to work with) and it is interesting to note that there is a good correlation
between the resistance of leaves and maturing berries (Stein and HOOS,
1984; Stein and Blaich, 1985). Midage berries are usually very resistant to
B. cinerea at the time they havethe highest potentialfor synthesizing resver-
atrol although the resveratrol potential decreases after vCraison (Creasy and
Coffee, 1988; Jeandet et al., 1991).
B. Plasmopara viticola
Relationship Between Stilbene Production Potential
and Disease Resistance
When 17 Vitis spp. were analyzed (Dercks and Creasy, 1989a), four differ-
ent groups were found (Table 8). Although the number of species tested
was smaller than for B. cinerea, and it cannot be said that all species will
follow the samepattern, the overall picture appears to be the same. Resis-
tance can be mediatedby factors other than stilbenes, but high concentra-
tions ofstilbenes are neverlinkedwithsusceptibility. Thus, for downy
Stilbenes 301
Table 8 Groups of RelationshipsBetweenStilbeneProductionPotential‘and

Level of Resistance toP. viticola in 17 Vitis spp.
Stilbene Level of
production resistance to
Group potential P. viticola Vitisspp.
1 low to low V. acerifolia, V. argentifolia
intermediate GBC-17
V. treleasei
V. vinifera CV. ‘Muller-Thurgau’
V. vinifera CV. ‘Chardonnay’
V. vinifera CV. ‘Riesling’
2 intermediate
intermediate V. andersonii
V. rupestris B-38, V. rupestris
6544
Interspecific cultivar Yignoles’
3 high high V. doaniana, V. riparia B-50
Interspecificcultivar ‘Castor’
Interspecfiiccultivar ‘Pollux’
4 low high V. cinerea 1-66, V. champini
CH-3-48
V. vinifera cv.‘Bacchus’
‘Induced by UV irradiation (see Table 3).
Source: Data from Dercks and Creasy (1989a).
mildew, the conclusions of stilbenes being important factors in the resis-

tance of Vitis spp. (similar results were obtained by Bavaresco
and Eibach,
1987) are in line with findingsfrom B. cinerea.
Role of Inoculum Density in the Elicitation of Stilbene Synthesis
and Plant Response During Early Stages of Infection
Since phytoalexins are only locally synthesized at the attempted site of
infection and the number of infection sites may be low a leaf
if is challenged
by onlya few fungal spores, the overall response of
the leaf may be difficult
to measure or even detect on a per-leaf base. To test whether or not the
stilbene response per sample unit can be optimized for quantitative research
by modifying the inoculum density, comparative studies were carried out
with Vitis spp. of different stilbene production potential and corresponding
levels of resistanceto P . viticola (Dercks and Creasy, 1989a). It was shown
that by increasing the number of infection sites it is possible to enhance
stilbene accumulation (Table9). Furthermore, it became obviousthat more
sporangia are produced (i.e., the attack is higher)when the inoculum den-
302 Dercks et al.
Table 9 Effect of Increasing the Inoculum Density” of P. viticola on Accumula-

tion of Stilbenesband Sporangia Production in Leaves of Three Vifisspp.
Stilbene Level of Effectcof increased inoculum
production stilbenes density on:
and level of actually
resistance to reached in Stilbene Sporangia
Vitis spp. P. vificola experiment accumulation production
V. vinifera Low Low + +d
CV. “Riesli@
V. rupesfris Intermediate Intermediate fd +
6544
Interspecific High High f +
CV. “Castor”
‘Inoculum densities tested (sporangia/ml):0; 30,000; 90,000,150,000.

bTestedbeforeinoculation; 2, 4, 6, 8, and 10 hrafterinoculation; 1, 2, 3. 4, 5 days after
inoculation.
c + ,positive; - ,negative.
dReduced effect again with the highest inoculum density tested.
Source: Data from Dercks and Creasy (1989a).
sity is increased.Thus, it was even possibleto break resistance ofthe inter-

specific cultivar “Castor.” With 150,000 sporangia/ml, reproduction was
obtained whereas this was not possible with lower inoculum densities.
These time course studies also revealedthat the most pronounced stil-
bene synthesis occurred duringthe first 2-10 hr following inoculation and
again later around the time of sporulation (Fig. 3). The absolute levels
reached depended on the Vitis spp. tested. They were highest in “Castor,”
intermediate in V. rupestris 6544, and lowest in “Riesling,” reflecting the
stilbeneproductionpotential of therespective Vitis spp.(Dercksand
Creasy, 1989a).
The implications of these findingsare twofold. First, they demonstrate
how important it is to take into account the severity of fungal challenge
when assessing the comparative degreeof phytoalexin response and corre-
sponding level of disease resistance in Vitis spp. This factor has frequently
been overlookedin studies with leavesfrom vines inthe field where disease
pressure is seldom homogenous. Second, the results show that phytoalexin
synthesis can occur very rapidly in plants. In most pathosystems this was
observed after days;here after hours.Whendrawingconclusionsone
should not only look at the plant’s internal “rhythm” of synthesis, but also
at the challenging pathogen’s development characteristics. The infection
process and colonization of host tissue by P. viticola are extremely rapid.
Stilbene synthesis
Oh 2h 4h 6 h 8 h 10h I d 2d 3d4d 5d
Time (hours: h; days: d) after inoculation
Figure 3 Schematicrepresentation of stilbene synthesis in three Vitis spp. V.

vinifera CV. “Riesling”; V. rupestris 6544; Interspecific cv.“Castor”(B-7-2) follow-
ing inoculation with P. viticola.
Hypersensitive reactions were seen after only a few hours by light and
electron microscopic studies,and there is strong evidence that stilbenes are
involved inthis reaction (Langcakeand Lovell, 1980; Langcake, 1981). Our
findings mayhaveimplications for studieswith other pathogens which
develop as rapid asP. viticola.
Comparative Relevance of Resveratrol and E- Viniferin
The antifungal properties of stilbenes ingenerd have already been reviewed
above. There isa lot of heterogeneityin these data concerning pathogens of
Vitis spp. from several research groups. One concept has been that resvera-
trol should not be considereda phytoalexin because of its low fungitoxicity
compared with that of winiferin (Langcake, 1981). In retrospect, it ap-
pears possible that the variability of the data can be attributed to the fact
that most studies were carried out with concentrations beyond the limits
of solubility of the compounds. When soluble concentrations of the two
substances were used (60-70 pg/ml) resveratrol was twice as inhibitory as
e-viniferintowardgerminationofsporangiaof P. viticola (Dercksand
Creasy, 1989a). These would, however, be equimolar concentrations. For-
mer comparisons hadbeen made on a weight basis. In addition, e-viniferin
was found to be less stablethan resveratrol in water. Inthe presence of P.
viticola sporangia, a rapid degradationof winiferin, but not of resveratrol,
was observed(Table10).Theseresults,togetherwith the fact that e-
viniferinhasneverbeendetectedbyusinleaftissuecolonizedby the
304 Dercks et al.
Table 10 Characteristic Features of Resveratrol and e-Viniferin with Regard to

Phytopathological Significance
Feature Resveratrol e-Viniferin
Short-term degradationin water -a +b

Degradation by sporangia
of P. viticola - +
Presence in diseased leavesof Vitis spp.
B. cinerea + +
P. viticola + -
Antifungal activity’
B. cinerea =e-Viniferin =Resveratrol
P. viticola >e-Viniferin < Resveratrol
’-, not observed.
+ ,observed.
‘Assessed on amolar basis.
Source: Data from Luczka (1982) and Dercks and
Creasy (1989a,b).
P. viticola isolate usedthroughout our studies, pointto an underestimation

of resveratrol’s regulatory role in the interaction between P. viticola and
Vitis spp. by several researchers.The significance of winiferin appears to
be more related to endo- and exogenous plant conditions than to active
resistantresponses of the host. In B. cinerea, resveratrol was equal to
eviniferin in termsof antifungal activity (Luczka,1982).
V. O C C U R R E N C E OF STILBENES IN G R A P E P R O D U C T S
A. Grapesand Juice
Resveratrol is found in all parts of the grape cluster (Creasy and Coffee,
1988). The component parts of “Riesling” berries, analyzed at harvest for
their residual resveratrol content, showed great differences. The highest
concentration was found in the seeds which, although contributing only7%
of berry weight, had73% of the total berry resveratrol. The skin and pulp,
composing half of berry weight, contained 18% of the resveratrol. The
small percentage remaining was found in the juice (Siemann and Creasy,
unpublished data).
B. Wines
Wines should reflectthe grapes from which they were made. Wine analyses
revealed unexpected large differences in resveratrol concentration and we
found that wine-making techniques greatly influenced the transfer of res-
Stilbenes 305
veratrol into wine (e.g., fermentation on or off skins) or removal from wine
(i.e., use of resins for fining; Siemann and Creasy, 1992). Research wines
made froma disease-resistant variety had a higher concentrationof resvera-
trol than one made from a more susceptible variety. In the same experi-
ment, wines from both varieties produced from vineyard plots not treated
with fungicides had higher concentrations than those from the fungicide-
sprayed plots (Siemannand Creasy, unpublisheddata).
VI. INVOLVEMENTOFSTILBENEPHYTOALEXINS IN THE

MODE OF ACTION OF THE FUNGICIDE FOSETYL-AL
AGAINST P. VITICOLA IN VITIS SPECIES
Fosetyl-Al (aluminum ethyl phosphite)ais systemic fungicide active against
fungi belonging to the Oomycetes(Cohen and Coffey, 1986). Its active
metabolite in plant tissue is phosphite (H,PO,), also referredto as phospho-
nate, phosphonic,orphosphorousacid(Fenn and Coffey, 1984,1985;
Luttringer and de Cormis, 1985; Saindrenan et al., 1985). Both fosetyl-Al
and phosphitehave a direct effect on fungalmetabolism and cellwall
synthesis (Dercks and Buchenauer, 1987; Dunstan et al., 1990; Smillie et
al., 1990) but the primary mode ofaction in fungiis still unknown. Recent
investigations suggestan interference with metabolismof phosphates (Bar-
chietto et al., 1990) and specifically with synthesis ofthe nucleotide adenyl-
ate (Griffith et al., 1990). There are also a number of reports which show
that accumulation of phenolic compoundsand phytoalexins in plant tissue
is associated with the action of phosphite in plants (e.g., Bompeix et al.,
1980; Durand and Sallt, 1980; Raynal et al., 1980; Dercks and Buchenauer,
1986; Guest, 1984,1986; Saindrenan et al., 1988; Afek and Sztejnberg,
1989; Nemestothy and Guest, 1990). These findings have led to the assump-
tion that there may bea combination of direct and indirect modes of action
of phosphite (Dercks and Creasy, 1989b; Smillie et al., 1989, 1990). The
following is a brief descriptionof the principles of fosetyl-Al-induced accu-
mulation of stilbenes in Vitis spp. It is based on the work of Dercks and
Creasy (1989b), where the results are described in detail.
Two basic ideas were pursued throughout the studies. Onewas to mea-
sure stilbenes at critical times during the infection processto see whether
stilbenes are specificallysynthesizedagainst P. viticola. Sooner or later
phytoalexins will always be synthesized unspecifically because of massive
cell death in any kind of reaction, but this need not necessarily be related
to
induction by fosetyl-Al. The other idea was to make a distinction of the
fosetyl-Al-induced effect in Witis leaf tissue between species of low, inter-
mediate, and high phytoalexin potential and corresponding levels of resis-
tance to P.viticola to see ifa stimulation of phytoalexin synthesis in suscep-
306 Dercks et al.
tible species might lead to a response equivalentto that in resistant species.

The results are schematically summarized in Table 11.
In noninoculated leaf tissue, fosetyl-Al caused only a moderate induc-
tion of stilbene synthesis irrespective the
of stilbene production potential of
the Vitis spp. tested. In inoculated tissue, the induced response depended
strongly on the inherent capacity for phytoalexin synthesis, i.e., the re-
sponseof Vitis spp.withlowstilbeneproductionpotentialwas not as
high as that in species with higher potential.Furthermore, the phytoalexin
response was greater with postinfectional than with preinfectional treat-
ments, probably because of the many more cells invaded bythe pathogen
and, hence, involved in the synthesis of stilbenes which occurs following
cell damageor death.
Despite the weak induction of stilbenes in Vitis spp. with low phyto-
alexin production potential,fosetyl-Alhad a pronounced effect on the
reduction of sporangiaformation following preinfectional treatments. This
was even stronger in species with higher potential. The effect of postinfec-
tional fungicide treatments at the same concentrations was weaker, but
depended heavily on the amounts of resveratrol synthesized. These were
higher in Vitis spp. with high than in species with low phytoalexin produc-
tion potential.
The results show that fosetyl-Al must have a direct influence on the
pathogen because reproduction ofP. viticola was reduced despite the lack
of significant concentrations of stilbenes being produced. This was obvious
with preinfectional treatments. On theother hand, it cannot be denied that
stilbenes are involved in the action of the fungicide. This became obvious
with postinfectional treatments where the efficacy was correlated with accu-
mulated levels of stilbenes which, in turn, were closely linked with phyto-
alexin production potential and level of inherent resistanceto P. viticola of
the Vitis spp. tested.
These findings are evidence for a combination of direct and indirect
modesof action of fosetyl-Al against P. viticola in Vitis spp. They are
consistent withthe hypothesis that the fungicide first interferes with metab-
olism ofthe fungus. This interference could then result in alterations of the
physiology of the pathogen-host interaction and thus lead to a response of
resistance in the host in which phytoalexins are involved. This view has
distinctly evolved in the more recent literature (e.g., Dunstan et al., 1990;
Smillie et al., 1989, 1990). On the other hand, claims of a primary indirect
mode of action of fosetyl-Al via stimulationnatural of defense mechanisms
in plantswhich were made initially when the potent directantifungal activ-
ity of phosphite had not yet been realized (e.g., Bompeix et al., 1981) are
no longer sustained.
+
+
+ +
+ +
+ + +
+ +
+ +
+ + +
+ + +
+ + +
+
+ +
+ + +
+ + +
+ + +
+ +
+ + +
+ + +
+ + +
U
c)
td
.e
L 2 s
S E E
U
c)
c
H
308 Dercks et al.
VII. EPILOG
A. Stilbenes and Disease Resistance in VifisSpecies
There is nodoubt today that stilbene phytoalexins are very important fac-
tors in the resistance of Vitis spp. to B. cinerea and P. viticola (Fregoni,
1983; Stein, 1984; Jeandet and Bessis, 1989). A leaf high in concentrations
of these compoundswill successfully ward off a challenge by these patho-
gens. There is evidencethat this might also be the case with berries (Pezet
and Pont, 1988a; Jeandet et al., 1991),but less information is available for
these organs, largely because berries are experimentally more difficult to
work with than leaves. With regard to B. cinerea, however, more data on
berries are urgently needed because this pathogen is eminently more impor-
tant on fruit than on leaves. The realization that at some developmental
stages there is a good correlation between the resistance of berries and
leaves, and that the conclusions from studies with leaves maybe extended
to berries, certainly needs more substantiation (Stein and Hoos, 1984; Stein
and Blaich, 1985).
With regardto Uncinula necatorthere is virtually noinformation in the
literature on the relationship between stilbene contents and resistance of
leaves or berries. The reason might bethat researchers maynot have found
a good correlation between resveratrol concentrations and disease resis-
tance but never reported on their negative findings.It is interestingto note
that stilbenes can only be induced by UV irradiation on the abaxial leaf
surfaces (Langcake and Pryce, 1977b; Pool et al., 1981) and that powdery
mildew is found more frequently on the adaxial than on the abaxial leaf
surfaces, although it is principally able to attack all green parts of the
plant (Bulit and Lafon, 1978). Furthermore, U.necator only penetratesthe
epidermal cells and not the deeper cell layers (Heintz and Blaich, 1990). If
it were generallytrue that a large part of the resveratrol synthesisis located
not just in but below the epidermal cells,which is suggested bythe findings
of Blaichand Bachmann (1980), then it would also be conceivable that only
insufficient amountsof the antifungal compound might come into physical
contact with the pathogen to account for inhibition of fungal growth. This
is certainly different from B. cinerea and P. viticola, which both penetrate
the outer cell layersand continue to grow inside the tissue of the inner cell
layers.
The antifungal properties of stilbenes initially stirred hopes that these
compounds might be used as biochemical markers of resistance in breeding
programs. Before this question can be finally resolved it is necessary to
overcome the difficulties discussed in Section 1II.C. A phenotype high in
stilbene production potential at any given point in time may not remain so
throughout its lifetime. Not all leaveson a plant are the same in potential;
Stilbenes 309
neither are different plants within a clone, nor different clones within a
species. Because ofthis, stilbenes have failedto serve as qualitative markers
of resistance in breeding and screening efforts (Luczka, 1982; Barlass et al.,
1987). Luczka’s work (1982) has also shown the difficulty of determining
traits of inheritanceof high stilbene accumulationthrough crosses between
parents of known stilbene production potential. None of the progeny fol-
lowed an obvious pattern and phenotypes belonging to all groups of stil-
bene production potential werefound irrespective ofthe parents’ potential.
It is of the utmost importance to characterize the genetics of stilbene syn-
thase expression in Vitis spp. to finally elucidate the involvement of stil-
benes in disease resistancethroughout the host’s life cycle.
In view of this, hopes for future disease resistance is some plant species
by introducing a gene from groundnut (Arachis hypogaea)which codesfor
stilbene synthase shouldmeet with cautious enthusiasm (Hain et al., 1990).
Even though the gene was expressed in transgenic tobacco plantsand res-
veratrol was identified, it is difficultto see whythe basic featuresof stilbene
production (see above) would be any more predictable or usable in trans-
genic plantsthan in its native location.
B. Detoxification of Stilbenes by Fungal Parasites of Vitis Species

Accumulation of stilbenes isthe result of synthesis, turnover of these toxic
metabolites by plant tissue(Hoos and Blaich, 1988), and active degradation
by the invading pathogen. Our work has shown that P. viticola may detox-
ify e-viniferin. The question whether this feature maybe important for
pathogenicity, asfound in someother parasite-host interactions (Van Etten
et al., 1989), deserves further study. At the moment, it is not known as to
whether detoxificationof winiferin is peculiar to some or common to all
isolates of P. viticola. However, Langcake’s observations(1978) contrasted
with ours, make it likely that the former may bethe case. It would also be
interesting to see whether or not some isolatesare able to degrade resvera-
trol. It has been reported that resveratrol and pterostilbene are subject to
detoxification by B. cinerea (Hoos and Blaich, 1990; Pezet et al., 1991).
Further research inthis area is needed.
C. Comparative Relevance and Distribution of

Stilbene Compoundsin DifferentPlant Organs
Our investigations have produced evidencethat resveratrol is more impor-
tant for the regulation of the interaction between Vitis spp. and P. viticola
than e-viniferin. We do not know muchabout a-viniferin and pterostilbene.
In which organs (leaves, berries, shoots, canes) are they heavily concen-
trated? Are they all phytoalexinsor are some just products of woody plant
310 Dercks et al.
parts (which seems to be the case for e-viniferin) with little relevance for
disease resistanceto pathogens that attack green organs?Also, there is little
information on the location of stilbenes withinthe cell and cell layersapart
from Blaich and Bachmann’s cytological studies(1980) which showed that
resveratrol appears to be deposited within small areas of the cytoplasm or
in the periplasm, probably near plasmodesmata, mostly in and below the
epidermal cells. Isthis commonly the case or were the deeper cell layersjust
not sufficiently challengedin this particular experiment?
D. Mode of Action of Stilbenes

Despite the studies of Pezet and Pont (1988b, 1990) and Pont and Pezet
a blank.
(1991) (seealso chapter14 by Pezetand Pont), this field is virtually
If we are to understand more about the overall significance of stilbenes,
more information on their mode of action at the biochemical and molecu-
lar, in addition to the histological, level is needed.
E. Can Synthesis of Stilbenes Be Induced Systemically?
Almost all research to date has been devoted to the elucidation of mecha-
nisms of local stilbene accumulation.To the best of our knowledge, nobody
has looked deeply into the question of whether synthesis of stilbenes can
also be part of a process of systemic acquired resistance, the signal for
in plant. If so, the upper leaves ofthe plant could
which is translocatible the
be “immunized”by challenging the lower leaves with an appropriate agent
or compound.
ACKNOWLEDGMENTS
This workwas supported by the New York State College of Agriculture and
Life Sciencesand the Deutsche Forschungsgemeinschaft.
The authors thank J. M. Babcock, J. A. Becker, and J. 0. Becker for
typing and editorial assistance with the manuscript. We are also grateful
to R. Pezet for providing a paperinpress and R. Blaich for valuable
discussions.
REFERENCES
Afek, U., and Sztejnberg,A. (1989). Effects of fosetyl-Al and phosphorous acidon
citrophthora. Phytopathology79 736-739.
Aguaniah, G. E., Langcake, P., Leworthy, D. P., Page, J. A., Pryce, R. J., and
istance
Disease
Stilbenes and 311
Strange, R. N. (1981). Isolation and characterization of novel stilbene phyto-

alexins from Arachishypogaea. Phytochemistry2 0 1381-1383.
Akeley, R. V., Mills, W. R., Cunningham, C. E., and Watt, J. (1968). Lenape: a
new potato variety high in solids and chipping quality. Am.Potato J. 45: 142-
145.
Amoldi, A.,Farina, G., and Merlini, L. (1989). Relazione struttura: attivitg anti-
fungina di fitoalessine isoflavonoidi.Notiziario sulle Malattie delle Piante 110
171-187.
Barchietto, T., Saindrenan, P., and Bompeix, G. (1990). Effects ofphosphonate on
phosphate metabolism in Phytophthora citrophthora. Pest. Sci. 3 0 365-366.
Barlass, M., Miller, R. M., and Douglas, T.J. (1987).Development of methodsfor
screening grapevines for resistance to infection by downy mildew. 11. Resvera-
trol production. Am. J. Enol. Viti. 3 8 65-68.
Battersby, A. R., and Greenock, A. I. (l%l). cis- and trans-3,3’, 4,4’-Tetrameth-
oxystilbenes. J. Chem. Soc. (part 2): 2592-2593.
Bavaresco, L., and Eibach, R. (1987). Investigations ofthe influence of N-fertilizer
on resistance to powdery mildew (Oidiumtuckeri), downy mildew(Plasmopara
viticola), and onphytoalexin synthesis indifferent grapevine varieties. Vitis26:
192-200.
Blackburn, E. V., and Timmons, C. J. (1969). The photocylisation of stilbene
analogues. Q.Rev. Chem. Soc. London 23: 482-503.
Blaich, R., and Bachmann,0 . (1980). Die Resveratrolsynthese bei Vitaceen; Induk-
tion und zytologische Beobachtungen. Vitis19: 230-240.
Blaich, R., Bachmann, O., and Stein, U. (1982). Causes biochemiques de la rksis-
tance dela vigne h Botrytis cinerea. Eur. Medit. Plant Prot. Org. (EPPO) Bull.
12: 167-170.
Bompeix, G., Fettouche, F., and Saindrenan, P. (1981). Mode d’action du pho-
skthyl-al. Phytiatrie-Phytopharmacie3 0 257-272.
Bompeix, G., RavisC, A., Raynal, G., Fettouche, F., and Durand, M. C. (1980).
ModalitCs de l’obtention des necroses bloquantes sur feuilles detachkes de to-
mate par l’action du tris-oCthyl phosphonate d’aluminium (phoskthyl d’alumi-
nium), hypothtses sur son mode d’action in vivo. Annales de Phytopathologie
12: 337-351.
Bulit, J., and Lafon, R. (1978). Powdery mildew ofthe vine. The Powdery Mildews
(D. M. Spencer, ed.), Academic Press, London, pp. 525-548.
Chappel, J., and Hahlbrock, K. (1984). Transcription of plant defense genes in
response to UV light or fungal elicitor.Nature 311: 16-18.
Cohen, Y., and Coffey, M. D. (1986). Systemicfungicides and the control of
oomycetes. Ann. Rev. Phytopathol. 24: 311-338.
Creasy, L. L., and Coffee, M. (1988). Phytoalexin production potential of grape
berries. J. Am. Soc. Hort. Sci. 113: 230-234.
Dercks, W., and Buchenauer, H.(1986). Untersuchungen zum Einfluss von Alu-
miniumfosetyl auf den pflanzlichen Phenolstoffwechsel in den Pathogen-Wirt-
Beziehungen Phytophthora fragariae-Erdbeere und Bremia lactucae-Salat. J.
Dercks, W., and Buchenauer, H.(1987). Comparative studieson the mode of action
312 Dercks et al.
of aluminium ethyl phosphite in four Phytophthora species. Crop. Prot. 6

82-89.
Dercks, W., and Creasy, L. (1989a).
L. The significance of stilbene phytoalexinsin
the Plasmopara viticola-grapevine interaction. Physiol. Mol. Plant Pathol. 3 4
189-202.
Dercks, W., and Creasy, L. L. (1989b). Influence of fosetyl-Al on phytoalexin
accumulation in the Plasmopara viticola-grapevine interaction. Physiol. Mol.
Plant Pathol. 3 4 203-213.
Dunstan, R. H., Smillie, R. H., and Grant, B. R. (1990). The effects of subtoxic
levels of phosphonate on the metabolism and potential virulence factors of
Phytophthora palmivora. Physiol. Mol. Plant Pathol. 36 205-220.
Durand, M. C., and SallC, G. (1981). effet du tris-o-&hylphosphonate d’aluminium
sur lecoupleLycopersicumesculentumMill.-Phytophthoracapsici LCon.
etude cytologique et cytochimique. Agronomie I : 723-732.
Fenn, M. E., and Coffey, M. D. (1984). Studies on the in vitro and in vivo antifun-
gal activity of fosetyl-Aland phosphorous acid. Phytopathology 7 4 606-61 1.
action of fosetyl-Aland phosphorousacid. Phytopathology 75: 1064-1068.
Fregoni, M. (1983). Fattori genetici ed agronomici predisponenti alla Botrytis. Vig-
nevini 10: 35-42.
Fritzemeier, K. H., and Kindl, H. (1981). Coordinate induction by ultraviolet light
ofstilbenesynthase,phenylalanineammonia-lyaseand cinnamate4hydroxy-
lase in leaves of Vitaceae.PIanta 151:48-52.
Griffith, J. M., Smillie, R. H., and Grant, B. R. (1990). Alterations in nucleotide
and pyrophosphatelevels in Phytophthorapalmivora following exposureto the
antifungal agent potassium phosphonate (phosphite). J. Gen. Microbiol. 136
1285-1291.
Guest, D. I. (1984). Modification of defense response in tobacco and capsicum
following treatment withfosetyl-A1[aluminiumtris(0-ethylphosphonate)].
Physiol. Plant Pathol. 25: 125-134.
Guest, D. I. (1986). Evidence from light microscopy of living tissuesthat fosetyl-Al
modifies the defense response in tobacco seedlings following inoculations by
Phytophthora nicotianae var. nicotianae. Physiol.Plant Pathol. 29 251-261.
Hain, R., Bieseler, B., Kindl, H., Schroder, G., and Stocker, R. (1990). Expression
of a stilbene synthase gene in Nicotiana tabacum results in synthesis of the
phytoalexin resveratrol.Plant Mol. Biol. IS: 325-335.
Hart, J. H. (1981). Role of phytostilbenes in decay and disease resistance. Ann.
Rev. Phytopathol. 1 9 437-458.
Hart, J. H., and Shrimpton, D. M. (1979). Role of stilbenes in resistance of wood
to decay. Phytopathology69: 1138-1 143.
Heintz, C., and Blaich, R. (1990). Ultrastructural and histochemicalstudies on
interactions between Vitis vinifera L. and Uncinula necator (Schw.) Burr. New
Phytologist 115: 107-117.
HOOS,G., and Blaich, R. (1988). Metabolism of stilbene phytoalexins in grapevines:
oxidation of resveratrolin single-cell cultures. Vitis27:1-12.
d Stilbenes 313
HOOS,G., and Blaich, R. (1990). Influence of resveratrolon germination of conidia

and mycelial growth of Botrytis cinerea and Phomopsis viticola. J. Phytopa-
thol. 129: 102-110.
Ingham, J. L. (1976). 3,5,4-Trihydroxystilbene as a phytoalexin from groundnuts
(Arachis hypogaea). Phytochemistry 15: 1791-1793.
Ingham, J. L. (1982). Phytoalexins from the leguminosae. Phytoalexins (J. A.
Bailey and J. W. Mansfield, eds.), Wiley,New York, pp. 21-80.
Ingham, J. L., and Harborne,J. B. (1976). Phytoalexin induction asa new dynamic
approach to the study of systematic relationships among higher plants. Nature
260 241 -243.
Jeandet, P., and Be&, R. (1989). Une rtflexion sur les mtcanismes morphologi-
ques et biochimiquesde l’interaction vignt-Botrytis. Bull. /‘Office International
de la Vigne et du Vin 703-704: 637-657.
Jeandet, P., Bessis, R., and Gauthtron, B. (1991). The production of resveratrol
(3,5,4’-trihydroxystilbene) by grape berries in different developmental stages.
Am. J. Enol. Vitic. 42: 41-46.
Langcake, P. (1978). Phytoalexin production by grapevines in relation to infection
by Plasmopara viticola(Abstr). Abstract of papers, 3rdInternational Congress
of Plant Pathology,Munich, p. 248.
Langcake, P. (1981). Disease resistanceof Vitisspp. and the production of the stress
metabolitesresveratrol, winiferin, a-viniferin and pterostilbene. Physiol.
Plant Pathol. 1 8 213-226.
Langcake, P., Cornford, C. A., and Pryce, R. J. (1979). Identification of pterostil-
bene as a phytoalexinfrom Vitis viniferaleaves. Phytochemistry 18: 1025-1027.
Langcake, P.,and Lovell, P. (1980). Light and electron microscopical studies of the
infection of Vitis spp. by Plasmopara viticola, the downy mildew pathogen.
Vitis 1 9 321-337.
Langcake, P., and McCarthy, W. V. (1979). The relationship of resveratrol produc-
tion to infection of grapevine by Botrytis cinerea. Vitis1 8 244-253.
Langcake,P., and Pryce, R. J. (1976). The production ofresveratrol by Vitis
vinifera and other members ofthe vitaceae as a responseto infection or injury.
Physiol. Plant Pathol.9 77-86.
Langcake, P., and Pryce, R. J. (1977a). A new class of phytoalexins from grape-
vines. Experientia33: 151-152.
Langcake, P.,and Pryce, R. J. (1977b). The production of resveratrol and the
viniferins by grapevines in response to ultraviolet irradiation. Phytochemistry
16: 1193-1196.
Langcake,P.,andPryce, R. J. (1977~). Oxidativedimerisation of4-hydroxy-
stilbenes in vitro: production of a grapevine phytoalexin mimic.J. Chem. Soc.
Chem. Commun. 7: 208-210.
Luczka, C. J. (1982). Stilbene phytoalexins and susceptibilityto Botrytis cinerea in
Vitis. M.S. Thesis, Cornel1 University, Ithaca, NY.
Luttringer, M., and De Cormis, L. (1985). Absorption, dtgradationet transport du
phostthyl-A1 et de son mttabolite chez la tomate (Lycopersicon esculentum
Mill.). Agronomie 5: 423-430.
314 Dercks et al.
Melchior, F., and Kindl, H.(1991). Coordinate and elicitor-dependent expression

of stilbene synthase and phenylalanine ammonia-lyase genes in Vitis CV. Op-
tima. Arch. Biochem. Biophys.288: 552-557.
Nemestothy, G . S., and Guest, D. I. (1990). Phytoalexin accumulation, phenylala-
nine ammonia lyase activity and ethylene biosynthesis in fosetyl-Al treated
resistant and susceptible tobacco cultivars infected withPhytophthora nicotia-
nae var. nicotianae. Physiol.Mol. Plant Pathol.37: 207-219.
Pezet, R., and Pont, V. (1988a). Mise en tvidence de pttrostilbtne dansles grappes
de Vitis vinifera. Plant Physiol. Biochem. 2 6 603-607.
Pezet, R., and Pont, V. (1988b). Activitt antifongique dansles baies de Vitis vin$
era: effets d’acides organiques et dupttrostilbtne. Revue Suksede Viticulture,
&Arboriculture, et &Horticulture20 303-309.
Pezet, R., and Pont,V. (1990).Ultrastructural observations of pterostilbene fungi-
toxicity in dormant conidiaof Botrytis cinerea Pers. J. Phytopathol. 129
19-30.
Pezet, R., Pont, V. and Hoang-Van,‘K. (1991). Evidence for oxidative detoxifica-
tion of pterostilbeneand resveratrol by a laccase-like stilbene oxidase produced
by Botrytis cinerea. Physiol.Mol. Plant Pathol.39441-450.
Pont, V., and Pezet, R. (1990). Relation between the chemical structure and the
biological activityof hydroxystilbenes againstBotrytis cinerea.J. Phytopathol.
130 1-8.
Pont, V., and Pezet, R. (1991). Un moyen de dtfense nature1 des vtgttaux: les
phtnols. Revue Sukse&Agriculture 23: 237-241.
Pool, R. M., Creasy, L. L., and Frackelton, A. S. (1981). Resveratrol and the
viniferins, their application to screening for disease resistance ingrape breeding
programs. Vitis20 136-145.
Pryce, R. J., and Langcake, P. (1977). a-Viniferin: an antifungal resveratrol trimer
from grapevines. Phytochemistry 16: 1452-1454.
Raynal, G., Ravist, A., and Bompeix, G . (1980). Action du tris-o-Cthylphosphonate
d’aluminium (phostthyl d’aluminium) sur la pathogtnie dePlasmopara viticola
et sur stimulation des rtactions de defense dea1 vigne. Annales de Phytopatho-
logic I 2 163-175.
Rupprich, N., and Kindl, H. (1978). Stilbenesynthases and stilbenecarboxylate
synthases, I. Enzymaticsynthesisof 3,5,4‘-trihydroxystilbene from p-coumar-
oyl coenzyme A and malonyl coenzyme A. Hoppe-Seyler’s Z. Physiol. Chem.
359 165-172.
Saindrenan, P., Barchietto, T., Avelino, J., and Bompeix, G . (1988). Effects of
phosphite on phytoalexin accumulation in leaves of cowpea infected with Phy-
tophthora cryptogea. Physiol. Mol. Plant Pathol.3 2 425-435.
Saindrenan, P., Darakis, G., and Bompeix,G . (1985).Determination of ethyl phos-
phite, phosphite, and phosphate in plant tissues by anion-exchange highperfor-
mance liquid chromatography and gas chromatography. J. Chromatogr. 347:
267-273.
Siemann, E. H., and Creasy, L. L. (1992).Concentrationof the phytoalexin resver-
atrol in wine. Am. J. Enol. Vitic. 43:49-52.
d Stilbenes 315
Smillie, R. H., Dunstan, R. H., Grant, B. R., Griffith, J. M., Iser, J., and Niere,
J. 0 . (1990). The mode of action of the antifungal agent phosphite.Eur. Med.
Plant Prot. Org. (EPPO) Bull. 2 0 185-192.
Smillie, R.H., Grant, B. R., and Guest, D. I. (1989). The mode of action of
phosphite: evidence for both direct and indirect modes of action on three
Phytophthora spp. in plants.Phytopathology 79: 921-926.
Stein, U. (1984). Untersuchungen uber biochemische und morphologische Merk-
male der Botrytisresistenz bei Vitaceen. PhD Thesis, University of Karlsruhe,
FRG.
Stein, U., and Blaich, R. (1985). Untersuchungen uber die Stilbenproduktion und
Botrytisanfalligkeit bei Vith-Arten.Vitis24: 75-87.
Stein, U.,and HOOS,G. (1984). Induktions- und Nachweismethoden fur Stilbene
bei Vitaceen. Vitis 23: 179-194.
Van Etten, H. D., Mathews, D. E.,and Mathews, P.S. (1989). Phytoalexin detoxi-
fication: importance for pathogenicity and practical implications. Ann. Rev.
Ward, E. W. B., Unwin, C. H., and Stoessl, A. (1975). Postinfectional inhibitors
from plants. XV. Antifungal activity of the phytoalexin orchinol and related
phenanthrenes and stilbenes. Can. J. Bot. 53: 964-971.
Zitnak, A., and Johnston, G. R. (1970). Glycoalkaloid content of B5141-6 potatoes.
Am. Potato J. 47: 256-260.
14
Mode of Toxic Actionof Vitaceae
Stilbenes on Fungal Cells
Roger Pezet and Vincent Pont
of Changins, Nyon, Switzerland
Swiss Federal Agricultural Research Station
1. INTRODUCTION
Though the toxicity of phytoalexins is well documented, their mechanism
of action on fungal cells ispoorly understood. However, many fragmentary
reports are available and it is possibleto propose a mechanism of fungitoxic
mode of action of the hydroxystilbenes. These compounds belong to the
large family of plant phenolics from which numerous other phytoalexins
are recognized. According to their basic common chemical structure, it is
expected that all phenolics act on similar biochemical pathways and that
their fungicidal mode of action is identical.
In a well-documented review, Smith (1982) discussed the toxicity of
phytoalexins. We can assumethat extensive membrane damages occur soon
after fungi are exposed to phytoalexins and suppression of exogenous respi-
ration may reflect insufficient uptake of substrate due to membrane dam-
ages. More precise is the toxic effect of rishitin to membranes; it acts by
increasing their permeabilityto small molecular weight compounds (Lyon,
1980).
Both phaseollin and pisatin have been shownto be toxic to plant and
fungal cells in a similar manner by disrupting either the structure or the
functioning of the plasma membrane (Hargreaves,1980). Laks and Prun-
ner (1989) suggest that pisatin and maackiain, as well as other flavonoid
phytoalexin analogues, function primarily against fungi as uncouplers of
oxidative phosphorylation. Accordingto O’Neil and Mansfield (1982), the
antifungal activity of flavonoids and isoflavonoids depends on some com-
mon physiochemical attributes, perhaps lipophilicity, and on the ability
to penetrate fungal cell wall and membranes rather than on a common
structure.
Onemaybesurprised that so manybiocidemolecules are phenols
possessing important conjugated systems. In such molecules, electrons of
chemical bondsare not immobile. They move inside the molecules, and this
317
Pont 318 and Pezet
movementofelectronsis due to alternate single and doublebonds

(Fig. 1).
Chemical structure of stilbenes constitutes a conjugated system with
aromatic character as described above. Hart (1981) reports the effects of
stilbenes on respiration of fungal cells. He suggests that they act as uncou-
pling agentsand they couldform protein-phenol complexes. Uncoupling of
electron transport and photophosphorylation were also observed on iso-
lated chloroplasts ofSpinacia oleraceain the presence of different stilbenes
(Gorham and Coughlan, 1980). The conjugated system of hydroxystilbenes
plays an important rolein the formation of charge transfer complexes
(CTCs), as described by Slifkin (1980), favoring contact and affinity with
proteins. Like hydroxystilbenes, the aromatic hydrocarbons, a group of
synthetic molecules,are able to constitute CTC with proteinstoo. Most of
these moleculesare synthetic fungicidesand for this reason their mechanism
of action was better known than that of phenolic phytoalexins. All these
informations are useful to explain how stilbenes kill fungal cells.
II. EFFECT OF HYDROXYSTILBENESON F U N G A L CELLS

Natural hydroxystilbenes that we have studied are phytoalexins produced
by Vitaceae, resveratrol (3,5,4’-trihydroxystilbene)and pterostilbene (33-
dimethoxy4’-hydroxystilbene). Other hydroxystilbenes with different phys-
icochemical properties were synthesized in our laboratory in order to ex-
plain relations betweenthe chemical structure and the biological activity of
these compounds (Table1) (Pont and Pezet, 1990).
A. Effect on Respiration
The first observed effect of hydroxystilbenes on conidia ofBotrytis cinerea
is a decrease of oxygen uptake. This effect may be slight or important,
depending on the substituents of the hydroxystilbenic basic structure (Table
2). Pterostilbene is the bestinhibitor of conidial respiration. We observed
that oxygen uptake was interrupted some minutes after the addition of
pterostilbene (Fig.2).
Figure 1 Conjugated system of p-coumaric acid. Arrows represent electron dis-

placement within the molecule (electronic delocalization).
Toxic Actionof Vitaceae Stilbenes 319
Table 1 Structure of trans-Hydroxystilbenes Tested on the Conidia of

Botrytis cinerea
R Ref.
~~ ~
H (1) -
Veschambre,
1-212
21 4-CH3
(2) (1967)
et al.
(3) 3&(OCH3)2 -
(4) 4-c1 Massarani (1957)
Veschambre etal. (1967)
3,5-(OCH,)z
86-86.5
King
(5) al. et (1953)
Spath and Schlager (1940)
3,5-(OH)z (6) 254-254.5 Nonomura et al. (1963)
Rupprich et al. (1980)
130-131
Veschambre
et(7) 3-Cl al. (1967)
(8) 3,4-C12 -
3,5-C12
148-150 (9) -
Table 2 Oxygen Uptake After10 min (nmol O2min)

by lo' 6-day-old Conidia of
Botrytis cinerea Treatedby the
Compounds of Table 1at 5 x lo-' Ma
R mol0 2
H 8.33 (1)
4-CH3 (2)
7.53
3,4-(OCH3)2 7.71
(3)
4-c1 5.40
~,~-(OCH~)Z (5) 2.79
3,5-(OH)z (6) 6.74
3-Cl (7)
5.51
3 ,4-C& (8)
2.92
3,5-C12 (9)
4.85
Conidia only: 10.4 nmol O,/min
*Cell respiration was measured polarographically witha Clark electrode
(PO, analyzer, Bachofer) at25 f 0.01 "C.
320 Pezet and Pont
50
Figure 2 Effect of pterostilbene on respiration of conidia of Botrytis cinerea. a,

Pterostilbene treated conidia. b, Control. Pterostilbene (50 pl/ml of an ethanolic
solution at 2.56 mg/ml) or ethanol for the control (50 pl/ml) areaddedsome
minutes (arrows) after the suspension of conidia (2.5 x lO’/ml) was placed into a
vessel of a Clark-type oxygen electrode (PO2Analyzer, Bachofer)maintained at 25
f O.0l0C.
B. Visible Effects on Conidia

When conidia of B. cinerea are placed on media containing different con-
centrations of hydroxystilbenes, germination is more or less affected de-
pending on the chemical structure of the hydroxystilbenes. The more dra-
matic effect is a total inhibition of germination and the destruction of all
organelles and cellmembranes(Pezet and Pont, 1990). Conidiaappear
under a light microscope to be structurally transformed (Fig. 3). We call
this effect the transformation of conidia. Their size decreased from a mean
of 10 x 8.5 pm to 8 x 6 pm and we observed a withdrawal of the cyto-
plasm from the wall. An outflow of cytoplasmic matter(arrow, Fig. 3) was
sometimes observed.
Figures 4 and 5 show the effects of concentration of different hydroxy-
stilbenes on theinhibition of germination and on the transformation ofB.
cinerea conidia. The most toxic compounds are pterostilbene and three
chlorinated hydroxystilbenes. Resveratrol remains totally inactive, even at
a concentration of M. Both inhibition of germination and transforma-
tion of conidiaand theeffect on cellular respiration design the pterostilbene
as the known most toxic natural hydroxystilbene producedthe byVitaceae.
C. Effect on Cellular Structures
Fungitoxic action of pterostilbene was studied through ultrastructural ob-
servations. Dormant conidia of B. cinerea are round to elliptical and rela-
tively smooth. The ultrastructure of healthy dormant conidia was similar to
6
! 9
!
Figure 3 Effect of pterostilbene on dormant conidia of Botrytis cinerea visible

under light microscopy. A and C, Untreated conidium. B and D, Pterostilbene
treatedconidia (5 X M). Observedunderdifferencialinterferencecontrast
(DIC) (Aand B). Observed under light microscope(C and D). Note the cytoplasmic
withdrawal from thesporewall.Someconidiapresent outflow of cytoplasm
(UWOWS).
that previously described (Buckley et al., 1966; Gull and Trinci, 1971).
Figure 6 shows that healthy conidia possess several nuclei; mitochondria
are usually round to ovoid with many cristae; numerous vacuoles, probably
containing glycogen, are located in the cytoplasm. Few round lipid bodies
are visible and the plasma membrane appears well defined with its three
layers-the external, dense to the electron, and the two internal layers,
more clear.
Applicationofpterostilbene at 5 X M ondormantconidiain-
duces very rapid modifications of their ultrastructure. Chronological obser-
vations, after 1-10 min following the addition of pterostilbene, show that
mitochondria and nuclear membrane are very rapidly (5 min) affected by
thephytoalexin.Thesemembranesbecomethickerandmanyelectron-
dense lipid bodies are visible in the cytoplasm. Soon after endoplasmic
reticula are disorganized and ribosomes tend to disappear. Before a com-
plete disorganization of cytoplasmic organelles and a disruption of the
cell membranes, large electron-dense lipid bodies are formed close to the
mitochondrial and nuclear membranes. The ultimate stage, where cyto-
Pont 322 and Pezet
80. 3
W.
“1
COMPOUND CONCENTRATION (M)
Figure 4 Effect of concentration of hydroxystilbenesof Table 1 on the inhibition

of germination of 6-day-old conidia of Botrytis cinerea (Pont and Pezet, 1990).
Inactive compound 6 (resveratrol)
is not represented.
Toxic Action of Vitaceae
Stilbenes 323
COMPOUNDCONCENTRATION (M)
Figure 5 Effect of concentration of hydroxystilbenes of Table 1 on the transfor-

mation of the aspect of 6-day-old conidiaBorryris
of cinerea(Pont and Pezet, 1990).
Compounds 1,2, 3,4, and 6 were inactive and are not represented.
plasmic materialseems to be dissolved and coagulated, correspondsto the

lethal transformation stage of conidia (Fig. 7). The degree of effect on
membranes appears to occur in the following order: mitochondrial mem-
brane 1 nuclearmembrane > endoplasmicreticulum > cytoplasmic
membrane.
The rapid modification of mitochondrial membranes corresponds to
the dramatic decrease of O2uptake of conidia in the presence of pterostil-
bene. Large electron-dense lipid bodies visible in mitochondrial
and nuclear
membranes are probably phospholipids released from disorganized mem-
branes. The presence of such lipid bodies in nuclear and mitochondrial
membranes of Sclerotinia fructigenahas been described by Najim and Tu-
rian (1979) and associated to phospholipids becauseof the high affinity of
osmic acidto this class of lipids.
111. PROPOSED MECHANISM OF ACTION

We have seen that hydroxystilbenes withdifferent substituents have differ-
ent fungitoxicity levels. The benzenic substituentsof these compounds can
be described bythe U values characterizingthe electron-attracting or donat-
324 Pezet and Pont
Figure 6 Healthy dormant conidia of Botrytis cinerea. N,nucleus; m, mitochon-

dria; v, vacuoles; ER, endoplasmic reticulum; W,wall; pm, plasma membrane; sb,
storage bodies; lb, lipid bodies.
ing power. A good linear relationship (r = 0.93) can be calculated between

the U value of 10 monosubstituted styryl groups of hydroxystilbenes (U,)
given by Veschambre et al.(1967) and theU value of benzenic substituents.
In consequenceU, values of disubstituted styryl groups of hydroxystilbenes
are deduced from these values.
The antifungal efficiency a ofcompound may be determined also by its
capacity to invade lipophilic membranesites. In this area, the relation be-
tween the biocidal characterof flavonoids, phytoalexins, and their lipophil-
ction of Vitaceae
Toxic Stilbenes 325
icity were related by Laksand Pruner (1989). The parameter of hydropho-

bicity, R,, of hydroxystilbenes was calculated from Rf obtained by thin
layer chromatography (TLC) on reverse phase CISutilizing the solvent sys-
tem MeOH-H,O (8 :2) as eluent (Rittich et al., 1980) according to the
formula R,,, = log (l/Rf - 1).Finally, the methoddescribed byBondi
(1964) permitted a calculation of the van der Waals volumes V,. These
chemical parameters ofthe tested hydroxystilbenesare given in Table3.
In view of the inhibition of germination and respiration, and transfor-
mation of conidia of B. cinerea, it can be concluded that the biological
activity of the tested hydroxystilbenes depended largelyon the U values. It
tended to increase withthe electron-attracting powerof the substituents. In
this view, the U, values were in better agreements with the apparent small
differences in the biological effects, for example, among the chloro com-
pounds 3-C1[7], 3,4,-C1,[8], and 3,5-C12[9]. The inhibition of germination
and the transformation of conidia for these compounds did not always
reach 100% beyond the “critical concentration” of 2.5 x IO-’ M. At high
concentrations, these lipophilic compounds crystallizedpartly in the agar
during the experiments. This fact resulted ainloss of activity at the appro-
priate concentration ofM.However, for the respiration, the difference
of activity between compounds with substituents 3,4-C1,[8] and 3,5-C12[9]
remained unexplained.
The weak activity of resveratrol[6], in spite of a U of +0.16, may be
the consequence of its hydrophilic character(R, = 0.60), which renders it
difficult to dissolve inthe membranes.
The real toxicity of pterostilbene is always higher than its theoretical
toxicity calculated from its electronic characteristics. On the other hand,
the important volume of the methoxy groups of this stilbene can perturb
the intermolecular bonds morethan the other substituents. Methoxy groups
give the possibility to form hydrogen bonding with their oxygen atoms.
This factor, including the possible formation of CTCs (Slifkin, 1980) and
hydrogen bonding with the phenolic-OH (Looms and Battaile, 1966), fa-
vors the contact of pterostilbene with membranous proteins. It is clearly
accepted that amino acids, particularlyaromatic amino acids, form CTCs
with electron acceptors (Slifkin, 1980). The delocalization of electrons in
the conjugated system representedby hydroxystilbenes enhancesthe polar-
ization of themoleculeswhen the substituents are electron-attracting
groups. Generally, the more electron-attractingare the stilbenes, the more
toxic are their effects. The affinity of hydroxystilbenes for proteins could
be the consequence of formation of CTCs between these compounds and
amino acids ofthe proteins.
Inhibition of respiration by pterostilbene is probably indirect.We have
described the effects of this stilbene on mitochondrial and nuclear mem-
326 Pezet and Pont
Figure 7 Aspect of dormant conidia of Botrytis cinerea induced by pterostilbene

(5 x M) afterdifferenttimes of incubation. A, 5 min; B, 15 min; C, 30 min;
D, 3 hr. (See legends in Figure6.)
C- l
r
Pont 328 and Pezet
Table 3 Aromatic Substituent Constants (U, usJ of Substituted Styryl Groups,

Parameter of Hydrophobicity (R,,,), and Van der Waals Volumes ( V w ) of
Substituted HydroxystilbenesPresented in Table 1
R
0 0.172 +0.07 114.16
-0.17 0.153 +0.23 125.31
-0.12 0.162 -0.09 142.85
+0.23 0.216 +0.23 123.64
+0.05 0.189 +0.09 142.85
+0.16 0.205 -0.60 125.20
+0.37 0.236 +0.27 123.64
+0.52 0.255 +0.43 133.12
+0.75 0.286 +OS5 133.12
branes and on endoplasmic reticulum. These effects on membranes are

identical to those reported by Lyr (1987)in connectionwith the mechanism
of action of aromatic hydrocarbon fungicides (AHFs). These compounds
possess electron-conjugated systems as stilbenes do. The U value was esti-
mated for one of them, dichloromethoxyphenol [DCMP], as +0.33. This
indicates that DCMP possesses an electron-attracting power as important
as 3-chloro-4’-hydroxystilbene[7].
Lyr (1987)proved that the primary toxic effectof AHFs is an induced
lipid peroxidation, especially inthe mitochondrial and nuclear membranes
and in the endoplasmic reticulum. The reason is the interaction of AHFs
with flavin enzymes, such as cytochrome C reductase, or other monooxy-
genases, located within the membrane system. AHF blocks the electron
transport from flavin to the substrate and induces, by generation of free
radicals, a pathological peroxidation of membranous phospholipids. Com-
paring AHF-resistantand sensitive strains of Mucor, Werner (1980)found
that the mitochondrial proteinsof resistant strains contained alow concen-
tration of tyrosine 0.2% of the total amino acids, compared to 2.9% in
sensitivestrains.Additionofantioxidant and anti-freeradicalsas a-
tocopherol acetateto a culture of Mucor, inhibited by AHF, nullifies lipid
peroxidation and counteracts the growth inhibition.
Amino acid tyrosine is an important site where electrons can enter
cytochrome to reach the iron-active center (Salemme et al., 1973). Aromatic
structure of tyrosine could favor CTCs (Slifkin, 1980) with all molecules
possessing conjugated system and electron-attracting capacities. Pterostil-
Toxic Action
Stilbenes
of Vitaceae 329
bene and other hydroxystilbenes having positiveU belong, as do AHFs, to

this category of compounds.
W. EPILOG
Many natural phenolics like hydroxystilbenes possessan important conju-
gated system. This electronic character, associated with the nature of the
substituents, determinethe U value relatingto the biological effects of hy-
droxystilbenes. We found among the high conjugated phenols not only
many phytoalexins but other active compounds, such as fungitoxins and
bactericidal compounds aswell (Pont and Pezet, 1991).
Pterostilbene is produced in a very low concentration by leaves and
immature berriesof grapevines comparingto resveratrol. Thelow fungitox-
icity of this last compound, even at high concentration, is explained by
its hydrophilic character incompatible with the lipophilicity of biological
membranes. Its direct role in the defense mechanism of Vitaceae against
fungal attacks is probablyvery poor. Pterostilbene, even at low concentra-
tions, displays, when associated to glycolic acid, a high toxicity toward
B. cinerea (Pezet and Pont, 1988). This organic acid, at a relatively high
concentration in immature berries, is toxic to B. cinerea conidia where it
provokes important damages to the organelle membranes. Microorganisms
can oxidize glycolic acidto glyoxylic acid (Corpeand Stone, 1960), proba-
bly as it has been described for plants, by glycolate oxidase witha concomi-
tant production of hydrogen peroxide (Metzler,1977).
The synergistic effects of pterostilbene and glycolic acid may increase
peroxide productionand consequent damagesto the biological membranes.
All these informations suggestthat the mechanism of action of hydroxy-
stilbenes involvesimportant lipid peroxidationby blocking flavin enzymes
such as cytochromec reductase and similar monooxygenases.
ACKNOWLEDGMENTS
We are grateful to Paul Parey, Editor for authorization to use previously
from the Journal ofPhytopathology, Vols. 129
published figures and tables
and 130.
REFERENCES
Bondi, A. (1964).Van der Waals volumesand radii. J. Phys. Chem. 158441451.
Buckley, P.M., Sjaholm, V. E., and Sommer, N. F. (1966). Electron microscopy
of Botrytis cinerea conidia. J. Bacteriol. 91(5):2037-2044.
Corpe, N. A., and Stone, R. W. (1960). Oxidation of glycolic acid by Penicillium
crysogenum. J. Bacteriol. 80452-456.
330 Pezet and Pont
Gorham, J., and Coughlan, S. J. (1980). Inhibition of photosynthesis by stilbe-

noids. Phytochemistry 19:2059-2064.
Gull, K., and Trinci, P. J. (1971). Fine structure of spore germination of Botrytis
cinerea. J. Gen. Microbiol. 68207-220.
Hargreaves, J. A. (1980).A possible mechanismfor the phytotoxicity ofthe phyto-
alexin phaesollin.Physiol. Plant Pathol.16351-357.
Hart, J. H. (1981). Role of phytostilbenes in decay and disease resistance. Ann.
King, F. E., Cotterill, C. B., Godson, D. H., Jurd, L., and King, T. J. (1953). The
chemistry of extractive hardwood. XIII. Colourless constituents of Pterocarpus
species. J. Chem. Soc. 3693-3697.
Laks, P. E., and Pruner, M. S. (1989). Flavonoid biocides: structure/activity rela-
tions of flavonoid phytoalexin analogues.Phytochemistry28( 1):87-91.
Loomis, W. D.,and Battaile, J. (1966).Plant phenolic compounds and the isolation
of plant enzymes. Phytochemistry 5:423-438.
Lyon, G. D.(1980). Evidence that the toxic effect of rishitin may be due to mem-
brane damage. J. Exp. Bot. 31(123):957-966.
Lyr, H.(1987).Mechanism of action of aromatic hydrocarbon fungicides. Modern
Selective Fungicides(H. Lyr, ed.), Longman, pp.75-89.
Massarani, E. (1957). Ulteriori ricerche sopra derivati stilbenici e difeniletanici. I1
Farmaco-Ed. Sc I2(5):380-386.
Metzler, D. E. (1977). Light in biology. Biochemistry: The Chemical Reactions of
Living Cells.Academic Press, New York, p. 787.
Najim, L., and Turian, G. (1979).Ultrastructure de l’hyphe vegetatif de Sclerotinia
fmctigena. Can. J. Bot. 521299-1313.
Nonomura, S., Kanagawa, H., and Makimoto, A. (1963). Chem. Abst. 604240~
(Abstr)
O’Neil, T. M., and Mansfield, J. W. (1982). Antifungal activity of hydroxyflavans
and other flavonoids. Trans. Br.Mycol. Soc. 79(2):229-237.
Pezet, R., and Pont, V. (1988).ActivitC antifongique dans les baies deVitis vinifera:
effets d’acides organiques et du ptCrostilbi?ne. Rev. Suisse Viti. Arbor. Hort.
20:303-309.
Pezet, R., and Pont, V. (1990).Ultrastructural observations of pterostilbene fungi-
toxicity in dormant conidia of Botrytis cinerea Pers. J. Phytopathol. 12919-
30.
Pont, V., and Pezet, R. (1990). Relation between the chemical structure and the
biological activityof hydroxystilbenes againstBotrytis cinerea.J. Phytopathol.
1301-8.
Pont, V., and Pezet, R. (1991). Un moyen de defense nature1 des vdgdtaux: les
phenols. Revue SuisseAgric 23(4):237-241.
Rittich, B., Polster, M., and Kralik, 0 . (1980). Reversed-phase high-performance
liquid chromatography. I. Relationship between RM values and Hansch’s pi
parameters for a seriesof phenols. J. Chromatogr. 19243-50.
Rupprich, N., Hildebrand, H., and Kindl, K. (1980).Substrate specificity in vivo
and in vitro in the formation of stilbenes. Biosynthesis of rhaponticin. Arch.
Biochem. Biophys.200:72-78.
ction Toxic Stilbenes
of Vitaceae 331
Salemme, F. R., Krant, J., and Kaman, M. D.(1973). Structural bases for function
in cytochrome c. J. Biol. Chem.248(22):7701-7716.
Slifkin, M. A. (1980). The significance of charge transfer interactions in biology.
Molecular Interactions,Vol. 2. H. Ratajezak and W. J. Orville-Thomas, eds.),
Wiley, London, pp. 271-304.
Smith, D. A. (1982). Toxicity of phytoalexins. Phytoalexins(J. A. Bailey and J. W.
Mansfield, eds.), Blackie, Glasgow, pp.218-252.
Spath, E.,and Schlager, J. (1940). Uber dieInhaltstoffe des roten Sandelholzes. 11.
8:881-884.
Mitteil.: Die Konstitution des Pterostilbens. Ber.
Veschambre, H., Dauphin, G., and Kerkomard, A. (1967). Transmission deseffets
Clectroniques dans les moltcules de trans-hydroxy4 stilbtnes et d’hydroxy-4
tolanes. Bull. Soc. Chim. 8:2846-2854.
Werner, P. (1980). Zum Wirkungsmechanismus des systemischen Fungicides Chlo-
roneb und zu den moglischen Ursachen erzeugter Resistenz gegenuber Mucor
mucedo L. Fres. Dissertation Martin-Luther-University, Halle-Wittemberg, 1-
142.
15
Inducible Compoundsin Phaseolus,
Vigna, and Dioscorea Species
S. A. Adesanya
Obafem' Awolowo University, lie-e-ife,Nigeria
M. F. Roberts
London University, London, England
1. INTRODUCTION
Phytoalexins are inducible compounds first observed by Muller and Borger
in 1940 and defined by Muller (1958) as antibiotics produced asa result of
biochemical interactions betweena host plant and a parasite. This defini-
tion hasbeenmodifiedbyseveralworkers (Harborne, 1977; Deverall,
1982). Essentially, they are chemical compounds that accumulate in the
living hypersensitive tissuesaround the infection sites (Bailey, 1973; Rahe,
1973; Mansfield et al., 1974). The cause of hypersensitive reaction may be
infection dueto microorganisms or to the toxic effects of their breakdown
products and metabolites (biotic inducers), or chemical and physical factors
(abiotic inducers) (Cruickshank and Perrin, 1968; Hadwiger and Schwo-
chau, 1971; Bailey et al., 1980). There are indications of different modes of
activity for these two groupsof inducers. Biotic inducersappear to activate
defense genes in the host, which leads to enzyme synthesisfor the produc-
tion of the compounds (Darvilland Albersheim, 1984; Dhawale et al.,1989;
Ellis et al.,1989; Preisig et al.,1991); the mode of action of abiotic inducers
is not clear since phytoalexins have been known to accumulate in tissues
treated with agents blocking genetic transcription and translation processes
(Yoshikawa, 1978). There are indications to suggest that on infection con-
stitutive compounds inthe host cause the parasite to release elicitors which
activate the phytoalexin productiongenes (Kiraly et al., 1972; Keen et al.,
1983; Ersek and Kiraly, 1986). The induced phytoalexins demonstrate toxic-
ity against nonpathogens as well as pathogens; pathogens, however, usually
have a way of avoiding their toxic effects by catabolism or the formation of
conjugates with glycosides (Taniand Mayama, 1982; Weltring et al., 1982;
Willeke and Barz, 1982; Denny and Van Etten, 1983; Tahara et al., 1987).
333
Roberts
334 and Adesanya
Plants which rapidly accumulate phytoalexinsare known to be more resis-

tant to infection. There are, however, doubts about their primary role in
disease resistance (Ersekand Kiraly, 1986).
The exploitation of phytoalexins relies on their inherent properties. The
genetic factor leads to the possibility of evolution of peculiar biosynthetic
pathways and compounds with restricted distribution that can be used in
chemotaxonomy. Interfamilial differences suchas the production of isofla-
vonoids in the Fabaceae (Leguminosae)and terpenoids in Solanaceae have
been observed together with intrageneric variations such as the production
of medicarpin (47), and maackiain (58) by some Trigonella species, and
medicarpin and vestitol (38) by others (Ingham and Harborne, 1976; Ing-
ham, 1981a; Ingham, 1990a; KuC, 1982). This type of study has been used
in clarifying the taxonomy of Trifolieae and recently Phaseolinae species
(Ingham and Harborne, 1976; Ingham, 1981b, 1990a,b) and intrageneric
variations inthe genus Glycine (Keen, 1986).
The major biological property attributed to phytoalexins is antifungal
and antimicrobial againsta range of organisms. It is noteworthy that such
chemotoxicitiesprincipallyinvolvebasicinterferencewithphysiological
processes and therefore possible activity againstother life forms;for exam-
ple, in a test of several isoflavonoids against zoopathogens it was found
that phaseollinisoflavan(43) a phytoalexin in several Phaseolus species, was
very active (Gordon etal., 1980; Smith, 1982). Other biological activities of
isoflavonoid phytoalexins have beenwell documented ina review by Smith
and Banks (1986).
It has been establishedthat previous inoculationof plants with a non-
pathogenprotects the plantagainstsubsequentinfection by pathogens
(Muller and Borger, 1940; Bell and Pressley, 1969; Mansfield, 1982). This
cross-protection property relies on the ability of the plant to accumulate
high levels of its phytoalexins more rapidly on infection and is the key
property for diseaseresistance(Mansfield,1982).Thus,monitoring the
levels of phytoalexins in induced plants can assist the selection
in and breed-
ing of resistantstrains for breeding.
In the present work, some aspects of elicited isoflavonoids inPhaseolus
and Vigna species and dihydrostilbenes in Dioscorea specieswillbede-
scribed. These two groups of compounds have common biosynthetic origin
in that both groups are formed from either cinnamoyl-CoApcoumaroyl- or
CoA and malonyl-CoA.
A. Phaseolus-Vigna Species
Thesespeciesbelong to the family Fabaceae (Leguminoseae, subfamily
Papilionoideae), tribe Phaseolae, and subtribe Phaseolinae. They constitute
the largest group andare the most important source of food inthe subtribe
Phaseolus, Vigna, and Dioscorea Species 33s
(Kay, 1979; Duke, 1981). The two dominant genera,Vigna (80 species) and
Phaseolus (28 species), are difficult to separate taxonomically. Many spe-
cies in one genus often have synonyms in the other, such as P . aureus
(Roxb.) syn. V, radiata (L.) R. Wilczek. This is complicated by their wide-
spread distribution in temperateand tropical zoneswhich might have led to
the evolution of regional characteristics and subsequent classificationinto
Asiatic and tropical groups (Maekawa, 1955; Verdecourt, 1970; Marechal
et al., 1978). This confusion provided a reason for the use of chemical
characteristics for a chemotaxonomic solution. Earlier workon the protein
and hybridization properties of Phaseolus species indicated intergeneric
differences inthe Phaseolus-Vigna complex that was largely in conformity
with the classification by Marechal et al. (1978) and also shows that P .
lunatus is possibly less evolved than other members of the genus (Kloz,
1962; Klozand Klozova, 1974).
Production of phytoalexinsinvolves the synthesisofindividualen-
zymes for the several steps in their biosynthesis. Various intermediates as
well as final products will accumulate and differences in these factors may
permit the separation of close species chemotaxonomically.The inducible
constituents of only a few species in the Phaseolus-Vigna complex have
beeninvestigated.These are P . vulgaris (20compounds)(Woodward,
1979a, 1979b, 1980; Ingham, 1982; Biggs et al., 1983), P . lunatus (2), P .
aureus (2), V. unguiculata (7) (Ingham, 1982), P . mungo (none) (Smith,
1971), and Voandozeia subterranea (syn. Vigna subterranea) (1) (Ingham,
1982). Our work has extended the investigation and data in four species, P .
lunatus, P . aureus, P . mungo, and P . coccineus (Adesanya et al., 1984a,
1985; O'Neill et al., 1983, 1986), while recent work by Ingham (1990b) has
provided additional data for V. subterranea, V. angularis, and V. umbel-
lata (Table 1).The large number of known and novel isoflavonoids isolated
(Fig. 1) providedan opportunity for a detailed analysis of their biosynthetic
pathways and their comparative toxicities toward the fungi Aspergilus niger
and Cladosporium cucumerinum.
B. Dioscorea Species
The major genus of the Dioscoreaceae isDioscorea, which is divided arbi-
trarily into several sections with several species spread overthe
allsections.
Other minor genera include Tamus and Rajunia (Coursey, 1967). Most of
the species are typified by twining stems and large storage tubers which in
some are reduced to rhizomes and in others grow on the aerial part as
bulbils. The tuberous species have a high carbohydrate content and are
cultivated in the tropics as food crops (Coursey, 1967). The nonpoisonous
species such as D..rotundata Poir., D. alata L., and D. cayanensis L. are
widely consumed;the poisonous ones suchas D. dumentorum Kunth. Pax.
336 and Adesanya Roberts
and D . hispida L. are only safeafter the removalof the poisonous alkaloid
dioscorine.
These tubers deteriorate rapidly in storage dueto microbial infection,
with an estimated loss of over half of the yearly harvests in Nigeria (Cour-
sey, 1967). Thus, concerted efforts are being made to find ways of increas-
ing their shelf life (dormancy period) and also of preventing microbial
attack.
The peel of D . rotundata has been shown to contain the constitutive
antifungal phenanthrenesbatatasin I (77), and hircinol (81) (Coxon et al.,
1982), and that of D . decipiens, 2,7-dihydroxy-l,3,5-trimethoxy-9,10-
dihyrophenanthrene (80) (Sunderet al., 1978). Constitutivecompounds
responsible for dormancy in the bulbils of D . opposita (D. batatas) have
been isolated and characterized as the phenanthrene batatasin I and the
dihydrostilbenes (bibenzyls)batatasin 11, I11 (72), IV (71), and V (73) (Fig.
1) (Hashimoto et al.,1972;Hashimoto andTajima, 1978). The distribution
of these substances in Dioscorea species varies, with some species having
none (Ireland et al.,1981). Chemical investigations in breeding experiments
have linked resistance to anthracnose disease in D . alata to the level of
phenolics (Alozie et al.,1987) indicating a possible role for inducible com-
pounds in the resistance to infection. A subsequent study of tissues of D .
batatas inoculated with bacteria led to the isolation of dihydropinosylvin
(68) as the major compound, together with batatasin I, batatasin IV, 3-
hydroxy-5-methoxybibenzyl(69), 6,7dihydroxy-2,4-dimethoxy-9,1Oaihydro-
phenanthrene (78), and 2,7~dihydroxy-4,6dimethoxy-9,l0-dihydrophenan-
threne (79) (Fig. 1) (Takasugi et al.,1987). This study has been extended to
the inducible compounds in D . rotundata, D . alata, D . dumentorum, D.
bulbifera, and D . mangenotiana, with the aim of evaluating their potential
in prolonging dormancy, disease resistance, crop protection, and chemotax-
onomy inthe various sections.
II. METHODOLOGY
Plant Materials and Microorganisms
Seedsof Phaseolus vulgaris L. var. Prince, P . aureus Roxb. (synonym,
Vigna radiata (L.)R. Wilczek), P . mungo L. (synonym V. mungo (L.)
Hepper.), and P. lunatus L. were purchased from Thompson and Morgan
Ltd., London and those of P . coccineus L., var. Scarlet Emperor, were
obtained from Northrop King Seeds (Minneapolis, MN). Seeds Sorghumof
bicolor (L.) Mench. and tubers of Dioscorea rotundata Poir., D . alata L.,
D . dumentorum (Kunth.) Pax. were broughtat the market in Ile-ife,Nige-
ria. Wild bulbils ofD . bulbifera L. and tubers of D . mangenotiana Meige.
were collectedat various sitesaround Ile-ife.
Phaseolus, Vigna, and Dioscorea Species 337
ISOFLAVONES
P3
1. R, = R, = R,= R, = R, = H Diadzein
2. & = R, = R, = R, = H. R, = OH Genistein
3. R, = R, = R, = = H, R, = OCH, lsoprunetin
4.R,=R2=R,=R5=H.R4=OH 2’-hydroxydiadzein
5. R, = R, = R, = H. R, = R, = OH 2”hydroxygenistein
6. R, = R, = R, = H, R, = OCH,, R, = OH 2’methoxygenistein
7. R, = R, = R, = H, R, = OCH,, R, = OH 2”hydroxyisoprunetin
8. R, = R, = H, R, = R, = R, = OH 8.2’dihydroxygenistein
9. R, = R, = R, = H.R, = R, = OH 8,2’dihydroxydiadzein
10. & = R, = H, R, = R, = OH, R, = CH,CH=C(CH,), 2.3dehydrokievitone
11. R2 = R, = H, R, = R, = OH, R, = CH,CH=C(CH,)CH,OH 2.3dehydrokievitol
12. R2 = R, = H. R, = R, = OH, R, = CH,CH=C(CH,), Phaseoluteone
13. R, = R, = H. R, = R, = OH, R, = CH,CH=C(CH,), Luteone
ISOFLAVONONES
P4
2”hydroxydihydrodiadzein
8.2’dihydroxy
dihydrodiadzein
Dalbergioidin
5.2”dimethoxydalbergioidin
Isofemirin
Sdeoxykievitone
5deoxykievitone hydrate
Figure 1 Structure of known and novel isoflavonoids,includingisoflavones,

isoflavonones, cyclo-derivatives of isoflavones
and isoflavonones, isoflavans,ptero-
carpans, furano-pterocarpans, cournestans, stilbenes, dihydrostilbenes, and dihy-
drophenanthrenes.
& = CHZ-CHZ-C(CH,)zOH
21. R,= R, = & = H, R, = & = R, = OH, 5-deoxykievitol
& = CHz-CH=C(CH,)CH,OH
22. = & = H, R,= R, = R,= R, = OH, Kievitone
& = CH,-CH=C(CH,),
23. R, = & = H, R,= R, = R, = R, = OH, Kievitone hydrate.
& = CH,-CH,-C(CH,),OH
24. R, = & = H, R,= R, = R, = R, = OH, Kievitol
R4 = CH,-CH=C(CH,)CH,OH
25.R, = & = H, R,= R, = R, = OH, R, = OCH,. 4“rnethoxykievitone
R, = CH,-CH=C(CH,),
26. R, = = H. R,= OH,R,= R, = R, = OCH, Trirnethoxykievitone
& = CHZ-CH=C(CH,),
27. R, = & = H, R,= R, = R, = R, = OCH,, Tetramethoxykievitone
& = CH&H=C(CHJ,
28. R, = H, R,= R, = R, = R, = OH, 3”(y,ydimethylallyl)
& = & = CH,-CH=C(CH,), kievitone
29.R, = H, R, = R, = R, = OH, R, = OCH, Isosophoranone
R2 = & = CHZ-CH=C(CH,),
30.R, = R, = & = H, R,= R, = OH, R, = & = OCH, Cajanol
31.R2 = R4= H. R,= R, = R, = OH, R, = OCH,. Sophoraisoflavanone A
& = CHZ-CH=C(CHJ,
CYCLO-DERIVATIVES OF ISOFLAVONES AND ISOFLAVONONES
32.Cyclo-2,3-dehydrokeivitonehydrate
Figure 1 Continued.
Phaseolus, Vigna,and Dioscorea Species 339
Cyclokievitone
Tritnethoxykievitone
1",2"dehydmyclokievitone
Cyclokievitone hydrate
ISOFLAVANS
Demethylvestitol
Vestitol
Isovestitol
Sativan
Laxifloran
2'-methoxyphaseollidin
isoflavan
43. R, = OH Phaseollin isoflavan

44. R, = OCH, 2'-methoxyphaseollin
isoflavan
340 Adesanya and Roberts
PTEROCARPANS
P3
45. R, = R, = R4 = R, = R, = H, R, = R, = OH Demethylmedicarpin
46. R, = R, = R4= k = R, = H, R, = OH, R, = OCH, Isomedicarpin
47. R, = R, = R4 = = R, = H, R, = OH, R, = OCH, Medicarpin
48. R, = R, = R4 = R, = H, R, = R, = & = OH Glycinol
49. R, = R, = & = R, = H, R, = R, = OH, Phaseollidin
h = CH,CH=C(CH,),
50. R, = R, = & = H, R, = R, = OH, R, = OCH,, l-methoxyphaseollidin
= CH,CH=C(CH,),
51. R, = & = R, = H, R, = R, = OH, 2(y,y-ilimethylallyl)
R, = = CH,CH=C(CH,), phaseollidin
52. R, = = R, = H, R, = R, = OH, 4(y,y-dimethylallyl)
R, = R4 = CH,CH=C(CH,), phaseollidin
53. R, = R, = H, R, = R, = = OH, 2,lO-(dimethylallyl)
R, = R, = CH,CH=C(CH,), glycinol
54. R, = R, = R, = H, R, = & = OH, R, = OCH,, Cristacarpin
R4 = CH,CH=C(CH,),
55. R, = R, = = R, = H, R, = R, = OH, Dolichin A
Dolichin B
Figure 1 Continued.
Phaseoh, Vigna, and Diosmrea Species 341
57. Phaseollin 58. Maackian
FTJRANO-PTEROCARPANS
59. Neodunol
COUMESTANS
60.R, = R2 = R, = R, = H Coumesterol
61. R, = R, = = H. R, = OH Aureol
62. R, = R, = R, = H. R, = CH2CH=C(CH,)2 Psoralidin
63. R, = R, = R, = H. R, = CH,CH=C(CH,), Phaseol
64. R, = R, = R, = H. R, = CH,CH=C(CH,), Isosojagol
342 Adesanya andRoberts
65. Vignafuran
STILBENES
66. R, = H Pinosylvin
67. R, = OH Trans-resveratrol
DIHYDROSTILBENES
68. R, = R, = R, = & = H,R, = R, = OH Dihydropinosylvin

69. R, = R, = R, = & = H,R, = OH, R, = OCH, 3-hydroxy-5-rnethoxy
dihydrostilbene
70. Rz = R, = & = H,R, = R, = R, = OH Demethylbatatasin IV
71. R2 = R, = & = H, R, = R4 = OH,R3 = QCH3 Batatasin IV
72. R, = R4 = & = H,R, = R, = OH,R, = OCH, Batatasin Ill
73. R, = & = H, R, = OH, R, = R, = R, = OCH, Batatasin V
74. R, = R, = & = H. R, = OH,R, = R, = OCH, 4-hydroxy-3,5dimethoxy-
dihydrostilbene
75. R5= & = H, R, = R, = OH,R, = RI = OCH, 2 & d i h ~ x y - 3 ~ m y
bibenzyl
76. R, = R, = R, = H. R, = R, = R+,= OH Dihydroresveratrol
Figure 1 Continued.
DIHYDROPHENANTHRENES
R2
3
‘ ‘qR5 R6
77. R, = R, = R, = R, = H,& = OH,R, = R, = R7 = OCH, Batatasin I

78. R, = R, = R, = R, = H,& = R, = OH,R, = R, = OCH,
79. R, = R, = R, = R, = H,R, = R, = OH,R, = & = OCH,
80. R, = & = R,, = H,R, = R7 = OH,R, = R, = R, = OCH,
81. R, = R, = & = R, = R,, = H,R, = & = OH,R, = OCH,
Cultures of bacteria, Bacillus cereus, Staphylococcus aureus, Pseudo-

monas aeruginosa, and Escherichia coli were obtained from the Depart-
ment of Microbiology, Obafemi Awolowo University, Ile-ife. The fungi
Cladosporium cucumerinum Ell. and Arth., and Aspergillus niger Van
Teigh. were obtained fromthe Commonwealth MycologicalInstitute, Kew,
London,while C. cladosporoides, A . niger, A . flavus (clinicalisolate),
Tricophyton mentagyrophytes (clinical isolate), and Botryodiplodia theo-
bromae Pat. were obtained from the Department of Microbiology, Oba-
femi Awolowo University, Ile-ife.
B. Induction, Extraction, Detection, and Isolation of Phytoalexins

Sterilized seeds ofthe Phaseolus species were germinated in running water
and grown for about 11 days in sterilized vermiculite before induction by
immersing theirroots in aqueous3 mM solution of CuC1, for 19 hr (control
seedlings in water alone). They were further grown for 4 days before analy-
sis (Adesanya, 1984).
Peeled tubers or bulbils were cut into small discs and washed with
sterile water before dipping in mycelia suspensionsof B. theobromae or 3
mM solution of HgC1, for 5 min (control specimens in water) before incuba-
tion for 4 days (Fagboun etal., 1987).
Induced and control specimenwereextracted into EtOAc, and the
extracts subjected to thin layer chromotography (TLC) bioassay using C.
cucumerinum (Phaseolus species) or C. cladosporoides (Dioscoreaspecies).
The anitfungal zones appearingas white on a green backgroundwas corre-
lated with duplicate TLC plates sprayed with vanillin/H,SO., or Fast Blue B
salt solution (Adesanya,1984).
For isolation of compounds in induced tissues, large-scale EtOAcex-
tracts of induced tissues werefractionated by column chromatographyand

purified by preparative TLC using various solvents. The pure compounds,
were characterized using their ultraviolet (UV), infrared (IR), mass spec-
trometry (MS), and nuclear magnetic resonance (NMR) spectroscopicdata
(Adesanya, 1984; Fagboun et al.,1987; Takasugi et al., 1987).
C. Assay of Induced Compounds and Enzymes

Phaseollin (57) and kievitone (22) in P. vulgaris, phaseollidin (49) and
kievitone inP. aureus, and kievitone inP. mungo were isolated by prepara-
tiveTLC and quantified by high-performanceliquidchromatography
(HPLC), while dihydropinosylvin and demethylbatatasinIV (70) and bata-
tasin IV were similarly isolated and quantified by UV spectroscopy. Sam-
pling was done over 120 hr in induced and control specimens, respectively,
and amounts produced extrapolated from standard curves obtained with
authentic compounds (Adesanya, 1984; Cline et al., 1989). The enzymes
phenylalanine ammonia-lyase (PAL), tyrosine ammonia-lyase (TAL), and
polyphenol oxidase (PPO) in D . alata were assayed simultaneously in in-
duced and control tubers as were the phytoalexins. The assays were done
using standard procedures (Cline et al.,1989).
D. AntimicrobialAssay
Paper discs were loaded with various graded concentrations of the dihy-
drostilbenes and placed on bacteria-seeded agar plates. Zones of inhibition
were measured for the active compounds after a 24-hr incubation period
(Takasugi et al. , 1989).
E. AntifungalAssay
Graded concentrations of the isoflavonoids were dispensed into multiwell
assay trays. The volume in each well was made up to 50 p1 by the addition
of ethanol and then 950 p1 of a spore suspension in Czapek Dox liquid
nutrient added. Controls without isoflavonoidsand test preparations were
incubated inthe dark for 3 days. MICwas taken as a range betweenthe last
well with mycelia development and the next concentration without growth
(Adesanya et al.,1986).
The dihydrostilbenes were assayed using agar plates with holes in a
circle containing various graded concentrations of the compounds and wa-
ter only inthe control. A mycelial plug was placedin the middle and zones
of limitation of mycelial growth measuredfrom the edge of each well.The
MIC was taken as a range between the well showing the least zone of
inhibition and the next concentration without a zone of inhibition (Adesa-
nya et al.,1989).
F. Seed Germination Tests

Twentysurface-sterilizedseeds of S. bicolor were put into sterilepetri
dishes. Fifty-milliliter aqueous solutions of dihydrostilbenes containing a
drop of Tween 20 were sterilized by membrane filtration and 5-m1 portions
of each concentration dispensed into separate Petri dishes. Controls con-
tained water and Tween 20 only. Samples and controls were incubated in
the dark and the number of germinated seeds counted at 24-hr intervals.
Experiments were performed in quadruplicate and the average percent ger-
mination calculated (Clineet al., 1989).
G. Seedling Root Elongation Tests
Twenty seedlings of S. bicolor grown in sterile distilled water and having
about 2 cm root length were distributed into sterile Petri dishes and 5-m1
test solutions prepared as in seed germination tests added to the Petri
dishes. The increase in root length of all seedlings in each dish was mea-
sured and the average readings from four dishes per concentration was
calculated (Cline etal., 1989).
111. RESULTS AND DISCUSSION

A. Induction of Phytoalexins
Fabaceae: Phaseolae: Subtribe Phaseolinae
Earlier reports on P . vulgaris had shownthat the major phytoalexins, kievi-
tone and phaseollin, accumulate after induction of the enzyme PAL in
tissues elicited with either biotic
or abiotic elicitors, with kievitone reaching
higher levels than phaseollin (Bailey, 1982; Adesanya, 1984). These com-
pounds similarly accumulated P. in aureus, while only kievitone was elicited
in P.mungo seedlings treated with CuCl,solution, contrary to the report of
Smith (1971) that P. mungo seed pods contain no phytoalexin diffusate
(Adesanya, 1984). These two compounds have been demonstratedto accu-
mulate after enzyme induction byan abiotic elicitor. Onlythe constituents
of infected tissues ofother species of the Phaseolinae subfamily have been
analyzed (Ingham, 1982); these analyses have to ledthe isolation of biosyn-
thetically related compounds in different plants.
The compounds isolated from large-scale CuCl, treated seedlings of P .
vulgaris, P. coccineus, P. aureus, P. lunatus, and P . mungo (Adesanya et
al., 1984a, 1985; O'Neill et al., 1983, 1986) together with those reported in
P.vulgaris and V. Unguiculata (Ingham 1982) and V. subterranea, V. angu-
laris, and V. umbellata (Ingham, 1990b) are presented in Table 1. Other
species ofthe Phaseolinae investigatedso far are Macroptilium atropurpur-
eum and M . marti whoseleafdiffusates contain genistein (2), 2'-
l $ l I
I I I I
+
1 + 1 I + I f + l + l = I I l + I I I I I
+
I I I I + + I I I I + f I I I + + I
+
I I +
+
I I I I I I I I I + + I l l 1 I I I I I
I I I I I I I I I + + I l l 1 I I I I I
I I I I I I I I I I I I I I I I I I I I
VI
346
+
I I I I I I I I I I I l l + l l l l l I I I
+
B
2
v
1 1 + 1 1 1 1 1 1 1 + 1 1 + 1 1 1 1 1 1 1 1
+ 1 + 1 1 1 1 1 1 1 + 1 1 + 1 1 1 1 1 1 1 1
+ 1 + 1 1 1 1 1 1 1 + 1 1 + 1 1 1 1 1 1 1 1
A I
347
hydroxygenistein(5),dalbergioidin (16), kievitone,demethylmedicarpin

(45), phaseollin, and phaseollidin. M . bracteatum and M . lathyroides con-
tain 2, 5, 16, and 22 and Macrotyloma axillare leaves accumulate 5, 16,
57, and 22 (Ingham, 1982, 1990b). Dolichos biforus produces genistein,
2 '-hydroxygenistein, dalbergioidin, isoferrierin (18), dolichin A (55)
and B
(59, Lablab niger produces 2' -hydroxygenistein, dalbergioidin, kievitone,
phaseollidin (49), demethylvestitol(37), isovestitol(39), laxifloran (41),
and
vignafuran (65),while Psophocarpustetragonolobus gives phaseollidin,
isomedicarpin (47), 1-methoxylphaseollidin (50), demethylmedicarpin,and
cristacarpin (54) (Ingham, 1982). Neorautanea mitis produces 45, neodunol
59, 49, and 37, while Oxyrhynchus volubilis has only glycinol (48). Steno-
phylis angustifolia and S. stenocarpa have the same compounds as M .
atropurpureum except for 2 and 57. Stophostyles helvola and Dipogon
lignosus produce the samephytoalexinsas M . atropurpureum withthe
addition of demethylvestitol(37) (Ingham,1982, 1990b). Kievitone appears
to be the most common inducible compound inthis subtribe, produced in
the greatest quantity, followed by phaseollidin, although their concentra-
tions differ in different species. Exceptions have been found in P. cocci-
neus, which produces larger amounts of phaseollin than kievitone (Ade-
sanya, 1984),D . lignosus and S. angustifoliawith higheramounts of phase-
ollidin, and V. subterranea, which is devoid of kievitone (Ingham, 1990b).
Dioscoreaceae
D . opposita (D. batatas),the first plant investigated in this family, produces
the phenanthrenes batatasin I (77), 6,7-dihydroxy-2,4-dimethoxy-9,10-
dihydrophenanthrene (78), 2,7-dihydroxy-4,6-dimethoxy-9,lO-dihydrophe-
nanthrene (79), and dihydrostilbenes, dihydropinosylvin (68), 3-hydroxy-5-
methoxybibenzyl (69), and batatasin IV (71) in response to the bacterium
Pseudomonas cichorri (Takasugi et al., 1987). Other compounds isolated in
our investigation ofother Dioscorea species infected with the fungi Botryo-
diplodia theobromaeare presented in Table 2 (Fagboun al., et 1987; Adesa-
nya et al., 1989; 1989; Cline et al., 1989; Kaganda and Adesanya, 1990).
Dihydropinosylvin (68) appears to be the principal induced compound in
the section Enantiophylum. On induction of one of the species, D . alata,
by the fungi,dihyropinosylvin,demethylbatatasinIV, and batatasin IV
accumulate in the infected yams (Fig. 1). By contrast, in tubers treated
withHgCl,, only dihydropinosylvin shows appreciable increase (Fig. 2).
Although the lower amounts of 68 accumulatein those induced withHgCI,,
in both cases increase inPAL activity precedesthe onset of dihydrostilbene
formation with less increase with the HgCl, treatment. TAL activity was
somewhat lower inboth treatments than in the controls. These results sug-
349
c
3
N
c
W
0 2b 4b 6b 8b 100
lime (hr)
Figure 2 Enzyme activities in relationto the accumulation of (a) dihydrostilbenes

1-3 in B. theobromae-induced tubers. M-., dihydropinosylvin (1); 0 - ,demethyl-
batatasin IV (2); A-A,batatasin IV (3); (b) 0-0, phenylalanine ammonia-lyase
(PAL); 0-0, tyrosine ammonia-lyase (TAL).
gest the synthesis ofthe dihydrostilbenesis viaa route involving phenylala-

nineasdemonstrated for the stilbenes(Rudloff and Jorgensen, 1963;
Schoppner and Kindl, 1979; Fritzemeier and Kindl, 1981; Stoessl, 1982).
Since these compounds are antimicrobial, they can be regarded as phyto-
alexins exceptbatatasin IV which has beenfound, without deliberate induc-
tion, in D . batatas (Hashimoto et al., 1972; Hashimoto and Tajima, 1978).
The lower levels of dihydropinosylvin resulting from HgClz elicitation
may also relate directlyto enzyme inhibition and rapid onset of cell death
(Hargreaves, 1979). However, the results obtained did not appear to sup-
port the suggestion that abiotic inducers do not affect phytoalexin synthesis
but only inhibit their degradation.P P 0 activity (Fig. 2) did not show any
correlation with the progressive browning which was observed in experi-
ments with controlsor in tuberstreated with HgC1,. Its transient increase in
HgClz induced tuberssupports the proposal for an altered rolefor PP0 in
stressed systems where its availability may assist in phenolic depositionin
cell walls as a mechanism to limit infections (Fearmanand Diamond, 1967;
Yoshikawa, 1978; Mayer and Hard, 1979; Cline et al., 1987).
B. Characterization of Isolated Compounds

The Phaseolus species investigated produced 42 isoflavonoids of the follow-
ing types: isoflavones, isoflavanones, isoflavans, pterocarpans, and coume-
stans (Table 1). In the Dioscorea species, six dihydrostilbenes were isolated
(Table 2). The identityof these compoundswas established from their UV,
IR, MS, and ‘H and I3CNMR spectral data, together with their color
reactions on TLC plates and by comparison with literature values (Adesa-
nya, 1984; Adesanya et al., 1984a,b, 1985, 1986; O’Neill, 1983; O’Neill et
al., 1983,1984,1986). These data conformed withthe established analytical
profile for flavonoids (Mabry et al., 1970) except ina few cases.The collec-
tion of a large number of isoflavonoids provided an opportunity for a
comparative study of their spectral characteristics in line with an earlier
study of spectral properties of flavonoids (Mabryet al., 1970). Diagnostic
differences were observed the in UV spectra of the isoflavones and isoflava-
nones, examples of whichare given in Table3.
C. UV Spectroscopy of lsoflavonoids
The UV spectrum of most isoflavones in either
EtOH or MeOH have princi-
pal absorption maxima between 245 and 270 nm as established in earlier
reports (Mabryetal., 1970; Ingham, 1982). However,isoflavoneswith
three or more oxygen atoms on the A ring absorb at 260-270 nm, while
those with fewer oxygen atoms have their maxima at 245-255 nm. Methyla-
tion of the 5-OH group or its absence causesa hypochromic shift of 5-10
xx x
X X
352
nm due to the absence of the normal hydrogen bonding betweenthe 5-OH

and carbonyl groups (Mabry et al., 1970). In the isoflavones the principal
maxima occurs at 285-300 nm with the 5-hydroxy types and 277-287 nm
with the 5-deoxy types. The formation of the benzofuran ring in pterocar-
pans producesan extended molecular conjugation resulting in characteristic
twin peaksor a peak with a shoulder at 280-290 m.
Mabry et al. (1970)and Voirin (1983) reported subtle differences the in
effect of shift reagents on the UV characteristics of flavones such that it
was possible to use these characteristicsto differentiatethe different classes
of flavonoids. A later study shows that isoflavones may be classified in a
similar manner by someshift reagents. NaOMe ionizes all phenolic groups
on the flavonoid and isoflavonoid nucleus leadingto a bathochromic shift
of 10-20 nm from the principal absorption maximum for isoflavones and
35-60 nm for isoflavanones and pterocarpans. AlCl, reagent forms a com-
plex between the 5-OH and the carbonyl groups that results in a bathoch-
romic shift for 5-hydroxyflavonoids and expectedly none for the 5-deoxy
types. Similarly, shifts of6-15nm and 20-30nmwere observed for 5-
hydroxyisoflavones and 5-hydroxyisoflavanones,respectively(Table 3)
(Mabry et al., 1970; Adesanya, 1984). Although no shift was observed for
5-deoxy-2’ -deoxyisoflavonoid derivatives, differences were observed when
free 2”hydroxy substituentsare present. Table 3 shows that in isoflavones,
but not in isoflavanones, when the 5-OH is absent or derivatized, a free
2 ’-OH does form a weak complex withthe carbonyl causinga shift of 4-7
nm. Thusan observed shift with AlCl, for isoflavones could indicate either
free 5-OH or 2 ’-OH in compounds wherethe 5-OH is absent incontrast to
the report for flavones (Mabry etal., 1970; Voirin, 1983).
Although Mabry et al. (1470) did not report any data for flavonoids
with isopentenyl side chains, the presence of such side chains at various
positions on the isoflavone molecule did not affect the AlC1,-induced shifts.
In the isoflavanones, no shifts were observed when the side chain is at
C-6(Table 3).DelleMonache et al.(1977)reported no AlCl, shift for
isosophoranone (29), a 5-hydroxy-6,3’-diprenylated isoflavanone. Similar
observations were reported for acetophenones prenylated at various posi-
tions on the benzene ring (de Lima et al., 1975). Thus it appears that the
presence of a bulky group at C-6 of isoflavanones ortho to the chelating hy-
droxyl group on a benzene ring prevents AlCI, complexation and consequently
results in no shift of the absorption maxima.
D. BiosyntheticConsiderations
Isoflavonoids
Enzymic studiesand radiolabeling experimentation have helped to elucidate
biosynthetic pathways and to establish a relationship between protein syn-
thesis and phytoalexin formation. Important in this sequence of events is

the overall increase in the rate of formation of RNA in plants following
infection. Results suggestde novo mRNA synthesis is required for aneffec-
tive defense response which enables the plant to resist infection (Bell et al.,
1984; Ryder etal., 1984; Schmelzer et al., 1985).
The biosynthetic pathways for flavonoids and isoflavonoids are well
established (Dewick,1992 and references therein)and many of the required
enzymes have been isolated and studied in relation to phytoalexin elicita-
tion. The modification of phenylalanine to cinnamoyl-CoA requires two
important enzymes,(PAL) and cinnamoyl-CoAligase, and increasesin
both these enzymes are observed in response to elicitation (Hahlbrock et
al., 1980; Whitehead et al., 1982; Dixon etal., 1983; Robbinset al., 1985).
The formation of isoflavonoids has been comprehensively studied by
Grisebach and collaborators, who demonstrated the biosynthesis of the
isoflavone skeletonfrom precursors of the general flavonoid pathway. The
key steps in isoflavonoid biosynthesis involve a l,2-aryl shift of the B ring
which takes placeafter the formation of the C-l5 chalcone intermediate,by
chalcone synthase and an NADPH reductase, and its subsequent conver-
sions to the isomericflavanone by chalconeisomerase(Hagmannand
Grisebach, 1984; Hakim and Dewick,1984;Koch and Grisebach, 1986;
Bless and Barz, 1988; Dewick, 1988; Hakamatsuka et al., 1991). All nine
enzymes requiredfor the biosynthesis of prenylated pterocarpans, i.e., gly-
ceollins, have now been identified in elicited soybean cellsand, as a result
of this and other studies, the pathways to the various groups of isoflavo-
noids are well established (Grisebach et al., 1989;Welle and Grisebach,
1991).
The differentiation of the 5-hydroxy- and 5-deoxyisoflavonoidshas
also been suggestedto occur at the chalcone stage (Woodward,1980). This
led to the proposal of different pathways for the 5-hydroxy- and 5-
deoxyisoflavonoidsisolated from P. vulgaris (Woodward,1979b, 1980).
These pathways can be extendedfor the compounds isolatedfrom various
Phaseolus species.
The apparent absence of some of the proposed intermediates in P.
a u r a s , V. Unguiculata (Table l), and other Phaseolae species is dueto the
analysis of either small amounts of infected tissues, compounds at zones of
antifungal activities in TLC bioassays, or diffusates from induced tissues
(Smith and Ingham, 1980; Ingham, 1982), which invariably leads to small
number of compounds. However the absence of phaseollin, amajor phyto-
alexin in the proposed 5-deoxy pathway inP. a u r a s , is most likely due to
the plant’s inability to synthesize this compound. Such deficiencywas also
observed for P. lunatus and P. mungo (Table 1). On the other hand, the
absenceofsomekey intermediates like 2’-hydroxydiadzein (4) and 2’-
Vigna,
Phaseolus, and Dioscorea Species 355
hydroxydihydrodiadzein (14) of the 5-deoxy pathway in P. coccineus and

P . lunatus, respectively, and dalbergioidin (16) of the 5-hydroxy pathway
in the two plants may be due to rapid turnover of these compounds during
biosynthesis of thefinal products.
In the 5-deoxy pathway, the primary products from the chalcone ap-
pear to bediadzein (l), 2”hydroxydiadzein (4), and2’-hydroxydihydro-
diadzein (14) progressively to give allthe related isoflavonoids except vigna-
furan (65), which may arise from the chalcone through a different route
(Martin and Dewick,1979;Stoessl,1982).The8-hydroxylatedproducts,
8,2’-dihydroxydiadzein(9) and 8,2’-dihydroxydihydrodiadzein(W), found
only inP. lunatus, can be derived directlyfrom the 2”hydroxy precursors.
The major pathways to the isoflavones, isoflavanones, isoflavans, pterocar-
pans, and coumestans may befound. Some of these constituents are preny-
lated, followed by the cyclization ofthe prenyl group.P . rnungo appears to
prenylate only isoflavanones, whileP . lunatus may not possess the ability
to cyclize its prenylated products (Table 1).
Following similar analysis, the primary products in the 5-hydroxy path-
way are genistein (2), 2”hydroxygenistein (S), and dalbergioidin (16). These
metabolitescangivethemethylatedcompounds,isoprunetin (3), 2’-
methoxygenistein (6),2’-hydroxyisoprunetin (7), 5,2’-dimethoxydalbergioidin
(17, and isoferreirin (18) found in P . coccineus and P . lunatus together
with 8,2‘-dihydroxygenistein(8) of P.lunatus (Table 1). Thispath seems to
produce only isoflavonesand isoflavanones (Table 1). The position of the
1-hydroxylated coumestan, aureol (61), is not clear, as it could be derived
fromthis path following the correspondingsteps for the synthesis of
coumestans in the 5-deoxy pathway (Stoessl, 1982) or by hydroxylation of
an appropriate coumestan in the 5-deoxy pathway. However, prenylation
of isoflavanones, especiallyto kievitone, the major phytoalexin in mostof
the species, seems common whilethe prenylation of isoflavones appearsto
be restrictedto P . lunatus (Adesanya etal., 1986).
The hydration of the isopentenyl side chain of kievitone or phaseollidin
was first observed as a result of fungal detoxification in cultures fed with
these compounds (Bailey et al., 1977; Kuhn et al., 1977; Smith al.,
et 1980).
Similar hydrates, 5-deoxykievitol (21) and 5-deoxykievitone hydrate (20),
derivable from 5-deoxykievitone (19); kievitol (24) and kievitone hydrate
(23), derivable from kievitone,2,3-dehydrokievitol (ll), derivable from
2,3-dehydrokievitone (lo), and cyclokievitone hydrate(X), derivable from
cyclokievitone (33), were isolated from CuCl,-induced tissues (Adesanya,
1984; O’Neill etal., 1986). Clearly, the host plant also has the capabilityto
carry out these conversions, which in all instances reduces the fungitoxicity
of the constituent andthus could be considered asa detoxifying process by
the hostcells.Toxicityofphytoalexins to hostcells and ability of the
cells to detoxify phytoalexins when applied at sublethal doses have been

demonstrated(Smith, 1982; Van Etten etal., 1982). Theseresultsalso
support the probable existence ofa phytoalexin-detoxifying capability in P.
mungo, P . lunatus, and possibly other Phaseolus species. Enzymes involved
in detoxification reactions have also been characterized in fungi and in-
duced plants; for example, kievitone hydratase has been isolated from the
cell-free culture filtrate of Fusarium solani f. sp. phaseoli and also from
bean tissues inoculated with the fungi (Kuhn et al., 1977b; Smith et al.,
1984).
Dihydrostilbenes
Stilbenes have also been shown to arise from phenylalanine and like flavo-
noids there is a requirement for the action of PAL to give cinnamic acid
(Rudloff and Jorgensen, 1963; Langcake and Price, 1977; Schoppner and
Kindl, 1979; Stoessl, 1982). Thisenzyme was also activated in D . alata
tissues induced withB. theobromae and HgClz indicatinga similar role in
the Dioscoreaceae for this enzyme. Cinnamic acid may be converted to
p-coumaric acid and these two acids, as their CoA derivatives react with
malonyl-CoA to give either pinosylvin or resveratrol. In recent years, the
enzyme responsiblefor the formation of pinosylvin from malonyl-CoAand
cinnamoyl-CoA has been isolated from Pinus sylvestris (Schoppner and
Kindl, 1979). The partially purified enzymefrom Rheum rhaponticum was
found to catalyze the formation of 3,5,4'-trihydroxystilbene (resveratrol
67) (Rupprichet al., 1980). Stilbenesynthases, whichoccurin a small
number of genera, have been grouped into two types: those which utilize
p-coumaroyl-CoA and malonyl-CoA and form resveratrol (Schoppnerand
Kindl, 1984;Liswidowati etal., 1991), and the Pinus or Vitaceae (Fitzmeier
and Kindl, 1981) enzyme, which utilizes cinnamoyl-CoAand malonyl-CoA
and leads to pinosylvin accumulation (Gehlert et al., 1990). It is likely that
the dihydrostilbenesdihydropinosylvin (68) and dihydroresveratrol (76)
may arise from pinosylvin (66) and resveratrol (67), respectively, by the
reduction of the bridge double bond and other dihydrostilbenes derived
from these products or appropriate stilbene precursors (Stoessl, 1982). It
has also been suggested that the biosynthesis of phenanthrenes and dihydro-
phenanthrenes proceeds via dihydro-m-coumaric acid in Combretaceae and
Dioscoreaceae(Fritzmeier and Kindl, 1983); it ispossible that the 2'-
hydroxydihydrostilbenes arisefrom this route to give demethylbatatasin IV
(70), whichissubsequentlymetabolized to batatasin 111, IV,and V in
Dioscorea species (Fig. 1).
The induced formation of pinosylvin synthase is an early, selective,
and sensitive process elicited by fungior environmental stress. Young pine
seedlings, especially, respond quickly to the attack of various fungi by

synthesizing the enzymes required for pinosylvin accumulation (Gehlert et
al., 1990). Pine stilbene synthase cDNAs have been isolated from P . sylves-
tris in young seedlings challenged with B. cinerea and the full-length cDNA
encoding pinosylvin-forming stilbene synthase has been sequenced and a
comparison made with resveratrol-forming stilbene synthases and chalcone-
forming synthases; there were found to be significant differences in both
instances (Schroder et al., 1988; Lanz et al., 1990; Melchoir and Kindl,
1991). Low concentrations of fungal spores during periods of high humidity
activate the expression of substantial amounts of stilbene synthase. The
quick responseof young seedlings to fungal attack seems to be an important
means of defense (Ebel, 1989; Hain et al., 1990). Enhanced concentrations
of ozone causea substantial increase in stilbene synthase activity (Roseman
et al., 1991) and in this respect the level of pinosylvin synthase mRNAis an
excellent indicator of environmental stress (Schwekendiek et al., 1992).
E. Chemotaxonomic Considerations
Phaseolus-Vigna Complex
The distinction betweenVigna and Phaseolus species has long been a matter
of taxonomic interest. Analysis of the morphological characteristics, pro-
tein content, and enzymic activities has indicated that the Asiatic beans
which had previously been classed as Phaseolus species are better placed in
the Vigna genus (Kloz, 1962; Casimir and LeMarchand, 1966; Kloz and
Klozova, 1974; Chrispeels and Baumgartner, 1978). Marechal’s (1978) clas-
sification of the Phaseolus-Vigna complex had also placed P . aureus, an
Asiatic sample, inVigna subsection ceratotropis. Comparison of the chemi-
cal constituents of the Phaseolus and Vigna species shown in Table 1 with
those which have been reported for other species in the Leguminoseae indi-
cates some differences (Ingham,1982, 1990b). In general, most of the com-
pounds, including those with isoflavan structures that occur in the genera
Phaseolus, Vigna, Lablab, and Erythrina, are of little chemotaxonomic
importance. Recent investigations clearly showed that the ability t o produce
methoxylated derivatives, once thought to be restricted to Vigna, is far
more widespread (Table1) (Lampard, 1974; Preston, 1975; Ingham, 1981b,
1990b). Phaseollinisoflavan (43), found only in the typical Phaseolus spe-
cies P . vulgaris and P . coccineus, may be of chemotaxonomic importance.
It is interesting to note thatP . lunatus produced no cyclized derivative
of its many prenylated compounds and has the novel 8-hydroxylation path-
way. These observations confirm earlier observations that P . lunatus is
different from other Phaseolus species in its protein characteristics (Kloz,
1962; Kloz and Klozova, 1974).
Dioscorea Species
Ireland et al. (1981) attempted to separate different sections of the genus
unsuccessfully by studying the distribution of naturally occurring dormancy
inducing batatasin I-V in tubers and bulbils of 13 tropical and 2 temperate
Dioscorea species. The major phytoalexins isolatedfrom Dioscorea species
are shown in Table2. The major phytoalexin inD . batatas (Takasugi et al.,
1987), D . rotundata (Fagboun etal., 1987), D. alata (Cline et al., 1989), and
minor in D . mangenotiana (Kaganda and Adesanya, 1990) in the section
Enantiophyllumisdihydropinosylvin. It was not found in D. bulbifera
(section Opsophyton) and D . dumentorum (section Lasiophyton). Dihy-
droresveratrol has only been found in D . dumentorum (Adesanya et al.,
1989).
F. BiologicalProperties
Antimicrobial Activities
Isofuvonoids. The isolation ofsignificantamountsofisoflavones,
isoflavanones, pterocarpans, and isoflavans allowed a further study of the
structure-activity requirements of isoflavonoids. Perrin and Cruickshank
(1969) had earlier suggested that a common three-dimensional shape was
important, but further studies didnot confirm this hypothesis (VanEtten,
1967). Harborne and Ingham (1978), as a result of work done with isoflavone
luteone (13), suggestedthat a lipophilic side chain was important for fungi-
toxicity and a similar suggestionwas proffered for kievitone with an addi-
tional requirement for phenolic hydroxyl groups by Smith (1978). Light
microscopic examination revealedthe gross effects of active isoflavonoids
on the development of hyphae as disorganized growth patterns involving
swollen hyphal tips, cellular granulation, and the production of septal plugs
(Smith, 1978; Adesanya et al., 1986). Comparative antifungal studies using
Aspergillus niger and Cladosporium cucumerinum were done on all com-
pounds with good yield, their chemical derivatives, together with demeth-
ylmedicarpin (45), medicarpin (47), and isomedicarpin (46) isolated from
Trifoliumrepens (Table 4).
Differential sensitivity of the two fungi used was observed and this
supports Van Etten's (1967) observation in the test of pterocarpans and
related isoflavonoids against Aphanomyces euteiches Drechs. and Fusarium
solani Mart. Sacc. f.sp. cucurbitae (Burk, F. R. Jones) Synd. and Hans.
The results in Table4 show that pterocarpans (45,46,47,49) and isoflavans
(37,43) are more fungitoxic than other classesof isoflavonoidsand support
the result obtainedby Van Etten (1967). He suggested that the comparable
fungitoxic activities of(-) vestitol(38), ("sativan (40) and (+)3-hydroxy-
9-methoxypterocarpan (47) were related to the fact that the isoflavans can
Phaseolus, Wgna, and Dioscorea Species 359
Table 4 Inhibition of Spore Germination in Cladosporium cucumerinum and

Aspergillus nigerby Isoflavonoids
Minimum inhibitory
concentration' (pg/ml)
Compound C. cucumerinum A . niger

Isoflavones
1Daidzein 50-75 50-75
2 Genistein 50-75 50-75
3 Isoprunetin 25-50 50-75
4 2"Hydroxydaidzein 50-75 25-50
5 2'-Hydroxygenistein 50-75 25-50
6 2'"ethoxygenistein 25-50 50-75
7 2'-Hydroxyisoprunetin50-75 25-50
10 2,fDehydrokievitone 25-50 25-50
11 2,3-Dehydrokievitol > 100 > 100
12 Phaseoluteone 10-25 10-25
Isoflavonones
16 Dalbergioidin 50-75 75-100
18 Isoferreirin 10-25 10-25
19 5-Deoxykievitone 50-75 25-50
22 Kievitone 25-50 10-25
23 Kievitonehydrate > 100 > 100
24 Kievitol > 100 > 100
26 2' ,4' ,5-Trimethyl kievitone
10-25 25-50
27 2',4',5,7-Tetramethyl kievitone > 100 > 100
28 3'(y,y-Dimethylally1)kievitone
25-50 25-50
31 Cyclokievitone 50-75 50-75
32 3',4' ,5-Trimethylcyclokievitone > 100 > 100
33 1" ,2 "-Dehydrocyclokievitone
50-75 25-50
Pterocarpans
45 Demethylmedicarpin 50-75 25-50
46 Isomedicarpin 50-75 10-25
47 Medicarpin 25-50 10-25
49 Phaseollidin 40-50 10-25
57 Phaseollin 25-50 50-60
Isoflavans
37 Demthylvestitol 50-75 25-50
43 Phaseollinisoflavan 50-75 10-25
Coumestans
61 Aureol 50-75 25-50
'Minimum inhibitory concentration = the range of isoflavonoid concentrations between
whichfungalspore(approx 2 x lo6 sporedml)germinationat 25OC in Czapeck Dox
liquid nutrient ceases. All isoflavonoids tested had shown fungitoxic activity in theTLC
bioassay.
assume a conformation similar to the three-dimensional shapeof pterocar-

pans. Although the coumestans have been generally considered not to be
fungitoxic (Smith, 1982), the planar coumestan, aureol (61), isolated from
P . aureus had significant fungitoxic activity.
The isoflavone phaseoluteone(12) ismore activethan its nonprenylated
analog 2’-hydroxygenistein (5); the isoflavanone kievitone is more active
than dalbergioidin and the pterocarpan phaseollidin (49) is more active
than demethylmedicarpin (45), showing that the dimethylallyl side chain
enhances fungitoxicity. This was noted earlier by Harborne and Ingham
(1978) and Smith (1978) using a more limited range of compounds, and it
wassuggested that the side chain improves lipophilicity and hence aids
penetration of the fungal cell wall. However, the presenceof a second
side chainas in 3 ’-(y,y-dimethyally1)kievitone(28) didnot further enhance
fungitoxicity when comparedto kievitone and cyclization of the side chain
shows a reduction in activitywhen cyclokievitone (33) and phaseollin (57)
are compared to kievitone and phaseollidin, respectively (Table 4).
The oxidative product of kievitone, kievitone hydrate (23), isolated
earlier as a product of fungal metabolism of kievtone is not fungitoxic
(Kuhn et al., 1977). Similar reduction in activity was observed inthe series
even when the side chain was oxidized to a primary alcohol as in kievitol
(U), and 2,3-dehydrokievitol(ll) (Adesanya et al., 1986). Fungal oxidative
metabolites of pterocarpan phytoalexins such as phaseollidinand phaseol-
lin also have similar reduced activity (Ingham, 1976; Bailey et al., 1977;
Smith et al., 1980). However, phenolic hydroxyl groups appear important
for activity, since the tetramethylated derivative of kievitone (2’,4’,5,7-
tetramethylkievitone,27)was inactive, while 2‘,4’ ,7-trimethylkievitone (26)
was active. This indicates that at least one phenolic group is required for
fungitoxicity. Inthe isoflavone series, compounds with C-7, (2-4‘hydroxyl-
ation have fungitoxic activityand further hydroxylation at C-2’causes only
slight changes in toxicity (Table 4). This is in agreement with the results of
Ingham etal. (1977) withC.cucumerinum and Phytophthora megasperma.
Hesuggests that it is the presenceofoxygen at these positions that is
important, since methoxylsubstitutionof these hydroxyls does not radically
alter fungitoxic activity. The lackof difference in activity between diadzein
(1) and genistein (2) suggests that hydroxylation at C-5 is relatively unim-
portant in isoflavones. However, some differences in fungitoxic activity are
observed for kievitone and 5-deoxykievitone (19) in the isoflavanone series.
Perrin and Cruickshank (1969) suggested that hydroxyl groups at C-3 and
C-9 are important for activity in pterocarpans, but this is not confirmedby
this work or by other workers (VanEtten, 1967). It would be of interest to
evaluate the contribution to activity of oxygenation at C-7 and C 4 (C-3
and C-9 for pterocarpans). Isoflavonoids lackingthis oxygenation pattern

are rare in nature.
The significant toxicityof isoferreirin (18) when compared to dalber-
gioidin (16) might seemto show that methylation increasesboth lipophilic-
ity and activity, but this is not so in the isoflavone and pterocarpan series,
since 2’-methoxygenistein (6) has an activity similar to that of 2”hydroxy-
genistein (5), while medicarpin (47) and isomedicarpin (46) are only slightly
more fungitoxicthan demethylmedicarpin (45) (Table 4).
DihydrostiZbenes. In this class, too few compounds have been tested
for antimicrobial activity to provide a detailed comparative study. How-
ever, Takasugi et al. (1987) have shown that induced phenanthrenes and
dihydrostilbenes inD. batatas (Table 2) are generally more effective against
fungi than bacteria, and the dihydrostilbenes more active than the phenan-
threnes. Our results are in agreement with this observation and also show
that dihydropinosylvin and batatasin IV are significantly more toxic(<25
hg/ml, respectively) to Aspergillus fumigatus than demethylbatatasin IV
(70) (> 100 pg/ml), and dihydroresveratrol(76)was also not active against
A. niger and A. flaws (Adesanya, unpublished; Adesanya et al., 1989).
This might suggestthe need for at least two free hydroxyl groupsfor activ-
ity. The role of demethylbatatasin IV in disease resistance is not yet clear.
It has only transient accumulation in fungal-inducedtubers and no signifi-
cant one in HgCI,-treated ones (Figs. 2 and 3), suggesting that it could be
a product of fungallplant detoxification process, or a phytoalexin with
yet-to-be-defined toxicity. It is also noteworthy that the stilbene analogof
68 and 76, pinosylvin (66), and trans-resveratrol (67) with antimicrobial
activities have been isolated from induced Pinus and Arachis species, re-
spectively (Ingham, 1976; Schoppner and Kindl, 1979).
Growth Inhibitory Activity
The growth inhibitory activities ofbatatasin IV and batatasin I11 in lettuce
seed germination, hypocotyl elongation,and wheat coleoptile section elon-
gation tests have been established (Hasimotoand Tajima, 1978). Batatasin
IV also showed strong persistent inhibition of the gemination of seeds of
Sorghum bicolor,and by comparison dihydropinosylvin (68) showed a simi-
lar activityat a higher concentration, with the effect wearing off after 24 hr
(Fig. 4) (Cline et al., 1989). Demethylbatatasin IV was not active (Fig. 4).
Batatasin IV also showed a strong and persistent inhibition of the elonga-
tion of roots of S. bicolor seedlings, with dihydropinosylvin having a simi-
lar activity and demethylbatatasin IV showing a lesser action, but was still
an effective inhibitor of root elongation (Fig. 5). In all experiments the
roots did not recover with time and root tips, at high concentrations of
- 0
20 40 60 80
Time (hr 1
Figure 3 Enzyme activities in relation to the accumulation of dihydrostilbenes 1-

3 in HgCl, induced tubers. (a) M-., dihydropinosylvin (1); 0 - 0 , demethylbata-
tasin IV (2); A-A, batatasin IV (3). (b) After elicitation: 0-0,
PAL; 0-0, TAL;
in controls: .-M, PAL; 0 - 0 ,TAL. (c) After elicitation:A-A, PPO; in controls:
A-A, PPO.
al
al
LA "1
Concentration x 1uSw
Figure 4 Inhibitory activities of dihydrostilbenes 1-3 on germination of S. bicolor

seeds. -
0 0,48 hr incubation; W-. ,24 hr incubation.
dihydropinosylvin and batatasin IV collapsed presumably dueto an uncon-

trolled leakage of metabolites and electrolytes from the cell due to severe
effects on the cell membrane (Tahara etal., 1972). The result suggeststhat
dihydropinosylvin and demethylbatatasin IV do not possessdormancy-
for batatasin IV and other bata-
inducing characteristics already established
tasins (Hashimoto et al.,1972), since theycannot prevent seed germination,
+
0
0
[L
Concentration x 10-SM
Figure 5 Inhibitoryactivities of dihydrostilbenes 1-3 on root elongation of S.

bicolor seedlings. 0 - 0 , 4 8 hr incubation; m-. , 2 4 hr incubation.
but like other phytoalexins are toxic to the parasites aswell as host cells at
high concentrations. The role of batatasin IV in disease resistance is not
clear, since it accumulatesto a low level in Dioscorea species, includingD .
batatas (Takasugi et al., 1987; Cline et al., 1989), and also occurs naturally
in D . batatas (Hashimoto et al.,1972).
IV. EPILOG
Selected species in the Phaseoleae and Dioscoreaceae have been investigated
for their inducible compounds. Large-scaleand detailed analysis of CuC1,
induced tissues ofPhaseolus species led to the isolation of many isoflavo-
noids that provided an insight into their possible biogenesis, structure-
antifungal activity relationships, and chemotaxonomy. The analysis of fun-
gitoxic spots in the TLC bioassay of fungus induced tissues of Dioscorea
species gave only a few dihydrostilbenes by comparison. Thus a detailed
large-scale analysis of induced tissues has more analytical potential than
isolation of fungitoxic compounds alone.
Although it was possible to speculate on the biosynthetic pathway to
most of the isoflavonoids and dihydrostilbenesconsidered, the .correct
pathway still needs to be established by isolating and characterizing the
terminal enzymes. This is being done for some compounds inother species
(Preisig etal., 1991). Phytoalexins have obvious chemotaxonomic potential
in these two families and so detailed analysis of induced compounds in
more species could assist in the taxonomy of these families. Such investiga-
tions could also yield more novel compoundsand new biological activities.
The bioassay of the dihydrostilbenes was extended to human fungal para-
sites. A. fumigatus and Tricophyton mentagyrophytes,in order to evaluate
their usefulness in treating human infections. Similar investigations have
shown that phaseollinisoflavan is very active against zoopathogens (Gordon
et al., 1980; Smith, 1982). Such screenings againstother test systems could
expose other potential uses of these compounds.
The role of many of these inducible compounds in disease resistance
needs to be established as they vary in composition, number, and amounts
in various species. In P. lunatus, P. mungo, and P . vulgaris, the largest
group of antifungal isoflavonoids was the isoflavanones, whereas in P .
aureus and P. coccineus there is a more even distribution of isoflavones,
isoflavanones, and pterocarpans. Only P . coccineus produces significant
levels of isoflavans, and the only fungitoxic coumestan isolated, aureol,
was found in bothP . aureus and P. mungo.
Although the work on the Dioscorea species is as yet not comprehen-
sive, variations are also noticeable in the type of major compounds pro-
duced, the seemingly inactive demethylbatatasin IV was the major inducible
compound inD. bulbifera, while D. dumentorum contains dihydroresvera-

trol and a yet-to-be-characterized compound, the other species produce
dihydropinosylvin asthe major compound, exceptD. mangenotiana, which
is more prone to rot than the other species considered. It is likely that the
plant uses all the active compounds it produces in a concerted effort to
combat fungal invasion, with each compound having different functions
(Mansfield, 1982; Smith, 1982).
The work on the species in the Dioscoreaceae, particularly the eco-
nomic yams, is aimed at breeding resistant species and finding ways of
protecting yams under storagefrom microbial attack while increasing their
dormancy period, thus increasing their shelf life. The results obtained so
far indicate that batatasin IV and to a lesser extent dihydropinosylvin might
be good subjects for further studies that would include the evaluation of
their toxicityto humans and metabolism inboth plants and humans.
ACKNOWLEDGMENTS
The authors with to thank the librarians at the Schoolof Pharmacy for help
in checking references and Mr. B. C. Homeyer for help with the compila-
tion of the figuresand tables.
REFERENCES
Adesanya, S. A. (1984). Inducible and constitutive isoflavonoids from Phaseolus

and Vigna species. PhD Thesis, University ofLondon.
Adesanya, S. A., Ogundana, S. K., and Roberts, M. F. (1989). Dihydrostilbene
phytoalexins from Dioscorea bulbifera and D. dumentorum. Phytochemistry
28(3):773-774.
Adesanya, S. A., O'Neill, M. J., and Roberts, M. F. (1984a). Induced and constitu-
tive isflavonoids in Phaseolus mungo. Z. Naturforschung. C, Biosciences 39:
881-893.
Adesanya, S. A., O'Neill, M. J., and Roberts, M. F. (1984b). Isoflavonoids from
Phaseolus coccineus. Phytochemistry 24(11):2699-2702.
Adesanya, S. A., O'Neill, M. J., and Roberts, M. F. (1986). Structure related
fungitoxicity of isoflavonoids.Physiol. Mole. Plant Pathol.2995-103.
Alozie, S. O., Nwankiti, A. O., and Oti, E. (1987). Source of resistance inanthrac-
nose blotch disease of water yam (Dioscorea alatu)caused by Colletotrichum
gloeosporioides Penz. 2. Relationship between phenol content and resistance,
Beitraege zur Tropischen Landwirtschaft und Veterinaermedizin 25H.155-59.
Arnoldi, A., and Merlini,L. (1990). Antifungal activityrelationships for some
isoflavonoid phytoalexins.J. Agric. Food Chem. 38:834-838.
Bailey, J. A. (1973). Production of antifungal compounds in cowpea (Vignasinen-
sis) and pea (Pisum sativum)after virus infection. J. Gen. Microbiol. 75: 119-
123.
Bailey, J. A. (1982). Mechanism of phytoalexin accumulation. Phytoalexins (J. A.
Bailey and J. W. Mansfield, eds.), Blackie, London, pp. 289-378.
Bailey, J. A. (1984). The relationship between symptom expressionand phytoalexin
concentration in hypocotyls of Phaseolus vulgaris infected with Collectotri-
chum lindemuthianim. Physiol. Plant Pathol. 4~477-488.
Bailey, J. A., Burden, R. S., Mynett, A., and Brown, C. (1977). Metabolism of
phaseollin by Septoria nodorumand other nonpathogens ofPhaseolus vulgaris.
Phytochemistry 161541-1544.
Bailey, J. A., Rowell, P. M., and Arnold,G . M. (1980). The temporalrelationship
between host celldeath, phytoalexin accumulationand fungal inhibition during
hypersensitive reactions of Phaseolus vulgaristo Colletotrichum lindemuthia-
num. Physiol. Plant Pathol.12329-339.
Bell, A. A. (1981). Biochemical mechanisms of disease resistance.Ann. Rev. Plant
Physiol. 32:21-81.
Bell, A. A., and Pressley, J. T. (1969). Heat inhibited or heat killed conidia of
Verticillium albo-atruminduce disease resistance in cotton. Phytopathology 5 9
1147-1151.
Bell, J. N., Dixon, R. A., Bailey, J. A., Rhowell, P. M., and Lamb, C. J. (1984).
Differential induction of chalcone synthase mRNA activity at the onset of
phytoalexin accumulation in compatible and incompatible plant pathogen in-
teractions. Proc. Nal. Acad. Sci. USA 91:3384-3488.
Biggs,D.R.,Shaw, J. G., and Yates, M. K. (1983). Identificationof 41,7-
dihydroxy -2’,3’-dimethoxyisoflavanin bean roots. J. Nat. Prod. 46(5):742-
744.
Bless, W., and Barz, W. (1988). Isolation of pterocarpan synthase, the terminal
enzyme of pterocarpan phytoalexin biosynthesis in cell suspension culture of
Cicer arietinum. Fed. Eur. Biol. Soc. Lett. 23247-50.
Casimir, J., and Le Machand, L. (1966). Bulletin Jardin Botanique de L’Etat a
Bruxelles. 3 623 3.
Chrispeels, M. J., and Baugartner, B. (1978). Seriological evidence confirming the
assignment ofPhaseolus aureusand P. mungo to the genus Vigna. Phytochem-
istry 12125-126.
Cline, E. I., Adesanya, S. A., Ogundana, S. K., and Roberts, M. F. (1989). Induc-
tion of PAL activity and dihydrostilbene phytoalexinsin Dioscorea data and
their plant growth inhibitory properties. Phytochemistry 28(10):2621-2625.
Coursey, D.G . (1967). Yams. Longmans, London.
Coxon, D. T., Ogundana, S. K.,and Dennis, C.(1982). Antifungal phenathrenes in
yam tubers. Phytochemistry21(6):1389-1392.
Cruickshank, I. A. M., and Perrin, D. R. (1968). The isolation and partial charac-
terization of monilicolin A, a polypeptide with phaseollin inducing activity
from Monilinia fructicola. Life Sci. 2449-458.
Darvill, A. G., and Albersheim, P. (1984). Phytoalexins and their elicitors: a de-
fense against microbial infections in plants.Ann. Rev. Plant Physiol. 35:243-
275.
De Lima, 0. G., DeMello, J. F., De Barros-Coeiho, J. S., De Andrade Lyra,

F. G., De Albuquerque, M. M., Marini-Bettolo, G. B., Delle Monache, G.,
and Delle Monache, F. (1975). New prenylated chalcones from Longicarpus
neuroscapha betth. II. Farmaco Ed. Sci. 30326-342.
Delle Monache, G., Delle Monache, F., Marini-Bettolo, G. G., De Albuquerque,
M., De Mello, J., and De Lima, 0 . G. (1977). Gazette Chimica Italiana107:
189.
Denny, T., and Van Etten, H. D. (1983). Tolerance of Nectria haematococcaMPVI
to phytoalexin pisatin inthe absence of detoxification. J. Gen. Microbiol.129
2893-2901.
Deverall, B. J. (1982). The concept of phytoalexins. Phytoalexins(J. A. Bailey and
J. W. Mansfield, eds.), Blackie,London, pp. 1-20.
Dewick, P. M. (1992). Biosynthesis of shikimate metabolites. Nut. Prod. Rep. 9
153-181.
Dewick, P. M. (1988). Isoflavonoids. The Flavonoids: Advancesin Research Since
1980 (J. B. Harborne, ed.). Chapman-Hall, New York, pp. 125-240.
Dhawale, S., Souciet, G., and Kuhn, D. N. (1989). Increase in chalcone synthetase,
messenger RNA in pathogen inoculated soybeans with race specific resistance
is different in leaves and roots. Plunt Physiol. 91(3):911-916.
Dixon, R. A. (1986). The phytoalexin response: elicitation signalingand control of
host gene expression.Biol. Rev. 61:239-291.
Duke, J. A. (1983). Handbook of Legumes of World Economic Importance.Ple-
num Press, London.
Ebel, J. (1986). Phytoalexin synthesis: the biochemical analysis of the induction
process. Ann. Rev. Phytopathol. 24:235-264.
Ellis, J. S., Jennings, A. C., Edwards, L. A., Mavandad, M., Lamb, C. J., and
Dixon, R. A. (1989). Defense gene expression in elicitor treated cell suspension
cultures of french bean cultivarImuna. Plant CellRep. 8(8): 504-507.
Ersek, T., and Kiraly, Z. (1986). Phytoalexins. Warding-off compounds in plants?
Physiol. Plantarum68:343-346.
Fearman, H., and Diamond, A. (1967). Phytopathology 15:279.
Fagboun, D. E., Ogundana, S. K., Adesanya, S. A., and Roberts, M. F. (1987).
Dihydrostilbene phytoalexins fromDioscorea rotundata. Phytochemistry26( 12):
3187-3189.
Fritzemeier, K. H., and Kindl, H. (1981). Coordinate induction byUV light of
stilbene synthetase, phenylalanine ammonia lyase and cinnamate-4-hydroxylase
in leaves of Vitaceae.Planta 151:48-52.
Fritzemeier, K. H., and Kindl, H. (1983). Two different pathways leadingto phen-
anthrene and 9,lO-dihydrophenanthrenesof the genus Dioscorea. Z. Naturfor-
schung. C, Biosciences 39217-221.
Gehlert, R., Schoppner, A., and Kindl, H. (1990). Mol. Plant-MicrobeInteract. 3:
444-449.
Gordon, M.A., Lappa, E. W., Fitter, M. S., and Lindsay, M. (1980). Susceptibility
of zoopathogenic fungi to phytoalexins. Antimicrob. Agents Chemother. 12:
120-123.
Grisebach, H.,Edelmann, L., Fischer, D., Kochs,G., and Welle, R. (1989). Biosyn-
Phaseolus, Vigna, and Dioscom Species 369
thesis of phytoalexins and nod-gene inducing isoflavones in soybean. Signal

Molecules in Plants andPlant-MicrobeInteractions (B. Lugtenberg,ed.),
NATO AS1 Series Vol H-36, Springer, Heidelberg, pp. 57-64.
Hadwiger, L.A., and Schwochau, M. E. (1971). Ultraviolet light inducedformation
of pisatinand phenylalanine ammonia lyase. Plant Physiol.42346-351.
Hagmann, M. L., Heller, W., and Grisebach, H. (1984). Induction of phytoalexin
synthesis in soybean: Stereospecific 3,9-dihydroxypterocarpan6a-hydroxalase
from elicitor-induced soybeancell cultures. Eur. J. Biochem. I42:127-931.
Hahlbrock, K., and Scheel, D. (1989). Physiology and molecular biologyof phenyl-
prionoid metabolism.Ann. Rev.Plant Physiol. PlantMol. Biol. 40347-369.
Hain, R., Bieseler, B., Kindl, H., Schroder, G., and Stocker, R. (1990). Plant Mol.
Biol. 15~325-335.
Hakamatsuka, T., Noguchi, H., Ebizuka, Y.,and Sankawa, U. (1988). Deoxychal-
cone synthasefrom cell suspension cultures ofPueraria lobata. Chem. Pharm.
Bull. 364225-4228.
Hakim, A.M., Ani, A., and Dewick, P. M.(1984). Isoflavonoidbiosynthesis:
concerning the aryl migration. J. Chem. Soc. Perkin Trans. I. 2831-2838.
Harborne, J. B. (1977). Introduction to Ecological Biochemistry. Pergamon Press,
London.
Harborne, J. B., and Ingham, J. L. (1978). Biochemical aspects of coevolution of
higher plants with their fungal parasites. Biochemical Aspects of Coevolution
in Higher Plants (J. B. Harborne, ed.), Academic Press,London, pp. 343-405.
Hargreaves, J. A. (1979). Investigation into the mercuric chloride stimulatedphyto-
alexin accumulation in Phaseolus vulgaris and Pisum sativum. Physiol. Plant
Pathol. 15~219-287.
Hashimoto, T., Hasegawa, K., and Kawarada, A. (1972). Batatasins: new dor-
mancy inducing substances ofyam bulbils. Planta 108369-374.
Hashimoto, T., and Tajima, M. (1978). Structures and synthesis of growth inhibi-
tors, batatasin IV and V and their physiological activities. Phytochemistry 12
1179-1 184.
Ingham, J. L. (1976). Fungal modification of pterocarpan phytoalexin from Melilo-
tus alba,and Trifolium pratense. Phytochemistry 15:1489-1495.
Ingham, J. L. (1981a). Phytoalexin induction and its chemosystematic significance
in the genus Trigonella. Biochem. Syst. Ecol. 9(4):275-281.
Ingham, J. L. (1981b). Phytoalexin induction and its taxonomic significance inthe
Leguminosae (subfamily Papillionoideae). Advances in Legume Systematics.
Proceedings of the International Legume Conference,Kew, 1978. (R. M. Pol-
hill, and P. G. Ravens, eds.), London, pp. 599-626.
Ingham, J. L. (1982). Phytoalexins in the leguminosae. Phytoalexins (J. A. Bailey
and J. W. Mansfield, eds.), Blackie,London, pp. 21-80.
Ingham, J. L. (1990a). A further investigation of phytoalexin formation in genus
Trifolium. Z. Naturforschung C, Biosciences 45(7-8):829-834.
Ingham, J. L. (1990b). Systematic aspects ofthe phytoalexin formation within the
tribe Phaseolae ofthe Leguminosae, subfamily Papillionoideae. Biochem. Syst.
EcoI. 18(5):329-344.
Ingham, J. L., and Harborne, J. B. (1976). Phytoalexin induction as a new dynamic
approach to the study of systematic relationshipsamong higher plants. Nature

260241-243.
Ingham, J. L., Keen, N. T., and Hymowitz, T. (1977). A new isoflavone phyto-
alexin from fungus inoculated stems of Glycine wightii. Phytochemistry 1 6
1943-1946.
Ireland, C. R., Schwabe, W. W., and Coursey, D. G. (1981). The occurrence of
batatasins in Dioscoraceae.Phytochemistry 20(7):1569-1571.
Kaganda, N. G., and Adesanya, S. A. (1990). A new dihydrostilbene phytoalexin
from diseased Dioscorea mangenotiana.J. Nut. Prod. 53(5):1345-1346.
Kay,D.E. (1979). Food, Legumes, Crop and Product Digest No. 3. Tropical
Product Institute, London.
Keen, N. T., Lyne, R. L., and Hymowitz, T. (1986). Phytoalexin production as a
chemosystematicparameter within the Genus Glycine. Biochem. Sys.Ecol. 14:
418-486.
of carbohydrates from Phytophthora megasperma f. sp., glycinea and other
sources. Plant Physiol.71:466-471.
Kiraly, Z., Barma, B., and Ersek, T. (1972). Hypersensitivityas a consequence, not
the cause ofplant resistance to infection. Nature 239456-472.
Kloz, J. (1962). An investigation ofthe protein character of four Phaseolus species
with special referenceto the question of their genus. Biologia Plantarum485-
90.
Kloz, J., and Lozova, E. (1974). Protein euphaseolin in Phaseolinae-a chemico-
taxonomical study.Biologia Plantarum16290-300.
Koch, G., and Grisebach, H. (1986). Enzymic synthesis of isoflavones. Eur. J.
Biochem. 155:311-318.
Koch, G., and Grisebach, H. (1989). Phytoalexin synthesis in soy bean:
purification
andreconstitutionofcytochrome P 4 5 0 3,9dihydroxypterocarpan6a-hydro-
xylase and separation from cytochromeP-24Ckcinnamaaate4hydroxylase. Arch.
Biochem. Biophys.
KuC, J. (1982). Phytoalexins in the Solanaceae. Phytoalexins (J. A. Bailey and
J W. Mansfield, eds.), Blackie,London, pp. 21-80.
Kuhn, P. J., Smith, D. A., and Ewing, D. F. (1977).5,7,2’,4’-Tetrahydroxy-8-(3”-
hydroxy-3”-methyl-butyl)isoflavanone, a metabolite of kievitone produced by
Fusaium solanifsp. Phaseoli. Phytochemistry 16296-297.
Kuhn, P. F., Smith, D. A., and Ewing, D. F. (1977b).Physiol. Plant Pathol. 1 4
179.
Lampard, J. F. (1974). Demethylhomopterocarpin; an antifungal compound in
Canavalia ensiformisand Vigna unguiculatafollowing infection. Phytochemis-
try 13:291-292.
Langcake, P., and Pryce, R. J. (1977). The production of resveratrol and the
viniferins by grapevines in response to ultraviolet radiation. Phytochemistry
16~1193-1196.
Lanz, T., Schroder, G., and Schroder, J. (1990). PIanta 181:169-175.
Liswidowati, Melchior,F., Hohmann, F., Schwer, B., and Kindl, H. (1991).Planta
183~307-314.
Mabry, T. J., Markham, K. R., and Thomas, M. B. (1970). TheSystematicIdentifi-

cation of Flavonoids. Springer, Berlin.
Maekawa, F. (1955). Topomorphological and taxonomical studies in Phaseoleae
Legiminosae. Jap. J. Bot.15:103-116.
(J. A. Bailey and J. W. Mansfield, eds.), Blackie,London, pp. 253-288.
Mansfield, J. W., Hargreaves, J. A., and Boyle, F. C. (1974). Phytoalexin produc-
tion by live cells in broad bean leaves infected with Botrytis cinerea. Nature
252~316.
Marechal, R., Mascharpa, J., and Stainer, F. (1978). Etude taxonomique d’un
group complexe d’especes des ageniesPhaseolus et Vigna (Papilionaceae) sur la
base de donnees morphologiques et polliniques, traites par l’analyse informa-
tique. Boisseira 28: 1-273.
Martin, M., and Dewick, P. M. (1979). Biosynthesis of 2-arylbenzofuran phyto-
alexin, vignafuran in Vigna Unguiculata.Phytochemistry 18:1309-1317.
Mayer, A. M., and Hard, E. (1979). Polyphenol oxidases inplants. Phytochemistry
18193-215.
Melchior, F., ana Kindl, H. (1991). Arch. Biochem. 288552-557.
Muller, K. 0.(1958). Studies on phytoalexins. 1. The formation and immunological
significance of phytoalexins produced by Phaseolus vulgaris in response to
infections with Sclerotinia fmcticola and Phytophthora infestans. Aust. J.
Biol. Sci. 11~275-300.
Muller, K. O., and Borger, H.(1940). Experimentelle utersuchungenuber die Phy-
tophthora resistenz der kartoffel. Arb. Biol. Anst. Reichsant 23:189-23 1.
O’Neill, M. J. (1983). Aureol and Phaseol, two new coumestans from Phaseolus
aureus Roxb., Z. Naturforschung C, Biosciences 38:698-700.
O’Neill, M. J., Adesanya, S. A., and Roberts, M. F. (1983). Antifungal phytoalex-
ins from Phaseolus aureus Roxb., Z. Natturforschung C, Biosciences 38693-
697.
O’Neill, M. J., Adesanya, S. A., and Roberts, M. F. (1984). Isosojagol, a coumes-
tan from Phaseolus coccineus. Phytochemistry 23(11):2704-2705.
O’Neill, M. J., Adesanya, S. A., Roberts, M. F., and Pantry, I. R. (1986). Novel
inducible isoflavonoidsfrom lima bean,Phaseolus lunatus. Phytochemistry 25:
1315-1322.
Perrin, D. R., and Cruickshank, I. A. M. (1969). The antifungal activity of ptero-
carpans towards Moniliniafmcticola. Phytochemistry 8:971-978.
Preisig, C. C., Van Etten, H. D., and Mareau, R. A. (1991). Induction of 6a-
hydroxymaackiain-3-0-methyltransferase activities during biosynthesis of pi-
satin. Arch. Biochem. Biophys.290(2):468-473.
Preston, N. W. (1975). 2’-O-methylphaseollidinisoflavin from infected tissue of
Vigna Unguiculata.Phytochemistry 14:1131-1132.
Preston, N. W., Chamberlain, K., and Skipp, R. A. (1975). A 2-arylbenzofuran
phytoalexin from cowpea (Vigna unguiculata). Phytochemistry 14: 1843-1844.
Rahe, J. E. (1973). Occurrence and levels of the phytoalexin phaseollin in relation
to delimitation at sites of infection of Phaseolus vulgaris by Collectotrichum
lindemuthinum. Can.J. Bot. 51~2423-2430.
Robbins, M. P., Bkolwell, G . P., and Dixon, R.A. (1985). Metabolic changes
in elicitor-treated bean cells. Selectivity of enzyme induction in relation to
Phytoalexin accumulation.Eur. J . Biochem. 148563-569.
Roseman, D., Heller, W., and Sandermann, H. (1991). Plant Physiol. 921280-
1286.
Rudloff, E.,and Jorgensen, E. (1963). The biosynthesis of pinosylvin in the sap-
wood ofPinus resinosaAIT. Phytochemistry 2:297-304.
Rupprich, N., Hildebrand, H., and Kindl, H. (1980). Substrate specificity in vivo
and in vitro in the formation of stilbenes, biosynthesis of rhaponticin. Arch.
Biochem. Biophys.200:72-78.
Ryder, T.B., Cramer, C. C., Bell, J. N., Robbins, M. P., Dixon, R. A., and Lamb,
C. J. (1984). Elicitor rapidly induces chalcone synthase mRNA in Phaseolus
vulgaris cells at the onset of the phytoalexin defense response.Proc. NI. Acad.
SC~.USA 81:5724-5728.
Schmelzer, E., Som sich, I., and Hahlbarock, K. (1985). Coordinated changes in
transcription and translation rates of phenylalanineammonia-lyase and 4-
coumarate:CoA ligase mRNAs in elicitor-treatedPetroselenium crispumcells.
Plant CellRep. 4:293-296.
Schoppner, A., and Kindl, H. (1979). Stilbene synthetase (pinosylvine synthase) and
its induction by ultraviolet light.Fed. Eur.Biochem Soc. Lett. 108(2):349-352.
Schoppner, A., and Kindl, H. (1984). Purification and properties of a stilbene
synthase from induced cell suspension cultures ofpeanut. J. Biol. Chem. 259
6806-681 1.
Schroder, G., Brown, J. W.S., and Schroder, J. (1988). Eur. J. Biochem. 172:161-
169.
Schwekendiek, A., Pfeiffer, G., and Kindl, H. (1992). Pine stilbene synthase of
cDNA, a tool for probing environmental stress. Fed. Eur. Biochem.Soc. 301:
41-44.
Smith, D. A. (1978). Observations on the fungitoxicity of kievitone. Phytopathol-
ogy 4881-87.
Smith, D. A. (1982). Toxicity of phytoalexins.Phytoalexins (J. A. Bailey and J. W.
Mansfield, eds.), Blackie,London, pp. 218-252.
Smith, D. A., and Banks, S. W. (1986). Biosynthesis,elicitation and biological
activity of isoflavonoid phytoalexins. Phytochemistry 25(5):979-995.
Smith, D. A., and Ingham, J. L. (1980). Legumes, Fungal Pathogens, and Phyto-
alexins. Advances in Legume Sciences (R. J. Summerfield, and A. H. Bonting,
eds.), HMSO, London, pp. 207-222.
Smith, D. A., Kuhn, P. J., Bailey, J. A., and Burden, R. S. (1980). Detoxification
of phaseollidin by Fusarium solani fsp. Phaseoli. Phytochemistry 191673-
1675.
Smith, D. A., Wheeler, H. E., Banks, S. W., and Cleveland, T.E. (1984). Association
between lowered kievitone hydratase activity and reduced virulence to bean in
variants of Fusarium solani f. sp. phaseoli.Physiol. Plant Pathol. 25135-147.
Smith, J. M.(1971). Physiological Plant Pathol.1:85.
Stoessl, A. (1982). Biosynthesis of Phytoalexins. Phytoalexins (J. A. Bailey and J.
W. Mansfield, eds.), Blackie,London, pp. 218-252.
Sun, Y.,Wu, Q., Van Etten, H. D., and Hrazdina, G. (1991). Stereoisomerism in
plant diseaseresistance: induction and isolation of 7,2 "dihydroxy-4'5'-
methylenedioxyisoflavone.Arch. Biochem. Biophys.284 167-173.
Sunder, R.,Rangaswami, S., and Reddy, G. C. (1978). A new dihydrophenanthrene
from Dioscorea decipiens. Phytochemistry 17:1037.
Tahara, S., Ingham, J., Nakahara, S., Mizutami, J., and Harborne, J. B. (1984).
Fungitoxic dihydrofuranisoflavones and related compounds in white lupin,
Lupinus albus. Phytochemistry23:1889-1990.
Takasugi, M., Kawashima, S., Monde, K., Katsui, N., Masamune, T., and Shirata,
A.(1987). Antifungal compounds from Dioscoreabatatas inoculatedwith
Pseudomonas cichorii. Phytochemistry 26(2):371-375.
Tani, T., and Mayama, A.(1982). Plant Infection:ThePhysiologicalandbiochemi-
cal Basis. Japan Science SocietyPress, Tokyo.
Van Etten, H. D. (1967). Antifungal activity of pterocarpans and other selected
isoflavonoids. Phytochemistry 15:655-659.
Van Etten, H. D.,Matthews,D.E., and Smith, D.A.(1982). Metabolismof
phytoalexins. Phytoalexins (J. A. Bailey and J. W. Mansfield, eds.), Blackie,
London, pp. 181-217.
Van Etten, H. D., Matthews, D. E., Matthews, P., Miao, V., Maloney, A., and
Straney, D. (1989). A family of genes for phytoalexin detoxification in the
plant pathogen Nectria haematococca. Signal Molecules in Plants and Plant
Microbe Interactions (B. J. J. Lugtenberg, ed.), NATO AS1 Series H36, pp.
219-228.
Verdecourt, B. (1970). Studies in the Leguminosae Papilionoideae for the flora of
tropical East Africa:IV. Kew Bull. 24507-570.
Viorin, B.(1983).UV spectral differentiation of5-hydroxy and 5-hydroxy-3-
methoxy flavones with mono-(4'), di-(3',4') or tri(3 ',4',5') substituted B-rings.
Phytochemistry 22:2107-2145.
Welle, R., and Grisebach, H. (1991). Properties and solubilization of phenyltransf-
erase of isoflavonoid phytoalexins biosynthesis in soybeans. Phytochemistry
30(2):479-484.
Weltring, K. M., Mackenbrock, K.,and Barz, W. (1982). Demethylation, methyla-
tion and 3'-hydroxylation of isoflavones by Fusarium fungi. 2. Naturfor-
schung. 371370-574.
Wengenmayer, H., Ebel, J., and Grisebach, H. (1974). Eur. J. Biochem. 50135.
Whitehead, I. M., Dey, P. M., and Dixon, R. A. (1982). Planta 154156.
Willeke, U., and Barz, W. (1982). Catabolism of 5-hydroxyisoflavonesby fungi of
the genus Fusarium. Arch. Microbiol. 132:266-269.
Woodward, M. D. (1979a). Phaseoluteone and other 5-hydroxyisoflavonoidsfrom
Phaseolus vulgaris. Phytochemistry18363-365.
Woodward, M. D. (1979b). New isoflavonoids relatedto kievitone from Phaseolus
vulgaris. Phytochemistry182007-2010.
Woodward, M. D. (1980). Phaseollin formation and metabolism in Phaseolus vul-
garis. Phytochemistry19921-927.
16
Involvement of Phytoalexins in the
Response of Phosphonate-Treated Plants
to Infection by Phytophthora Species
P. Saindrenan
Institut de Biologie Mol4culaire des Plantes,C.N.R.S., Strasbourg, France
David 1. Guest
University of Melbourne, Parkville, Victoria, Australia
1. INTRODUCTION
The rapid and localizedpathogen-inducedsynthesisofphytoalexins is
thought to play an important defensive r d e in the early response of many
plants to infection by potential parasites (Dixonand Lamb, 1990). In sus-
ceptible hosts, the growth of biotrophic fungal pathogens is unrestricted,
and a compatible host-parasite interaction may develop given favorable
environmental conditions. Responses usually associated with resistance are
not entirely absent, but rather are not expressed soon enoughor with suffi-
cient magnitudeat the infection site to be effective (KuC and Rush, 1985).
Presumably the necessary elicitor signals are either not released in sufficient
magnitude or are not recognized sufficiently rapidly the by plant.
In terms of strategies for plant disease protection, the ability to trans-
form a compatible interactioninto an incompatible interactionby modify-
ing the pathogenic abilityof the parasite (Kiraly etal., 1972) or the defense
response of the host (Langcake, 1981) offers many exciting possibilities.A
number of chemicals appearto have this ability.The antipenetrant tricycla-
zole actsby reducing the pathogenic competence ofthe parasite (Woloshuk
etal., 1980). Probenazole(Sekizawa and Mase,1981),2,2-dichloro-3,3-
dimethylcyclopropane carboxylic acid (Langcake et al., 1983), and methyl-
2,6-dichloroisonicotinicacid (MCtraux et al., 1991) alter the host physiology
so that disease resistance is increased.
Phosphonates,including the phosphonate anion (H,PO,? and fo-
setyl-Al (aluminum tris-0-ethyl phosphonate, Aliette), are systemic com-
pounds that control diseases caused by oomycetes, particularly Phytophth-
375
376 and Saindrenan Guest
ora species (Bertrand et al., 1977; Pegg et al., 1985; Cohen and Coffey,
1986). Compared to other systemicfungicides,theyexertlittleactivity
on mycelial growth in vitro at the submillimolar concentrations usually
found in vivo in protected plants (Bompeix et al., 1981; Guest, 1984a).
Conversely, their effects in vitro frequently do not match their perfor-
mance in vivo. This paradox has been explained by a hypothesis propos-
ing a complex mode of action, involving the active participation of the
hostdefensereactions,mainlyphytoalexinaccumulation(Bompeix et
al., 1981; Guest, 1984a). However, at higher application rates, and with
certain host-parasite interactions,a direct modeof action is more plausible
(Fenn and Coffey, 1984; Dercks and Buchenauer, 1986; Ye and Deverall,
1986).
Bompeix and coworkers (Bompeix et al., 1980;Vo-Thi et al., 1979)
made the original observationthat tomato leaves produced more phenolic
compoundsfollowinginoculationwith P . capsici when the leaveswere
floated on a solution of fosetyl-Althan they did when floated on distilled
water. The response was associated withthe formation of necrotic blocking
zones and hypersensitive cell death (Durandand SallC, 1981) similar to that
observed in the incompatible interaction between Cladosporium fulvum
and tomato (Lazarovits and Higgins, 1975a,b). Although the involvement
of phenylpropanoid metabolism in the resistance responseof tomato has
not yet been clearly demonstrated @U&,1982), it was suggested that en-
hanced production of phenolic compounds might be a component of the
resistance mechanism of phosphonate-treated leaves. These observations
were subsequently confirmed in other plant-pathogen interactions (Guest,
1984b; Khan etal., 1986; Saindrenan et al., 1988a,b; Afekand Sztejnberg,
1989; Smillie etal., 1989; Nemestothyand Guest, 1990).
In this chapter, we will discuss the effects of phosphonates (disodium
phosphonate salt, NazHPO3.5H20,and fosetyl-AI)on the stimulation of the
plant’s defense reactions via phytoalexin accumulation in two well-defined
model systems, cowpea (Vigna unguiculata (L.) Walp. CV. Tvu 645)-Phy-
tophthora cryptogea and tobacco (Nicotiana tabacum L.)-Phytophthora
nicotianae var. nicotianae. Some other plant-pathogen interactionswill be
described briefly. Indeed, it has been shown that the protection given by
submillimolar concentrations of phosphonates is higher in plants having
highly active dynamic defense systems (Guestand Grant, 1991). Moreover,
experimental treatments that suppress phytoalexin accumulation provide
useful cluesto assess the role of phytoalexins and other defense reactions in
the mode of action of phosphonates. A range of biosynthetic inhibitors
have been used to identify the metabolic pathways that are important to
resistance and to the effect of phosphonate on these pathways.
Response of Plants to Phytophthora 377
II. PLANT-PATHOGEN INTERACTIONS

A. The Cowpea-Phytophthora Cryptogea Interaction
Phytoalexins are the best candidates as biochemical markers for the re-
sistance ofcowpea to fungalparasites (Lampard, 1974; Preston, 1975;
Preston et al., 1975; Partridge and Keen, 1976). Cowpea phytoalexins are
isoflavonoids, withthe exception ofthe 2-arylbenzofuran phytoalexin,vig-
nafuran (Preston et al., 1975). Bioassays on detached leaves ofthe suscepti-
ble cultivar Tvu 645 were usedto determine whether local concentrations of
phytoalexins at the infection site in compatibletreated leaves reach inhibi-
tory levels by the time the pathogen is inhibited. Inoculationof untreated
leaves produced light brown expanding lesions and hyphae had ramified
throughout the leaf tissues48 hr after inoculation (Fig. 1). In leaves floated
on buffered phosphonate solution (2.44 mM), dark brown necrotic lesions
500
a 300
%
m
100
0 12 24 36 48
Time after inoculation ( h )
Figure 1 Lesion areas caused by Phytophthora cryptogeain cowpea leaves ofthe

susceptible cultivar "Vu645 untreated (0)or treated ( 0 ) with phosphonate (2.44
mM). Data arerepresentative ofthree separate experiments; bars represent *SE; 12
infected siteswere observed in each experiment.The primary leaves weredetached,
inoculated on their adaxial side with a mycelial plug, and floated on a solution
containing 40 mM MES buffer adjusted to pH 6.5 and 50 mg/liter benzimidazole
with or without phosphonate. The assessment of the extent of infection was done
after clearing leaf discs in chloroform-ethanol-acetic acid (30:60:10) and staining
the tissues in a 0.002% Rose Bengalsolution.
378Guest and Saindrenan
developed. Fungal growth ceased about 24 hr after inoculation. In leaves

floating on buffer alone, small amountsof kievitoneand phaseollidin were
produced, while no vignafuran was detected (Fig. 2). A s infection pro-
ceeded, the concentration of both phytoalexins decreased. In phosphonate-
treated leaves, larger quantities of phytoalexins were produced in two dif-
ferent patternsof accumulation. Kievitone accumulation reacheda maximum
(134.5 pg/g freshwt) 24 hr after inoculation, decreasing afterward. Phase-
ollidin and vignafuran accumulated graduallyto levels of 160pg/g freshwt
and 9 pg/g fresh wt, respectively. These different patterns of the appear-
ance of the5-hydroxylatedisoflavonekievitone and the 5-deoxyiso-
flavonoid phaseollidin reflects a typical feature of induction of isoflavonoid
phytoalexinsinlegumes(Whiteheadet al., 1979; Robbins et al., 1985).
Cessation of fungal growth coincides with levels of kievitone which exceed
the EDwvalue (100 pg/ml)for fungal growth (Saindrenan et al., 1988a), so
that the involvement of kievitone in the restriction of P.cryptogea is likely.
Phaseollidin reaches inhibitory concentrations (EDw = 130pg/ml) at a
later stage of infection, while vignafuran, which has been reported to be
induced specifically in cowpea leaves infected with Colletotrichum linde-
muthianum (Preston et al., 1975), doesnot accumulate to a sufficient level
in treated leaves and cannot on its own be involved in the restriction of
fungal growth. Furthermore, cowpea leaves do not appear to produce the
cycloderivativeofphaseollidin,phaseollin,inresponse to infection by
P. cryptogea. This contrasts with cowpea hypocotyls infected by tobacco
- 10
-5
Hours after inoculation
Figure 2 Time-course accumulation of kievitone (a), phaseollidin @), and vigna-

furan (c) in untreated(0and treated ( 0 )cowpea leaves following inoculation with
P. ctyptogea. Hatched bars represent ED, values of phytoalexins determined in
vitro by radial growthbioassays.
From these results it was clear that phosphonate treatment enhanced

the rate and magnitude of phytoalexin accumulation in the necrotic cells of
the lesions. Whether a causal relationship existed between enhanced host
defenses and disease protection still remainedan open question. The first
answers came from studies showing that the levels ofphosphonate in lesions
where the pathogen was completely restricted were only sufficient to inhibit
mycelial growth by 25% in a medium containing a phosphatelevel similar
to that found in infected tissues (Saindrenan etal., 1988a). Phosphate, by
its structural analogy with phosphonate, has been shown to interfere with
the activity of phosphonates inplanta (Bompeix et al., 1980; Smillie et al.,
1989). Moreover, the infection did not create a metabolic sink leading to
the accumulation of the toxophore, which might then have caused a direct
inhibition of the fungus. Indeed, although phosphonate is poorly inhibitory
to P. cryptogea in vitro, especially in nonlimiting phosphate conditions
(Bompeix and Saindrenan, 1984), as it is the case duringthe infection court
in cowpea leaves where phosphate content is about 7 mM (Saindrenan et
al., 1988a), accumulation of the oxyanion at the site of infection might be
expected. ,
A second line of evidence was presented using inhibitors of biosynthetic

pathways of phytoalexins. a-Aminooxyacetate (AOA) is a competitive in-
hibitor of phenylalanine ammonia-lyase (PAL; EC 4.3.1.5) (Amrhein et al.,
1976),whichis thought to play an important role in the regulation of
isoflavonoid phytoalexin biosynthesis (Dixon et al., 1983). It was shown
that 5 mM AOA annulled the effect of phosphonate in the detached leaf
bioassay and that leavesexhibitedsymptoms of completesusceptibility
(Saindrenan et al.,1988b). Using tritiated phosphonic acid, Fenn andCof-
fey (1985) argued that the aminooxy compound might interfere with the
uptake of phosphonate by P. capsici, thus reducing the efficacy of the
toxophore in tomato-treated plants. However, no such effects were found
in either the mycelium of P. cryptogea or cowpea leaf (Saindrenan et al.,
1988b). It was subsequently shownthat under the experimental conditions
used by Fenn and Coffey (1985) an exchange of the phosphorus-bound
tritium of phosphonic acid withthe hydrogen of water occurred, making it
uncertain as to what was really measured (Bompeix et al., 1989).In a very
simple way, Bompeix (1989) established unequivocally that AOA did not
affect the activity of phosphonate against three different species of Phy-
tophthora in axenic culture. Because the reversal effect of AOA was not as
a result of the inhibition of phosphonate uptake, it might be expectedthat
its effect was through the modification of host defense reactions. In vitro
experiments showedthat AOA inhibitedPAL extracted from cowpea leaves
and the inhibition closely paralleledthe increase in susceptibility of leaves
to P. cryptogea (Saindrenan et al., 1988b):Five millimolar AOA reduces
Table 1 Influence of a-Aminooxyacetate (AOA) on Phosphonate-Induced

Accumulation of Phytoalexins and Levelsof Extractable Phenylalanine
Ammonia-lyase (PAL) in Infected Cowpea Leaves
Phytoalexins (pg/g fresh PAL activity'

(nmol cinnamate
Kievitone
Treatment' Phaseollidin h" mg" protein)
Control 40.9 k 13.9 45.2

75.1 f 12.2 f 15.5
+
Control AOA (5 mM) 12.3 f 8.4 1.3 f 4.4 19.5 f 2.4
Phosphonate (2.44mM) 96.1 f 24.5
90.1 * 18.1
92.9 & 20.9
Phosphonate (2.44 mM)
+ AOA (5 mM) 24.4 * 10.4
6.2 * 35.1
5.5 f 16.4
'Leaves were floated for4 hr in darkness on control or 5 mM AOA solutions before inocula-
tion. The leaves were then inoculated and placed on again
the same mediumbut with 2.44 m M
phosphonate.
bPhytoalexin contents of infected area were determined 24 hr after inoculation. Data are the
mean of two separate experimentsi SD.
T A L activity was assayed radiometrically. Data are the mean of two separate experimentsf
SD.
Source: From Saindrenan et al. (1988b).
kievitone and phaseollidin accumulation in phosphonate-treated leaves as

well as in untreated leaves (Table 1). These results indicatethat the decrease
of phytoalexin accumulation may be caused by inhibition of PAL activity
in vivo. Likewise, the reduction in phytoalexin content coincides with a
decrease in PAL activity. Indeed, AOA does not cause an increase in ex-
tractable PAL activity as reported for another inhibitor L-a-aminooxy-P-
phenylpropionic acid (AOPP) (Amrhein and Gerhardt, 1979), but rather a
substantial decrease which may be the consequence of the inhibition of
transaminase by AOA(John et al., 1978), resulting in loweredPAL synthe-
sis (Havir, 1981). Alternatively, AOA mayremaintightlybound to the
enzyme during extraction (Hoaglandand Duke, 1982).
The effectof AOA in reducing phytoalexin content is positively corre-
lated with the increase in susceptibility of cowpea leaves. These results
clearly demonstrate the active participationof the host defense reactions in
the mode of action ofphosphonate and, particularly in this model system,
of the phenylpropanoidswhich play an important role in the resistance of
some legumesto pathogens.
The primary target of phosphonate seems to be the pathogen. Indeed,
it was shown in some casesthat the in vivo performanceof the compound
paralleled its in vitro activity (Dolan and Coffey, 1988). Moreover, phos-
phonate was shown to enhance the capacity of P . cryptogea to overproduce
Response of Plantsto Phytophthora 381
molecules with elicitor activity in culture filtrates (Saindrenan et al., 1990)

and therefore to alter indirectlythe outcome of the plant-pathogen interac-
tion through a modification of fungalpathogenicity.Barchiettoet al.
(1992) suggested that phosphonate may interact with phosphate for the
catalytic site of phosphorylating enzymes, as well as for the regulation of
several enzymes, inducing a physiological state similar to that induced by
a limitation of phosphate, thus leading to the overproduction of elicitor
compounds.
B. The Tobacco-fhytophfhora nicotianae var. nicofianae Interaction

Resistant and susceptible cultivars of tobacco differ in the magnitude and
timing ofthe phytoalexin response following infection by P.nicotianae var.
nicotianae (Nemestothy and Guest, 1990). Sesquiterpenoid phytoalexins ac-
cumulate earlierand faster inthe resistant cultivar(NC 2326) than they do
in the susceptible cultivar (Hicks) (Fig. 3). Inhibition of sesquiterpenoid
biosynthesis by the specific inhibitor of HMG CoA reductase, mevinolin
150 c
0 12 24 36 48
Hours after inoculation
Figure 3 Dynamics of total sesquiterpenoidphytoalexin (capsidiol, phytuberin,

rishitin, and phytuberol) accumulation in tobacco stems inoculated with zoospores
of Phytophthoru nicotianae var. nicotiunae: CV.“Hicks” (0); +
CV.“Hicks” 10
mg/plant fosetyl-AI (W); CV.“NC 2326” (0) and CV.“NC 2326” + 10 mg/plant
fosetyl-Al(0).
Table 2 Effect of Fosetyl-Al(10 mg/plant) and Inhibitors on the Total

Sesquiterpenoid Phytoalexin Accumulation in Tobacco Stems (mg/g fresh wt.) 24
hr After Inoculation with Zoospores ofPhytophthora nicotianaevar. nicotianae
“Hicks” Cultivar + “NC 2326” +

fosetyl
“Hicks”
treatment ‘WC 2326” fosetyl
134 107
Control 50 114
Inhibitor:
8
Mevinolin 9 8
52 AOA 48 46 60
08 AHPP78 66 121
15 120 AVG 101 65
SED (6 replicates) = 6
AOA, d-aminooxyacetate; AHPP, L-a-aminohydrazino-&phenylpropionic acid; AVG, ami-
noethoxyvinylglycine.
Source: From Nemestothy and Guest(1990).
(Chappell and Nable,‘1987), and the nonspecific inhibitor or aromatic bio-

synthesis, AOA (Amrhein et al., 1976; John et al., 1978) inhibits phyto-
alexin accumulation and increases susceptibility in a leaf disc bioassay of
the resistant cultivar (Tables2 and 3) (Nemestothy and Guest,1990). These
results provide strong circumstantial evidence that sesquiterpenoid phyto-
alexin biosynthesis and accumulationis involved in the expressionof race-
specific resistance of tobacco to this pathogen. Specific inhibition of PAL
Table 3 Lesion Diameters (mm) 60 hr After Inoculation with Phytophthoru

nicotianae var. nicotianae of Leaf Discs Floated on40 mM MES Buffer Containing
1 mg/liter Benzimidazole, and, Where Indicated, Fosetyl-AI (282 PM), Mevinolin
(10 PM), AOA (100 PM), AHPP (100 FM), or AVG (30 PM)’
“Hicks” Cultivar + ‘WC 2326” +

treatment ‘%licks” fosetyl ‘WC 2326” fosetyl
6 12
Control 27 4
Inhibitor:
35 25
Mevinolin 28
AOA 2914 24 29
18 AHPP14 22 5
AVG11 22 4
2 POX25 25
SED (6 replicates) = 2
‘Propylene oxide (POX) treated discs were exposed
to 1 ml/liter for 1 hr before inoculation.
by L-a-aminohydrazino-0-phenylpropionic acid (AHPP; Munier and Bom-

peix, 1985) or ethylene by aminoethoxyvinylglycine (AVG;Paradies et al.,
1980) did not affect the susceptibility or resistance of leaf discs (Tables 2
and 3), suggesting at most a minor role for phenylpropanoids, such as
lignin, or for ethylene, in the early stages of resistance of tobacco to this
pathogen (Nemestothyand Guest, 1990).
We also demonstrated that, coincidingwithitsability to protect a
susceptiblecultivaragainst the blackshankpathogen, the phosphonate
compound fosetyl-Al enhances sesquiterpenoid phytoalexin accumulation
(Table 2) and other responses following inoculation (Guest,1986; Guest et
al., 1989; Nemestothy and Guest, 1990). Two pieces of evidence suggest a
causal link between enhanced host defenses and disease protection. First,
mevinolin and AOA inhibit sesquiterpenoid phytoalexin accumulation (Ta-
ble 2) and reduce the efficacy of fosetyl-Al (Table 3) in the susceptible
cultivar.AOAwouldalsoreducelignindeposition(Hammerschmidt,
1984). Second, at the rates applied and at the phosphate concentrations
present, the measured concentration of fosetyl-A1 or its toxophore, phos-
phonate, at the infection siteis only partially inhibitoryto mycelial growth
in vitro (Smillieet al., 1989).
Our results support the hypothesis that in this and some other host-
parasite interactions, enhanced speedand magnitude of phytoalexin accu-
mulation is involved, along witha direct obstruction of normal growth of
the pathogen, in the complex mode of action of phosphonates (Saindrenan
et al., 1988a; Smillie et al., 1989; Guest and Bompeix, 1990; Guest and
Grant, 1991). The relative importanceof the “direct”and “indirect” compo-
nents of the mode of action varies withthe characteristics of the particular
host-parasite pair studied, the application rates, and possibly environmen-
tal factors (Guest and Grant, 1991). Our hypothesis is supported by the
evidence that fosetyl-Al had no additive effect onthe resistance, or on the
phytoalexin, PAL, or ethyleneresponsesof the resistant cultivar, “NC
2326.” When phytoalexin biosynthesis was blockedthe in resistant cultivar,
lesion growth was still partially inhibited in fosetyl-Al-treated leaf discs,
indicating that factors other than phytoalexins are involved in the total
resistant response inthis cultivar. There is little significant directeffect of
phosphonate on fungal growth in planta, because there was no residual
inhibition of lesion growth in leaf discs of CV. “Hicks” taken from fo-
setyl-Al-treated plants followingtreatment with mevinolin, or in leaf discs
from treated plants subsequently killed by exposure to propylene oxide
(Nemestothy and Guest, 1990).
A primary effectof phosphonates on the pathogen may beto interfere
with its ability to elude phytoalexin elicitation (Guestand Bompeix, 1984),
perhaps through alterations to water-soluble cell wall componentsof Phy-
tophthora (Dunstan et al., 1990).Many other elementsof the resistant

response may be triggered by a phosphonate-affected pathogen in a resis-
tant host. By disturbing the pathogen’s recognition-avoidance mechanisms,
fosetyl-Al could enable the activation of an “idle capacity” the
in susceptible
cultivar, supporting the hypothesis that all plants have the potential to
express resistance (Kuk, 1983).
Phosphonates enhance plant responses normally associated with stress
and disease resistance. The suppression of sesquiterpenoid phytoalexin bio-
synthesis and, to a lesser degree, lignin deposition reduces the activity of
phosphonates. We propose that rapid phytoalexin accumulation,and possi-
bly lignin deposition, is involved in the mode of action of phosphonates
against this pathogen in tobacco, in combination with, and following, the
direct disturbance of the pathogen. Identification of the precise site@) of
metabolic disturbance caused by phosphonates may provide some insight
into the regulation of the host-parasite interaction.
C. Other Interactions
Fosetyl-Al was shown to control Phytophthoraparasitica infections of cap-
sicum fruits (Guest, 1984b). The control was associatedwithenhanced
hypersensitivity and increased capsidiol accumulation similar to that found
in genetically incompatible interaction. The Citrus-P. citrophthora interac-
tion is also well documented (Khanand RavisC, 1985; Afek and Sztejnberg,
1986; Khan etal., 1986). It was found that low concentrations of fosetyl-A1
and phosphonate act againstP. citrophthora by increasing defense mecha-
nisms against the pathogen and accumulationof the 6,7dimethoxycoumarin
phytoalexin, scoparone, while higher concentrations of phosphonates act
directly as a fungistat (Afek and Sztejnberg, 1989). Dercks and Buchen-
auer’s (1986)study of the strawberry-P.fragariae and lettuce-Bremia Iactu-
cae interactions led themto claim that increases in phenolic compounds in
plants treated with fosetyl-Al were of secondary importance in disease con-
trol. However, the high application rates of fosetyl-A1 used by theseauthors
may explain the restriction of fungal growth via a direct action of the
compound on the pathogen, possibly throughthe toxicity of the aluminum
ion (Muchovejet al., 1980). Smillie et al. (1989) demonstrated that in lupin,
which lacks a highly active dynamic defense system, phosphonate, even
at high concentrations in the plant, is unable to stop the growth of P.
cinnamomi.
111. EPILOG
The evidencethat phytoalexins contribute to plant disease resistance comes
from two types of investigations that provide either circumstantial and
correlative evidenceor experimental evidence.
hytophthora
Response
to of Plants 385
There is a large body of circumstantial evidence that resistant plants

accumulate more phytoalexins at the site of attempted infection, faster than
susceptible plants (Dixon, 1986; Preisig and Kud,1987; Nemestothy and
Guest, 1990). The most convincing evidence comes from studies of near
isogenic cultivars ofthe same species suchas soybeans, cowpea,or tobacco.
There are now several lines of experimental evidencethat phytoalexins
are central to disease resistance in some plants. Van Etten et al. (1989)
demonstrated an absolute correlation between virulence and phytoalexin
detoxification by the necrotrophic Nectria haematococca. They were able
to transform avirulent fungi withthe pisatin demethylase gene, which also
conferred virulence.
Treatments that inhibit phytoalexin biosynthesisor accumulation, such
as those described in this chapter, increase the susceptibility of the plant to
disease (Moesta and Grisebach, 1982; Massala et al., 1987; Saindrenan et
al., 1988b; Nemestothy and Guest, 1990). Conversely, treatments that en-
hance phytoalexin accumulation increase disease resistance (Ward, 1984;
Saindrenan et al., 1988a; Nemestothy and Guest, 1990). A model for the
mode of action ofthe phosphonates has been proposed (Guest and Grant,
1991; Barchietto et al., 1992). In this model, sublethal concentrations of
phosphonate, such as those commonly encountered in treated plants, in-
duce a stress physiology in oomycetes. One consequence of this is that
factors that normally suppress the host recognition system are not pro-
duced, or fail to function, and that elicitors are either overproduced or
become more active.The net result isthat the plant recognizes the pathogen
as an incompatible invader and invokes its normal array of defense re-
sponses asa consequence.
These experiments open several avenues of investigation. First, it will
be instructive to learn the precise mechanism by which phosphonates dis-
turb the virulence mechanisms of Phytophthora spp. The loose selectivity
of phosphonates for the oomycetes suggesta distinctive metabolism,proba-
bly phosphate metabolism. Thefact that interference with this metabolism
affects the virulence of these pathogens suggests that further exploration
may provide information about the virulence factors of these organisms,
and thus the factors that control phytoalexin accumulation inthe host. It
may also opennew avenues of diseasecontrol based on the exploitation of
preexisting but dormant mechanisms.
Despite the accumulatedevidence,phytoalexins are not a universal
resistance mechanism, nor should they be viewed in isolation from other
plant disease resistance mechanisms, with which they certainlyact in syn-
ergy. Phytoalexins have not been found in some plants which have well-
described disease resistance mechanisms, such as wheat or cucumber. Al-
though there is often a strong link between gene-for-gene resistance and
rapid phytoalexin accumulation, no mechanism for this linkage has been

described. There has beenno report of a gene encoding rapid phytoalexin
accumulation, nor has there been a report of a “phytoalexin-minus”mutant
plant that is consequently susceptible
to disease.
REFERENCES
Afek, U., and Sztejnberg,

A. (1986). A Citrus phytoalexin, 6,7-dimethoxy-
coumarin, as a defense mechanismagainst Phytophthora citrophthora,and the
influence of Aliette and phosphorous acid on its production. Phytoparasitica
14:26.
Afek, U.,and Sztejnberg, A. (1989). Effects of fosetyl-Al and phosphorous acid on
citrophthora. Phytopathology79736-739.
Amrhein, N., and Gerhardt, J. (1979). Superinduction of phenylalanine ammonia
lyase ingherkinhypocotyls caused by the inhibitor, h-aminooxy-&phenyl-
propionic acid. Biochimica and Biophysica Acta 583:434-442.
Amrhein, N.,Godeke, K. H., and Kefeli, V. 1. (1976). The estimation of relative
intracellular phenylalanine ammonialyase (PAL) activities and the modulation
in vivo and in vitro by competitive inhibitors. Berichte des Deutschen Botani-
schen Gesellshaft89247-259.
Bailey, J. A. (1973). Production of antifungal compounds in cowpea (Vignasinen-
sis) and pea (Pisum sativum)after virus infection. J. Gen. Microbiol. 75:119-
125.
Barchietto, T.,Saindrenan, P., and Bompeix, G. (1992). Physiological responses of
Phytophthora citrophthora to a subinhibitory concentration of phosphonate.
Pest. Biochem. Physiol.42:151-166.
Bertrand, A., Ducret, J., Debourge, J. C., and Horrikre, D. (1977). Etude des
proprittts d’une nouvelle famillede fongicides: les monotthyl phosphites mtt-
alliques. Caracttristiques physicochimiques et proprittts biologiques. Phytid-
trie-Phytopharmacie263-1 8.
Bompeix, G . , and Saindrenan, P. (1984). In vitro antifungal activity of fosetyl-A1
and phosphorous acid on Phytophthora species. Fruits 39777-786.
Bompeix, G., Ravist, A., Raynal, G., Fettouclie, F., and Durand, M-C. (1980).
Modalites de l’obtention des ntcroses bloquantes sur feuilles dttachtes de to-
mate par l’action du tris-0-ethylphosphonate d’aluminium (phostthyl d’alumi-
nium), hkothtses sur son mode d’action in vivo.Annales de Phytopathologie
12:337-351.
Bompeix, G., Fettouche, F., and Saindrenan, P. (1981). Mode d’action du phos-
tthyl-Al. Phytidtrie-Phytopharmacie30:257-272.
Bompeix, G. (1989). Fongicides etrelations plates-parasites: cas des phosphonates.
Comptes Rendusde I’AcadPmie #Agriculture6 183-189.
Bompeix, G., Burgada, R., d’Arcy-Lameta, A., and Soulit, M-C. (1989). Phospho-
nate hydrogen exchangein aqueous solutions of *Hand 3H-labelledphosphonic
acid. Pest. Sci. 25:171-174.
hytophthora
Response
to of Plants 387
Chappell, J., and Nable, R. (1987). Induction of sesquiterpenoid biosynthesis in

tobacco cell suspension culturesby fungal elicitor.Plant Physiol. 85469473.
Cohen, Y., and Coffey, M. D. (1986). Systemicfungicides and the control of
oomycetes. Ann. Rev. Phytopathol. 24:311-338.
Dercks, W., and Buchenauer, H.(1986). Investigations on the influence of alumin-
ium ethyl phosphonate on plant phenolic metabolism in the pathogen-host
interactions, Phytophthora fragariae-strawberry and Bremia lactucae-lettuce.
J. Phytopathol. 115:37-55.
Dixon, R.A. (1986). The phytoalexin response: elicitation, signallingand control of
host gene expression. Biol. Rev. 61:239-291.
Dixon, R. A., and Lamb, C. J. (1990). Molecular communication in interactions
between plants and microbial pathogens.Ann. Rev. Plant Physiol. Plant Mol.
Biol. 41:339-367.
Dixon, R. A., Dey, P. M.. and Lamb, C. J. (1983). Phytoalexins: enzymology and
molecular biology. Adv. Enzymol. Rel. Areas Mol. Biol. 550:1-135.
Dolan, T. E., and Coffey, M. D. (1988). Correlative in vitro and in vivo behaviour
of mutant strains of Phytophthora palmivoraexpressing different resistances
to phosphorous acidand fosetyl-Na. Phytopathology 78:974-978.
Dunstan, R. H.,Smillie, R. H., and Grant, B.R. (1990). The effects of sub-toxic
levels of phosphonate on the metabolism and potential virulence factors of
Phytophthora palmivora.Physiol. Mol.Plant Pathol. 36205-220.
Durand, M.-C., and SallC, G. (1981). Effet du tris-0-6thyl phosphonate d’alumi-
nium sur le couple Lycopersicum esculentum Mill.-Phytophthora capsici Leon.
Etude cytologique et cytochimique. Agronomie 1:723-731.
Fenn, M. E., and Coffey, M. D. (1984). Studies on the in vitro and in vivo antifun-
gal activity of fosetyl-Aland phosphorous acid.Phytopathology 74x36-611.
action of fosetyl-A1and phosphorous acid.Phytopathology 75:1064-1068.
Guest, D.I. (1984a). The influence of cultural factors on the direct antifungal
activities of fosetyl-Al, Propamocarb, Metalaxyl, SN75196 and Dowco 444.
Phytopathol. Z.111:155-164.
Guest, D. I. (1984b). Modification of defense responses in tobacco and capsicum
following treatment with fosetyl-Al. Physiol.Plant Pathol. 25: 125-134.
Guest, D. I. (1986). Evidence from light microscopy of living tissues that fosetyl-AI
modifies the defence response in tobacco seedlings following inoculation with
Phytophthora nicotianae var. nicotianae. Physiol. Mol. Plant Pathol. 29151-
161.
Guest, D. I., and Bompeix, G. B. (1984). Fosetyl-Al as a tool in understanding the
resistant response in plants.Phytophthora Newslett. 12:62-69.
Guest, D. I., and Bompeix, G. (1990). The complex mode of action of phospho-
nates. Aust. PlantPathol. 19113-115.
Guest, D. I., and Grant, B. R. (1990). The complex action of phosphonates in
plants. Biol. Rev. 66159-187.
Guest, D. I., Upton, J. C.R., and Rowan, K. S. (1988). Fosetyl-Al alters the
respiratory response in Phytophthora nicotianae var. nicotianae-infected to-
bacco. Physiol. Mol.Plant Pathol. 34257-265.
388 Saindrenanand Guest
Hammerschmidt, R. (1984). Rapid deposition of lignin in potato tuber tissue as a

response to fungi non-pathogenicon potato. Physiol. Plant Pathol.2433-42.
Havir, E. A. (1981). Modification of L-phenylalanine ammonia lyase in soybean
cellsuspensionculturesby2-aminooxyacetateandL-2-aminooxy-3-phenyl-
propionate. Planta 152:124-130.
Havir, E. A., and Hanson, K. R. (1968). Phenylalanine ammonia lyase.I. Purifica-
tion and molecular size of the enzyme from potato tubers. Biochemistry 7:
1896-1 899.
Hoagland, R. E., and Duke, S. E. (1982). Effects of glyphosate on metabolism of
phenolic compounds. VIII.Comparison of the effects of aminooxyacetateand
glyphosate. Plant Cell Physiol.23:1081-1088.
John, R. A., Charteris, A., and Fowler, L. J. (1978). The reaction of aminooxyace-
tate with pyridoxal phosphate-dependent enzymes. Biochem. J. 171:771-779.
Khan, A. J., and Ravist, A. (1985). Stimulation of defence reactions in Citrus by
fosetyl-AI and fungal elicitors against Phytophthora spp. Br. Crop Prot. Coun-
cil Monogr. 31:281-284.
Khan, A. J., Vernenghi, A., and RavisC,A.(1986). Incidence of fosetyl-A1 and
elicitors on the defence reactions of Citrus attacked by Phytophthora spp.
Fruits 41587-595.
Kiraly, Z., Barna, B., and Ersek, T. (1972). Hypersensitivity as a consequence, not
the cause, ofplant resistance to infection. Nature 239456-458.
KuC, J. (1982). Phytoalexins from the Solanaceae. In Phytoalexins (J. A. Bailey and
J. W. Mansfield, eds.), Blackie,London, pp. 81-105.
KuC, J. (1983). Induced systemic resistanceof plants to disease caused by fungiand
bacteria. The Dynamics of Host Defence, (J. A. Bailey and B. J. Deverall,
eds.), Academic Press, Sydney, pp. 191-221.
KuC, J., and Rush, J. S. (1985). Phytoalexins. Arch. Biochem. Biophys. 236455-
472.
Lampard, J. F. (1974). Demethylhomopterocarpin: an antifungal compoundin
Canavalia ensiformis and Vigna unguiculatafollowing infection.Phytochemis-
try 13:291-292.
Langcake, P. (1981). Alternative chemical agentsfor controlling plant disease. Phil.
Trans. R . Soc. London B295:83-101.
Langcake, P., Cartwright, D. W., and Ride, J. P. (1983). The dichlorocyclopro-
panes and other fungicides withan indirect modeof action. Systemische Fungi-
zide und Antifungale Verbindungen (H. Lyr and C. Polter, eds.), Akademie-
Verlag, Berlin, pp. 199-210.
Larazovits, G., and Higgins, V. J. (1975a). Histological comparison of Cladospor-
ium fulvum race 1on immune, resistant and susceptible tomato varieties. Can.
J. Bot. 54224-234.
Larazovits, G., and Higgins, V. J. (197513). Ultrastructure of susceptible, resistant
and immune reactions of tomato to races of Cladosporiumfulvum. Can. J.
Bot. 54~235-249.
Legrand, M., Fritig, B., and Hirth, L. (1976). Enzymes of the phenylpropanoid
pathway and the necrotic reaction of hypersensitive tobacco to tobacco mosaic
virus. Phytochemistry 15:1353-1359.
Massala, R., Legrand, M., and Fritig, B. (1987). Comparative effects of two com-
petitive inhibitors of phenylalanine ammonia lyase on the hypersensitive resis-
tance of tobacco to tobacco mosaic virus.Plant Physiol. Biochem. 25:217-225.
Mktraux, J. P., Ahl-Goy, P., Staub, T., Speich, J., Steinemann, A.,Ryals, J.,
and Ward, E. (1991). Induced systemic resistance in cucumber in response to
2,6-dichloro-isonicotinicacid and pathogens. Advances in Molecular Genetics
ofPlant-MicrobeInteractions, Vol. 1 (H.Hennecke and D. P. S. Verma,
eds.), Kluwer, Dordrecht, pp. 432-439.
Moesta, P., and Grisebach, H.(1982). L-2-Aminooxy-3-phenylpropionicacid inhib-
its phytoalexin accumulation in soybean with concomitant loss of resistance
against Phytophthora megasperma f. sp. glycinea. Physiol. Plant Pathol. 21:
65-70.
Munier, R. L., and Bompeix, G. (1985). Inhibition de la L-phenylalanine ammonia
lyase de Rhodotorula glutinis par des phenylalanines N-substitudes. Comptes
Rendus de I'Acadgmie des Sciences Paris300203-206.
Muchovej, J. J., Maffia, L. A., and Muchovej, R. M. C. (1980). Effects of ex-
changeable soil aluminiumand alkaline calcium saltson the pathogenicity and
growth of Phytophthora capsici from green pepper. Phytopathology 701212-
1214.
Nemestothy, G. N., and Guest, D. I. (1990). Phytoalexin accumulation, phenylala-
nine ammonia lyase activity and ethylene biosynthesis in fosetyl-A1 treated
resistant and susceptible tobacco cultivars infected withPhytophthora nicotia-
nae var. nicotianae. Physiol.Mol. Plant Pathol.32207-219.
Paradies, I., Konze, J. R., Elstner, E. F., and Paxton, J. D. (1980). Ethylene:
Indicator but not inducer of phytoalexin synthesis in soybean.Plant Physiol.
661106-1 109.
Partridge, J. E., and Keen, N. T.(1976). Association of the phytoalexin kievitone
with single-gene resistance of cowpeas to Phytophthora vignae. Phytopathol-
ogy 66426-429.
Pegg, K. G., Whiley, A. W., Saranah, J. B., and Glass, R. J. (1985). Control of
Phytophthora root rot of avocado with phosphorous acid. Aust. Plant Pathol.
14~25-29.
Preisig, C. L., and KuC, J. A. (1987). Phytoalexins, elicitors, enhancers, suppressors
and other considerations in the regulation of R-gene resistance to Phytophthora
infestans in potato. Molecular Determinants of Plant Disease (S. Nishimura et
al., eds.), Japan Science SocietyPresdSpringer-Verlag, pp. 203-221.
Preston, N. W. (1975). 2'-O-methoxyphaseollidinisoflavanfrom infected tissue of
Vigna unguiculata. Phytochemistry 14:1131-1132.
Preston, N. W., Chamberlain, K., and Skipp, R. A. (1975). A 2-arylbenzofuran
phytoalexin from cowpea (Vigna Unguiculata). Phytochemistry 141843-1844.
Robbins, M. P., Bolwell,G. P., and Dixon, R. A. (1985). Metabolicchanges
in elicitor-treated bean cells.Selectivity of enzyme induction in relation to
phytoalexin accumulation.Eur. J. Biochem. 148563-569.
Saindrenan, P., Barchietto, T., Avelino,J., and Bompeix,G. (1988a). Effects
of phosphite on phytoalexin accumulation in leaves of cowpea infected with
Phytophthora cryptogea. Physiol. Mol. Plant Pathol. 32425-435.
Saindrenan, P., Barchietto, T., and Bompeix,G. (1988b). Modificationof the

phosphite induced resistance responsein leaves of cowpea infected with Phy-
tophthora cryptogeaby a-aminooxyacetate. Plant Sci.58245-252.
Saindrenan, P., Barchietto, T., and Bompeix, G. (1990). Effects of phosphonate on
the elicitor activity of culture filtrates of Phytophthora cryptogea in Vigna
Unguiculata. Plant Sci.62245-251.
Sekizawa, Y., and Mase, S. (1981). Mode of controlling action of probenazole
against rice blast disease with reference to the induced resistance mechanism in
rice. J . Pest. Sci. 691-94.
Smillie, R. H.,Grant, B. R., and Guest, D. I. (1989). The mode of action of the
fungicide phosphite: evidence for both direct and indirect modes of action in
plants. Phytopathology 79921-926.
van Etten, H. D.,Matthews, D. E., and Matthews, P. S. (1989). Phytoalexin
detoxification: Importance for pathogenicity and practical implications.Annu.
Rev. Phytopathol. 27: 143-164.
Vo-Thi, H.,Bompeix, G., and Ravist, A. (1979). R61e du tris-0-tthylphosphonate
d'aluminium dans la stimulation des rtactions de dtfense des tissus de tomate
contre le Phytophthora capsici. Comptes Rendus de I'AcadPmie des Science
Paris2881171-1174.
Ward, E. W. B. (1984). Suppression of metalaxyl activity by glyphosate: evidence
that host defence mechanisms contribute to metalaxyl inhibition of Phytophth-
ora megasperma f. sp. glycinea in soybeans. Physiol. Plant Pathol. 25:381-
386.
Whitehead, I. M., Dey, P. M., and Dixon, R. A. (1979). Differential patterns of
phytoalexin accumulation and enzyme induction in wounded and elicitor-
treated tissues ofPhaseolus vulgaris. PIanta154156-164.
Woloshuk, C. P., Sisler, H., Tokousbalides, M. C., and Dutky, S. R. (1980).
Melanin biosynthesis in pVricularia oryzae: site of tricyclazole inhibition and
pathogenicity of melanin-deficient mutants. Pest. Biochem. Physiol. 14:256-
264.
Ye, X. S., and Deverall, B. J. (1986). Effects of Aliette on Bremalactucae on
lettuce. Trans. Br. Mycol. Soc. 86597-602.
17
Phytoalexins in Forage Legumes
Studies on Detoxification by Pathogens
and the Role of Glycosidic Precursors
in Roots
V. J.Higgins and J.Hollands
Universityof Toronto, Toronto, Ontario, Canada
Dallas K. Bates
Michigan Technological University, Houghton,Michigan
1. INTRODUCTION
Until recently, our research on phytoalexins concentrated on various as-
pects of therole of thephytoalexinsmedicarpin(3-hydroxy-9-methoxy-
pterocarpan)(l),ormedicarpin and maackiain (3-hydroxy-8,9-methylene-
dioxypterocarpan(2), in foliage of alfalfa and in foliage and roots of red
clover,respectively.Withfoliage, the initial objective was to determine
whether phytoalexins might explain differences in host specificity and also
whether they mightbe involved in limiting lesion size in leaf spot diseases.
The work with roots arose initially out of our curiosity about the origins
and relationships of phytoalexins inthe roots as compared withthe leaves.
Unfortunately, because of our dissatisfaction with the genetics of these
crops, our interest gradually turned to leaf mold of tomato leaving many
interesting questionsabout these forage legume systems unanswered; how-
ever, in thischapter, we take the opportunity to briefly review some ofthat
work and to present some as-yet unpublished results.
The general picture that emerged initially from work with alfalfa was
that pathogens of alfalfa generally could rapidly convert medicarpin to
nontoxic components (Higgins and Millar, 1969; Higgins, 1972) whereas
391
392 Higgins et al.
nonpathogens lacked that ability or produced a derivative that was still

toxic (Higgins and Millar, 1970). Further work with the alfalfa pathogen
Stemphylium bottyosum showed that it could also metabolize the phyto-
alexinmaackiain from redclover, a host on which this fungus is only
weakly virulent (Duczek and Higgins, 1976a), and pisatin and phaseollin
from peas and beans, two nonhosts of S. bottyosum (Heath and Higgins,
1973). Although on peas and beans the slowerrate of phytoalexin metabo-
lism appeared to allow enough ofthe compounds to accumulate to account
for the inhibition of S. botryosum, this was not the case on red clover.
Thus,medicarpin and maackiain did not appear from thiswork to be
involved in the low virulence of this fungus on red clover. Interestingly,
phaseollin, maackiain, and medicarpin were all reduced to their respective
isoflavans (Higgins et al.,1974; Higgins, 1975) bya system, which at least
for maackiain, could be induced by any of the three compounds.
As pointed out by Van Etten et al. (1982, 1989), any means of interfer-
ing with the pathogen’s ability to detoxify phytoalexins could affect the
pathogenicity of the pathogen.In this regard,we tested S-containing ptero-
carpans for their ability to inhibit the reduction of maackiain to dihydro-
maackian (DHMaa) (7,2’-dihydroxy-4’ ,5 ’-methylenedioxyisoflavan)(3).
Our hypothesis was that reduction of the S-containing molecule by the

pterocarpan oxidoreductase would create an -SH derivative which might
bind irreversiblyto the enzymeand thus act as a “suicide substrate” (Walsh,
1983). Alternatively, even if the S-containing pterocarpans simply proved
to be good competitive inhibitors, they could be useful tools for studying
the importance of phytoalexin detoxification planta.
in As well, knowledge
of such potential enzyme inhibitors might be of potential commercial value
in diseasecontrol. The choice ofthis system was influenced,not only by the
fact that we had a ready source of maackiain as the result of a gift of
trifolirhizin from Albert Stoessl (formerly Agricultural Canada Research
Institute, London, Ontario), butalsobecauseusingthe drop diffusate
method with this funguson clover leaflets wouldbe a convenient meansof
eventually testing inhibition in planta. Some results from this project are
discussed below.
An intriguing aspectof maackiain production in red clover is the fact
that the glycoside of maackiain, trifolirhizin, occurs in roots in relatively
high concentration yet no more than 1-2 pg/g fresh wt of tissue was found
n Phytoalexins Forest Legumes 393
in foliage (Duczek and Higgins, 1976a). Previously, we reported (McMur-

chy and Higgins, 1984) that trifolirhizin appeared to be a source of the
phytoalexin maackiain in red cloverroots infected with Fusarium roseum.
The production of pterocarpanoid phytoalexins from preformed glycosides
or other isoflavone conjugates has recently received renewed attention, par-
ticularly in soybean (Graham et al., 1990; Morris et al., 1991). Below are
summarized results of a study designed not only to clarify the source of
maackiain in red cloverroots but to investigate the presence of a compara-
ble glycosideof medicarpin.
II. METHODOLOGY
Fungal Metabolism of Phytoalexins byStemphyrium bofryosum:
Inhibition by S-ContainingPterocarpanoid
The chemicals testedas inhibitors of maackiain conversionto DHMaa were
synthesized by D. K. Bates, D. A. Hay, andJ. A. Fleming (Michigan Techno-
logical University, Houghton, Michigan) and were derivatives of 6a,l la-
dihydrobenzo[3,2-c][llbenzopyrans (1 la-thioptercarpans) including 11-
thiapterocarpan (TP) (4), 3-methoxy-l l-thiapterocarpan (3-OMe-TP) (9,
3-hydroxy-ll-thiapterocarpan (3-OH-TP) (6), and dihydrobenzothiophene
(DHBTP)(7).
4)R=
5)'R =HOCH3 \ %
R \
6)R=OH 7
Maackiain conversion assays were basically carried

out as in Higgins (1975).
Maackiain,DHMaa, TP, and 3-OMe-TPwereanalyzedbyhigh-perform-
ance liquidchromatopraphy (HPLC) using a C,*column and 70% acetoni-
trile in water. Bioassays of each compound were on
made
germ tube growth
as described in Duczek and Higgins (1976b).
Phytoalexin Production from Preformed Glycosides

Maackiain and medicarpin were measured by HPLC as above, or by UV
absorbance of compounds elutedfrom thin layer chromatography plates as
describedinMcMurchyandHiggins(1984).Maackiain and medicarpin
were removed fromroot extracts by partitioning with CC&, which does not
remove the glycosides; the extracts were then treated with a 0-glucosidase
and the maackiain and medicarpin released by hydrolysis of the glycosides
were extractedand measured.
394 Higgins et al.
111. METABOLISM O F PHYTOALEXINS BY

STEMPHYLlUM BOTRYOSUM INHIBITION
BY S-CONTAINING PTEROCARPANOIDS
As all inhibitor tests were to be done in situ with germinated spores, this
required that the S-pterocarpan derivatives tested have no or very little
toxicity to fungalgrowth; this requirementeliminated3-OH-TP and
DHBTP which were as toxic as maackiain (Fig. 1). Fortunately, TP and
3-OMe-TP were less toxic (Fig. 1) and the toxic effect did not change
as the
concentration increased.
The effect of TP, the compound chosen for detailed study, on the
conversion of maackiain(at 10 pg/ml) by germinated conidiaof S. botryo-
sum was monitoredover an 8-hrperiodduring which about 50%
loss of maackiain normally occurred in untreated controls (Fig. 2). Treat-
ment of conidia withTP generally reducedthe maackiain lossto about 20-
30% but again this effect was not significantly concentration-dependent
(Fig. 2). Maackiain loss over the 8-hr period occurred at a relatively con-
stant rate for untreated controls (Fig. 3) with DHMaa detectable by 2 hr.
TP-treated conidia showed a less uniform loss of maackiain (Fig. 3) and
M M DHBTP 3-OMe-TP TP 3-OH-TP
Figure 1 Effect of maackiain (Maa), dihydrobenzothiophene(DHBTP), 3-methOxy-

1 l-thiapterdcarpan (3-OMe-W). 1 l-thiapterocarpan (TP), and 3-hydroxy-l l-thiap-
terocarpan (3-OH-TP) on germ tube growth of Stemphylium botryosum conidia
pregerminated before treatment with 5, 10, or 20 pg/ml of each compound. Germ
tubes (n = 100) were measured after a 4-hr incubation and the average increase in
growth is presented as a percentageof the untreated control.
Phytoalexins in Forest legumes 395
100
l
401
30
0 10 20 30 40 50 60
1 l -MIA!JTEROCARPAN (yg/rnl)
Figure 2 Effect of increasing concentrations of 11-thiapterocarpan (TP) on the

conversion of maackiain to dihydromaackiain by Stemphylium bottyosum after 8
hr incubation with 10 pg/ml of maackiain. Recovery of maackiain is shown as a
percentage of that recovered from conidial suspensions immediatelyafter maackiain
addition (zero time).
J-
O T
0 2 4 6 8
TIME (HOURS)
Figure 3 Kinetics of maackiain (added to give 10 pg/ml) conversion to dihydro-

maackiain by Stemphylium bottyosum during an 8 hr period in the presence or
absence of1l-thiapterocarpan (TP) at 10 pg/ml. Recovery of maackiain is shown as
a percentage of the maackiain recovered immediatelyafter maackiain addition (zero
time).
396 Higgins et al.
DHMaa was first detected at 4 hr.(Otherexperimentsindicated that

DHMaa is further metabolizedby this fungus; thus thekineticsof
DHMaa concentration couldnot be used to monitor the rate of maackiain
conversion.) About 90% of addedTP was recovered from cultures through-
out the incubation period indicating that TP was not metabolized by the
fungus and that it was not, or at least not irreversibly, changed duringthe
interaction with the enzyme. Thus, it does not appear to act as a typical
suicide substrate.
In anin situ experiment, suchas the above, it is impossibleto differenti-
ate between inhibition of enzyme synthesis vs. enzyme activity. One ap-
proach to answering this question took advantage of our knowledge about
the inducibilityof the S.botryosurn-maackiain conversion system which is
assumed to be dueto induction of the synthesis ofan oxidoreductase (Hig-
gins, 1975). Maackiain (to give 10 pg/ml) or an equivalent volume of sol-
vent was addedto pregerminated spores whichwere incubated for a further
17 hr before the addition of more maackiain (to give 20 pg/ml) with or
without TP. If TP inhibits enzyme synthesisrather than activity, the effect
of TP on maackiain conversion should be decreased. Inhibition of maacki-
ain by TP still occurred in induced conidia (Fig.4) suggesting that enzyme
TIME (hours)
Figure 4 Effect of preinducing Stemphylium botryosum with maackiain on the

ability of 11-thiapterocarpan (TP) to inhibit the conversion of maackiain to dihy-
dromaackiain. Pregerminated conidia were incubated with maackiain (10 pg/ml)
for 8 hr before the addition of more maackiain (20 pg/ml) and TP. Recovery of
maackiain is shown as a percentage of the maackiain recovered immediately after
the additionof the second maackiain treatment (zero time as graphed).
legumes
Phytoalexins in Forest 397
activity is affected; however, interpretation of these data are complicated

by an array of factors including the inhibitory effects of both compounds
on growth. Confirmation of a direct interaction of TP with the enzyme
must await purificationand characterization of the enzyme. A soluble con-
stitutivelyproducedpterocarpan:NADPHoxidoreductaseisolatedfrom
Ascochyta rabiei has been shownto reduce both medicarpin and maackiain
to their respective isoflavans(Hohl and Barz, 1987; Hohl et al., 1989). The
enzyme from S. botryosurn, although induced, is probably very similar in
activity and merits further investigation.
Despite our lack of understanding of the mode of inhibition, some
preliminary tests were made to determine if TP was effective in planta.
Maackiain and medicarpin usually do not accumulate beyonda few pg/ml
when red clover leaves are inoculated with S. botryosurn, presumably be-
cause of the rapid metabolism of these compounds. Preliminary experi-
ments to test the effectof TP on maackiain and medicarpin accumulation
on S. botryosurn-inoculated clover leaflets indicated that TP was not effec-
tive in planta, i.e., the levels of maackiain and medicarpin in drop diffu-
sates containing TP and the fungus were not markedly different from the
combined total for either alone. Such data must be treated with caution,
however, because of the possibility that TP might also inhibit enzymes
involved inthe final steps of pterocarpanoid synthesis. One such candidate
enzyme is NADPH:isoflavone oxidoreductase which has been isolated from
several plants including chickpea (Tiemann et al., 1991). In situ inhibition
of a plant enzyme could be explored using abiotic elicitors or fungi that do
not metabolize maackiain.
W. PHYTOALEXINPRODUCTIONFROMPREFORMEDGLYCOSIDES
Challenge of roots of axenically grown red clover seedlings by the abiotic
elicitor CuClzwas used to clarify the role of trifolirhizin, and a comparable
glycoside of medicarpin, in the production of maackiain and medicarpin,
respectively. The kinetics (Fig. 5A) of the loss of trifolirhizin and increase
in maackiain inthe 24-hr period followingCuC12treatment were compara-
ble to those seen(McMurchy and Higgins, 1984) following inoculation
with F. roseurn. Although a glycoside of medicarpin was not isolated and
characterized, techniques similarto those used to monitor trifolirhizinlev-
els indicated the presence of such a glycoside. The kinetics (Fig. 5B) of
glycoside loss and appearance of medicarpin following CuC1, treatment
were very similarto those for trifolirhizin and maackiain.
To further test if maackiain and medicarpin were originating from their
respective glycosides as a result of &glucosidase activity in the injured
tissue, as suggested by the kinetic data, glyphosate, an inhibitor of the
398 Higgins et al.
1.4,
-W
L
2 1.0
z,
'5 0.8
4
0 4
0 .C
0
- ..""_
0.6l b . . -
- ..-..- ..
2 $
iial
0.4
5o *E 0.2 - ..-.._
e.3
-
z 0.0
0 12 24
TIME (hours)
Figure 5 (A) Changes in trifolirhizin (measured as the aglycone maackiain) and

maackiainconcentration in redclover roots treatedwithCuCl,M)andincu-
bated 12 or 24 hr. (B) Comparable changes in a glycoside (Med gly) of medicarpin
(measured as the aglycone) and medicarpin inthe same roots.
shikimate pathway (Hollander and Amrhein, 1980), was used on CuC1,-

treated roots. If maackiain and medicarpin are formed by de novo synthe-
sis, then treatments with glyphosate should show reduced amounts of these
compounds by 24 hr. The data (Fig. 6A,B) for both maackiain and medic-
arpin indicate that the presence of glyphosateat 10 pg/ml didnot affect the
relative glycoside/phytoalexin ratio normally seen after CuCl, treatment.
This suggests that the contribution of de novo synthesis to phytoalexin
synthesis in the initial 24-hr period is at most very limited. Similar experi-
ments withthe phenylalanine ammonia lyase (PAL) inhibitor aminooxyace-
Phytoalexins in Forest Legumes 399
1.4 -
n
c
3
A D Moockiain -
Z 1.2 - m Trifolirhizin
?TI
5O\ 1.0
-
"- .1 -
W 3 0.8
<E
z c
3.5 0.6 -
E '3
-
3; 0.4
-
0.2
v
0.0 n m n
rB
ll D Medicarpin
m Medicarpin glycoside
A con
trol CuC12 control
+.
GP
cu+
c1.2
GP
Figure 6 (A) Effect of the glyphosate (GP) (10 pg/ml) on CuC12-inducedchanges

intheconcentration of trifolirhizin(measured as theaglyconemaackiain)and
maackiain in red clover roots. (B) Comparable effects of glyphosate on concentra-
tions of medicarpin glycoside and medicarpin the
in same roots.
tic acid (AOA), in which only trifolirhizin/maackiain levels were moni-

tored, gave similar results.
Interestingly, and although difficult
to substantiate statistically because
of the high variation in glycoside levels in unelicited plants, the level of
glycoside inAOA- or glyphosate-treatedcontrols appeared to decrease over
the 24-hr incubation period. This suggests that there may be a constant
turnover of glycosideand that in the presence of an inhibitor such asAOA
or glyphosate this loss is not replaced and results in a diminishing pool of
glycosides. Johal and Rahe (1990), after showing a strong positive correla-
400 Higgins et al.
tion between the effects of glyphosate on loss of resistance of bean to

Colletotrichum Iindemuthianum and loss of phytoalexin production, ex-
pressed concerns about the possible effects on plant health of residues of
preplant herbicides containing glyphosate. For plants such as red clover, in
which glycosides appear to be an important initial source of phytoalexins
following pathogen attack, such residues might also result in depletion of
the glycoside pooland further reduce the resistance of the crop to root rot
pathogens.
Increasing evidence shows that the role of phytoalexin glycosides in
resistance is not unique to clover. The studies of Graham and coworkers
(1990) suggest that in all organs of soybean seedlings glyceollin biosynthesis
following infection by Phytophthora is not solely dependenton the induc-
tion of enzymes of the early phenylpropanoid pathway; instead glyceollin is
partly produced from daidzein released by hydrolysis of daidzein conju-
gates. In contrast, in soybean leaves challenged with Phytophthoru, such
glycoside precursors did not appear to play a role in biosynthesis of glyceol-
lin (Morris etal., 1991). More intriguingis the recent work of Mackenbrock
and Barz (1991) who showed that maackiain and medicarpin are partially
(30%) produced de novo by chickpea suspension cells despite high levels of
the potential precursor formonetin 7-0-glycoside-6"-0-malonate (FGM);
in contrast,cells treated with the PAL inhibitor L-cr-aminooxy-&phenylpropi-
onic acid(AOPP) totally utilized FGMto produce these same phytoalexins.
(Interestingly,the pool size of FGM, although not discussed by theauthors,
appeared to decline gradually in the presence of AOPP as would be ex-
pected if some turnover normally occurs as discussed above.) Obviously,
the role of preformed precursors in the biosynthesis of phytoalexins de-
serves more study but the results can be expected to vary from plant to
plant, organ to organ, and pathogen to pathogen.
V. EPILOG
Some ofour past questions regarding the biosynthesis, role,and detoxifica-

tion of phytoalexins will soon be answered by the use of molecular tech-
niques. There isno doubt that antisense technologywill be used to experi-
mentally prevent phytoalexin detoxification by pathogens. This will be an
important means of clarifying the importance of detoxification in those
many host-pathogen systems which are not amenable to the types of genetic
analysis used by VanEtten and coworkers for Nectriu (see VanEtten et al.,
1989). Nonetheless, commercial exploitation of antisense technology re-
lated to detoxification mechanisms inthe fungus seems unlikely as doesthe
use of chemical inhibitorsof such enzymes.The latter are unlikelyto be so
Phytoalexins in Forest legumes 401
specific that they will not also affect the biosynthetic enzymes usedby the
plant in thefinal stages of phytoalexin synthesis.
Also, our increasing knowledge of the enzymes involved in the final
biosynthetic stepsof phytoalexin biosynthesis will allow prevention of the
final reactions via antisense techniques. Understanding the role of pools of
glycosidic precursors in phytoalexin production will be very important in
such experiments asthe appropriate antisense gene may blockthe develop-
ment of such pools. If these glycosidic compounds have a function other
than defense, phenotypic changes may be detected inthe transgenic plants.
Most experiments which have relied on blocking PAL or other enzymes
occurring early in the biosynthetic pathway of pterocarpanoid phytoalexins
have generally recognizedthat many other plant functions relyon the same
pathway, but have failed to adequately consider the effects of preformed
precursors on the experiments.
Increasingly, as bioengineering of plantsand fungi becomes more rou-
tine, we need to investigate the relative contribution of each of the plant’s
defense componentsto the final disease interactionand to understand the
interactions between components. For example, it is possible that, in at
leastsomeplants,phytoalexinschangethegrowth of the fungus just
enough that the hyphal tips become amenable to total inhibition by the
activity of chitinase and/or glucanases. Indeed, the effect of phytoalexins
on fungal development is poorly researched. Duczek and Higgins (1976b)
showed that maackiain and medicarpin inhibited formation of appressoria
by Cochliobolus (Helrninthosporium) carbonum; although appressorium
formation is a developmental response that usually proceeds phytoalexin
production, comparable developmental changes, e.g., haustorium forma-
tion in the rusts, that occur after penetration might be similarly affected.
Our reliance on in vitro bioassays of hyphal growthand our failure to
reproduce the gradual buildup of phytoalexins that occurs in plants may
prove to havemisleadus about their relative importance in resistance.
Similarly, our failure to treat plants with gradually increasing amountsof
fungal elicitors have undoubtedly created misleading data on the role of
many elicitors, e.g., in tomato, a cell wall-derived (nonspecific) elicitor
from Cladosporiumfulvum appears to be degraded by enzymes produced
by infected or stressedplants(Peever and Higgins,1989). As well, the
elicitor activity is inhibited by other low molecular weight components of
the apoplast (Lu and Higgins, unpublished). Under normal conditions of
infection by C. fulvum, it is assumed that these “suppressors” build upat
about the same rate as the nonspecific elicitor, so that nonspecific elicita-
tion of the defense response never occurs. How many other elicitors under
active study are likewise “artifacts” of our assay systems? Molecular dissec-
tions of defense responses will eventually reveal how much we have been
402 Higgins et al.
led astray by our failure to properly mimic the pathogen and plant in our
experimentation.
REFERENCES
Duczek, L. J., and Higgins, V. J. (1976a). The role of medicarpinand maackiain in
the response of red clover leaves to Helminthosporium carbonum, Stemphy-
lium botryosum, and S. sarcinaeforme. Can. J. Bot. 542609-2619.
Duczek, L. J., and Higgins, V. J. (1976b). Effect of treatments with the phytoalex-
ins medicarpin and maackiain on fungal growth in vitro and in vivo. Can. J.
Bot. 54~2610-2619.
Graham, T. L., Kim, J. E.,and Graham, M. Y. (1990). Role of constitutive isofla-
vone conjugates in the accumulation of glyceollin in soybean infected with
Phytophthora megasperma.Mol. Plant-Microbe Interact.3:157-166.
Heath, M.C., and Higgins, V. J. (1973). In vitro and in vivo conversionof phaseol-
lin and pisatin by an alfalfa pathogen Stemphylium bottyosum. Physiol. Plant
Pathol. 3:107-120.
Higgins, V. J. (1972). Role ofthe phytoalexin medicarpin inthree leaf spot diseases
of alfalfa. Physiol. PlantPathol. 2:289-300.
Higgins, V. J. (1975). Induced conversion of the phytoalexin maackiain by the
alfalfa pathogen Stemphylium botryosum. Physiol. Plant Pathol. 65-18.
Higgins, V. J., and Millar, R. L. (1969). Comparative abilityof Stemphylium
bottyosum and Helminthosporium turcicum to induce and degrade a phyto-
alexin from alfalfa. Phytopathology 5 9 1493-1499.
Higgins, V. J., and Millar. R. L. (1970). Degradation of alfalfa phytoalexin by
Stemphylium lotiand Colletotrichumphomoides. Phytopathology 60269-271.
Higgins, V. J., Stoessl, A., and Heath, M. C. (1974). Conversion of phaseollin to
phaseollinisoflavan by Stemphylium-botryosum. Phytopathology64:105-107.
Hohl, B., Arnemann, M., Schwenen, L., Stockl, D., Bringmann, G., Jansen, J.,
and Barz, W. (1989). Degradationof the pterowpan phytoalexin (-)-
maackiain by Aschochyta rabei.2.Naturforschung. 44c:771-776.
Hohl, B., and Barz, W. (1987). Partial characterization of an enzyme from the
fungus Ascochyta rabieifor the reductive cleavage of pterocarpan phytoalexins
to 2'-hydroxyisoflavans. 2.Naturforschung. 42~897-901.
Hollander, H., and Amrhein, N. (1980). The site of inhibition of the shikimate
pathway by glyphosate I. Inhibition by glyphosate of phenylpropanoid synthe-
sis in buckwheat (Fagopyrum esculentum Moench.). Plant Physiol. 66823-
829.
Johal, G. S., and Rahe, J. E. (1990). Role of phytoalexins in the suppression of
resistance of Phaseolus vulgaris to Colletotrichum lindemuthianum by glyo-
phosate. Can. J . Plant Pathol.12225-348.
Mackenbrock, U.,and Barz, W. (1991). Elicitor-induced formation of pterocarpan
phytoalexins in chickpea (Cicer arietinum L.) cell suspension cultures from
constitutiveisoflavone conjugatesupon inhibition of phenylalanine ammonium
lyase. 2.Naturforschung. 46c:43-50.
Phytoalexins in Forest Legumes 403
McMurchy, R. A., and Higgins, V. J. (1984). Trifolirhizin and maackiain in red

clover: changes in Fusarium roseum “Awnaceum”4nfected roots and in vitro
effects on the pathogen. Physiol. Plant Pathol.25229-238.
Morris, P. F., Savard, M.E., and Ward, E. W.B. (1991). Identification and
accumulation of isoflavonoidsand isoflavone glucosides in soybean leaves and
hypocotyls in resistance responses to Phytophthora megasperma f. sp.glycinea.
Physiol. Mol. Plant Pathol.39:229-244.
Peever, T., and Higgins, V. J. (1989). Suppression of the activity of non-specific
elicitor from Cladosporiumfulvum by intercellular fluids from tomato leaves.
Physiol. Mol. Plant Pathol.34:471-482.
Tiemann, K.,Inze, D., Van Montagu, M., and Barz, W. (1991). Pterocarpan phyto-
alexin biosynthesisin elicitor-challenged chickpea(Cicer arietinumL.)cell cul-
tures: purification, characterization and cDNA cloningof NADPH:isoflavone
oxidoreductase. Eur. J. Biochem. 200751-757.
Van Etten, H. D.,Matthews,D. E., and Smith,D. A. (1982). Metabolismof
phytoalexins. Phytoalexins (J. A. Bailey and J. W. Mansfield, eds.), Wiley,
Van Etten, H.D.,Matthews,D. E., and Matthews, P. S. (1989). Phytoalexin
detoxification: importance for pathogenicity and practical implications.Annu.
Walsh, C. T. (1983). Suicide substrates: mechanism based enzyme inactivators with
therapeutic potential. Trends Biochem. Sci. 8(7):254-257.
18
Induction of Phytoalexin Synthesis in
Medicago sativa(Lucerne)- Verticillium
albo-atrum Interaction
Christopher J.Smith, J.Michael Milton,
and J.Michael Williams
University of Wales, Swansea, Wales
1. INTRODUCTION
The first report of a wilt disease of Medicago sativa L. (lucerne, alfalfa)
causedby Verticilliumalbo-atrum was from Sweden in 1918 (Hedlund,
1923), but during the period 1938-1950 the disease was reported in a num-
ber of European countries, including Germany (Richter and Klinkowski,
1938), Holland (Hansenand Weber, 1948), and France (Krietlow, 1962). In
the .United Kingdom, where there had been a large postwar increase in
lucerne cultivation, 12,800 hectares in 1942 to 44,000 hectares in 1954, the
first recorded outbreak was in 1952 (Noble et al.,1953). Following this first
occurrence the disease spread rapidly, resulting in a decline in the area
under cultivationto 15,000 hectares by 1970. Subsequently the disease was
recorded in Canada in 1964 (Aube and Sackston, 1964) and in Washington
State (Graham et al., 1977), and today the disease is of major significance
in North America.
Isaac and coworkers carried out extensive studies on a number of as-
pects of the disease in the United Kingdom (Isaac,1957a; Isaac and Lloyd,
1959; Isaac and Heale, 1961). From these studies it became apparent that a
degree of specificity existed in the interaction of V . albo-atrum with the
host plant. For example, from a large number of strains in the fungus
isolated from a range of different host plants, only those strains isolated
from lucerne were virulent when inoculated into this particular host. Sucha
reaction is unusual in Verticillium infections; the only other such casesthat
have been reportedare from peppermint (Nelson,1947) and Brussels sprout
infection causedby V . albo-atrum (sic) (V. dahliae) (Isaac, 1957b).
In common with other plant-pathogen combinations, incompatibility
405
406 Smith et al.
between lucerne and V. albo-atrum is frequently associated with the hyper-

sensitive response (HR). This expression of resistance, which is accompa-
nied by induction of a number of active defense mechanisms, results inthe
formation of a barrier of defense-related molecules around the invading
microorganism. In roots of lucerne, for example, synthesis of a lipid-su-
berin conjugate is induced (Newcomb and Robb, 1989), with a resulting
limitation to the spread of the fungus, while synthesis of inhibitors of
pathogen-secreted polygalacturonases has been observed in some lucerne
tissues in responseto V. albo-atrum (Degra et al.,1988).
No doubt such mechanisms haveimportant influences on the develop-
ment of infection under a variety of circumstances but there is now good
evidence to indicate that in many plant-fungal interactions accumulation of
phytoalexins is a significant determinant of resistance (Smith, 1994). For
example, plants withan established resistanceto a pathogen produce phyto-
alexins to higher concentrationsthan susceptible genotypes (Mayama et al.,
1981), and phytoalexins accumulate to sufficiently high concentrations in
the vicinity of the invading microorganism to inhibit its further growth
(Pierce and Essenberg, 1987). Further evidence is provided by demonstra-
tions that inhibition of phytoalexin synthesis can leadto loss of resistance
(Waldmuller and Grisebach, 1987) as well as by related observations con-
cerning the relationship betweenthe ability of a microorganism to detoxify
phytoalexins and its virulence (Van Etten et al., 1989). More comprehensive
discussions of the evidencefor the role of phytoalexins as determinants of
disease resistance will be found elsewhere in this volume and in a number
of reviews (see,for example, Dixon, 1986; Keen, 1990).
In lucerne, however, Higgins and Millar (1968) observed that when
droplets of a suspension of spores from Helminthosporium sativum, a fun-
gus nonpathogenic to lucerne (incompatible reaction),were placed on the
upper surfacesof the leaf, the phytoalexins medicarpin,a pterocarpan, and
sativan and vestitol, both of which are isoflavanols (Fig. l), diffused into
the droplets. These three compounds are the major phytoalexins that can
be detected in tissues of lucerne in response to a variety of invading micro-
organisms. They arise from the reactions of the central phenylpropanoid
pathway (Fig. 1) by which L-phenylalanine is converted to 4-coumaroyl-
CoA, and which provides precursorsfor a number of different isoflavanoid
phytoalexinsin a variety oflegumes(Hahlbrock and Scheel, 1989).4-
Coumaroyl-CoAis subsequently convertedto medicarpin, vestitol,and sat-
ivanby a seriesofreactions that wereelucidatedinlucernetissuesby
Dewick and Martin (Dewick, 1975; Dewick and Martin, 1976, 1979).
The interaction between V. albo-atrum and lucerne has been studied in
our own laboratories for a number of years, and Khan and Milton (1975,
1978, 1979), using the drop diffusate technique, examined the response of
Synthesis in Lucerne4 alboatnrm Interaction 407
P
408 Smith et ai.
lucerne leaf tissueto a range of different isolates of V. albo-atrum and V.

dahliae. Their studies showedthat the concentration of sporesin the drop-
let, the length of exposure to spores, and the presence of light all contribute
to the outcome of the plant-fungal interaction and hence influence the
pathogenicity ofthe particular fungal isolate. Of particular significance was
their conclusion that the differential pathogenicity of the isolates used in
the study was closely correlated with the concentration of phytoalexin in-
duced in and accumulated by the host tissues. V. albo-atrum isolated from
tomato plants, for example, was not pathogenic to lucerne and was found
to induce higher concentrations of phytoalexins in lucerne leaf tissuethan
did the fungus that had originally been isolatedfrom lucerne. In addition,
the lucerne isolate appearedto be relatively insensitiveto the lower concen-
trations of phytoalexins it induced in lucerne leaf tissue.
At first glance it mayappear strange that these studieswere performed
with leaf tissuesand a fungus, the normal pointof entry ofwhich would be
the root. Induction of a defense response in tissues that are remote fromthe
point of infection is not an unusual occurrence, however, and phytoalexin
accumulation in leaf tissues could play a significant role in preventing
spread of the fungus throughout the rest of the plant.In any casethe results
of these studies clearly indicatedthe cellular interactionsthat occur between
plant and fungus and have since been confiied in the relatively undiffer-
entiated tissues of suspension culture where morphological differentiation
is not a problem.
This conclusion, that the differential pathogenicity observed was close-
ly related to the concentration of phytoalexins accumulated inthe host, is
one that has been reached by other workers studying a variety of other
host-pathogen interactions. It highlighted the potential of this particular
defense response to be a determinant of resistance in the lucerne-V. albo-
atrum interaction and gave riseto three relatedareas of investigation inour
laboratories: 1.) determination of the kinetics of phytoalexin accumulation
in the incompatible interaction;2.) characterization of the elicitor compo-
nent ofthe pathogen responsiblefor induction of phytoalexin accumulation
in the host; and3.) characterization of the intracellular mechanism regulat-
ing induction ofthe response inthe host.
II. MATERIALS A N D M E T H O D S
A. Growth of Plants
Seeds of lucerne, cvs. Maris Kabul and Europe, were soaked overnight in
running tap water prior to sowing in a peat-based compost. Once germi-
nated, seedlings were transferred
to a soil-based compost and maintainedat
inSynthesis Lucerne+. albostrum Interaction 409
22 *
2OC under a 16-hr photoperiod at an intensity of 12,000 lux, when
plants were intended for use in bioassay, or transferred to a greenhouse
when usedin pathogenicity studies.
B. Development and Maintenance of Lucerne,

Callus, Culture, and Cell Suspension Cultures
Leaveswereremoved from 6-week oldplants of lucerne and surface-

sterilized by,washing in a 5% aqueous hypochlorite solution for 2 min.
Squares, 1 x 1 cm, were cut aseptically and transferred to the surface of
Murashige and Skoog (1962) agar medium. Growth of callus was allowed
*
to develop for 6 weeks ,at 25 1OC in the dark. Once established the
undifferentiated callus tissue
was maintained on the same medium by trans-
ference of a small (-0.5 cm3) piece of callus to fresh medium at 6-week
intervals.
Liquid cell suspension cultures ofM.sativa were initiated from callus
cultures maintained in this 1 cm3 solid
laboratory, by aseptic transference of
callus to 100 m1 Murashige and Skoog (1962) liquid medium at pH 6.7.
Cultures were incubated inthe dark at 25 f 1OC on an orbital shaker (100
rpm) and subcultured every2 weeks. For experiments, cultures were used 5
days after subculturing and additions were made aseptically by passing
solutions through sterile membrane filters (Minisart,NML 0.2 pm, Sarto-
rius, UK) from a sterile hypodermic syringe. At the end ofthe experimental
incubation period cells were harvested by centrifugationfor 2 min at 500g
prior to further analysis.
C. Preparation of Elicitor
The isolate of V. albo-atrum pathogenic to M. sativa (designated V,) was

obtained from field-grown plants ofM. Sativa and the nonpathogenic (des-
ignated V*) from Lycopersicon esculenturn. Elicitors from both were iso-
lated from 6-week-old liquid cultures grown DOX’S in medium. The medium
was filtered to remove mycelia and spores, dialyzed exhaustively against
distilled water, lyophilized, and the resulting powder dissolved in distilled
water at a concentrationof 4 mg/ml.
Elicitor activity was assessed by the ability to induce phytoalexin syn-
thesis, or increases in phenylalanineammonia-lyase(PAL)activity,in
shoots of M . sativa, placed for 16 hr with their cut ends in 50 m1 distilled
water containing putative elicitor. Where cell suspension cultures were em-
ployed, addition of putative elicitorwas carried out as described above,and
phytoalexin contentwas determined at 16 hr and PAL activity after 4 hr.
410 Smith et al.
D. Cut-Shoot Assay for Elicitation of Phytoalexin Accumulation

and Phenylalanine Ammonia-lyaseActivity
The ability of various components to elicit phytoalexin accumulation and
increases inPAL activity in lucerne leaf tissuewas determined usingshoots
of 6-week-old plants. Shoots were harvested and their cut ends placed in
water. The lower 0.5 cm of each stem was removed while remaining under
the surface of the water and four shoots were transferred to each conical
flask containing50 m1 of a solution of the compound under test in water.A
few drops of oil were addedto the surface of the solution to reduce evapo-
ration. Flasks were transferred to a controlled environmental chamber,23
f 2OC, 7000 lux, for 16 hr before leaves were removed for extraction of
phytoalexins or for determination of PAL activity.
E. Extraction of Phytoalexins
Phytoalexins were extracted from leaf tissue using a modification of the
method described by Keen(1978).Leaveswereremoved from the stem
tissue of experimentalshoots, weighed, and transferred to flasks containing
40% ethanol-water (v/v), (16 ml/g fresh wt tissue). Leaf tissue was sub-
jected to vacuum infiltration for 5 min at the end of which timethe vacuum
was removed and the process repeated. Each flask was incubated in the
dark at 25 * l 0 C , on an orbital shaker (100 rpm) for 2 hr. At the end of
this time the leaf tissue was removed by filtration, washed (5 m1 distilled
water/g fresh wt), and the aqueous washings combined withthe ethanolic
filtrate. The resulting solution was partitioned three times against ethyl
acetate or diethyl ether (0.5 ml solvent/ml extract). The solvent extracts
were combined and the solvent was removed byrotary film evaporationat
3OoC under reduced pressure until 1 m1 remained. The residue was trans-
ferred to a l-dram vial and the remaining solvent removed undera stream
of filtered air at 4OOC. Samples were storedat -2OOC until required.
F. High-Performance liquid Chromatography of Phytoalexins

Samples were dissolved in100 p1 of acetonitrile (AnalaR,BDH) containing
0.01% (w/v) dibutylphthalate (DBP), which was used as an internal stan-
dard. The samples were filtered(0.45 pm Acro LC3S, Gelman Sciences)to
remove any particulate matter and the resultant filtrate was subjected to
high-performanceliquidchromatography (HPLC) using an instrument
(Milton Roy, UK) fitted with an integrator (C1 lOB, Milton Roy, UK).
Typically, a 10-pl sample was injected, viaa rheodyne valve,onto a reverse
phase column (SSODS, Spherisorb, 5 pm, 250 mm x 4.9 mm (i.d.), Hich-
rom, UK) and was eluted isocratically with acetonitrile-water(3 :2 v/v) at
Synthesis in Lucerne-V. albo-atrurn Interaction 411
a flow rate of 1.3 ml/min. Compounds eluting from the columnwere

detected bytheir absorption at 240 nm. A standard ratio of retention times
for each phytoalexin, compared to the retention time of DBP, was deter-
mined by the injection of authentic medicarpin, sativan, and vestitol to-
gether with0.01‘YO DBP in acetonitrile. Using this standard ratio, the reten-
tion time of each phytoalexin could be determined in unknown samples.
The quantity of medicarpin ina sample was determined usingthe ratio
of the integrated areas of the phytoalexin and the DBP peaks, correctedfor
a response factor that adjusts for the difference in absorbances of the
phytoalexins and DBP. Response factors were determined for each of the
phytoalexins by injectionof known quantitiesof each phytoalexin and the
same quantity of DBP followed by comparison the of integrated areas.
C. Experimental Use of Cell Suspension Cultures

and Extraction of Phytoalexins
Where suspension cultures were used to determine the effects of elicitor, or
of various agonists and antagonists of the elicitation process, on the pro-
duction of phytoalexinsor PAL, the cultures were used 5 days after subcul-
turing. Effectors in solution were added to the cultures aseptically by pas-
sage through a sterile membrane filter (Minisart, NML 0.2 pm, Sartorius,
UK) from a syringe. The volume added never exceeded2 m1 and where it
was necessary to use ethanol to solubilize the effector the final concentra-
tion in the culture was never greater than 0.18 M. At this concentration
ethanol was found to have no effect on the phytoalexin content or the
activity’of PAL of cell cultures. Equivalent volumes of water or ethanol
were used in control cultures. For extractionof phytoalexins at the end of
the experimental treatment period, ethanol was added to the suspension
culture to a final concentration of 40% (v/v). The culture was then sub-
jected to vacuum infiltration and the extraction procedure described for
leaf tissue.
H. Extraction and Assay of PAL

Cell-free homogenates were prepared from tissues of M. sativa by grinding
in 50 mM Tris-HC1, pH 8.6, containing 10 mM ascorbic acid(3 m1 buffer/
g fresh wt), with a pestle and mortar and withthe aid of a little acid-washed
sand. The homogenate was filtered through two layers of Miracloth (Calbi-
ochem, UK), centrifuged(2O,OOOg, 20 min, 4OC), and the supernatant dia-
lyzed against 50 mM Tris-HC1, pH 8.6 (20 m1 buffer/ml sample) for 16 hr
at 4OC.
Thedialyzate was assayedspectrophotometrically for PAL activity
(Bolwell et al., 1985) by the addition of 1 m1 30 mM L-phenylalanine in50
412 Smith et al.
mM Tris-HCI, pH 8.6 (final concentration inthe assay 10 mM), to a 2-m1

sample of dialyzate (protein content inthe range 0.7-1 .O mg). Absorbance
at 290 nm was measured after 2 hr incubation at 3OoC, against a control
incubation containing10 mM D-phenylalanine in place ofthe L isomer.
111. PHYTOALEXIN ACCUMULATION IN THE

LUCERNE-V. ALBO-ATRUM INTERACTION
A. Development of an HPLC Method for Quantitation
of Lucerne Phytoalexins
In the earlier studies of Khan and Milton (1975), the drop diffusate tech-
nique of Cruickshank and Perrin (1960) was employed. In this method
lucerne leaves were treated with a spore suspension preparedfrom V. albo-
atrum and the phytoalexins that were synthesizedby the leaf in responseto
the challenge accumulated in the droplets. Following collection of
the drop-
lets from leaf surfaces the phytoalexins were extracted into a suitable or-
ganic solvent, purified by thin layer chromatography (TLC) using silica
gel plates developed with a chloroform-carbon tetrachloride solvent (3 : 1
v :v), and the areas corresponding to phytoalexins were scraped from the
plate. The phytoalexins which were eluted from the adsorbent were quanti-
tated from their absorbance at characteristic wavelengths. Using this ap-
proach, Flood et al. (1978) identified medicarpin, sativan, and vestitol as
the major phytoalexins present and the presenceof at least four other
components with the properties of phytoalexins was indicated. A number
of errors are inherent in these methods, however, and HPLC is now widely
used in the analyses of isoflavanolsand related components (Williamsand
Harborne, 1989). Until fairly recently, few HPLC methods had been de-
signedspecifically for phytoalexins, the majority of those in usebeing
general methodsfor separation of isoflavanoid compounds with little adap-
tation for the specific phytoalexinsand host tissues involved. As
a prerequi-
site for our quantitative studies of phytoalexin accumulation, therefore,we
undertook development of an efficient extraction and separation method,
based on HPLC, for the phytoalexins of lucerne.
The current method in use in our laboratory is already described in,
Section 11, Materials and Methods, but some comments relating to general
features and rationale of the approach will be made here. Except where
studies are performedwith the drop diffusate technique the extraction
method that has been found to be the most satisfactory with tissues of
lucerne, including cell suspension cultures, is a modification of the facili-
tated diffusion method (Keen, 1978). In terms of the efficiency of extrac-
tion of phytoalexins and the lower content of interfering compounds coex-
Synthesis in Lucerne-V. albwfrum Interaction 413
tracted, this technique has advantages over homogenization ofthe tissue,

and routinely as little as 1 gm fresh wt tissue can be handled, with the actual
amount depending on such factors as source of the tissue, duration of
exposure to the eliciting stimulus, sensitivity of the tissue to the stimulus,
and lower limit ofdetection of the HPLC system. Where leafor stem tissue
is involved 40%ethanol is added directlyto the tissue but in the case of cell
suspension cultures absoluteethanol is added to the whole culture, includ-
ing medium, untila concentration of 40% is reached. The growth medium
is includedin the extraction procedure because lucerne cultures secrete phy-
toalexins into the medium and these would be lost if the cells were removed
from the medium before extraction takes place.
It is sometimes possible to quantitate the phytoalexin content of an
ethanolic extract directly by HPLC, and this may be appropriate for cell
suspension cultures with their lower content of phenolics and absence of
chlorophyll. In our experience, however, a clearer separation of compo-
nents is achieved and column life is extended if a partitioning step is in-
cluded in the procedure. Routinely we have found ethyl acetate to be a
more satisfactory solventthan the tetrachloromethane that was used origi-
nally with lucerne extracts (Higgins, 1972). However, more recently we have
used diethyl ether because it is as efficient as ethylacetate for extraction of
medicarpin, but with its lower boiling point it can be removed with a stream
of filteredair rather than by rotary film evaporation. This permits simulta-
neous handling of multiple samples.
Liquid chromatography on a column of reverse phase adsorbent of-
fered the best possibility for resolution of the three phytoalexins vestitol,
medicarpin and sativan, and trials were carried out using a 250 x 4.9 mm
(i.d.) column of MODS Spherisorb silica (Hichrom, UK), 5-pm particle size
and elution with combinationsof acetonitrile and water. Gradient elution
has been used to separate the phytoalexins extracted from a variety of
sources by a number of workers, both before and since our initial experi-
ments (see, for example, Edwards and Strange, 1991), and we found that a
gradient from 47% to 85% aqueous acetonitrile overa 15-min period (1.3
ml/min flow rate) achieved a satisfactory separation of a mixture of the
three authentic phytoalexins. The same conditions were used to achieve
separation of an extract of lucerne leaf tissue that has been challenged for
16 hr with elicitor derivedfrom V. albo-atrum. The three phytoalexins are
separated from each other by this gradient; however, leaf extracts contain a
number of components that are more polarthan the phytoalexins and these
are not fully resolved from vestitol, which is the first of the phytoalexins to
be eluted.
Accurate quantitation of any phytoalexin requires a knowledge of the
absorption coefficient at the monitoring wavelength, establishment of the
414 Smith et al.
linearity of the response of the detector, and inclusion in the extract of an

internal standard so that peak areas may be expressed relative to the known
amount of standard that has been added.In oursystem, absorption coeffi-
cients were determined by spectrophotometry of authentic samples of the
phytoalexins and linearity of the response was established by injection of a
series ofstandards ranging inquantity up to 10 pg.
Selection of an internal standard was initially more difficultand DBP
was chosen because of its availability, absorption
its spectrum in relation to
that of the phytoalexins, and because out of the compounds tested it was
the only oneto be fully resolved fromthe phytoalexins and the other com-
ponents present in extracts of lucerne. Because ofthe differences between
the wavelength of maximum absorption of the three phytoalexins and of
the internal standard, however, monitoringof the eluant was carried out at
a “compromise” wavelength of240 nm where absorption of the phytoalex-
ins is reduced comparedto absorption ‘at their maxima, but absorption of
the internal standard while not maximal is still significant.
Routinely, then, DBP has been includedas an internal standard in the
extracts, but, using gradient elution, analysis time was prolonged because
of the time taken at the end of each separation to reequilibrate the column
to the starting conditions. However, vestitol, the earliest eluting phyto-
alexin of those present in lucerne, has rarely been encountered our studies
in
and since application ofa gradient is more criticalto separation in thisarea
of the chromatogram rather than for resolving medicarpin and sativan,
isocratic elution was tested as a means of separating the phytoalexins of
lucerne. Elution with60% aqueous acetonitrileat a flow rate of 1.3 ml/min
was found to give the optimum resolution with retention times, relative to
DBP, of 0.25, 0.30, and 0.40, for vestitol, medicarpin,and sativan respec-
tively. While separation of vestitol from other components is not entirely
satisfactorywith isocratic elution, it has rarely been detected ourinsamples
and so we routinely use this method (Fig. 2a,b).
B. Chemical Synthesis of Medicarpin, Sativan, and Vestitol
Samples of the authentic phytoalexins were required for the development

of the HPLC assay and racemic vestitol, sativan, and medicarpinwere
synthesized (Evans, 1986) as follows. The most efficient synthesis of these
compounds involved, as the key step, the thallium(II1)-induced oxidative
rearrangement ofthe intermediate chalcones(1) to form the 1,2-diaryl-3,3-
dimethoxypropan-l-ones (2). The latter were converted, without isolation,
into isoflavones (3) (Farkas et al.,1974).
Synthesis in Lucerne-V. a/bcMrnrm Interaction 415
phcH20% CH(0Me)z
/ Me
(1) R = Me or CH,Ph
(2)
(3)
Sativan (Dewick and Martin, 1979) and vestitol (Farkas et al., 1974)
have been synthesized by this route but full details were not given for
sativan. The chalcone (1, R = Me),preparedbycondensationof 4-
benzyloxy-2-hydroxyacetophenone and 2,4-dimethoxybenzaldehyde, was
converted to the isoflavone (3, R = Me), mp 137-139OC in 73% yield by
reaction with thallium(II1)nitrate followed by acid hydrolysis. Reaction of
an acetone solution of the isoflavone with hydrogen at atmospheric pres-
sure over 10% palladiumon charcoal gave sativan (36%after recrystalliza-
tion). The synthesis of medicarpin by the same route (Dewick, 1977) was
complicated bythe need to include a I4C-labeled methylgroup at C-4’, and
the isoflavone (3, R = CH,Ph)was not reported.Thisisoflavone was
readily prepared in 70% yield (mp 134-135OC) from the known chalcone
(1, R = CH,Ph) and debenzylation by hydrogenolysis over 10% palladium
on charcoal under controlled conditions (the reaction being terminated
after 2 eqofhydrogenhadbeenconsumed)gave7,2’-dihydroxy-4‘-
methoxyisoflavone, mp 218-220OC. Reduction of this isoflavone with so-
dium borohydride in ethanol followed by the action of l M hydrochloric
acid gave medicarpin (mp 200-2Ol0C), isolated by extraction into ethyl
acetate.
An alternativeroute to medicarpin was also used. This hadthe advan-
tage that the same isoflavone intermediate (3, R = Me) could be used to
synthesize sativanand medicarpin. The route to medicarpin involved selec-
tive aluminum chloride-catalyzed demethylation of(3, R = Me) in reflux-
ing acetonitrile, conditions which also removedthe benzyl protectinggroup
(Cockeretal.,1965).Theresultingcrudeproductcontaining 7,2‘-
dihydroxy-4’-methoxyisoflavonewas converted as above to medicarpin,
416 Smith et al.
Figure 2 Chromatograms of the separation of the phytoalexins vestitol, medicar-

pin, and sativan by HPLC. (a) Authentic samples of the phytoalexins and the
internal standard di-n-butylphthalate. (b) Separation of an extract of lucerne leaf
tissue that has been challengedfor 16 hr with the V, elicitor preparation. Details of
conditions will be found in Section11, Materials and Methods.
which was isolated by chromatographyon a chromatotron (silica gel, sol-

vent: chloroform). Vestitol was preparedfrom isoflavone (3, R = CH2Ph)
by hydrogenation/hydrogenolysis over 10% palladium on charcoal in ace-
tone (cf Farkas et al., 1974). All three phytoalexins were characterized by
mass spectrometry.
C. Future Development on Methodology

Two improvements to the HPLC method are at present under consider-
ation. The choice ofDBP as an internal standard is not entirely satisfactory
to sativan, which has
because of its relatively long retention time compared
Synthesisin Lucerne-V. albwtrum Interaction 417
the longest retention time of the phytoalexins, and because it is necessary

to
monitor at 240 nm in order to detect DBP (A- 225 nm) while the of
vestitol, medicarpin, and sativan are 282, 287, and 284 nm, respectively.
The 4-alkoxyacetophenones are being examined as potential internal stan-
dards partially because their absorbance characteristics are more suited
to this application, but also because the retention times of the different
alkoxyacetophenones vary with the length of the alkyl chain. So, for in-
stance, 4-butoxyacetophenone has a&= of 270 nm and a retention timeof
1.32 times that of sativan on the ODS column (250 x 4.9 mm i.d.) when
80% aqueous methanol is used as eluant. Its performance as an internal
standard in extracts of lucerne is currently being assessed.
The second area where an improvement to the current methodology
may be achieved isthe detection system. All three phytoalexins fluoresce in
the region of 310 nm when excitedat their absorption maxima, closeto 280
nm (Williams, unpublished). Potentiallythis is a more selective method of
monitoring becausefew ofthe other components present in lucerne extracts
fluoresce at that wavelength in response to incident light of 280 nm. De-
pending on the circumstances and equipmentavailable,afluorescent
418 Smith et al.
method may havea far greater sensitivitythan the absorbance methodand

the technique is in the process of evaluation.
D. Phytoalexin Accumulation in Response to Fungal Elicitors
Khan and Milton’s observation that the differential pathogenicity of V.
albo-atrum isolates was related to phytoalexinaccumulation (1975)
prompted us to examine the isolates to determine what elements of the
fungi could affectthe host in sucha way that a difference in expression of
the defense response resulted. The same host tissues, and thus the same
metabolic potentials, were involved in each case, so that some aspect of
the fungus must have been responsible for the different outcomes of the
interactions. Thereare a number of possible explanations including differ-
ences in structure of the isolates such that the host is unable to detect
the pathogen or does not detect it quickly enough to activate phytoalexin
production, while the nonpathogen is detected at an early stage; secretion
by the pathogen of suppressors of phytoalexin synthesis that prevent the
host from producing phytoalexins;and a greater efficiency of some isolates
to metabolize phytoalexins, so that in vivo their concentrations inthe host
do not reach antimicrobial levels.
Our approach has beento focus on the pathogen factor@)of the fungus
responsible for induction of phytoalexin synthesis inthe host, and for this
purpose two isolates were used, one recovered from lucerne (V,) and the
other from plantsof tomato (Lycopersicon esculentum) (V&.Pathogenicity
studies had already established that the V, isolate is capable of successfully
infecting lucerne plants CV. Europe (compatible interaction) while a high
degree of resistance is shown to isolate V,.
Phytoalexin accumulation canbe induced not only bythe living micro-
organism but also by components derived from it, termed elicitors (Keen
1990), molecules which by virtue of the structural information they contain
or the functional activity they possess are characteristic of the organism.
Mostly such elicitors are not race-specific, i.e., they are found to be pro-
duced by avirulentand virulent racesof the pathogen alikeand elicit phyto-
alexin synthesisin both resistant and susceptible cultivars ofthe host.
Nevertheless they providea useful starting point to the investigation of
the differential responseof lucerne sinceat least one possibility, metabolism
of phytoalexins by the pathogen, is eliminated and the complexity of the
interaction is greatly reduced. Elicitors of phytoalexin synthesis have been
previously isolated from V. albo-atrum by several workers. For example,
Zaki et al. (1972) identified a protein-lipopolysaccharide component that
induced formation of gossypol-related phytoalexins in stem tissue of cot-
ton. A glycoprotein elicitor also is knownto elicit synthesis of medicarpin
in lucerne(Onuorah, 1987).
in Synthesis lucerne-V. albo-atrum Interaction 419
In our present studies we have used elicitors derived from the two
isolates of V. albo-atrum, V, and V2,prepared as described in Section 11,
Materials and Methods. In the initial experimentsthe elicitors were used in
their crude state because there is some evidence to indicate that induction
of synthesis of the phytoalexin phaseollin in Phaseolus vulgaris, by an
elicitor from Colletotrichum lindemuthianum, relies on the presence of a
number of related polysaccharides, none of which by themselvescan induce
synthesis of the phytoalexin (Hamdan and Dixon, 1989). Fractionation in
the early stages may have led to the loss of a component without which
elicitation of the phytoalexin response may not have taken place.
Leaf tissues respond to treatment with V, elicitor by production of
phytoalexins (Fig. 3), accumulation being dependenton the time of treat-
mentwith the maximum concentration occurring 16 hr after the initial
challenge with elicitor(data not shown). The concentration of phytoalexin
accumulatedalsodepends on the concentration of elicitoremployed,
though the concentration of the elicitor that achieves maximum synthesis
of the phytoalexin is different for each of the phytoalexins. Only low con-
centrations of vestitol are accumulated by lucerne whatever may be the
elicitor concentration, though the optimum concentration of elicitor, 0.1
mg/ml carbohydrate, is the same asthat achieving maximum accumulation
of medicarpin. In contrast, sativan accumulation continues to riseeven
at the highest concentration of elicitor used, the rise in its concentration
corresponding with the decline in medicarpin accumulation. This shift to-
wards sativan accumulation is interesting and must reflect differential ef-
fects of the elicitor preparation on the separate branches of the pathways
leading to sativan and medicarpin (Fig. 1).
The effect on phytoalexin accumulation of elicitor from the V, (patho-
genic) isolate was also determined in leaf tissue,and although synthesis of
medicarpin was induced pig. 3), the concentration of elicitor required was
significantly higherthan was the case withthe V, elicitor, and little accumu-
lation occurred with elicitor concentrations below 1.0 mg/ml. In control
tissue (distilled water only) none the of phytoalexins could be detected.
The crude elicitorpreparations contained both carbohydrate and pro-
tein but the composition of the two elicitors isdifferent, the ratio of carbo-
hydrate to protein being 1 : 1 for the V, elicitor and 15 : 1 for the V,. These
ratios are consistent and have been observed over 15 different batches of
elicitor prepared using several reisolatesof each fungus. The difference in
the abilities ofV, and V, elicitors to induce phytoalexin accumulation in the
host tissue must reflect differences either in the structure of the two elicitors
or in the quantity of the structural determinant of elicitor activity each
preparation contains. This difference theirin composition led usto examine
whether it is the carbohydrate or the protein moiety that is the determi-
Smith et al.
14.0
12.0
10.0
8 .O
6 .O
4.0
2.0
0.0
0.00 0.10 0.20 0.30 0.40
Elicitorconcentration (me carbohydratelml)
8.0
6.0
4.0
2.0
0.0
0.0 0.5 1.0 1.5 2.0
Elicitorconcentration (mg carbohydratelml)
Figure 3 Phytoalexinaccumulationinleaftissue of lucerne (CV.Kabul) in re-

sponse to treatment for 16 hr with variousconcentrations of (A) V, elicitor prepara-
tion, (B) V, elicitor. A-A, vestitol; B-., medicarpin; 0 -0,sativa.
Synthesis in Lucerne-V. albo-afrum Interaction 42 1
nant of activity. We established that treatment of V, elicitor with periodate

leda 50% increase
led to a loss of elicitor activity while trypsin treatment to
in activity compared to untreated elicitor (Smith and Milton, 1992). This
indicated that it is the carbohydrate rather than the peptide moiety that
determines activity, whilethe effect of trypsin in increasing elicitor activity
suggests that a peptide inhibitor of phytoalexin synthesis is normally pres-
ent in the preparation, and trypsin destroys it, or that the elicitoris a
glycopeptide and trypsin cleavesthe peptide backbone, makinga carbohy-
drate determinant of activity more available to interact with the plant tis-
sues.
The difference in compositionof the two elicitorsand their effects on
phytoalexin accumulation is reflected in their effects on the activityof
PAL, a key enzymefrom the phenylpropanoid pathway leading to synthesis
of medicarpin, sativan, and vestitol in lucerne (Fig. 1). In common with
other tissues that have been characterized (see, for example, Daniel et al.,
1988) induction of phytoalexin synthesis in lucerne is associated with in-
creases inthe activities of enzymesfrom the biosynthetic pathway(see later
section for further details). In cell suspension cultures of lucerne, induction
of phytoalexin synthesis by the V, elicitor (Fig. 4) is accompanied by and
3.0 I
0.00 0.05 0.10 0.15

Elicitorconcentration (mg carbohydrate/ ml)
Figure 4 Sativan accumulation incell suspension culture of lucerne (CV.Kabul) in

response to 16-hr treatment with various concentrations of elicitor prepared from
the V, isolate of V. albo-atrum.
422 Smith et al.
dependent on an increase in activity ofPAL, the enzyme that catalyzes the

committed step on the phenylpropanoid pathway (Fig. 1). Treatment of cell
cultures with cordycepin, an inhibitor of mRNA transcription, completely
inhibits the increase inPAL activity normally associated with elicitor treat-
ment (Fig. 6) (Little, 1989) and leads to failure of the fungal elicitor to
induce phytoalexin synthesis. It wasof interest, therefore, to determine
whether the reduced rate of induced phytoalexin synthesis observed with V,
elicitor resultedfrom a failure to increase the activity of thiskey enzyme.
Bothleaftissues and cellsuspensionculturesrespond to treatment
with V, elicitor with an increase in activity of PAL (Figs. 5 and 6). The
concentration of elicitor required to give the maximum response in cell
cultures (0.05 mg carbohydrate/ml) is,however,less than that for leaf
tissue (0.1 mg/ml), presumably because access to the cells is easier in sus-
pension culture than in leaf tissue. In leaf the concentration of elicitor that
produces the maximum increase inPAL activity (0.1 mg/ml) is the same as
that producing the maximum accumulation of medicarpin (Fig. 3), while
in cell suspensiona threefold higher concentration of elicitor is necessaryto
induce the maximum increase inPAL activity (Fig.6) compared to that re-
quired to achieve maximum accumulation of sativan (0.15 mg/ml) (Fig.4).
Surprisingly, treatment of both leaf and cell suspension cultures with
V, elicitor caused a significant increase in PAL activity (Figs. 5 and 6) and
for both tissues treatment with V, elicitor broughtabout a greater percent-
age of increase in activity than did the V, (Table 1). The data of Table 1
also indicate that whether V, or V, elicitor is used, suspension culture ap-
pears more responsive than leaf tissue.
Another feature of the response induced by both V, and V, elicitor is
the presence inboth dose-response curvesof two peaks of activity,the one
correspondingto the lower increase occurringat the lower concentration of
elicitor in each case. Interestingly, when cordycepin at 100 km is included
in the cell culture to inhibit transcriptionof mRNA, the V, elicitor-induced
increase in PAL activity is almost totally prevented 'at the higher elicitor
concentration (Fig.6), while there is onlya small reduction in the induction
achieved at the lower concentrations.
It is obvious from these resultsthat the requirement for a greater con-
centration of V, elicitor to achieve induction of phytoalexin synthesis in
lucerne compared to V, elicitor, and the lower quantities of phytoalexin
accumulated inV, treated tissue, cannot be explained bya failure of the V,
elicitor to induce the increase inPAL activity. In both leaf and cell suspen-
sions, V, elicitor is actually more efficient in percent terms than the V,
elicitor at increasing PAL activity (Table l), and the absolute levelsof
activity induced byV, are higher than in tissuestreated with V, elicitor (100
Synthesis in Lucerne-V. albo-afrurn Interaction 423
0.00 0.10 0.20 0.30 0.40

Elicitor concentration (me carbohydratelml)
v
I
6o
40
-1
I I I I I
d 0 '
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Elicitor concentration (mg carbohydratelml)
Figure 5 Activity of PAL in leaf tissue of lucerne (CV. Kabul) induced by 16-h
treatment with various concentrationsof (A) V, elicitor preparation,(B) V, elicitor.
Smith et al.
200
A
150
l00
l
0
0.00 0.02 0.04 0.06 0.08 0.10
Elicitor concentration (carbohydrate mglml)
400
B
300
I
200
100
0 B
0.0 0.2 0.4 0.6 0.8 1.0 1.2
Elicitor concentration (carbohydrate mg/ml)
Figure 6 Activity of PAL in cell suspension cultures of lucerne induced by 16-hr

treatment with various concentrations of (A) V, elicitor, (B) V, elicitor. In (A) the
effects on the induction of activity of adding cordycepin at 100 p M are indicated,
0-0.
Synthesis in Lucerne+. alboafnrm Interaction 425
Table 1 PAL Activity in Leaf and Cell Suspension Cultures of Lucerne in

Response to Treatmentwith V, and V, Elicitors
Cell Leaf suspension
PAL
activityl Vo Increaseb PAL activityl To Increaseb
Control 20
Elicitor
v1 100 320 400 611
288 v2 175 85 325
*Expressedin nmol cinnamic acid/mg proteidhr.

bComparedto control.
and 320 nmol cinammic acid/mg protein/hr for leaf and cell suspension
cultures, respectively, treated withV, elicitor. Corresponding figuresfor V2
elicitor are 85 and 175 nmol cinammic acid/mgproteidhr). Since the lower
levels of activity induced by V, elicitor lead to synthesis of phytoalexins,
then the greater levelsof activity induced bythe V, elicitor should certainly
be adequateto support phytoalexin synthesis.
There are a number of possible explanations that could account for
these results. In lucerne cell suspensions at least three different isoforms of
PAL have been identified and the evidence indicates that PAL activity is
encoded by a multigene family, as is the case in bean (Jorrin and Dixon,
1990). The three isoforms each have different K,,, values and in responseto
treatmentwith an elicitor from Colletotrichumlindemuthianum a large
relative increase inthe form with the lower K, value for phenylalanine was
observed. In responseto an elicitor from yeast, however, a different pattern
of induction occurred and one of the other isoforms predominated. It is
possible thenthat in the case ofthe elicitors from Verticillium the V, elicitor
induces a different isoformof PAL, with a higher K, than that induced by
the V2elicitor, and that this isoformis lessefficient at feeding phenylalanine
substrate into the phytoalexin biosynthetic pathway. The result would an be
increased PAL activity as determined underthe assay conditions employed
where the concentrationsof phenylalanine is high but there is a failure to
synthesize phytoalexins in vivo.We have no evidence at present that could
evaluate this possibility.
It is also possiblethat both elicitor preparations from V. albo-atrum are
a mixture of componentsand that different enzymes from the biosynthetic
pathway are induced by different components of the mixture. This is cer-
tainly the case for a crude elicitor preparation from Colletotrichum linde-
muthianum in which a number of polysaccharides, each containing galac-
Smith et al.
tose and mannose, were identified. Whenthe mixture wasfractionated each

fraction was found to induce similar increases in activity of several of
the enzymes from the phenylpropanoid pathway in cultures of Phaseolus
vulgaris, but none of them by themselves were able to elicit synthesis of
the phytoalexins phaseollin or kievitone, in contrast to the unfractionated
mixture which did (Hamdan and Dixon, 1987). An integrated response in
the host may depend on the presence in the elicitor preparation of several
components from the microorganism, and if the V, elicitor preparation
used in these present studies lacksone or more of the components that are
present in the V, preparation, this may account for the differential phyto-
alexin synthesis observed.
Our studies relating to the purification ofthe elicitor determinantsfrom
the V, and V, elicitor preparations have so far indicated that each contains
a number of different components, both peptide and carbohydrate (Evans,
1986; Smith et al., unpublished). Initial studies have failed to identify one
component that could alone be responsible for the induction and phyto-
alexin elicitor activity has been located in several components, of various
molecular weights (Evans,1986). This isnot surprising inview of the results
with other elicitors,e.g., the glucan from Phytophthoramegasperma,
where the determinant of elicitor activity has beenfound to be distributed
throughout the heterogeneous mixture of glucans presentthe in unfraction-
ated mixture (Sharp et al.,1984). However, if any component ofa mixture
that is necessary for the induction of the complete response was missing
from the V, preparation, then phytoalexin synthesis at the higher concentra-
tions of elicitor employed (Fig.4) would not have been achieved. This may
mean that there is a lower quantity of a particular component present rather
than a complete absence, but whatever the factor is, obviously it must affect
a part of the pathway beyond the PAL-catalyzed step. Since that part of
the pathway is unaffected, itwill be interesting to assay these enzymes from
the later part of the pathwaythoughmany of them are membrane-
associated and are difficult to assay.
We are not in a position to evaluate the other possibility, i.e., that the
V, preparation contains a suppressor molecule similarto that identified in
chickpea (Kessmann and Barz, 1986) and which could inhibita part of the
pathway subsequent to the PAL-catalyzed step. If such a suppressor is
present its inhibitory effects must reacha saturation point before the elici-
tor of PAL activity achieves its maximum effect and its effect must be at
less than 100% inhibition of the pathway, because at the higher concentra-
tions of V, used in our experiments medicarpin synthesis is eventually in-
duced in leaf (Fig. 4). The possibility ofa suppressor molecule being present
is currently being examinedby fractionation of the V, elicitor preparation
as well as by assays in which the potential of the V,preparation to moderate
Synthesis in lucerne-V. a/bo-atrum Interaction 427
the phytoalexin response induced bythe V, elicitor is determined.If the V,

preparation contains an inhibitor of the later stages of the biosynthetic
pathway, then addition of the V, preparation would be expectedto lead to
inhibition of the elicitor activity ofthe V, preparation.
E. Future Prospects
There are a number of possible explanations for the difference in concentra-
tion of phytoalexins accumulated by lucerne in response to the nonpatho-
genic (V,) and pathogenic (V,)isolates of V. albo-atrum. In this respect, we
have so far examined one of these possibilities, that the difference lies in
differences inthe structure of the component of the pathogen, the elicitor,
that is perceived by the host and leads to phytoalexin production. Com-
pared to the whole organism, working with an elicitor preparation, even an
unfractionated one suchas we haveused,represents a simplification
whereby the physiological and metabolic capabilities of the pathogen are
absent. While this has been helpful in identifying a difference in oneof the
to induce PAL activity by
characteristicsof the two isolates, i.e., the ability
both elicitors whereas onlythe V, elicitor is efficient'at inducing phytoalexin
synthesis, aspects such as the capacity of the fungus to metabolize the
phytoalexins are not included in this approach.
In addition to studies aimed at further characterizing the two elicitor
preparations in order to determine the structural determinant of elicitor
activity and to examine whether a suppressor of phytoalexin synthesis is
present in the V, preparation, experiments will haveto be carried out with
the fungus as the inducer. Although the complexity of the interaction will
be increased greatly bythis approach, it will take into account the respon-
siveness ofthe fungus to the defense mechanisms ofthe plant and therefore
give a more detailed pictureof the host-pathogen interaction.
IV. SIGNALTRANSDUCTION
A. Background
Phytoalexins are not normally present in healthy plant tissues; they are
synthesized following perception ofthe potential pathogenby the host ina
process that requires increases in the activities of the enzymes responsible
(Dixon and Harrison, 1990). Generally the induction process leading to
phytoalexin synthesis is veryrapid, and the increases in enzyme activity that
are observed are knownto result from increases in the rates of transcription
of the specific genes encoding the enzymes of the particular biosynthetic
pathway. Many legumes synthesize isoflavanoid phytoalexinsand in these
tissues increases occur inthe activities of several enzymes of the phenylpro-
Smith et al.
panoid pathway. Without the induced increases in the activities of these

enzymes phytoalexin synthesis does not occur (Dixon and Harrison, 1990).
In bean, for example, accumulation of mRNA for PAL, the first enzyme
of the phenylpropanoid pathway, and for chalcone synthase (CHS), the
first enzyme from the branch leading to isoflavanoid phytoalexins (Fig.l),
occurs within 10 min of treatment of cells withan elicitor from Colletotri-
chum lindemuthianum (Edwards et al., 1985). Such increases result from
activation of the transcription rate of the genes which, in the case of the
PAL and CHS genes, occurs within 5 min of the first exposure of cells to
elicitor (Lawtonand Lamb, 1987).
B. Receptor Sites
Such studies indicatethat activation of gene transcription is one of the later
events in a process that must begin with perception of the pathogen or
elicitor by the cell. The perception process must involve interaction of a
molecule that is characteristic of the pathogen, with a receptor in the host.
Since these two events-binding of the elicitor and the subsequent activa-
tion of gene transcription-do not appear to occur on the same molecule,
they must be linked by an intracellular signal transduction system that is
capable of carrying specific information between the two leading to the
biochemical response. The nature of the signal mechanism that mediates
this induction of phytoalexin synthesis has not been characterizedso far for
any plant but the kinetics of the response in bean indicate that it is .a
very rapid process. Our own studies with lucerne have indicated that the
mechanism is complexand that two signal systems may interactto control
the induced synthesis of phytoalexins.
The location and nature of a receptor for the elicitor, whether located
on the plasma membraneor at an intracellular site,will have consequences
in any considerationof the mechanism of signal transduction involved. If
the receptor is located on the plasma membrane, as the is case for the amine
and peptide hormones of mammalian systems, then binding must leadto
the generation of a second messengerthat subsequently interacts withother
elements of the cell. In contrast, if the elicitor is able to cross the plasma
membrane and interact with a receptor located inthe cytoplasm, ina man-
ner similarto that established for the steroid hormones, thena more subtle
form of transduction mechanism may operate, one in which the elicitor-
receptor complex entersthe nucleus and interacts withthe DNA directly. In
such a case as this, second-messenger molecules are not a feature of the
signal transduction mechanism.
A comprehensive discussion of the experimental evidence relating to
the location of receptor sitesfor elicitors of phytoalexin synthesis is beyond
Synthesis in Lucerne-V. albo-atrum Interaction 429
the scope of this chapter. It is relevant to note, however, that the most
detailed studiesto date, carried out in soybean usinga specific heptagluco-
side elicitorof glyceollin synthesis, have indicatedthe presence of a protein
or glycoproteinreceptor on the plasmamembrane(Cosioet al., 1990)
rather than at an internal receptor site. Hadwiger and his colleagues(1981),
however, obtained some evidence that fungal elicitors can be transported
into cells of pea, and Kendra et al.(1987) demonstrated a direct interaction
of a chitosan elicitor with DNA.
When we first started our investigations, using cell suspension cultures
of lucerne and elicitor from the V, isolate of V. albo-atrum, there had
already been some studies otherin systems aimedat establishing the role of
second messengers in the induction process. The results, at least of some of
them, appearedto indicate that the signal, representedby the interaction of
elicitor and receptor, was transduced prior to activation of gene transcrip-
tion rather than there being a direct interactionof elicitor withthe genome.
Although we do not have direct evidence of the location of the elicitor-
receptorinteractioninlucerne, our studieshaveshown that a second-
messenger system, similar to that operating in the caseof the mammalian
amineandpeptidehormones,must operate to mediate the phytoalexin
response.
So while none of these early studies ofother workers had been carried
out with lucerne tissues, their findings appeared generally applicable and
we considered the possibility that a second messenger could be involved in
mediating induction of phytoalexin synthesis in lucerne.
C. SignalSystems
Among the second-messenger molecules that have been identified so far,
three are encountered frequently: adenosine 3',5 "cyclic monophosphate
(cyclic AMP), Ca", and diacylglycerol (DAG). Cyclic AMP is generated in
signal systems bythe activity of a receptor-regulated adenylyl cyclase, while
Ca2+and DAG are both products of the phosphatidylinositide signal path-
way and are known to function in combination in specific agonist-induced
responses (Berridge, 1987). All three of these second-messenger molecules
have now been implicated part as of the mechanism mediating the induction
of phytoalexin synthesis in lucerne.
D. Cyclic AMP in the Inductionof Phytoalexin Synthesis

As long ago as 1976 cyclic AMP was implicated in the signal transduction
mechanisms mediating phytoalexin synthesis when it was demonstrated that
application of cyclic AMP to cells of sweet potato led to the synthesis of
terpenoid phytoalexins (Oguni et al.,1976), and our own results are consis-
430 Smith et al.
tent with such a second-messenger role for cyclic AMP. For example, in
lucerne cell suspension cultures inthe absence of fungal elicitor, dibutyryl
cyclic AMP, an analog of cyclicAMP that can cross membranes, was found
to induce a fivefold increasein PAL activity that was also accompanied by
accumulation of the phytoalexin sativan (Cooke et al., 1989). Normally
neither medicarpin or sativan can be detected in lucerne cell suspension
cultures unless they are challenged with the fungus or the fungal elicitor.
The dependency of the elicitor-induced synthesis of phytoalexins upon an
increase in activity of PAL was discussed earlier, and this demonstration of
a direct effect of the cyclic AMP analog upon both the increase in PAL
activity and phytoalexin accumulation indicates a potential for cyclic AMP
to have a role as a second messenger inthe response induced bythe fungal
elicitor.
Until recently the question of the existence of cyclicAMP in plants has
beencontentious, so that itsrole as a secondmessengerinagonist-
controlled systems in plants hasn’t always received appropriate consider-
ation. However, there is now much evidence to indicate that cyclic AMP
and itsassociatedenzymes are presentinplants(Brown,1991), and in
lucerne we have useda sensitive radioimmunoassay to detect cyclicAMP at
a concentration of 2.5 pmol/g fresh wt cell suspension cultures (Cooke et
al., 1989). That identification was subsequentlyconfirmed bymass-
analyzed ion kinetic energy mass spectroscopy. In response to treatment of
lucerne cells with the V, elicitor the intracellular concentration of cyclic
AMP rose to reach a maximum concentration of 29.5 pmol/g fresh wt
within 4 minof the first challenge with elicitor, and the concentration
returned to the basal figure within 7 min of the start of the treatment
(Cooke et al.,1989) (Fig. 7).
The rapidity of the rise in cyclicAMP concentration that we observed
is comparable to those that occur in responses in mammalian tissues that
are controlled by the fast-acting hormonesand in which cyclicAMP acts as
a second messenger. Kurosaki et al. (1987a), working with suspension cul-
tures of carrot, observed a similar rise inthe intracellular concentrationof
cyclic AMP accompanying elicitor-induced phytoalexin synthesis, though
in this caseit only reacheda maximum after 30 min of treatmentof the cells
with elicitor.
These same workers were ableto demonstrate synthesis incarrot cells
of the phytoalexin 6-methoxymellein in responseto treatment with cholera
toxin, an agonist of adenylyl cyclasethat would be expectedto increase the
intracellular cyclic AMP concentration. The fact that cholera toxin stimu-
lates phytoalexin synthesis implies both that adenylyl cyclase activity is
present in carrot suspension cultures andthat cyclic AMP is a second mes-
senger in the induction of phytoalexin synthesis. In fact, adenylyl cyclase
Synthesis in Lucerne-V. albo-atmm Interaction 43 1
30.0
20.0
10.0
0.0 '
0.0
I
2.0
I
4.0
I
6.0
I
8.0
I
10.0
Time (mins)
Figure 7 Change in intracellular cyclic AMP concentration in response to treat-

ment of lucerne cell suspension cultures withV, elicitor (0.05 mg carbohydrate/ml).
has previously been partially purifiedfrom lucerne (Carricarteet al., 1988)

and since that report we have also identified adenylyl cyclase activity in
lucerne, in a plasma membrane-enriched fraction from cell suspension cul-
tures. In vivo the activity of this enzyme shows a rapid transient rise shortly
after treatment of cell suspensions withthe V, elicitor (Cooke et al.,1994),
a finding that, together with its location in a plasma membrane-enriched
fraction, lends strong support to a model for signal transduction that fea-
tures an elicitor-regulated adenylyl cyclase generating second-messenger cy-
clic AMP.
The activity of the degradative enzyme cyclic nucleotide phosphodies-
terase was also observed in cell free homogenates of lucerne (Cooke et al.,
1994; Robinsonetal., 1992), so that both themeansofsynthesis and
degradation of cyclic AMP have now been identified in lucerne. Both cyclic
2',3 '- and cyclic 3 ' ,5 '-AMP are hydrolyzed by the preparation, but we
have shownthat only the cyclic 3 ',5 '-AMP activity is regulated by calmod-
ulin (CaM) sinceit is inhibited by CaM antagonists suchas trifluoperazine,
calmidazolium, andEGTA, and is stimulated by addition of Ca2+and CaM
(Robinson et al.,1992). We have isolated CaMfrom lucerne cell suspension
cultures, using hydrophobic chromatography on phenylsepharose (Robin-
son et al.,1991), and have shownthat it is capable of regulatingthe activity
432 Smith et al.
of the 3’,S ‘-cyclic nucleotide activity whilethe 2‘ ,3 ”clyclic nucleotide ac-

tivity is unresponsive.
These studies have provided evidence that the enzymatic meansfor the
synthesis and degradation of cyclicAMP are present in lucerne,and that in
response to the fungal elicitor, the biological inducer of the phytoalexin
response, a flux of cyclic AMP is generated within the cell. These elements,
and their interaction with a cellular agonist, are essential characteristics, of
course, of a second-messenger system based on cyclic AMP. The demon-
stration that a cyclic AMP analog (dibutyryl cyclic AMP) can induce a
qualitatively similar responseto that induced by the fungal elicitor addsto
the evidence that cyclic AMP is involved in mediating the elicitor-induced
phytoalexin response.
E. Ca2+andPhytoalexin Synthesis
The results ofour studies with lucerne have so far been .entirely consistent
with a second-messenger role for cyclic AMP in the induction of phyto-
alexin accumulation. However, some of our earlier work had indicated that
induction of the response also depends on an elicitor-stimulated flux of
Ca” across the plasma membrane (Little, 1989). Various related observa-
tions, including inhibitionby Ca” antagonists such as La”and EGTA, of
arachidonic acid-induced lubimin synthesis in potato (Zook et al., 1987)
and inhibition of elicitor-induced 6-methoxymellein synthesis carrot in cells
by the plasma membrane Ca” channel blocker verapamil (Kurosaki et al.,
1987b) have implicated an involvement of intracellular Ca” fluxes in the
induction process.
In many animal systems elevation of intracellular Caz+acts as amajor
second messenger in signaltransduction systems and many effectors bring
about their cellular responses though the generation of raised concentra-
tions of cytosolic Ca” (Berridge, 1987). There are also some responses in
which cyclic AMP and Ca” are known to operate in concert to control a
single response (Rasmussen, 1981), and in this respect identification of a
Ca’+/CaM-responsive 3’,5 “cyclic AMP phosphodiesterase may proveto
be a link between the signal molecules cyclic AMPand Ca2+(Robinson et
al., 1992).
In fact, Ca” has been shown to be an effective inducer of both an
increase in PAL activityand phytoalexin accumulation in lucerne, with the
response demonstrating a dose dependency on the ion with a 10-fold in-
crease in PAL activity observed at a Ca” concentration of 3 mM (Little,
1989; Smith et al., 1989) (Fig. 8). Like the elicitor-induced increase inPAL
activity, the effect of Ca” can be inhibited by the mRNA transcription
inhibitor cordycepin indicating that Ca” causes an increase in gene tran-
Synthesis in Lucerne-V. albo-afmm Interaction 433
0 2 4 6 8
EXOQenOUS ca2+ concentration (mM)
i
0 2 4 6 8 10
La3+ concentration (mM)
Figure 8 (A) Induction of PAL activity (W-.) and sativan production(0 - 0 ) in

lucerne cell suspension cultures by treatment for
4 hr with variousconcentrations of
Ca”. (B) Effect of the Ca” antagonist lanthanumon the inductionof PAL activity
by the elicitor from V. ulbo-utrum. The elicitor was present in each treatment at
0.05 mg of carbohydrate/ml, andPAL activity was measured after4 hr.
434 Smith et al.
scription and production of mRNA for PAL. By itself an effect of Ca2+on

the induction of phytoalexin accumulation does not provide direct evidence
that CaZ+has a second-messenger function in the fungal elicitor-induced
response. However, compoundsthat are antagonist to Caz+and the phyto-
alexin response induced byCa”were also found to be antagonistic to
the effect of fungal elicitor in inducing the increase in PAL activity and
phytoalexin accumulation. For example,La3+, anion which competitively
inhibits Ca” movement through plasma membrane channels, effectively
prevents the induction brought about by Ca” and the induction by fungal
elicitor when it is added to cell suspension cultures (Fig. 8). Similarly, the
metal ion chelator EGTA, or the Ca” channel blocker verapamil, will
prevent the increase in PAL activity and the phytoalexin accumulation
normally associated with elicitor treatmentof cell suspension cultures (Lit-
tle, 1989;Smith etal., 1989).
On the other hand, increasing the cytosolic Ca” concentration, by the
addition of lucerne cell cultures ofthe calcium ionophore A23187 at 5 PM,
enhances the ability of the fungal elicitor to increase PAL activity, the
concentration of elicitor required to give the maximum increase in its pres-
ence reducingto only 10% of that required in the absence of the ionophore
(Little, 1989). The concentration of exogenous Ca” requiredfor maximum
effect, in the absence of fungal elicitor, is likewise reduced in the presence
of the ionophore, decreasingfrom 3 mM to less than 1 mM.
Such results provide indirect evidence that a flux of Ca”, generated
across the plasma membrane in responseto the elicitor, is a necessary part
of the induction process, but they do not provideinformation regarding the
duration or the sizeof the flowrequired.Suchdetails will require the
further development of intracellular Ca” imaging systems that are capable
of measuring changes the in concentration of Ca”.
Neither do our present results establish what relationship exists between
the requirement for an elicitor-generated flux ofCaz+on the one hand and
on the other the activation of gene transcriptionthat is the resultof elicitor
treatment on the other. Nor is it possibleto state whether the change in flux
is the “primary” trigger inthe sequence. We do have some evidence, how-
ever, to indicate that the flux of Ca” across the plasma membrane must be
sustained for a period of about 50 min in order to achieve the maximum
attainable elicitor-induced increase in PAL activity, but the way in which
such a sustained Caz+flux can leadto activation of gene transcription will
only become evident from further detailed biochemical studies. So, in our
current studies, we are using a protoplast system to determine the flux of
Ca” induced across the plasma membrane in responseto the elicitor, and
the effects of various agonistsand antagonists of the signal system uponit,
Synthesis in Lucerne-V. alboarrum Interaction 435
to determine the role of Ca2' cycling acrossthe membrane inthe induction

process.
F. The Phosphoinositide Signal System

The involvement of a Ca2' flux in the induction process is clearly indicated
by such resultsand, although there is morethan one way by which a flux of
Ca2' across the plasma membrane may be generated, in many animal sys-
tems Ca"-mobilizing hormones generate second-messenger Ca" through
the phosphoinositide signal transduction system outlined in Fig. 9. Two
intracellular signal molecules are generated by this pathway: 1,2-diacylgly-
cero1 (DAG) and inositol-l,4,5-trisphosphate(IP,) (Berridge, 1987). Both
result from the activity of a receptor-mediated phosphoinositidase Cthat
catalyzes the cleavage of phosphatidylinositol4,5-bisphosphatein the
plasma membrane. IP, is linked to Ca2' as a second messenger because of
its ability to releaseCa" from internal sites, while DAG is capable of
C8
2'
sequeilcred
,"""""""""""""""""-c
,:'processes
: P t d l n s4 . 5 P 2 Cellular
response
..".-" 8 , .
processes
""""""""2
8
':
, .
-" -"
"
A""""""""
::
: :,
Figure 9 Phosphatidylinositidesignal pathway. Whenan agonist bindsto a recep-

tor (R), it activates a G protein (G) which simulates phosphoinositidase C (PIC).
Two intracellular signal molecules are generated from the hydrolysis of phosphati-
dylinositol4,5-bisphosphate (PtdIns 4,5-P3, diacylglycerol(DAG), and inositol
1,4,5-trisphosphate (Ins 1,4,5-P,). Dashed lines represent the interaction of Ca"
and protein kinase C (PKC) with various elements of the system, +
indicates a
promotion, "Ca" sequestered'' refers to calcium ions whichare removed to cellular
compartments (vacuole, endoplasmic reticulum,mitochondria) and outside the cell
via the activity ofCa" ATPases.
436 Smith et al.
activating the membrane-associatedproteinkinase C (Nishizuka, 1986).

Although the Ca2'releasingactivityofIP,appears to be restricted to
internal sites (endoplasmic reticulum, tonoplast membrane), the product
of its phosphorylation, inositol 1,3,4,5-tetrakisphosphate (IPJ, has been
reported to stimulate entryof Ca" into the cell by wayof its effect onCa2+
channels inthe plasma membrane (Irvine et al., 1988).
The number of studies in whichthe role of the phosphoinositide signal
pathway in the activation of defense response genes has been examined is
somewhat limited, though many of the elements of the system have now
been identified in plants including phosphatidylinositol (PI), phosphatidyl-
inositol4-phosphate (PIP), and phosphatidylinositol4,5-bisphosphate
(PIP,); Ca2+-dependent phosphoinositidases (enzymes capable of hydrolyz-
ing PIP&; protein kinases similar to protein kinase C of mammalian sys-
tems; an IP,-stimulated release of Ca" from intracellular sites; and the
kinases and phosphates involved
in
phosphatidylinositideturnover
(Drsbak, 1992).
Labeling experiments with lucerne in which cell suspension cultures
have been incubated with [3H]myoinositol and [32P]orthophosphate have
established the presence of PI, PIP, and PIP,, and in responseto challenge
with V, elicitor we have been able to demonstrate a rapid turnover of
'*P-labeled PIP and PIP, (Walton etal., 1993). The response israpid, with
turnover occurring within1 min of addition of elicitor. Thisis not part of a
general turnoverof membrane phospholipid since phosphatidylcholine and
phosphatidylserine are unaffected by elicitor treatment. Such a turnover of
PIP and PIP2in responseto an agonist is typically observed in physiological
responses mediatedby the phosphoinositide signaltransduction system and
in these cases it represents an agonist-induced hydrolysis ofPIP2, catalyzed
by the receptor-regulatedphosphoinositidase C referred to earlier, and
which results inthe generation of the signal molecules DAG and IP,. Kuro-
saki et al. (1987b) detected a breakdown of phosphatidylinositol lipids in
carrot cells in response to a fungal elicitor, and the result with lucerne is
significant since it impliesthat, in addition to the presence of the phospho-
inositides in the membrane, there is also present a phosphoinositidase re-
sponsive to the fungal elicitor.Further, it pointsto the possibilitythat both
DAG and IP3are generated by elicitortreatment.
In fact it has now been possible, using a specific binding protein assay,
to identify IP3 in extracts from lucerne suspension cultures (Cooke et al.,
1991a; Walton et al., 1993), and the role of the phosphoinositide signal
system in mediating elicitor-induced phytoalexin synthesis has beenfurther
indicated by measurement of the intracellular IP, concentration in response
to the fungal elicitor.
When lucerne cells were treated with V, elicitor, the intracellular con-
Synthesis in Lucerne4 albo-atrum Interaction 437
centration of IP, rose within1 min of treatmentto reach a maximum of 60

pmol/g fresh weight (Cooke et al., 1991b). The occurrence ofthis transient
increase (the concentration returned to basal level within 15 min) in IP,
concentration together with the demonstrated turnover of PIP and PIPz
is consistent with an agonist-induced phosphoinositide hydrolysis system
operating during the early phase of the induction process. In this context
compound 40180, an inhibitor of phospholipase C, effectively inhibits the
elicitor-induced increase inPAL activity and phytoalexin synthesis (Little,
1989), implying that induction of the response by the elicitor depends on
the activity ofthe phosphoinositide signaltransduction system.
A central transducer of the Caz' message, especially in the type of
responsedescribedhereinwhichsustainedcyclingof CaZ+across the
' plasma membrane appears to be a feature, is protein kinase C, an enzyme
that is activated by the increase in intracellular Ca" concentration and that
subsequently affects the rate of cellular reactions by phosphorylating key
enzymes(Nishizuka,1986). l-Oleoyl-2-acetylglycerol,a syntheticdiacyl-
glycerol that can activate protein kinase C in the same way that endogenous
DAG produced by the phosphoinositide pathway does in vivo, at a concen-
tration of 1 p M stimulated an increase inPAL activity and accumulation of
phytoalexin to levels comparable with those achieved by the V2 elicitor
(Little, 1989). Similar increases were obtained when cells were treated with
phorbol 12-myristate-l3-acetate, a potent activator of protein kinase C.
Induction of the phytoalexin responseby activators of protein kinaseC in a
manner similar to that achieved by fungal elicitor impliesthat this type of
kinase is involved in the induction process. In contrast, inhibition of the
elicitor-induced response in lucerne by l-(5-isoquinoline sulfonyl)-2-methyl-
piperazine and calphostin C, both of which are potent and selective inhibi-
tors of protein kinase C, reinforces the results obtained with the protein
kinase C activators (Smithet al., unpublished).
Collectively these results offer considerable evidence that an agonist-
induced phosphoinositide hydrolysis operates duringthe elicitor-activated
synthesis of phytoalexinsin lucerne. We have been ableto demonstrate the
presence of the essential phosphoinositidesand their turnover in response
to fungal elicitor,while elicitor treatment of lucerne cells has been shown to
cause a rapidtransientincreaseinIP, concentration, characteristic
of phosphoinositide signaltransduction systems. The fact that compound
40/80 effectively inhibits elicitation ofPAL activity and phytoalexin accu-
mulation by the fungal elicitor, together with the fact that agonists and
antagonists of protein kinase C are also agonists and antagonists of the
elicitation by fungal elicitor, offers good support for a role for the phos-
phoinositide pathway in signal transduction mediatingthe phytoalexin re-
sponse.
438 Smith et al.
C. Future Development
The resultsfrom lucerne and the results of others have givenus a good deal
of encouragement in our efforts to characterize the signal system involved
in the phytoalexin response. Evidence now exists that two signal systems,
the cyclic AMPand the phosphoinositide pathways,are involved in mediat-
ing the phytoalexin response induced by the fungal elicitor. While at first
glance there may appear to be a conflict in a model for signal transduction
that features two signal systems, there are physiological responses in animal
cells that are known to be regulated by the two systems operating together
(Bolander, 1989). In these cases cross-regulation is known to occur and one
of the areas whichwill continue to be of interest to us will be characteriza-
tion of the way in which the two systems interact and examination of the
system for evidence of regulatory elements between and within the path-
ways. This will require isolationof the relevant enzymesand determination
of their kinetic parameters as well as identifying endogenous agonistsand
antagonists of their activities.
In this respect since protein kinases, those controlled by Caz+/CaMas
well as the C- and cyclie AMP-dependent type,are central mediatorsof the
second messengers Ca”, DAG, and cyclic AMP, we are currently examin-
ing lucerne tissuefor the presence of these kinases. Our initial investigations
have revealed the presence of both cyclic AMP and Ca’+/CaM-dependent
kinases, and it will be interesting to determine what their endogenous pro-
tein substrates are since these may be the ultimate targets of the transduc-
tion system.
In the future too it will be necessary to establish how the various ele-
ments of the transduction system are focused to affect the rate of gene
transcription and to this end a number of studies ofthe molecular biology
of the induction processare being carried out (Dixon and Harrison, 1990).
The role of Ca” fluxes, an essential feature of the response, will continue
to be an area of interest in which it will be important to determine the
mechanism by which the fungal elicitor can induce such a change and how
the flux affectsthe cell biochemistry.
V. EPILOG
Defense against disease in plants involvesa number of mechanisms, some
structural, some takingthe form of preexisting chemical barriers, and some
only being calledinto play oncethe attack by a pathogen has been detected
by the plant. Of this latter category, the inducible defense responses, syn-
thesis of phytoalexins has attracted a great dealof attention, while thereare
many cases of resistance in which phytoalexin synthesis is the not key event
in Synthesis Lucerne-V. albo-atnrm Interaction 439
. equally there are instances of resistance to a potential pathogen in which

the ability of a host to rapidly synthesize phytoalexins is themajor factor
determining the outcome of the interaction. Various aspects of the response
have been studiedso far, including characterizationof elicitor components;
the kinetics of accumulation inthe host; the biochemistry of the pathway;
the ability of the pathogen to suppress synthesisor detoxify the phytoalex-
ins; and, latterly, the molecular biologyof the response. The complexity of
host-pathogen interactions is such that no singleapproach can be expected
to evaluate the role phytoalexin synthesis plays in the resistance of the
host. In combination, however, particularly with the newer techniques of
molecular biology available,that task is becoming easier.
REFERENCES
AubC, C., and Sackston, W. E. (1964). Verticillium wilt of forage legumes in Can-
ada. Can. J. Plant Sci.44:427-432.
Berridge, M. J. (1987). Inositol trisphosphate and diacylglycerol: two interacting
second messengers.Ann. Rev. Biochem. 56159-193.
Bolander, F. M. (1989). Molecular Endocrinology.Academic Press, New York.
Bolwell, G . P.,Bell, J., Cramer C., Schuch, W., Lamb, C. J., and Dixon,R.
A. (1985). L-Phenylalanine ammonia-lyase from Phaseolus vulgaris. Eur. J.
Biochem. 149:411-419.
Brown, E. G. (1991). Purines, pyrimidines, nucleosides and nucleotides. Methods
in Plant Biochemistry, Vol. 5 (L. J. Rogers, ed.), Academic Press, London,
pp. 53-90.
Carricarte, V. C., Bianchini, G. M., Muschietti, J. P., Tellez-Inon, M. T., Perticari,
A., Torres, N., and Flawia, M. M. (1988). Adenylate cyclase activity in a higher
plant, alfalfa (Medicagosativa). Biochem. J. 249807-81 1.
Cocker,W.,McMurry, T. B. H., and Staniland, P. A. (1965). A synthesisof
demethylhomopterocarpin. J. Chem. Soc. 1034-1037.
Cooke, C. J., Newton, R. P., Smith, C. J., and Walton, T. J. (1989). Pathogenic
elicitation of phytoalexin in lucerne tissue: involvement of cyclic AMP in the
intracellular mechanism.Biochem. Soc. Trans. 12919-920.
Cooke, C. J., Smith, C. J., Newton, R. P., and Walton, T. J. (1991a). Binding
saturation analysisofinositol1,4,5-trisphosphateinsuspensioncultures of
lucerne cells. Biochem. Soc. Trans. 19593.
Cooke, C. J., Smith, C. J., Newton, R. P., and Walton, T. J. (1991b). Effects of
phytoalexin elicitor on levels of inositol phosphates in lucerne cells in suspen-
sion culture. Biochem. Soc. Trans.19948.
Cooke, C. J., Smith, C. J., Walton, T. J., and Newton, R. P. (1994). Evidencethat
cyclic AMP is involved in the hypersensitive response of Medicago sativa to a
fungal elicitor.Phytochemistry, 35:889-895.
Cosio, E. G., Frey, T., Verduyn,R.,vanBoom, J., and Ebel, J. (1990).High
affinity binding of a synthetic heptaglucoside and fungal phytoalexin elicitorto
soyabean membranes.FEBS Lett. 271:223-226.
440 Smith et al.
Cruickshank, I. A. M., and Perrin, D. R. (1960). Isolation of a phytoalexin from

Pisum sativumL. Nature 187:779-800.
Daniel, S., Hinderer, W., and Barz, W. (1988). Elicitor-induced changesof enzyme
activities relatedto isoflavone and pterocarpan accumulation in chickpea (Cicer
arietinum L) cell suspension cultures. 2. Naturforschung. C Bioscience 43c:
536-544.
Degra, L., Salvi, G., Mariotti, D.,De Lorenzo, D., and Cervone, F. (1988). A
polygalacturonase-inhibiting protein in alfalfa callus cultures.J. Plant Physiol.
133~364-366.
Dewick, P. M. (1975). Pterocarpan biosynthesis: chalcone and isoflavone precur-
sors of demethylhomopterocarpinand maackiain inTrifolium pratense. Phyto-
chemistry 14979-982.
Dewick, P. M. (1977). Biosynthesis of pterocarpan phytoalexins in Trifolium pra-
tense. Phytochemistry16:93-97.
Dewick, P. M., and Martin, M. (1976). Biosynthesis of isoflavanoid phytoalexins
in Medicago sativa: the biosynthetic relationship between pterocarpans and
2'-hydroxyisoflavans. J. Chem. Soc. Chem. Commun.637-638.
Dewick, P., and Martin, M. (1979). Biosynthesis of pterocarpan and isoflavanoid
phytoalexins inMedicago sativa.The biochemical interconversion ofpterocar-
pans and 2'-hydroxyisoflavans. Phytochemistry 18591-596.
Dixon, R. A. (1986). The phytoalexin response:elicitation, signalling and control of
host gene expression.Biol. Rev. 61:239-291.
Dixon, R. A., and Harrison, M. J. (1990). Activation, structure and organisation
of genes involved in microbial defense in plants. Adv. Genet. 28165-234.
Drsbak, B. K. (1992). Plant signal perception and transduction: the role of the
phosphoinositide system. Essays in Biochemistry, Vol. 26 (K. F. Tipton, ed.),
Portland Press, Colchester, pp. 27-37.
Edwards, C., and Strange, R. N. (1991). Separation and identification of phytoalex-
ins from leaves of groundnut (Arachis hypogaea) and development of a method
for their determination of reversed-phase highperformance liquid chromatog-
raphy. J. Chromatogr. 542185-193.
Edwards, K.,Cramer, C. L., Bolwell, G. P., Dixon, R. A., Schuch, W., and Lamb,
C. J. (1985). Rapid transient induction of phenylalanine ammonia-lyase mRNA
in elicitor-treated beancells. Proc. Nat. Acad.Sci. USA 82:6731-6735.
Evans, D. J. (1986). PhD thesis, University of Wales.
Farkas, L., Gottsegen, A., Nogradi, M., and Antus, S. (1974). Synthesis of sopho-
rol, violanone, lonchocarpan, claussequinone, philenopteran, leiocalycin and
some other natural Isoflavanoids by the oxidative rearrangement of chalcones
with thallium I11 nitrate. J. Chem. Soc. Perkin k305-308.
Flood, J., Khan, F. Z., and Milton, J. M.(1978). The role of phytoalexins in
Verticilliumwilt of lucerne.Ann. Appl.Biol. 89329-332.
Graham, J. H., Peaden, R. N., and Evans, D.W. (1977). Verticilliumwilt ofalfalfa
found in the United States.Plant Dis. Rep. 61:337-340.
Hadwiger, L. A., Beckman, J. M., and Adams, M. J. (1981). Localisation of fungal
components in the pea-Fusarium interaction detected immunochemically with
antichitosan and antifungal cell wallantisera. Plant Physiol.62170-175.
Synthesis in Lucerne-V. albo-afrum Interaction 441
Hahlbrock, K., and Scheel, D. (1989). Physiology and molecular biology of phenyl-
propanoid metabolism. Ann. Rev. Plant Physiol. Plant Mol. Biol. 40347-369.
Hamdan, M. A. M. S., and Dixon, R. A. (1987). Differential patterns of protein
synthesis in bean cells exposed to elicitor fractions from Colletotrichum linde-
muthianum. Physiol.Mol. Plant Pathol.31:105-121.
Hansen, H. R.,and Weber, A. (1948).Plant diseases and pests in Denmark in 1945.
Tidsskrvt for PIanteavl51:434.
Hedlund, T. (1923). Om Nagra sjukdomar och skador pa vara lantbruksvaxter
1922. Sveriges Allmanna Jordbrukstidskrift 5:167-168 (cited by Lundin, P.,
and Jonsson, H., 1975,Agri Hortique Genetica33:17-32).
Higgins, V. J. (1972).Role of the phytoalexin medicarpin inthree leaf spot diseases
of alfalfa. Physiol. Plant Pathol.2:289-300.
Higgins, V., and Millar, R. (1968). Phytoalexin production by alfalfa in response to
infection by Colletotrichumphomoides, Helminthosporium turcicum, Stem-
phylium lotiand S. botryosum. Phytopathology58:1377-1383.
Irvine, R. F., Moor, R.M., Pollock, W.K., Smith, P. M., and Wregett, K. A.
(1988). Inositol phosphates: proliferation, metabolism and function. Phil.
Trans.R. Soc. London B320281-298.
Isaac, I. (1957a).Wilt of lucerne causedby species ofVerticillium.Ann. Appl. Biol.
45:550-558.
Isaac, I. (1957b). Verticilliumwilt of Brusselsprouts. Ann. Appl. Biol. 45:276-283.
Isaac, I., and Heale, J. B. (1961). Wilts of lucerne caused by speciesof Verticillium.
111. Viability of V. albo-utrum carried with lucerne seed;effects of seed dress-
ings and fumigants. Ann. Appl.Biol. 49675-691.
Isaac, I., and Lloyd, A. T.E. (1959). Wilt of lucerne caused by speciesof Verticil-
lium. 11. Seasonal cycle of disease; range of pathogenecity; host-parasite rela-
tions; effects of seed dressings.Ann. Appl.Biol. 42673-684.
Jorrin, J., and Dixon, R. A. (1990).Stress responses in Alfalfa (Medicago sativaL.)
11. Purification characterisation and induction of phenylalanine ammonia-lyase
isoforms from elicitor treated cell suspension cultures.Plant Physiol. 92447-
455.
Keen,N. (1978). Phytoalexins:efficient extraction from leavesbyafacilitated
diffusion technique. Phytopathology 6 81237-1239.
Keen, N. T. (1990). Phytoalexins and their elicitors. American Chemical Society
Symposium, Vol. 439,pp. 114-129.
Kendra, D. F., Fritensky, B., Daniels, C. H., and Hadwiger, L. A. (1987). Disease
resistance response genes in plants: expression and proposed mechanisms of
induction. Molecular Strategies for Crop Protection (C. J. Arntzen and C.
Ryan, eds.), AlanR. Liss, New York, pp. 13-24.
Kessmann, H., and Barz, W. (1986). Elicitation and suppression of phytoalexin and
isoflavone accumulation in cotyledons of Cicer aerietinum L. as caused by
wounding and by polymeric components from the fungus Ascochyta rabiei.J.
Khan, Z. F., and Milton, J. M. (1975). Phytoalexin production by lucerne (Medic-
ago sativa L.) in response to infection by Verticillium. Physiol. Plant Pathol.
2179-187.
442 Smith et al.
Khan, Z. F., and Milton, J. M. (1978). Phytoalexin production and theresistance

of lucerne (Medicagosativa L.) to Verticilliumalbo-utrum.Physiol.Plant
Pathol. 13:215-221.
Khan, Z. F., and Milton, J. M. (1979). Some factors affecting the production of
medicarpin and sativan by lucerne leafletsin response to V. albo-utrum. Phys-
iol. Plant Pathol.1 4 11-17.
Kreitlow, K. W. (1962). Verticilliumwilt of alfalfa. A destructive disease in Britain
and Europe not yet observed in the United States. United States Department of
Agriculture, Agricultural Research Service, ARS 34-20: 1-1 5.
Kurosaki, F., Tsurusawa, Y., and Nishi, A. (1987a). The elicitation of phytoalexins
by Ca" and cyclicAMP in carrot cells. Phytochemistry2 619 19-1 923.
Kurosaki, F., Tsurusawa, Y., and Nishi, A. (1987b). Breakdown of phosphatidyl-
inositol duringthe elicitation of phytoalexin production in cultured carrot cells.
Plant Physiol.85:601-604.
Lawton, M. A., and Lamb, C. J. (1987). Transcriptional activation of plant defense
genes by fungal elicitor, woundingand infection. Mol. Cell. Biol.2335-341.
Little, J. P.(1989). PhD thesis, University of Wales.
Mayama, S., Tani, T., Matsuura, Y., Veno, T., and Fukami, H.(1981). The pro-
duction of phytoalexins by oat in response to crown rust. Puccinia coronota f.
sp. avenue. Physiol. Plant Pathol. 19:217-226.
Murashige, T., and Skoog, F. (1962). A revisedmedium for rapid growth and
bioassays with tobacco tissue cultures. Physiologica Plantarum15:473-497.
Nelson, R. (1947). The specific pathogenesis of the Verticilliumthat causes wilt of
peppermint. Phytopathology3217.
Newcombe, G., and Robb, J. (1989). The chronological development of a lipid-to-
suberin response at Verticilliumtrapping sites in alfalfa. Physiol. Mol. Plant
Pathol. 3455-73.
Nishizuka, Y. (1986). Studies and perspectives of Protein kinase C. Science 233:
305-312.
Noble, M., Robertson, N. F., and Dowson, W. J. (1953). Verticillium wiltof
lucerne in Britain. Plant Pathol.2:31-33.
Oguni, I., Suzuki, K., and Uritani, I. (1976). Terpenoid induction in Sweet Potato
roots by Adenosine 3 ' , S "cyclic monophosphate. Agric. Biol. Chem.401251-
1252.
Onuorah, 0 . (1987). Isolation of pathogen-synthesised fraction that elicits phyto-
alexins in lucerne (Medicago sativaL,). Acta Biologica Hungaria38247-256.
Pierce, M.,and Essenberg, M. (1987). Localization of phytoalexins in fluorescent
separated from other cells by fluorescence-activated cell sorting. Physiol. Mol.
Plant Pathol.31:273-290.
Rasmussen, H.(1981). Calcium and CAMP as Synarchic Messengers. Wiley, New
York.
Richter, H., and Klinkowski, M.(1938). Wirtelpilz-Welkekrankheit an Luzerne und
Exparsette (Erreger: Verticillium albo-atrum Rke. et Berth). Nachrichtenblatt
fur den Deutschen Pfanzenschutzdienst Berlin 1857-58.
Robinson, P. S., Newton, R. P., Walton, T. J., and Smith,C. J. (1991). Calmodu-
Synthesisin Lucerne-V. albo-afrum Interaction 443
lin from lucerne: its potential role in signal transduction. Biochem. Soc. Trans.
19190s.
Robinson, P. S., Newton, R. P., Walton, T. J., and Smith, C. J. (1992). Cyclic
nucleotide phosphodiesterase activity in Medicagosativa L. Biochem. Soc.
Trans.20:3555.
Sharp, J. K., McNeil, M., and Albersheim, P. (1984). Host pathogen interactions
27. The primary structures of one elicitor active and seven elicitor inactive hexa
(P-D-glucopyranosy1)-D-glucitolsisolated from the mycelialwallsof Phy-
tophthora megasperma f. sp. glycinea. J. Biol. Chem.2591 1321-1 1336.
Smith, C. J. (1994). Phytoalexin accumulation: a defence response. New Phytolo-
gist (in press).
Smith,C. J., Newton, R. P., Mullins, C. J., and Walton, T. J. (1989). Plant
host-pathogen interaction: elicitation of phenylalanine ammonia-lyase activity
and its mediation byCa2'. Biochem. Soc. Trans.161069-1070.
Smith, C. J., and Milton, J. M. (1992). Phytoalexin accumulation in the forage
crop. Medicago sativa: a defence against the soil-borne wilt pathogen Verticil-
lium albo-atrum. Indian Phytopathol. 45:l-6.
Van Etten, H. D., Mathews, D. E., and Mathews, P. S. (1989). Phytoalexin detoxi-
fication: importance for pathogenicity and practical application. Ann. Rev.
Waldmuller, T., and Grisebach, H. (1987). Effects of R-(l-amino-2-phenylethyl)
phosphonic acid on glyceollin accumulation and expression of resistance to
Phytophthora megasperma f. sp. glycinea in soyabean. Planta 172:424-430.
Williams, C. A., and Harborne, J. B. (1990). Isoflavanoids. Methods in Plant
Biochemistry Vol. 1, Plant Phenolics. (P. M. Dey and J. B. Harborne, eds.),
Academic Press, New York.
Zaki, A., Keen, N., and Erwin, D. (1972). Implication of vergosinand hemigossy-
pol in the resistance of cotton to Verticillium albo-atrum. Phytopathology62:
1402-1406.
Zook,M., Rush, J., and Kuc, J. (1987).A role for Ca2' in the elicitation of Rishitin
and Lubimin accumulation inpotato tubertissue. Plant Physiol.84520-525.
19
Stereoselective Synthesis of
Spirovetivane-Type Phytoalexins
Chuzo lwata and Yoshiji Takemoto
Osaka University, Osaka, Japan
I. INTRODUCTION
Phytoalexins are common in all the natural product groups, e.g., isoflavo-
noids, furanoterpenoids , sesquiterpenoids, polyacetylenes, and dihydro-
phenanthrenes (Tomiyama, 1971; Grisebach and Abel, 1978). They were
defined in 1940 as defensive substances possessing antimicrobial properties
which were produced by host plants after infection (Muller and Borger,
1940), but now they are recognized to be induced by products of microbial
origin or stress treatment (injury, heat, UV light, etc.) as well as living
microorganisms (Masamune et al. , 1978). For example, representative ses-
quiterpene phytoalexins isolated from the Solanaceae family (potato, to-
bacco, etc.) are listed in Fig. 1 (1-7) (Stoessl et al., 1976). These compounds
involve unique structures. In particular, phytoalexins (4-7) belong to spiro-
vetivane sesquiterpenes bearing a spiro[4.5]decane skeleton.
Spirovetivanes are structurally characterized by the presence of a
methyl group at C-10 and an isopropyl group at C-2 in the spiro[4.5]decane
skeleton, and are furthermore divided into two groups depending on the
relative configuration between the C-1-C-5 bond and methyl group at C-10.
One group, represented by hinesol (8), and 0-vetivone (9), has the cis con-
figuration (C-10: @-methyl), which was isolated from vetiver oil as a
fragrant principle. The other, as illustrated in Fig. 1 (4-7), has the trans
configuration (C- 10: a-methyl), which is known as spirovetivane-type phy-
toalexin. Since rishitin (1) was first isolated from tuber tissue of diseased
white potatoes as phytoalexin in 1968 (Katsui et al., 1968), spirovetivane-
type phytoalexins such as solavetivone (3, lubimin (6), and oxylubimin (7)
have been successively discovered (Murai et al., 1982d).
As plants do not possess an immune system, these findings might be an
answer to the question of how plants can protect themselves against patho-
genic agents. In order to identify defense mechanisms of plants, much
effort has been made for structural determination of these phytoalexins,
445
I
446 lwata and Takemoto
rishitin (1) rishitinol (2) phytuberin (3)
an hydro-@- solavetivone (5) lubiminol (6) oxylubimin (7)

rotunol (4)
hinesol (8) (3-vetivone (9)
Figure 1 Sesquiterpene phytoalexins isolated from Solanaceae.
total synthesis, and elucidation of biosynthetic pathway and their inductive

mechanism. As a result, several groups succeeded in total synthesis of spiro-
vetivane-type phytoalexins (5-7), which helps to reveal that these phytoalex-
ins are biosynthetic intermediates of rishitin (l),exhibiting the strongest
antifungal activity (Stoessl and Stothers, 1983; Coolbear and Threlfall,
1985). However, much remains unknown and further investigations should
be able to reveal defense mechanisms of plants in detail. Major goals for
b
Spirovetivane-Type Phytoalexins 447
phytoalexin research are the identification of new analogs, which are more
effective and less toxic with a broader spectrum of antifungal or antibacte-
rial activity. The aim is to use phytoalexins as agricultural chemicals and to
elucidate the inductive or regulative mechanism of phytoalexins against
pathogenic factors.
This chapter deals mainly with the organic synthesis of spirovetivane-
type phytoalexins. For the purpose of preparing phytoalexin analogs and
elucidating the biosynthetic pathway, it is imperative to synthesize all the
highly oxygenated spirovetivane-type phytoalexins. Many synthetic strate-
gies , especially spiroannulation techniques, have been developed in more
than a decade by many synthetic chemists (Marshall et al., 1974; Krapcho,
1976, 1978; Hiroi, 1977; Murai, 1981). Herein we describe our racemic
syntheses of solavetivone (5) and oxylubimin (7)and also some elaborations
for asymmetrical synthesis of ( -)-solavetivone in detail together with oth-
ers’ synthetic approaches. For a review of other phytoalexins and biochemi-
cal aspects, see Friend and Threlfall, 1976; Keen and Bruegger, 1977; Na-
kajima, 1978; and Tomiyama, 1981.
II. SYNTHETIC STUDIES ON SPIROVETIVANE PHYTOALEXINS

A. Synthesis of (*)-Solavetivone
Solavetivone (5) is a representative member of the spirovetivane-type phyto-
alexins, which was obtained from diseased potatoes in 1974 (Coxon et al.,
1974) as well as from air-cured tobacco leaves in 1977 (Fujimori et al.,
1977). Solavetivone has been proven to play an important role in the bio-
synthetic pathway leading to formation of rishitin (Murai et al., 1982a). Its
structure was synthetically determined by Yamada et al. (1977). The crucial
problem in the synthesis of (&)-solavetivone is the construction of the
spiro [4.5]decane system, and thus far, employing different spirocyclization
techniques, three groups have succeeded in its total synthesis.
The first synthesis was accomplished via utilization of intramolecular
acid-catalyzed cyclization of acetal (10) (Fig. 2) (Yamada et al., 1977).
The obtained compound (11) is a useful intermediate, which has also been
converted to (&)-a-vetispirene and (&)-hinesol (Yamada et al., 1973b).
Although incorporation of the three-carbon unit at C-2 of 12 proceeded
nonstereoselectively to afford 13 and 14, successive transformation of the
minor adduct (14) via 15 produced (*)-solavetivone (5) as a sole product.
The second approach involves an acid-promoted ?r cyclization of bicy-
clic compounds (endo-18 and ex0-18)~which were prepared from Diels-
Alder adducts (endo-16 and exo-16), respectively (Murai et al. , 1981a,b,
1984a,b). Endo-18 and exo-18 showed similar reactivity under the acid
448
Takemoto and lwata
COOH COOH
10
p 0
l1 12
+
13
0
14
(13:14=51 : M )
15
Figure 2 (a) Ph,P=CHOMe; (b) (CH,OH),,H';(c) Li, liq. NH,, tert-BuOH,

THF; aq. (CO,H),; (d) 6 M HCl, DME, reflux; (e) LiAlH,, THF; ( f ) 2 M HCl,
DME, r.t.; (g) Ac,O, pyridine; (h) KOH, MeOH; (i) CrO,, pyridine; (j) LDA, THF;
PhSSPh; mCPBA, CH,Cl,; NaHCO,, toluene, reflux; (k) CH,=CMeMgBr, Cul,
THF; (l) H2NNH2-H20,KOH,DEG,reflux;(m) OsO,, pyridine,. THF, r.t.;
(C02H),, aq. MeOH, r.t.; (n) (PhO),P.Mel, BF,.Et,O, MeCN, r.t.; Zn, NH,Cl, aq.
EtOH, r.t.; (0) diimidazol-l-yl thioketone, MeCOEt; (MeO),P, reflux. r.t., room
temperature.
conditions, yielding the aimed compound (19) and its dehydrated.com-

pound (20), respectively. The major product (19) was heated with pyridine-
modified alumina to furnish the desired compound (5) in 13.7% overall
yield from the starting material (Fig.3).
At the beginning of research,our first concern was to develop a general
and more stereoselective methodology which could generate the spirocyclic
Spirovetivane-Type 449
system. Our firstattempt,aphotolysisofsodium salts of p[2-(2-halo-

genomethyl-l,3-dioxolan-2-yl)ethy1]phenol,gave desired spirodienone, but
only in moderate yields (Iwata etal., 1974). In the second stage, we tried a
strategy utilizinga carbenoid intermediate (Iwata et al., 1981b, 1987).Start-
ing from phenolic carboxylic acid (21), the diazoketone (22) was prepared
in 65% overall yield. When22 was heated in the presence of cuprous chlo-
-
Ye0
a, b
C
&
o CN
16
+ h0
16 : 17 = 3.5 : 1
CN
(endo : exo = 2.7 : 1)
Figure 3 (a) LDA, HMPA, THF; ClCH,CN (94%); (b) Ph,P-Mel, NaH, DMSO
(71010); (c) MeCOCH=CH2, dichloromaleic anhydride, C a 6 , reflux (16 59% and
17 17%); (d) MeLi, EhO, -78OC; DIBAL, Et,O, OOC; NaBH4, THF, H,O,O°C
(endo 96'70, ex0 75%); (e) MsCl,pyridine;(COzH)2, 33% aq. acetone [endo:
19(63%),20(23%), exo: 19(69%),20(26010)];(f)pyridine-Al,O,, 22OOC (60%).
ride in benzene, cyclizationtook place smoothlyto give the aimed spirodie-

none (23)in 56% yield. This carbenoid-mediated cyclization proceeds under
mild conditions and turns out to be applicable to many other phenolic
diazoketones (Iwata et al.,1977, 1980). Inspection of minorproduct, dihy-
droazulene derivative (a),indicates that the catalytic decomposition of
phenolic a-diazoketones involves norcaradiene intermediate (25). Reduc-
tion of the prochiral ketone (23) by lithium tri-tert-butoxyaluminum hy-
dride gave monoalcohol (26), which was subjected to Birch reduction pro-
viding possible four diastereoisomers (27a, 27b, 27c,in 78% 27d) yield in
the ratio of 91 :7 : 1 : 1 (Fig. 4). This stereoselectivity can be explained by
the intramolecular proton migration from the hydroxyl group to formed
dienolate dianion (Fig. 5). That is, two possible epimeric dianion (or radical
anion) transition states, A (chair form) and B (boat or half-chair form),
might be considered inthe reduction process. Thereis a strong interaction
between the alcohol oxygenand the methyl group at C-10 in the transition
A but not in B. Therefore, the protonation would proceed predominantly
via the transition state B, thus producing the isomer 27a as a major product.
In spite of numerous examples of the Birch reduction of a,P-unsaturated
carbonyl compounds (Caine, 1976), there are few examples of neighboring
group participation in the protonation at the /3 position of enones (Mc-
Murry et al., 1978), and so this is the first example of controlling the stereo-
chemistry at the P carbon of enolate dianionby 1,4-asymmetrical induction
of the hydroxyl group (Iwata et al., 1981a). Introduction of three carbon
unit at C-2 of the major product (27a) was best achieved as follows: Thi-
oacetalization of the ketone and mesylation ofthe alcohol were followed by
the substituted reaction with sodium diethyl malonate anion to give 28
bearing desired stereochemistry in55% overall yield. Conversion of malo-
nyl group of 28 into allylic alcoholwas carried out by a three-step sequence
to afford allylic alcohol(29). Removal ofthe hydroxyl group and sequential
deprotection ofthe thioketal group of 29 provided (*)-solavetivone (5).
B. Synthesis of (*)-Oxylubimin
Oxylubimin (7), which was isolated from tuber tissue of white potatoes
infected by Phytophthora infestans, is a most highly oxygenated spiroveti-
vane-type phytoalexin bearing six asymmetrical carbon centers (Katsui et
al., 1974; Stoessl et al., 1975). This
natural product is not only a representa-
tive from the viewpoint of its strong biological activity and the complex
stereochemistry, but it is also a promising biosynthetic intermediate inthe
major pathway to rishitin via solavetivone in vivo (Sat0 et1978). al.,
There have been tworeports on the total synthesis of (*)-7. The first
synthesis was achieved by Murai et al. in 1982 through a related approach
SpirovetivaneType Phytoalexins 45 1
9"
-p-""
?H
'COOH
21
,Q1-9 25 23
26
J (27a:2m:27c:27d = 91:7:1:1
,\\"
&Et '7-
28 5
Figure 4 (a) CuCl, C&, reflux (56010); (b) LiAlH(tert-OBu),, THF, r.t. (80%);
(c)Li,liq. NH,, THF, toluene, -85OC (78%); (d) 1,2-ethanedithiol, BF,.EhO,
MeOH, r.t. (84010); (e) MeS02C1, pyridine, r.t. (98%); ( f ) diethyl malonate, NaH,
DME, reflux (67070); (g) KOH, EtOH, r.t. (99010); (h) formarin, diethylamine, re-
flux; NaOAc, AcOH, reflux (93070); (i) DIBAL, toluene, -7OOC(87010); (j) hexa-
chloroacetone, triphenylphosphine, THF, r.t. (90070); (k) Zn, c&&, EtOH, AcOH,
reflux (l) Mel, CaCO,, MeCN,HzO, reflux (63%).
Figure 5 Intramolecularprotonmigration from the hydroxylgrouptoformed

dienolate dianion. (A) Chair form, (B) Boat or half-chair form.
to solavetivone synthesis (Fig.6) (Murai et al., 1982c, 1984c,d). The prob-

lem of stereoselectivity in a Diels-Alder reaction was nicely resolved by
using bicyclic diene (30) to afford antiadduct (31) diastereoselectively as a
mixture of end031 and exo-31. However,the acid promoted T cyclization
of 31 and cy' hydroxylation of the major product (32) resulted in proceeding
in low diastereoselectivity (32133 = 35 :25 and 34/35 = 51 :25). Subse-
quently, introducing two asymmetrical centers at C-6 and C-8 through hy-
drocyanation (36/37 = 1 : 1) and borane reduction, enone (34) was con-
verted to (*)-oxylubimin (7) and (*)-epioxylubimin (38). Consequently,
two natural products (7 and 38) which are epimerized by each other under
basic conditions, were synthesized from 34 in the same way. They also
succeeded in total synthesis of (*)-lubimin (6) and its diastereoisomers
from 32.
Our synthetic route for (*)-oxylubimin (7), is shown in Fig. 7 (Iwata
et al., 1985b, 1990). As the starting material, we employed the common
intermediate (27a) in solavetivone synthesis. Our synthetic plan involved
two key steps: 1.) the stereoselective Q ' hydroxylation of a,&unsaturated
ketone (39) into 40a, and 2.) regio-and stereoselective hydride reduction of
the cy'-hydroxyenone (40a or 41) into trans-diol (42a). The first problem
was overcome with a new hydroxylation method developed in our labora-
tory (Iwata et al., 1985a). By treatment of silyl enol ether derivative of
enone (39) with triphenylphosphite ozonide(TPPO) and then triphenylphos-
phine, a diastereo mixtureof 40a and 40b was obtained in 71% yield with
high stereoselectivity (40a/40b= 8 : 1). Onthe other hand, oxidationof 39
with reagents such as mchloroperbenzoic acid (MCPBA) (Rubottom and
Gruber, 1978), MoOPH (vedejs et al., 1978), and Mn(OAc)3 (Dunlap et
al., 1984) gavea mixtureof 40a and 40b in moderate yields,but the stereo-
selectivity turned out to be unsatisfactory (Table 1). This high stereoselec-
tivityusing TPPO should be attributable to the reactionintermediate.
Namely, in the TPPO oxidation the oxidant is consideredto react simulta-
neously at two carbon centers, C-6 and C-9, in the silyl enol ether and
OH
31
32 33
(32:33 = 3525)
c
34 35
I
1
(2)-epioxylubimin (38)
Figure 6 (a) ClCH,COCl, AlCl,, CS, (84%); (b) NaBH4; Hz, Pd-C (100%); (c) Li,
liq. NH,, ether; EtOH (8lV0); (d) methyl acrylate, dichloromaleicanhydride, l5OoC
(70010);(e) (CO,H),; AGO, pyridine (70010); (f) TsNHNH,; MeLi (71%); (g) MsCl;
(CO,H),, H20, methyl isobutyl ketone(60%); (h) LDA; TBSC1; PhC03H (34: 51%,
35: 25%); (i) HCN, Et,Al (36: 40%, 37: 39%); 6) BH,-NH,, aq. MeOH; TBSCI,
imidazole, DMF (80% for 36, 58% for 37); (k) pyridine-Al,O,,220OC; aq. HF,
THF,MeCN(67%,70%);(l)DIBAL(57%,55%);(m)5% KOH,MeOH.
453
-*
a
\\"'
- b, C y:,&
l
1-BUOCO '\'Q
t-BuOCO
40a:X=OH,Y=H
39 40b:X=H,Y=OH
41:X=OMOM,Y=H
t-BuOCO
/
42a:X=OH,Y=H 44
42b:X=H,Y=OH
43:X=OMOM,Y=H
4s 46
Figure 7 (a)t-BuCOC1,pyridine(92%); (b) LDA, DME, -18OC;TMSCI;

TPPO, CH,C12.-5OOC; Ph3P, ether, r.t. (71%); (c) MOMCI, (i-Pr),NEt, CH,Cl,
(84% for 40a, 91% for 42a); (d) NaBH,,CeCI,.7H,O,MeOH,O°C (90070); (e)
MeLi, ether (95%);(f)MsCl, pyridine;NaH, CH,(CO,Et),, DME, reflux (97Vo); (g)
NaH, DME;Red-Al,DME(74'70); (h) NaBH,, CoCl2*6H2O,EtOH (78010); (i)
MsCl,pyridine (79Vo); G) SeO,,xylene, reflux;NaBH,,MeOH (70Vo); (k) Hz,
RaneyNi(W2), EtOH (90010); (l) PCC, NaOAc, CH,Cl, (82%); (m) (CH,SH),,
BF3-Et20,CHzClz(879'0); (n) Nal, DBU, DME (66%);(0) Mel, CaC03, aq. MeCN,
reflux (43%).
Spirovetivane-Type 455
Table 1 a' Hydroxylation of the Enone (39) with Various Oxidants

Ratio
Reaction
conditions
Entry X Yield (070) ( h40b)
:
1 LDA, DME, - 18OC; TMSCl;

H 71 8:l
TPPO, CHZC12, -50°C,
Ph,P, Et,O, room temp.
2 LDA, DME, - 18OC; TMSCl; TMS 63 1:l
MCPBA, hexane, - 15O C
3 LDA, THF, -78 'C; TBS 52 (83)" 2: 1
MoOPH, -22OC; TBSCl,
imidazole, DMF
4 C,H,,
Mn(OAc),, reflux Ac 62 2: 1
'Yield based on the consumed starting material.
form an endoperoxide (C) (Fig.8). Because of severe 1,3-diaxial interaction

between TPPO and the methylene unit at C-l and C-4, the silyl enol ether
was attacked from the top side of conformation A by the oxidant, leading
to a predominant formation of 40a. Conversion of a'-hydroxyenone (40a)
in hand into trans-diol, the second crucial step, was achieved as follows.
Since all attempts to reduce 40a under various conditions resulted in the
formation of a mixture oftrans- and cis-diol withoutsatisfactory stereosel-
ectivity,methoxymethyl(MOM)ether (41) was reducedwithNaBH,in
methanol in the presence of cerium chloride at O°C to give desired trans-
diol (42a)as a major product (42a/42b = 6.4 : 1) (Table 2). Introduction
of a three-carbon unit at the C-2 position in 43 was accomplished in a
similar manner as described in solavetivone synthesisto afford 44 in 92%
overall yield. Subsequently, by modified Marshall's method (Marshall et
al., 1967), diester (44)was transformed into corresponding allylic alcohol
in one step, followed by cobalt hydride reduction and mesylation to give
mesylate (45). Successive reactions by allylic oxidation, catalytic hydroge-
nation of the C-6-C-7 double bond, and PCC oxidation providedthe satu-
rated aldehyde(46) as a single isomer concerningat C-6. Finally, derivation
of 46 into (*)-oxylubimin (7) was straightforward, i.e., thioacetalization
accompanied with deprotection the of bis-MOM ether, elimination of meth-
anesulfonic acid with base, and deprotection of thioacetal group led to
stereoselectiveformation of (*)-oxylubimin (7).
C. Other Phytoalexins
Based on our strategy described above,we attempted to synthesize a series
other than solavetivone (S), and oxylubi-
Of spirovetivane-type phytoalexins
0
8
a
0
C
.I
SpirovetivaneType 457
Table 2 Reduction of a'-Hydroxyenone (41) Under Various Conditions

Ratio'
Entry conditions
Reaction Yield (Vo) (42a/42b)
1 Zn(BH,),,
Et,O, O°C 79 1.9
2 LiBH,(n-Bu),
toluene,
hexane, -78OC 84 1.2
3 NaBH,CH,
MeOH, 2 N HCl, 71 5.6
room temp.
4 NaBH,, O°C
CeCI,.7H20,
MeOH, 90 6.4
'Ratio basedon isolated yields.
min (7). Consequently, total syntheses of 3-hydroxysolavetivone (aglycone

A*) (47) (Iwata et al.,1986), lubiminol(48)(Iwata et al.,1984)and aglycone
A3 (51) (Iwata et al., 1981b)togetherwithformalsynthesesof10-epioxy-
lubimin (38) (Murai et al., 1982c), isolubimin (49) (Katsui et al., 1982),
solanascone (50) (Fujimori et al., 1978),and aglycone 4 (52) (Anderson et
al., 1977) were accomplished. The results obtained in the synthetic studies
of spirovetivane-type phytoalexins were briefly depicted in Fig.9 (Iwata et
al., 1988a,b; 1989).
D. Asymmetrical Synthesis of (- hSolavetivone

Therehasbeen no report about asymmetrical synthesis oftrans-spiroveti-
vanes, although a few groups completed in total synthesis of cis-spiroveti-
vanes, e.g., (+)-hinesol(8), (-)P-vetivone (9), starting from chiral natural
products (Buddhasukh and Magnus, 1975; Deighton et al., 1975). Herein
we would like to describe some synthetic studiesfor optically activetrans-
spirovetivanes.
Our synthetic strategyfor asymmetrical synthesis of spirovetivane-type
phytoalexins relieson vinylic sulfoxide(53) possessing chiral sulfinyl group
(Fig. 10).The key reactions to construct chiralquaternary carbon centers in
our route consist of the formation of chiral cyclopropane ring(53 + 54),
the regioselective cleavage ofthe cyclopropane ring(54 55), and spiroan-
+
nulation reactionof alkylmercury chloride(55 + 56 and 57). The obtained

ketoester (56) can be transformed into solavetivone (5) and hinesol (8) via
58. Another ketoester (57) is also a good intermediate for the synthesis of
agarospirol. According to the strategy, we first examined a Michael reac-
tion of Grignard reagents into chiral vinylic sulfoxides. From the prelimi-
nary studies, we know that a reaction of a vinylic sulfoxide with allyl
magnesium bromideaffords a mixture of monoallylated product and dially-
lated compound through a additive Pummerer-type reaction (Iwataet al.,
oxMubimin (71 lbpioxylubimin (38)
I
N
\, *. $"
OH
\*.c$
OR
_.)
L
isolubirnin (49)
lubiminol (48)
solavetivone (5)
our synthesis 1
formal synthesis
aglycone A, (51)
Figure 9 Results obtained in synthetic studiesof spirovetivane-type phytoalexins.

SpirovetivaneType Phytoalexins 459
-Lo,Et
54 55
g9
53
,V'*
*'''<~
58 * ' I f
hlnesol (8)
solavellvone (5)
57 sgarospirol
Figure 10 Synthetic strategy for opticallyactivespirovetivane-type sesquiter-

penes.
1991). Therefore, we undertook the above reaction with vinylic sulfoxide

(53) bearing a chloromethyl group, hoping for a different modeof reaction
(Fig. 11). The vinylic sulfoxide (53) was prepared from 59 by a four-step
sequence, which wastreated with allyl magnesium bromidetetrahydrofu-
in
ran at -78OC to providecyclopropylsulfoxide (54) in 66% yield as a
single diastereoisomer along with coupling compound (60). The absolute
configuration (Ss, R, S) of 54 was determined by an x-ray crystallographic
study of its derivative. This highly diastereoselective cyclopropanation can
be explained by the comparisonof the energetic stability between two che-
lated transition models (A and B). Considering that the magnesium center
of Grignard reagent can coordinate both to the oxygen atom of sulfinyl
group and to the chloride atom at the y position, conformerA in Fig. 12 is
much more stablethan conformer B because ofthe severe steric interaction
between the tolyl and chloromethyl groups, and then the allyl group of the
reagent is introduced exclusively from the a face. In order to resolve the
second problem, i.e., regioselective cleavage of cyclopropane ring, sulfox-
ide (54) was transformed into cyclopropyl sulfide(61) in four steps. Among
various electrophiles, reaction of 61 with mercury(I1) trifluoroacetate in
methylene chloride (CH,CI,) in the presence of sodium acetate at room
temperature gave rise to the desired a,&unsaturated y-sulfenylalkyl mer-
a-d
(x S(0)TOl
e 54
+
\\
53
&l
(x S(0)Tol
Figure 11 (a)n-BuLi; TolS(0)O-(l)-mentyl (9107'0); (b) p-TsOH, acetone, H 2 0

(97%); (c)NaBH,,MeOH (100%);(d)MsCl,LiCl,2,4,6-collidine, n-Bu4NC1,
DMF, O°C (94%); (e)(Allyl)MgBr, THF, -78OC (82%); (Q (CF,CO),O, Nal,
acetone, O°C (100%); (g) 9-BBN, THF, OOC;3M NaOH, 30% H20z,r.t. (95C70);(h)
(COCI),, DMSO, CH,Cl,, -5OOC; Et3N, r.t. (87%); (i) (EtO),POCH2CO2Et,NaH,
THF, r.t. (82010);G) Hg(OCOCF,),, NaOAc, CH2C12, r.t.; saturated aq. NaCl, r.t.
(89%); (k) n-Bu,SnH, CH,Cl,,-4OOC + r.t. (86%); (l) Li2PdC14, DMF, THF,
reflux (91Oi'o); (m) 10% HCI, MeCN, 60°C (84% for 62 and 63); (n) H,, 10% Pd-C,
EtOH, r.t. (100%).
Spirovetivane-Type Phytoalexins 461
A B
Figure 12 Plausible reaction mechanism of the asymmetric cyclopropanation.
cury chloride(55) in a highly regioselective manner without affecting a,&

unsaturated ester. Various reaction conditions were attempted to develop a
new type of spiroannulation method. Sequential treatment of 55 by the
radical cyclization (n-Bu,SnH, AIBN) and acid-catalyzed hydrolysis gave
ketoester (56 and 57) in 72% overall yield in the ratio of 11 :89. On the
other hand, palladium(I1)-assisted cyclization followed by acid-catalyzed
hydrolysis and catalytic hydrogenation alsoafforded ketoester (56 and 57)
with the former as the major product in 76% overall yield (56/57 = 83 :
17) (Imanishi et al., 1992). These obtained products (56 and 57) are useful
intermediates for importantnaturalproducts, e.g., natural(-)-solavetiv-
one from 56 and unnatural (+)-agarospirol from 57. We are now further
elaborating our research withan aim for total synthesis of (-)-solavetivone
(5) utilizing 56.
111. EPILOG
As was detailed in the foregoing sections, considerable progress has been
made inthe total synthesis of naturally occurring spirovetivanes. Numerous
strategies have been developed for construction of the spiro[4.5]decane ring
system, and some routes have enabled the efficient regio- and/or stereospe-
cificsynthesisoftrans-spirovetivanephytoalexins.However, structural
modifications have not been reported in synthetic spirovetivanes except for
a minor one, i.e., dideuterated (*)-solavetivone prepared by Murai et al.,
(1982a), who extended the research for establishing the biosynthetic pro-
duction of rishitin from solavetivone via lubimin and oxylubimin on the
basisofexperimentswith the ( -+)-[2,2-ZH,]solavetivone.Thesynthetic
methodology now available should allow the preparation of analogs having
more profound structural modification, which provides us much opportu-
nity to get more efficient drugs or prodrugsand to elucidate inductiveand
regulative mechanism of phytoalexin biosynthesis, including the isolation
of enzymes responsiblefor the formation of spirovetivane phytoalexins.
462 lwata and Takernoto
An attempt has been made in this chapter to review the most significant
synthetic advances reported overthe last two decades. There can be little
doubt that isolation of new types of phytoalexins from natural sources as
well as developmentof new synthetic approacheswill continue, and the next
decade should seefurther advances in synthetic methodologyand synthetic
analogs, as well as the utilization of the enzymes involved in phytoalexin
biosynthesis for the preparation of spirovetivane phytoalexins, resulting in
improved plant disease therapy.
REFERENCES
Anderson, R. C., Gunn, D. M., Murray-Rust, J., Murray-Rust, P., and Roberts,
J. S. (1977). VetiSpirane sesquiterpene glucosidesfrom flue-cured Virginiato-
bacco: structure, absolute stereochemistry, and synthesis. X-Ray structure of
the p-bromobenzenesulphonateof one of the derived aglycones.J. Chem. Soc.,
Chem. Commun.(1): 27-28.
Buddhasukh, D., and Magnus, P. D. (1975). Synthesis of (+)-hinesol and lo-epi-
(+)-hinesol. J. Chem. Soc. Chem. Commun.(23): 952-953.
Caine, D. (1976). Reduction and related reactions of a,D-unsaturated compounds
with metals in liquidammonia. Org. React.23: 1-258.
Coolbear, T., and Threlfall, D. R. (1985). The biosynthesisoflubimin from
[l-'4C]isopentenyl pyrophosphate by cell-free extracts of potato tuber tissue
inoculated with an elicitor preparation from Phytophthora infestans. Phyto-
chemistry 24(9): 1963-1971.
Coxon, D. T.,Price, K. R., Howard, B., Osman, S. F., Kalan, E. B., and Zacha-
rius, R.M. (1974). Two new vetispirane derivatives: stress metabolites from
potato (Solanum tuberosum)tubers. Tetrahedron Lett. (34): 2921-2924.
Deighton, M., Hughes, C. R., and Ramage, R. (1975). Stereospecific synthesis of
(-)-agarospirol and (-)-0-vetivone. J. Chem. Soc., Chem.Commun. (16):
662-663.
Dunlap, N. K., Sabol, M.R., and Watt, D. S. (1984). Oxidation of enones to
a'-acetoxyenones using manganesetriacetate. TetrahedronLett. 25(51): 5839-
5842.
Friend, J., and Threlfall, D. R. (1976). BiochemicalAspects of Plant-Parasite
Relationships. Academic Press, London.
Fujimori, T., Kasuga, R., Kaneko, H., and Noguchi, M. (1977). Isolation of sola-
vetivone from Nicotiana tabacum. Phytochemistry16(3): 392.
Fujimori, T., Kasuga, R., Kaneko, H., Sakamura, S., Noguchi, M., Furusaki, A.,
Hashiba, N., and Matsumoto, T. (1978). Solanascone: a novel sesquiterpene
ketone from Nicotiana tabacum. X-Ray structure determination of the corre-
sponding oxime.J. Chem. Soc., Chem. Commun.(13): 563-564.
Grisebach, H., and Ebel, J. (1978). Phytoalexins, chemical defense substances of
higher plants? Angewandte ChemieZnt. Ed. (English)Z7(9): 635-647.
Hiroi, K. (1977). Recent synthetic studies on carbocyclic spiro compounds. Yuki
Gousei Kagaku Kyokaishi35(12): 1029-1044.
Imanishi, T., Ohra, T., Sugiyama, K., Ueda, Y., Takemoto, Y., and Iwata, C.
(1992). Novel asymmetriccyclopropanation utilizing sulfinyl chirality: applica-
tion to construction of a spiro[4S]decane system. J. Chem. Soc., Chem. Com-
mun. (3): 269-270.
Iwata, C., Nakashita, Y., and Hirai, R. (1974). Studies on the syntheses of spiro-
dienone compounds.IV. Photolysis of sodiumsalts of p-[2-(2-halogenomethyl-
1,3-dioxolan-2-yl)-ethyl]phenolin N,Ndimethylformamide. Chem.Pharm.
Bull. 22(1): 239.
Iwata, C., Yamada, M., Shinoo, Y., Kobayashi, K., and Okada, H. (1977). Intramo-
lecular cyclisation of phenolic a-diazoketones. Novel synthesis of the spiro-
[4.5]decane carbon framework. J. Chem. Soc., Chem. Commun.(23): 888-889.
Iwata, C., Yamada, M., Shinoo, Y., Kobayashi, K., and Okada, H. (1980). Studies
on the synthesesofspiro-dienonecompounds.VII.Novelsynthesisof the
spiro[4.5]decane carbon framework. Chem. Pharm.Bull. 28(6): 1932-1934.
Iwata, C., Miyashita, K., Ida, Y., and Yamada, M. (1981a). Effects of neighbour-
ing hydroxy-groups in metal-ammonia reductions of a,&unsaturated carbonyl
compounds. J. Chem. Soc., Chem. Commun.(10): 461-463.
Iwata, C., Fusaka, T., Fujiwara, T., Tomita, K., and Yamada, M. (1981b). Total
synthesisof(*)-solavetivone;X-raycrystal structure of 2-hydroxy-6,lO-
dimethylspiro[4.5]decd-en-8-one. J. Chem. Soc., Chem. Commun. (10): 463-
465.
Iwata, C., Kubota, H., Yamada, M., Takemoto, Y., Uchida, S., Tanaka, T.,
and Imanishi, T. (1984). Stereocontrolled total synthesis of (*)-lubiminol, a
spirovetivane phytoalexin.TetrahedronLett. 25 (31): 3339-3342.
Iwata, C., Takemoto, Y., Nakamura, A., and Imanishi, T. (1985a). Oxidation of
2-trimethylsilyloxy-1,3-dienswith triphenylphosphite ozonide. A regioselective
a’-hydroxylation of a,&unsaturated ketones. TetrahedronLett. 26(27): 3227-
3230.
Iwata, C., Takemoto, Y., Kubota, H., Kuroda, T., and Imanishi, T. (1985b). A
stereoselective total synthesis of (*)-oxylubimin. Tetrahedron Lett. 26(27):
3231-3234.
Iwata, C., Nakamura. A., Takemoto, Y., and Imanishi, T. (1986). Stereoselective
total synthesis of(*)-3-hydroxysolavetivone. Chem. Ind. (20): 712-713.
Iwata, C., Yamada, M.,Fusaka, T., Miyashita, K.,Nakamura, A., Tanaka, T.,
Fujiwara, T., and Tomita, K. (1987). Total synthesis of (i)-solavetivone and
aglycone A,. Regio- and stereo-selective Birch reduction of 6,10dimethyl-2-
hydroxyspiro[4.5]dec6,9-dien-8-one.Chem. Pharm.Bull. 35(2): 544-552.
Iwata, C., Takemoto, Y., Kubota, H., Yamada, M.,Uchida, S., Tanaka, T., and
Imanishi, T. (1988a). Synthetic studies on spirovetivane phytoalexins. IV. A
stereoselective synthesis of (~)-3-hydroxysolavetivone.Chem. Pharm. Bull.
36(10): 2643-2646.
Iwata, C., Takemoto, Y., Kubota, H.,Yamada, M., Uchida, S., Tanaka, T.,
and Imanishi, T. (1988b). Synthetic studies on spirovetivane phytoalexins. 11.
Stereoselective
synthesis
of (2RS,5RS,6SR,8RS,lOSR)-6-hydroxyrnethyl-8-
methoxymethoxy-lO-methyl-2-pivaloyloxyspiro[4.5]decane.Chem.Pharm. Bull.
36(11): 4581-4584.
464 Takemoto lwata and
Iwata, C., Takemoto, Y., Kubota, H., Yamada, M., Uchida, S., Tanaka, T., and
Imanishi, T. (1989). Synthetic studies on spirovetivane phytoalexins. 111. A
total synthesis of(&)-lubiminol. Chem. Pharm. Bull.37(4): 866-869.
Iwata, C., Takemoto, Y., Kubota, H., Kuroda, T., and Imanishi, T. (1990). Syn-
thetic studies on spirovetivane phytoalexins.V. A stereoselective total synthesis
of (*)-oxylubimin. Chem. Pharm. Bull.38(2): 361-365.
Iwata, C., Maezaki, N., Kurumada, T., Fukuyama, H., Sugiyama, K., and Iman-
ishi, T. (1991). New carbon-carbon bond formation by the Pummerer-type
reaction of vinylic sulfoxides with allylmagnesium bromide. J. Chem. Soc.,
Chem. Commun. (19): 1408-1409.
Katsui, N., Murai, A., Takasugi, M., Imaizumi, K., Masamune, T., and Tom-
iyama, K. (1968). The structure of rishitin, a new antifungal compound from
diseased potato tubers. J. Chem. Soc., Chem. Commun. (1): 43-44.
Katsui, N.,Matsunaga, A., and Masamune, T. (1974).The structure of lubiminand
oxylubimin, antifungal metabolites from diseased potato tubers. Tetrahedron
Lett. (51/52): 4483-4486.
Katsui, N., Yagihashi,F., Murai, A., and Masamune, T. (1982). Structure of
epilubimin, epioxylubimin, and isolubimin, spirovetivane stress metabolites in
diseased potato. Bull. Chem. Soc. Jpn. 55(8): 2424-2427.
Keen, N. T., and Bruegger, B. (1977). Hostplant resistance to Pests. ACS Sympo-
sium Series,No.62 (P.A. Hedin, ed.), p. 1.
Krapcho, A. P. (1976). Synthesis of carbocyclicspiro compounds via rearrangement
routes. Synthesis (7): 425-444.
Krapcho, A. P.(1978). Synthesis of carbocyclic spiro compounds via cycloaddition
routes. Synthesis (2): 77-126.
Marshall, J. A., Anderson, N. H., and Hochstetler, A. R. (1967). The reduction of
malonic enolates with lithium aluminum hydride. J. Org. Chem. 32(1): 113-
118.
Marshall, J. A., Brady, S. F., and Anderson, N. H. (1974). The chemistryof
spiro[4.5]decane sesquiterpenes.Fortschritte Der Chemie Organischer Naturs-
toffe, Vol. 31. Springer-Verlag. New York, pp. 283-376.
Masamune, T., Murai, A., and Katsui, N. (1978).Phytoalexin no taisha. Kagaku to
Seibutsu 16(10): 648-660.
McMurry, J. E., Blaszczak, L. C., and Johnson, M. A. (1978). On the stereochemis-
try of reduction of 9-methyl-l-carboxy-A4-3-octalone: a remarkable lithium-
ammonia reduction. TetrahedronLett. (19): 1633-1634.
Muller, K. O., and Boger, H. (1940). Arb. Biol. Reichsanstalt. Landw. Forstw.
Berlin 23: 189.
Murai, A. (1981). Recent advancesin the synthesis of spirovetivane sesquiterpenes.
Yuki Gousei Kagaku Kyokaishi 39(10): 893-908.
Murai, A., Sato, S., and Masamune, T. (1981a). Efficient synthesisof (*)-
solavetivone. J. Chem. Soc., Chem. Commun. (17): 904-905.
Murai,, A. Sato, S. and Masamune, T. (1981b). n-Cyclization: the synthesis of
(i)-solavetivone and (*)-hinesol. TetrahedronLett. 22(11): 1033-1036.
Murai, A., Sato, S., Osada, A., Katsui, N., and Masamune, T. (1982a). Biosynthe-
sis from solavetivone of the phytoalexin rishitin in potato. Implicit role of

solavetivone as an activator. J. Chem. Soc., Chem. Commun.(1): 32-33.
Murai, A., Sato, S., and Masamune, T. (1982b). Total synthesis of (+)-15-
norsolavetivone. J. Chem. Soc., Chem. Commun.(9): 511-512.
Murai, A., Sato, S., and Masamune, T. (1982~).Total synthesis of (*)-lubimin and
(Et)-oxylubimin. J. Chem. Soc., Chem. Commun.(9): 513-514.
Murai, A., Katsui, N., and Masamune, T. (1982d). Significant role of syntheses of
phytoalexins. Kagaku no Ryoiki36(10): 673-685.
Murai, A., Sato, S., and Masamune, T. (1984a). A general synthetic approach of
spirovetivanes. The synthesis of (f)-solavetivone, (*)-hinesol, and related
compounds. Bull. Chem. Soc. Jpn. 57(8): 2276-2281.
Murai, A., Sato, S., and Masamune, T. (1984b). Efficient synthesisof (*)-
solavetivoneand ( f)-[8,8-2H2]solavetivone.Bull. Chem. Soc. Jpn.57(8): 2282-
2285.
Murai, A., Sato, S., and Masamune, T. (1984~).Total synthesis of (*)-lubimin and
(*)-oxylubimin. I. Synthesisof(*)-15-norsolavetivone and relatedcom-
pounds. Bull. Chem. Soc. Jpn. 57(8): 2286-2290.
Murai, A., Sato, S., and Masamune, T. (1984d). Total synthesis of (*)-lubimin
and (*)-oxylubimin. 11. Transformation of (*)-15-norsolavetivone into ( f)-
lubimin, and (*)-oxylubimin, and related compounds. Bull. Chem. Soc. Jpn.
57(8): 2291-2294.
Nakajima, M. (1978). Phytoalexin no seisei to tenkan in seibutsu no seigyo kikou.
Kagaku Zokan 75: 31-45.
Nystrom, J.-E., and Helquist, R. (1989). Short intramolecular Diels-Alder approach
to spirovetivanes. Total synthesis of dl-hinesol. J. Org. Chem. 54(19): 4695-
4698 and references cited therein.
Rubottom, G. M., and Gruber, J. M. (1978). m-Chloroperbenzoic acid oxidation
of 2-trimethylsilyloxy-1,3-dienes.Synthesisof a-hydroxy and a-acetoxy en-
ones. J. Org. Chem. 43(8): 1599-1602.
Sato, K., Ishiguri, Y.,Doke, N., Tomiyama, K.,Yagihashi, F., Murai, A., Katsui,
N., and Masamune, T. (1978). Biosynthesis of the sesquiterpenoid phytoalexin
rishitine from acetate via oxylubimin in potato. Phytochemistry 17(11): 1901-
1902.
Stoessl, A., Stothers, J. B., andWard, E. W. B. (1975). A 2,3-Dihydroxygerm-
acrene and other stress metabolites of Datura stramonium. J. Chem. Soc.,
Chem. Commun.(11): 431-432.
Stoessl, A., Stothers, J. B., and Ward, E. W. B. (1976). Sesquiterpenoid stress
compounds of the Solanaceae. Phytochemistry Z5(6): 855-872.
Stoessl, A., and Stothers, J. B. (1983). Biosynthesis ofantifungal stress metabolites
from potato: observation of hydride shifts via the “P-hop” from incorporation
of [%‘H,, 2-I3C]acetateand ‘,Cmr. Can. J. Chem. 61(8): 1766-1769.
Tomiyama, K. (1971). Phytoalexin: chemistryand biology. Kagaku no Ryoiki31(7):
658-668.
Tomiyama, K. (1981). Shokubutsukansenniokeru Kabinshosaiboushino Seirigaku.
Seibutsukagaku33(1): 17-25.
466 Takemoto lwata and
Vedejs, E., Engler, D. A., and Telschow, J. E. (1978). Transition-metal peroxide

reactions, synthesis of a-hydroxycarbonyl compounds from enolates. J. Org.
Chem. 43(2): 188-196.
Yamada, K., Nagase, H., Hayakawa, Y., Aoki, K., and Hirata, Y. (1973a). Syn-
thetic studies on spirovetivanes. I. spirocondensationof a 443 ’-formylpropyl)-
3-cyclohexenone and stereospecific total synthesis of dl-8-vetivone. Tetrahe-
dron Lett. (49): 4963-4966.
Yamada, K., Aoki, K., Nagase, H., Hayakawa, Y., and Hirata, Y. (1973b). Syn-
thetic studies on spirovetivanes. 11. Stereospecific total synthesis of dl-hinesol,
dl-a-vetispirene, and dl-8-vetispirene(dl-8-isovetivenene). Tetrahedron Lett.
(49): 4967-4970.
Yamada, K., Goto, S., Nagase, H., and Christensen, A. T. (1977). Total synthesis
and stereostructure of (&)-solavetivone, a stress metabolite from infected po-
tato tubers; X-ray crystal and molecular structure of an intermediate in the
synthesis. J. Chem. Soc., Chem. Commun.(16): 554-555.
20
Enzymic Conversion of
Furanosesquiterpene in Ceratocystis
fimbriata-Infected Sweet Potato
Root Tissue
Masayuki Fujita
Kagawa University, Kagawa, Japan
Hiromasa lnoue
Kyushu Dental College, Kitakyushu, Japan
K. Oba and lkuzo Uritani*
Nagoya Women's University, Nagoya, Japan
1. INTRODUCTION
Sweet potato (Ipomoea batatas Lam.) root tissues accumulate bitter and
oily substances in the mold-damaged regions when they are infected with
pathogenic fungus such asCeratocystisfimbriata Ell. and Halst. They are
antifungal and Ehrlich's reagent-positive furanosesquiterpenes, which con-
sist of several major components such as ipomeamarone [3 + 4' + 2"]
and ipomeamaronol[3 + 4' 3"1 and many kinds of minor family com-
+
pounds derived from them. Ipomeamarone was first isolated and deter-
mined in the structure among phytoalexins in the plant kingdom (Hiura,
1943; Kubota andMatsura, 1953). Its chemical structure is shown in Fig.1.
These furanosesquiterpenes are produced not only by infection of patho-
genic fungibut also by treatment with poisonousheavy metal ions (Uritani
et al., 1960) and mycotoxins (Fujita and Yoshizawa, 1987) in sweet potato
root tissues. Therefore, various kinds of poisons, including mercuric chlo-
ride, have also been used as artificial phytoalexin inducers (elicitors)
for the
biochemical investigation on synthesis of furanosesquiterpenes inthe root
tissues (Oba and Uritani, 1981; Fujita and Yoshizawa, 1989; Fujita, 1985).
Using various I4C-labeled intermediates, it was elucidated that the furan-
*Current affiliation:Aichi Konan Gakuen School Corporation, Konan City,Japan
467
468 Fujita et al.
I Farnesol I
IlpomeamarondI
I
I
I I
Figure 1 Chemicalstructures of farnesolandipomeamarone. (I), Furan ring
moiety; (11), tetrahydrofuran ring and the nextethylene moieties; (111), the rest (the
isobutyl endof side chain) of the moiety.
osesquiterpenes were biosynthesized via acetyl CoA (Oba et al., 1970), 3-

hydroxy-3-methylglutaryl-CoA(HMG-CoA),mevalonate(MVA)(Aka-
zawaet al.,
1962; Akazawa, 1964; OshimaandUritani, 1968), MVA
phosphate, MVA pyrophosphate, isopentenylpyrophosphate (Oshima and
Uritani, 1969), geranyl pyrophosphate (Oguni and Uritani, 1969), farnesyl
pyrophosphateandfarnesol [l +
1' +
1 "](OguniandUritani, 1970,
1971b), initiated from pyruvate, which was supplied by glycolysis. The
metabolic pathway between acetyl-coA and farnesylpyrophosphateis des-
ignated the first half-step and the pathway from farnesol to the phytoalex-
ins the second half-step. They functionally bear their sharesof responsibili-
ties in the biosynthesis of the furanosesquiterpenes. The former playsa role
in supplying farnesol as a material for carbon skeletons of ipomeamarone
and ipomeamaronol. It has been clear that the amount of the supply is
Conversion
Enzymic of Furanosesquiterpene 469
enzymatically controlled by HMG-CoA reductase as the rate-limiting en-

zyme in the pathway (It0 et al., 1979). Therefore, the amount of accumula-
tion of furanosesquiterpenes is primarily controlledby that as quantitative
pathway. In contrast to that, the latter contributes to the formation of
the characteristic carbon skeleton of ipomeamarone and also confers the
diversity of the minor family compounds which are derivatively produced
by the bypathof synthesisof ipomeamarone (qualitative pathway).
Thirty-seven species ofthe sesquiterpenes and their decomposed prod-
ucts, which are considered to be metabolically derivedfrom farnesol, have
been isolated and identified as stress compounds inC. fimbriata-infected
and mercuric chloride-treated sweet potato root tissues (Schneider et al.,
1984). Among them, at least five kindsof sesquiterpenes, suchas 9-hydroxy-
+ +
farnesol [l 2' l "1, 9-oxofarneso1 [l + +3' l"], B-hydroxydendro-
lasin [3 + 2' + 1"l, 6-oxodendrolasin(component A2) [3 + 3' + l " ]
(Burka et al., 1981; Ito et al., 1984), and dehydroipomeamarone [3 + 4'
+ 1"l, are regarded as intermediates inthe most direct pathway between
farnesol and ipomeamarone on the basis ofthe information from I4C-tracer
experiments. However, there may have been a few biochemical interpreta-
tions about the metabolic conversions among them, since few enzymologi-
cal approachesto them have been carriedout by other workers apart from
Uritani's group. Therefore, it is not clear what kinds of enzymic systems
regulatively catalyze the formation of characteristic carbon skeletons of
ipomeamarone and ipomeamaronol and influence the composition of the
minor family compounds produced by their bypath,and at what cell organ-
elle their biosynthesis is done. In this
chapter, we describe enzymologically
the second half-step on the basis of some of the results obtained fromour
investigations andso forth.
As a matter of convenience, the structures of the furanosesquiterpenes
described in this text were stated in combination ofthe numbers added to
the partial ones in Fig.2.
II. ENZYMIC FORMATION OF SKELETON OF IPOMEAMARONE

IN THE SECOND HALF STEP
Ipomeamarone possesses one furan ring and one tetrahydrofuran ring in
the skeleton. In order to interpret the metabolic formation of the character-
istic carbon skeleton downstreamfrom farnesol, it is significant to consider
separately the expected enzymic conversions occurring the in three indepen-
dent parts, namely, furan ring (I), tetrahydrofuran ring and the next ethyl-
ene (11), and the rest (the isobutyl end of side chain) (111) (see Fig. 1). We
describe the possible enzymic pathways inthe three parts successively. As
470 Fujita et al.
described above, the chemical structures of sesquiterpenes are originally

designated withthe numbers shown in Fig. 2.
A. FuranRing Moiety
The most plausible pathway of the carbon skeleton formation occurring in
the furan ring moiety is shown in Fig. 2a. The authors considered that it
wasnecessary that the hydroxylated C-l be enzymically oxidized to the
aldehyde for the closing ofthe furan ring. Therefore,a farntsol dehydroge-
nase catalyzing the oxidation of farnesol to farnesal [2 + 1' + l " ] was
highly purified and characterized in detail from cut-injured sweet potato
root tissues (Inoue et al., 1984). This enzyme isa homodimer consisting of
subunits with molwt 47,000 and needs NADP+ as a cofactor for the activ-
ity. The activity was considerably decreasedby some SH group inhibitors.
The maximal activity was detected when t,t-farnesol was used as a sub-
strate. The activitywas reduced inthe following order: c,t-farnesol(C,J >
Figure 2 (a) Formation of furan ring of ipomeamarone. * ,Enzymic conversion

proven; 3 , enzymic conversion presumed; +, nonenzymic conversion presumed;
[ 1, undetected intermediates. (b) Ring closing and ring opening of tetrahydrofuran
ring of ipomeamarone. a , Enzymic conversion presumed; 6 ,nonenzymic conver-
sion presumed; [ 1, undetected intermediates. (c) Reduction and oxygenation in the
end of the side chain of ipomeamarone. S , Enzymic conversion proven;k , chemi-
cal equilibrium reaction.
Enzymic Conversionof Furanosesquiterpene 471
1
2:
the other
Cl
structure
-----
detected
+on
the other
structure
(4')
472 Fujita et al.
geraniol(Clo) > citronerol(Clo) > nerol(Clo) > decanol(Clo).Ben-

zylalcohol, glycerol, and ethanol showed no activity. On the other hand,
NADPH-dependent geraniol dehydrogenase was detected in orange juice
vesicles and oxidizedsomemonoterpenealcoholsin the presenceof
NADP', but at a much slower rate than farnesol (Potty and Bruemmer,
1970). Time courses ofthe enzyme activityand amount of furanosesquiter-
penes accumulated in sweet potato root tissues after cut injury and C.
fimbriata infection are shown in Fig. 3a. Though a little enzyme activity
was detected in the fresh tissue (without cutinjury and C. fimbriata infec-
tion), it was dramatically increased with a little lag time (2-3 hr) after
infection by C. fimbriata and about 20 hr before accumulation of furan-
osesquiterpenes, suggestingthat this enzyme playsan important role inthe
biosynthesis of furanosesquiterpenesin diseased sweet potato root tissues.
The above results suggest that oxidation at the C-l site of farnesol by the
farnesol dehydrogenase participated in formation and accumulation of var-
ious furanosesquiterpenes. On the other hand, substitution of air phase
with nitrogen gas suppressedthe rate of the synthesis of ipomeamarone in
vivo (Oguni and Uritani, 1971b). Thereafter, Burka and Thorsen (1982)
proved that all three oxygen atoms of ipomeamarone were introduced from
molecular oxygenbut not from farnesol or water by tracer experiment using
(4 Incubation(days)
time
Figure 3 Changes in the activities of farnesol dehydrogenase (a), terpene reduc-
tase (b), ipomeamarone 12-hydroxylase(c), and the contentof furanosesquiterpenes
in responseto fungal inoculationor cut injury. The enzymic activities
in the casesof
fungal inoculation( 0 ) and cut injury(0),and the contentsof furanosesquiterpenes
in the casesof fungal inoculation(M) and cut injury (U).
Enzymic Conversionof Furanosesquiterpene 473
(b) Incubation time (days)
0 I 2 3
(c) Incubation time (days)
"02.This evidence indicates that introduction of one oxygen atom from

molecular oxygen by oxygenase occurs duringthe formation of the furan
ring. The above evidence strongly suggests that C-l oxidized to aldehyde
with farnesol dehydrogenase nonenzymically binds to the oxygen atom of
the hydroxyl group introduced to C-l4 with oxygenase, so that hemiacetal
is formed as a precursor of the furan ring. As soon as this hemiacetal is
formed, one dehydration is expected to occur at C-l and C-l4 nonenzymic-
ally, so that furan ring is synthesized. These over all reactions show
that an
474 Fujita et al.
oxygen atom from farnesol is eliminated as a water molecule and another

oxygen atom from molecular oxygen is utilized during cyclization of furan
ring. Recently, two butenolides such as 6-oxodendrolasinoide [4 + 3’ +
1 ”1 and ipomeamaronolide [5 + 3 ’ + 1 ”1 have been isolated and identi-
fied in C. fimbriata-infected sweet potato root tissues (Schneider et al.,
1984). The former compound might support the above hypothesis for the
ring closing, since it appears to be a byproduct, to which the hemiacetal
intermediate is immediately dehydrogenated but not dehydrated.
B. Tetrahydrofuran Ring and the Ethylene Moieties

There has been no reaction of terpene conversion elucidated enzymologi-
cally among ones speculated to occur in the tetrahydrofuran ring and the
next ethylene moieties. However, the most reasonable metabolic pathway
has been considered on the basis of some results from isotopic tracer experi-
ments and organic chemical knowledges (Fig. 2b).
It has been proven with 14C-tracerexperiments that farnesol was con-
verted to dehydroipomeamarone via 6-oxodendrolasin (Burka et al., 1981;
Ito et al., 1984). As described above, since all three oxygen atoms of ipo-
meamarone originate from molecular oxygen (Burka and Thorsen, 1982),
two oxygen atoms in the moieties are considered to be introduced by oxy-
genases. C-9 is first hydroxylated and then oxidated to ketone. The five
kinds of sesquiterpenes having hydroxyl group at C-9 -9-hydroxyfarnesol
(4), 6-hydroxydendrolasin (Burka et al., 1981), 6-myoporol [3 + 6’ + 2 ” ]
(Burka and Iles, 1979), 6-dihydro-7-hydroxymyoporone[3 + 6’ + 4 ” ]
(Burka, 1978), and 9-hydroxyfarnesoic acid [6 + 2’ + l ” ] (Schneider et
al., 1984)-have been isolated and determined in their chemical structures.
Burka et al. (1981) deduced that C-4 was hydroxylated and isomeric transi-
tion from the 6,7 double bond to the 7,8 one occurred, followed by Michael
addition (1,4 addition) during the closing of the tetrahydrofuran ring. The
most direct proof was recently given with experiments using deutrioin-
termediates by Schneider et al., (1984). The introduction of oxygen atom
by oxygenase seems to be a rate-limiting reaction for the closing of the ring,
since no compound containing a hydroxyl group at C-4 has been detected
yet.
Using 14C-labeledprecursors, Burka and Kuhnert (1977) demonstrated
that the tetrahydrofuran ring of ipomeamarone was cleaved so that ipo-
meamarone was converted to 4-hydroxymyoporone [3 + 5’ + 2 ”1. Dehy-
droipomeamarone seems to be converted in a similar manner to 4-hydroxy-
dehydromyoporone [3 + 5 ‘ + l “ ] too. Though the ring openings are
expected to be triggered by the hydroxylation of C-4 by oxygenases, it has
not been proven that an oxygen atom newly introduced originated from
Conversion
oxygen molecule and no corresponding intermediate in their reactions has

also been detected yet. Therefore, the reaction of introduction of oxygen
atom is also presumedto be rate-limitingstep on these ring openings.
C. The Isobutyl End of Side Chain

The reactions for formation of the carbon skeleton in the rest moiety are
shown in Fig.2c. This pathway has been established enzymologically.
The 10,ll double bond originatingfrom farnesol is saturated after the
closing of the tetrahydrofuran ring, suggestingthat it might playa role for
the cyclization like C-9 ketone. The authors elucidated the properties of
the enzyme catalyzing the saturation of dehydroipomeamarone using the
cell-free system (Inoueand Uritani, 1980; Inoue et al., 1984). This enzyme
was located in microsomalfraction and required NADPH for the activity.
SH group-inhibiting reagents such as Cu2+,monoiodoacetamide, and N-
ethylmaleimide strongly reduced activity. It is likely that the SH group of
the enzyme participates inthe activity. On the other hand, addition of the
same concentrationof 4-hydroxydehydromyoporoneas that of dehydroipo-
meamarone to the reaction mixture resulted ina 50% decrease in activity,
suggesting that this enzyme could practically reduce both compounds as
substrates in parallel. In contrast to the first half step forming a linear
pathway, it has been elucidated through a lot of tracer experiments that the
second half step was made up of complicated network systems for the
biosynthesis of furanosesquiterpenes in sweet potato root tissues. The flexi-
bility of the reductase in recognitionof the substrate seems to support the
complexity inthe second half step. Time courses of the enzyme activityand
accumulation of furanosesquiterpenes after C. fimbriata infection and cut
injury are shown in Fig. 3b.The enzyme activity isnot detected inthe fresh
root tissues but is found in the C. fimbriata-infected and cut-injured ones.
The maximal activityof the former is two- to threefold higherthan one of
the latter. After infection with C. fimbriata, while furanosesquiterpenes
accumulate witha lag time of1 day, this enzyme activity appearsat 0.5 day
and then drastically increases in advance of the terpene accumulation. The
above results suggestthat the enzyme participates inthe biosynthesis of the
furanosesquiterpenes.
After the saturation of the double bond, C-l2 of the side chain end
is hydroxylated. The authors observed the hydroxylation at C-l2 in vivo
experiments using I4C-labeled ipomeamarone (Oba et al., 1982). Further-
more, the biochemical properties in enzymological manner using the cell
free system have been elucidated (Fujita et al., 1981, 1982). The enzyme
catalyzing the reaction was located in the microsomal fraction, required
NADPH as a cofactor for the activity, and was an SH enzyme. The inhibi-
al. 476 et Fujita
Table 1 Inhibition of Ipomeamarone12-HydroxylaseActivity by Car-

bon Monoxideand Its Reversal by Light
activity Relative
Gas mixture (Vo)'
Air 100
NZ(100%) 25
02 (15%)+ NZ (85%) 90
CO (15%) + O2(15%) + NZ(70010) 36
CO (15%) + 0, (15%) + NZ(70Vo) + light 70
of the activityof air control.

'Relative activity was expressed as a percentage
tion by cytochrome c indicated the participation of a microsomal electron

transport system in the activity. The enzyme activity required molecular
oxygen and was severely inhibited by CO. The inhibition was efficiently
suppressed by radiation of visible light during the reaction (Table 1). Ac-
cording to the above results, it was concluded that this enzyme reaction
proceeded from a mixed function monooxygenase system containing cyto-
chrome P450 as a terminal constituent. Time courses of the enzyme activity
after cut injury and C. jimbriata infection are shown in Fig.3c. The enzyme
activity was not detected in the fresh tissues but appeared and increased
with little lag timeafter wounding and infection. However,the latter maxi-
mal activity was several fold higher than the former, suggesting that this
enzyme also closely participated in the biosynthesis . of furanosesquiter-
penes. Tanaka et al. (1976) reported that cinnamic acid 4-hydroxylasewas
a cytochrome P-450-dependent enzyme system in sweet potato root tissues.
However, nosubstrate competition between ipomeamarone 12-hydroxylase
and cinnamic acid 4-hydroxylase strongly indicated that the two enzyme
reactions were catalyzed by individual cytochromes P 4 0 (Fujita et al.,
1982). The hydroxyl group introduced at C-l2 is considered to contribute
to the subsequent formation of a five-member ring (Schneider et al., 1984).
111. PARTICIPATION OF CYTOCHROME P 4 5 0 IN THE

SECOND HALF STEP
As described above, at least three oxygenations are required in the forma-

tion of ipomeamarone (Burka and Thorsen, 1982) and then the introduc-
tion of one more oxygenatom by oxygenase is considered to facilitate the
split inthe tetrahydrofuran ring. However,there has been no enzymological
evidence to'this effect yet. In
order to examine to what extent cytochrome(s)
Conversion
P-450 participates in such oxygenations as ipomeamarone hydroxylation,

furanosesquiterpene production and cytochrome(s) P-450 synthesis were
determined in sweetpotato root tissues treated with various chemicals(Fuj-
ita, 1985). Content of cytochromeP-450was calculatedfrom the CO differ-
ence spectrum of the microsomes under reduced condition (Fig. 4). The
amount of production of furanosesquiterpenes and content of cytochrome
P-450are shown in Fig. 5 and in Table 2, respectively. These results indicate
an intimate mutual relation between both factors. Therefore, there is no
doubt that cytochrome(s) P-450participates in multiple oxygenationsdur-
ing ipomeamarone biosynthesis and in the following oxygenative conver-
sions including ipomeamarone 12-hydroxylation. As described above, the
most plausible intermediates (hydroxylated forms) takingpart in ring clos-
ing or ring opening ofthe furan and tetrahydrofuran rings of ipomeamar-
one have not yet been detected. This strongly indicatesa possibility that the
I I
400 450
Wavelength(nm)
Figure 4 Difference spectrum of CO-bound - CO-unbound forms of dithionite-

reduced microsomal fraction from diseased tissue.
Fujita et al.
0
HgC12 or CdS04 (mM)
Figure 5 Terpene accumulation in discs of root tissue of sweet potato treated with
HgCl, or CdSO.,, with or without phenyl isocyanide. 0 , HgCl,; 0,HgCl, + phenyl
+
isocyanide; A,CdS04; A, CdS04 phenyl isocyanide.
multiple monooxygenases minutely regulate partin the biosynthesis rate of

ipomeamarone asthe rate-limiting enzymesfor the formation and degrada-
tion of the rings, in contrast to the HMG-CoA reductase controlling
that in
whole.Hence,eachcytochromeP-450 and NADPH-cytochrome P450
reductase should be purified as constituentsfor microsomal monooxygen-
ase systems and the reconstitution system could allow examination of the
individual furanosesquiterpene conversions in vitro enzyme systems. The
authors have purified NADPH-cytochrome P-450 reductase to homoge-
neous on polyacrylamide gel electrophoresis and characterized (Fujita and
Asahi, 1985b).
W. INTRACELLULAR LOCALIZATION OF THE SECOND HALF STEP

Intracellular localization of ipomeamarone 1Zhydroxylase has been exam-
ined with linear sucrose density gradient centrifugation and microscopic
observation in detail, resulting in its being located in rough-surfaced endo-
Conversion
Table 2 Effects of Treatment of Tissue Discs with Chemicals on Cytochrome

P450 and Microsomal Protein During Incubation ofDiscs of Root Tissue of
Sweet Potato
Specific
cytochrome Microsomal
protein
Cytochrome P450 P450 amount
Discs (mg/g fresh wt) (pmol/g fresh wt) (pmol/mg
protein)*
12.9discs 2.74
Freshly prepared 0.213
discsb 74.1
Incubated 37.0 0.500
Discs treated with'
9.2 Phenyl 42.9
isocyanide(PI) 0.541
5 HgCl,d88.6 0.538
198 8.0 and PI 11
HgC1,d 0.597
CdS0,d 57.1 0.481
7 CdS0,d60.4
and PI 0.565
The specific amount in the microsomes.

-he discs were incubated for2 days without treatment.
1 more day.
The discs incubated for1 day were treated with the chemicals incubated for
dThe concentration was3.68 mM.
plasmic reticulum (Fujita and Asahi, 1985a). The localization was appar-
ently different from one of cinnamic acid 4-hydroxylase participating in the
polyphenol metabolism in sweet potato root tissues. As described above,
the terpene reductase was also located in postmitochondrial membraneous
fraction.Multipleoxygenationsinthefuranosesquiterpenemetabolism
must also occur in microsomal fraction since they are considered to be
dependent on participation of cytochromeP-450.This information strongly
suggests that a lot of processes in the second half step are performed on
microsomes. On the other hand, electron-dense vesicles containing furan-
osesquiterpenes were observed by electron microscopy and were recovered
in 2% Ficoll fraction after discontinuous Ficoll density gradient centrifuga-
tion (Oba et al., 1984). These vesicles are enveloped in a single membrane
in 7-10 nmthickandcontainingonly a littleantimycinA-insensitive
NADPH-cytochrome c reductase activity, suggesting that they might origi-
nate from endoplasmic reticulum. The endoplasmic reticulum with hetero-
geneity in C. fimbriata-infected sweet potato root tissues was previously
demonstrated by the distribution of RNA and various organelle marker
enzymes after sucrose density gradient centrifugation (Fujita and Asahi,
1985a). Therefore, there might be highly differentiated endoplasmic reticu-
480 Fujita et al.
lum participating in
the biosynthesis, transport, and secretion of furanoses-
quiterpenes inC.fimbriata-infected sweetpotato root tissues.
V. EPILOG
The biosynthesis of the furanosesquiterpenes in mold-damaged sweet po-

tato root is divided into two parts; the first half step from acetyl-coA to
farnesyl ppophosphate and the second halfstep after farnesol. The former
is comparatively linear and the rate is strictly controlled by HMG-CoA
reductase as a rate-limiting enzyme. The regulation in this step strongly
influences the ebb and flow of metabolism inthe latter step. Therefore, the
former may be called the quantitative pathway. The latter forms network
systems and produces ipomeamarone as well as various kinds of family
compounds. The composition of the furanosesquiterpenesis variable under
various conditions such as kinds of elicitors (Fujita, 1985; Fujita and Yoshi-
zawa, 1987, 1989). Therefore, in contrast to the former, this step may be
designated the qualitative pathway. In this chapter, the second half step
was describedenzymologically.In the step, it appears that cytochrome
P-450-dependent mixed-function monooxygenationlurks as a rate-limiting
reaction in ring closingand ring opening of thefuran and tetrahydrofuran
rings. This is supported by intimate mutual relation between the synthesis
of cytochrome P-450 and the production of furanosesquiterpenes under
diverse conditions.The hydroxylation at the end of side chain of ipomeam-
arone is the only reactionthat has been enzymologically provedto depend
on cytochrome P-450 in the furanosesquiterpene metabolism ofsweet po-
tato. Hence, it is necessary to isolate cytochromes P-450 and to examine the
substrate specificities using the reconstitution systems. Since cytochromes
P-450 participating indrug metabolism possess broadsubstrate specificities
in animals, it is likely that the cytochrome P-450 catalyzing multiple sub-
strates participates inthe biosynthesis of furanosesquiterpenes C. infimbri-
ata-infected sweet potato root tissues. It was elucidated that the enzyme
reducing dehydroipomeamaroneto ipomeamarone also converted Chydroxy-
dehydromyoporone to 4-hydroxymyoporone, suggesting that many kinds
of enzymesparticipating inthe second halfstep might havebroad substrate
specificities. For example, the farnesol dehydrogenase might actively oxi-
dize 9-hydroxyfarnesol and 9-oxofarnesol to the corresponding aldehydes
like farnesol. According to information on localization of the enzymes
participating in the step, it seems to be located in endoplasmic reticulum.
On the other hand, furanosesquiterpene-containingvesicleswhichwere
considered to be derived from endoplasmic reticulum were observed. The
vesicles mightparticipate in the transport of furanosesquiterpenes fromthe
Conversion
region adjacent to infected region) to the site

site for synthesis (noninfected
for accumulation (infected necrotic region).
REFERENCES
Akazawa, T. (1964). Biosynthesis of ipomeamarone. 11. synthetic mechanism.Arch.
Biochem. Biophys. 105(3): 512-516.
Akazawa, T., Uritani, I., and Akazawa, Y. (1962). Biosynthesis of ipomeamarone.
I. the incorporation of acetate-2-CI4and mevalonate-2-CI4into ipomeamarone.
Arch. Biochem. Biophys.99(1): 52-59.
Burka, L. T. (1978). 1-(3’-Furyl)-6,7-dihydroxy-4,8-dimethylnonan-l-one, a stress
metabolite from sweet potatoes (Ipomoea batatas). Phytochemistry I7(2): 317-
318.
Burka, L. T., and Iles, J. (1979). Myoporone and relatedketoalcohols from
stressed sweet potatoes.Phytochemistry 18(5):873-874.
Burka, L. T., and Kuhnert, L. (1977). Biosynthesis of furanosesquiterpenoid stress
metabolites in sweet potatoes (Ipomoea batatas).Oxidation of ipomeamarone
to 4-hydroxymyoporone. Phytochemistry 16(12): 2022-2023.
Burka, L. T., and Thorsen, A. (1982). Biosynthesis of furanosesquiterpenoid stress
metabolites in Ipomoea batatas:isotopic oxygen incorporation into ipomeam-
arone. Phytochemistry 21(4): 869-870.
Burka, L. T., Felice, L. J., andJackson, S. W. (1981). 6-Oxodendrolasin, 6-
hydroxydendrolasin, 9-oxofarnesol and 9-hydroxyfarnesol, stress metabolites
of the sweet potato. Phytochemistry 20(4): 647-652.
Fujita, M. (1985). Stimulation of cytochromeP450 synthesis in sliced sweet potato
root tissue by chemicals applied to surface. Agric. Biol. Chem. 49(10): 3045-
3047.
Fujita, M., and Asahi, T. (1985a). Different intracellular localization of two cyto-
chrome P-450 systems, ipomeamarone 15-hydroxylase and cinnamic acid 4-
hydroxylase, in sweet potato root tissue infected with Ceratocystisfimbriata.
Plant Cell Physiol.26(3): 389-395.
Fujita, M., and Asahi, T. (1985b). Purification and properties of sweet potato
NADPH-cytochrome c(p-450) reductase. Plant Cell Physiol.26(3): 397-405.
Fujita, M., and Yoshizawa, T. (1987). Induction of sweet potato phytoalexins by
trichothecene mycotoxins.Proc. Jpn. Assoc. Mycotoxicol. 25: 29-30 (in Japa-
nese).
Fujita, M., and Yoshizawa, T. (1989). Induction of phytoalexins by various myco-
toxins and metabolism of mycotoxins in sweet potato root tissues. J. Food
Hyg. Soc. Jpn. 36(6): 501-505 (in Japanese).
Fujita, M., Oba, K., and Uritani, I. (1981). Ipomeamarone 15-hydroxylase from
cut-injuredand Ceratocystis fimbriata-infectedsweet potatoroot tissues.
Agric. Biol. Chem.45(8): 1911-1913.
Fujita, M., Oba, K., and Uritani, I. (1982). Properties of a mixed function oxygen-
ase catalyzing ipomeamarone 15-hydroxylation in microsomes from cut-injured
482 Fujita et al.
and Ceratocystis fimbriata-infected sweet potato root tissues. Plant Physiol.

70(2): 573-578.
Hiura, M. (1943). Studies on storage and rot of sweet potato. 2, Rep. Gifu Agric.
Coli. 50(1): 1-5.
Inoue, H., and Uritani, I. (1979). Biosynthetic correlation of various phytoalexins
in sweetpotato roottissues infectedby Ceratocystisfimbriata. Plant Cell Phys-
iol. 20(7): 1307-1314.
Inoue, H., and Uritani, I. (1980a). Conversion of 4-hydroxydehydromyoporoneto
other furanoterpenes in Ceratocystis fimbriata-infected sweet potato. Agric.
Biol. Chem. 44(8): 1935-1936.
Inoue, H., and Uritani, -I. (1980b). Furano-sesquiterpene reductase from fungal-
inoculated sweet potato root tissue. Agric. Biol. Chem.44(9): 2245-2247.
Inoue, H., Oba, K., Ando, M., and Uritani, I. (1984a). Enzymatic reduction of
dehydroipomeamarone to ipomeamarone in sweet potato root tissue infected
by Ceratocystisfimbriata. Physiol. Plant Pathol. 25: 1-8.
Inoue, H., Tsuji, H., and Uritani, I. (1984b). Characterization and activity change
of farnesol dehydrogenase in black rot fungus-infected sweet potato. Agric.
Biol. Chem. 48(3): 733-738.
Ito, R., Oba, K.,and Uritani, I. (1979). Mechanism for the induction of 3-hydroxy-
3-methylglutaryl coenzymeA reductase in HgC1,-treated sweet potato roottis-
sue. Plant Cell Physiol.20(5): 867-874.
Ito, I., Kato, N., and Uritani, I. (1984). Biochemistry of two new sesquiterpenoid
phytoalexins from sweetpotato roots. Agri. Biol. Chem. 48(1): 159-164.
Kubota, T., and Matsuura, T. (1953). Chemical studieson the black rot disease of
sweet potato. J. Chem. Soc. Jpn. 7 4 101-109,197-199,248-251,668-670.
Oba, K., and Uritani, I. (1981). Mechanism of furano-terpene production in sweet
potato root tissue injured by chemical agents. Agric. Biol. Chem. 45(7): 1635-
1639.
Oba, K., Shibata, H., and Uritani, I. (1970). The mechanism supplyingacetyl-coA
for terpene biosynthesis in sweet potato with black rot: incorporation of ace-
tate-2-I4C, pyruvate-3-14C, and citrate-2,4-I4C into ipomeamarone. Plant Cell
Physiol. 11(3): 507-510.
Oba, K., Oga, K., and Uritani, I. (1982). Metabolism of ipomeamarone in sweet
potato root slices beforeand after treatment with mercuric chlorideor infection
with Ceratocystisfimbriata. Phytochemistry 21(8):1921-1925.
Oba, K., Nakamura, A.,andIwaikawa, Y. (1984). Isolationoffuranoterpene-
containing particles from Ceratocystis fimbriata-infected sweet potato root
tissue. J. Biochem. 96(6): 1951-1954.
Oguni, I., and Uritani, 1. (1969). Biochemical studies on the terpene metabolism in
sweet potato root tissue with black rot. Effect of Cl0- and C,,-terpenols on
acetate-2-I4Cincorporation into ipomeamarone. Agric. Biol. Chem. 33(1): 50-
62.
Oguni, I., and Uritani, I. (1970).The incorporationof farnes01-2-'~Cinto ipomeam-
arone. Agric. Biol. Chem. 34(1): 156-158.
Oguni, I.and , Uritani, I. (1971a). Effect of monofluoroacetate on~yrubate-3-'~C
Conversion
and gl~cose-U-'~Cincorporation into ipomeamarone. Agric.Biol.Chem.

35(12): 1980-1983.
Oguni, I., and Uritani, I. (1971b). Participation of farnesol in the biosynthesis of
ipomeamarone. Plant Cell Physiol.12(4): 507-515.
Oguni, I., and Uritani, I. (1974). Dehydroipomeamarone as an intermediate in the
biosynthesis of ipomeamarone, a phytoalexin from sweet potato root infected
with Ceratocystisfimbriata. Plant Physiol.53(4): 649-652.
Oguni, I., and Uritani, I. (1974). Effect of (-)-hydroxycitrate on ipomeamarone
biosynthesis from pyruvate in sweet potato with black rot. Plant Cell Physiol.
15(1): 179-182.
Oshima, K., and Uritani, I. (1968). Participation of mevalonate in the biosynthetic
pathway of ipomeamarone.Agric. Biol. Chem.32(9): 1146-1 152.
Oshima, K., and Uritani, I. (1969). Enzymatic synthesis of isopentenyl pyrophos-
phate in sweet potato root tissue in responseto infection by black rot fungus.
Plant Cell Physiol.lO(4): 827-843.
Potty, V. H., and Bruemmer, J. H. (1970). Oxidation of geraniol by an enzyme
system from orange. Phytochemistry 9(5):1001-1007.
Schneider, J. A., Lee, J., Naya, Y., Nakanishi, K., Oba, K., and Uritani, I. (1984).
The fate of the phytoalexin ipomeamarone: furanoterpenes and butenolides
from Ceratocystis fimbriata-infected sweet potatoes. Phytochemistry 23(4):
759-764.
Schneider, J. A., Lee, J., Yoshihara, K., Mizukawa, K., and Nakanishi, K. (1984).
Biosynthetic studies of ipomeamarone. J. Chem. Soc. Chem. Commun.6: 372-
374.
Tanaka, Y., Kojima, M., and Uritani, I. (1976). Properties, development and cellu-
lar-localization of cinnamic acid 4hydroxylase in cut-injured sweet potato.
Plant Cell Physiol.15(5): 843-854.
Uritani, I., Uritani, M., and Yamada, H. (1960). Similarmetabolic alterations
induced insweet potato by poisonous chemicalsand by Ceratocystisfimbriata.
Phytopathology 5ql): 30-34.
21
Defense Strategies of Tea (Camellia
sinensis) Against Fungal Pathogens
B. N. Chakraborty, Usha Chakraborty, andA. Saha
University of North Bengal, Darjeeling, India
I. INTRODUCTION
The most widely consumed and the cheapest hot beverage in the world
today is tea. Recent available statistics show that the total amount of world
tea production reaches2.5 million tons annually, and an average of2 billion
cups of tea are drunk every day (Yamanishi,1991). Tea is made from the
young leaves and unopened leaf buds of Camellia sinensis (L.)0,Kuntze,
a species which includes some very distinct varieties. The cultivated varieties
are classifiedinto two main groupson the basis of foliar and growth charac-
teristics: China tea, Camellia sinensis var. sinensis, and Assam tea or C.
sinensis var. assamica. The former is a slow-growing dark tree with small,
erect, comparatively narrow, dark green leaves whichare resistant to cold;
the latter is a rapidly growing, taller tree, with large drooping leaves and
resistance to cold, butadapting well to tropical conditions.
The pests and blights which preyon the leaves are of vital importance,
since any damageto the leaves defeatsthe whole purpose of its cultivation.
The monocultural conditions under which tea is grown commercially makes
the transmission of disease more difficult to control than where mixed or
rotational crop cultivation is followed. The danger of disease is further
aggravated by the fact that in many tea districts virtually noother form of
cultivation is practiced and an epidemic outbreak of an easily transmitted
disease can affect very large areas in a short period. The aerial surfaces
of tea plants, like any other plant, are usually inhabited by a variety of
microorganisms, many of which are capable of influencing the growth of
foliar pathogens (Chakraborty and Chakraborty, 1988). The interaction
between these microorganisms might result inthe suppression of pathogen
activity. Besides,it is likelythat the tea plant, in the course of its adjustment
to varying environments, has evolveda very effective defense mechanism
which successfully wards off most the of fungal pathogens.
The common plant pathogens (fungi, bacteria, viruses) and pests (in-
485
486 Chakraborty et al.
sects and other animals) induce some typeof resistance in plantsto subse-
quentchallenges, both to the original as well as to other biotic agents
(Chessin and Zipf, 1990). In general, the defenses of higher plants against
any form of stress, whether biotic or abiotic, fall under two categories:
preformed and induced. The inducible defense systems have been much
more elaborately studied than the preformed ones. These have been termed
as alarms. According to Chessin and Zipf (1990), alarm is a complex physi-
ological phenomenonthat comprises a series of steps, i.e., perturbation by
a particular biotic or abiotic stress, followed by production of a systemi-
cally transmitted signal (electrical, chemical,or both) in the stressed tissue
and finally the induction of a new morphological or physiological state
which protects the target tissues from subsequent exposure to the same or
other stresses. Among the inducible defenses photoalexin accumulation is
believed to be an important early defense response in several plant-patho-
gen interactions (Ward, 1986; Van Etten et al., 1989). The preformed de-
fense mechanism of plants,on the other hand, deals with internal chemical
defenses which are already fully expressed in host tissues before infection
and do not rise to higher levels in response to invading microorganisms
(Schlosser, 1980). Various types of antifungal plant constituents such as
saponins, unsaturated lactones, mustard oils,and cyanogenic and phenolic
glycosides are widespread in the plant kingdom. Occurrence, distribution,
and possible function of preformed antifungal compounds have been re-
viewed (Schonbeck and Schlosser, 1976; Schlosser, 1988). The role of pre-
formed chemical barriers as an important part of the defense mechanisms
of plants is now gaining wide acceptance and needs to be carefully looked
into, before one can get a complete picture of the defense mechanisms of
plants.
The study of defense strategies in tea was promoted by a few important
considerations. First, tea, which forms the backbone of the economy of
north eastern India, is subject to attacks by foliar pathogens likeCorticium
invisum, C. theae, Exobasidum vexans, Colletotrichum camelliae, Pestalot-
iopsis theae,and Botrytis sp. (Sarmah,1960), resulting ina reduction of the
quality and quantity of tea produced. Second, almost no work has been
done on the defensemechanismsof tea plant. Our studies werebased
mainly on three foliar diseases: brown blight caused by Colletotrichum
camelliae, grey blight caused byPestalotiopsis theae, and leaf spot caused
by Bipolaris carbonum (first reported by Chakraborty, 1987). The main
objectives have been to determine the mechanism of defense of tea plant
against these pathogens, with special emphasis on both preformed and
induced chemical defenses. Since polyphenolsare major constituents of tea
leaves, their involvement inthe defense mechanism either as preformedor
induced chemicals seemed highly probable. All four forms of catechin (Fig.
Fungal
Defense of Tea
Against Pathogens 487
l), i.e., epicatechin(EC)epicatechingallate(ECG),epigallocatechin

(EGC), and epigallocatechin gallate (EGCG), have already been reported
from tea (Wang,1991). It is also known that catechin is oxidatively cleaved
to simpler phenols and phenolic acids like catechol, phloroglucinol, and
protocatechuic acid (Fig. 2A-C). The enzymecatechin-2,3-dioxygenasewas
isolated from Chaetorniurn cupreurn, which cleaved catechin into simpler
\
O a
----
-"H
'W-----@
OH
\ /
H H
"-on
H0 OH
Figure 1 Chemical structures of catechin and its different forms. A, catechin; B,

(-) epicatechin; C, gallocatechin; D, (-)epigallocatechin; E, (-)epigallocatechin gal-
late; F, (-)epicatechingallate.
0
Figure 2 Chemical structures of some commonly occurring phenols and phenolic
acids of teaplants. A, catechol; B, phloroglucinol; C, protocatechuicacid; D,
p-coumaric acid;E, ellagic acid;F, caffeic acid; G, gallic acid;H, chlorogenic acid;
I, quinic acid.
Pathogens
Fungal
Defense
Against
of Tea 489
phenols as mentioned (Sambandam, 1982). A numberother of phenols and

phenolic acids have also been reported from tea, some of which are pre-
sented in Fig. 2D-I). Kawamura and Takeo (1989) demonstrated the anti-
bacterial activity of tea catechin to Streptococcus mutans. Hamaya et al.
(1984) established the presence of characteristic antifungal compounds in
the leaf extracts of Camellia japonica. Conidial germinationor growth of
hyphae of many fungi, namely,Pyricularia oryzae, Cochliobolus miyabea-
etc.
nus, Pestalotia longiseta, Gloeosporium theae-sinensis, Diaporthe citri,
was inhibited. They isolated two triterpenoid saponins from the extract,
which they named camellidin I and camellidin 11, with molecular formulas
CSsHB602s (MW = 1146) and CS3HUOM (MW = 1104), respectively. These
were composed of 3/3-hydroxy, 18/3-acetoxy, 28-nor-olean-l2-en-16-one,or
38,18/3-dihydroxy, 28-nor-olean-12-en-16-0neas the sapogenin and D-
glucuronic acid, D-glucose and 2 mol of D-galactose as the sugar moiety
(Nagata et al, 1985). Nishino et al. (1986)further determined the sequence
and configuration of the glycosides in the sugar moiety.
A.PlantMaterial
Tea plants (18-month-old) of seven clonal varieties (TV-l, TV-9, TV-14,
TV-16, TV-17, TV-18, and TV-26) were obtained from the clone house of
Mohurgong and Gulma TeaEstate, Sukhna, Siliguri; and five clonal varie-
ties (TV-19, TV-20, TV-23, TV-25, and Teen Ali-17/1/54) and two seed
varieties (TS-449 and CP-1) were collected from the nursery of Dey’s Tea
Plants, Khaprail, Siliguri, West Bengal.
B. FungalCulture
Colletotrichum camelliae Mass. and Pestalotiopsis theae Sawada were ob-
tained from Tocklai ExperimentalStation, Jorhat, Assam. A virulentstrain
of Bipolaris carbonum (syn. Helminthosporium carbonum), anamorph of
Cochliobolus carbonumNelson, was obtained from the Departmental stock
culture collection.
C. Inoculation Technique and Disease Assessment

A detached leaf inoculation technique as described by Dickens and Cook
(1989) was used for artificial inoculation of tea leaves. Assessment of dis-
ease intensitywas done onthe basis of percent drops that resulted in lesion
production after 24,48, and 72 hr as described by Deverall and Wood
(1961).
D. Collection and Bioassay of leaf Diffusate
Leaf diffusates were obtained following the drop diffusate procedure of

Muller (1958) with modifications. Dropsof spore suspension were collected
from leaf surfaces, combined, and centrifuged. Similarly incontrol, water
drops werecollected,combined, and centrifuged(exudate).Finally, the
diffusates(spore-free supernatant) and exudates werepassed through a
sintered glass filter and bioassayed following the slide germination proce-
dure as describedby Trivedi and Sinha (1976). Usually 1.9 m1 of diffusates
was mixed with 0.1 m1 of spore suspension of known concentration. Two
drops (0.02 ml/drop) of suspensionwere placed separately at two ends ofa
grease-free glass slide. The slides were incubated in moist Petri dishes for
15 hr at 25 f 1OC. Finally,germinated and ungerminatedsporeswere
stained with cotton blue in lactophenol and examined under the micro-
scope.
E. Extraction of Antifungal Compound
Tea leaves were collectedfrom the experimental gardenand kept in plastic

trays. Half of the total number of leaves were inoculated with conidial
suspension of B. carbonurn while theother half were uninoculated control.
For the extraction of antifungal compound from infected tea leaves, the
method of Keen (1978) was followed with modifications. Leaves werehar-
vested after 6, 12, 24, 36,48, and 72 hr of inoculation,weighed, and
extracted with 40% aqueous ethanol in a rotary shaker for 24 hr. The
filtrates were vacuum-concentratedto approximately one-half volumeand
extracted thrice with ethylacetate and the organic layers were pooled and
dehydrated with MgS04. Both the ethyl acetate and water fractions were
taken to dryness separatelyand residues were dissolved in methanol.
F. ChromatographicAnalysis
The water and ethyl acetate fractions of the extracts of both healthy and
infected tea leaves were analyzed by thin layer chromatography (TLC)on
silica gel G. The development of the chromatograms was carried out at
room temperature and using a chloroform-methanol solvent system (9 : 1
v/v). Following evaporation of the solvent, the thin layer plates were ob-
served under W light and sprayed separately either with diazotized p-
nitroaniline (Van Sumere et al., 1965), FeC1,-K,Fe(CN),,vanillin-H,SO,
(Stahl, 1967), or Folin-Ciocalteau phenol reagent (Harborne, 1973). Color
reactions and Rf values were noted.
Defense of Tea Against Fungal Pathogens 491
C. Fungitoxicity Assay
The extracts from leaf tissue of tea plants (TV-l8 and TV-26) were bioas-
sayed on TLC plates usingB. carbonurn as the test organism followingthe
method of Hofmans and Fuchs (1970). Fungitoxicity was ascertained by the
presence of inhibition zones,which appeared as whitespots surrounded by
a blackish backgroundof mycelia.
The regions ofthin layer chromatograms correspondingto these inhibi-
tory zones were scraped and eluted in methanol. The eluants were rechro-
matographed on TLC plates to purify them and then eluted again. The
eluants were tested for antifungal activities following spore germination
testing as described by Werder and Kern (1985).
H. Preparation of Mycelial Wall Extract

Mycelial wall extract of B. carbonurn was prepared following the method
of Anderson-Prouty and Albersheim (1975). Mycelia (10-day-old culture)
were collected on filter paper ainBuchner funneland 20 g of fresh, packed
cells were homogenizedin an electric blender with 100 m1 water. The result-
ing slurrywas filtered andthe residue was again homogenized in water. The
mixture was centrifuged (1500g) for 5 min, the supernatant discarded, and
the sedimented walls washed with 200m1 water and pelleted by centrifuga-
tion at least six times or until the supernatant was visually clear. Subse-
quently isolated cell walls were homogenized once in a 50-m1 mixture of
chloroform and methanol (1 : 1 v/v) and finally 50 m1 acetone. This prepa-
ration when air-dried represented mycelial wall fraction. One gram of my-
celial wallfraction was suspended in 100m1 of water and autoclaved for 30
min at 15 psi pressure. The autoclaved suspension was filtered, the filtrate
centrifuged, and the supernatant concentrated to 10 m1 under reduced pres-
sure.
111. ANTIFUNGAL COMPOUNDS OF TEA

A. Varietal ResistanceTest
Pathogenicity of the three selected foliar pathogens, i.e., Colletotrichurn
camelliae, Pestalotiopsis theae, and Biopolaris carbonurn, were tested sepa-
rately on different clonal and seed varieties of tea certified by Tocklai
Experimental Station, Jorhat, Assam. Results (Table 1) revealed that, of
the different TV varietiestested,TV-9 was moderatelyresistantwhile
TV-18 was highly susceptibleto both C.carnelliae and P. theae. TV-18 and
TV-26 were highly susceptibleand resistant, respectively, to B. carbonurn.
With all three pathogens tested, it was seen that the younger leaves were
Table 1 Pathogenicity Test of Three Different Foliar Pathogens on Selective Tea

Varieties Releasedfrom Tocklai Experimental Station
070 Lesion formed after inoculation with'.'

Varieties Character' C. camelliae P. theae B. carbonum
Clones:
TV- 1 35.0 f 2.6 44.6 * 3.2 38.7 f 2.4
TV-9 25.9 f 1.8 18.3 f 1.5 56.2 f 2.7
TV-l4 63.8 f 2.2 27.8 * 1.6 37.3 f 1.6
TV-l6 49.5 * 1.9 16.4 f 1.2 06.0 f 0.8
TV-l7 38.6 f 2.8 27.1 f 2.4 52.6 * 3.5
TV-l8 77.8 f 2.6 70.5 f 3.6 58.7 f 2.8
TV-l9 44.8 f 1.6 52.3 f 2.5 38.8 f 1.6
TV-20 59.4 f 3.4 21.9 f 2.2 25.4 f 2.5
TV-23 36.3 f 1.8 63.6 f 3.8 21.6 f 1.5
TV-25 30.6 f 1.6 19.7 f 1.2 03.8 0.2
TV-26 38.0 * 2.4 22.3 f 1.9 02.5 f 0.3
Teen Ali 17/1/54 43.9 * 2.5 28.7 f 2.4 31.9 f 1.2
Seed stocks:
TS-449 a,b,c 51.6 f 1.8 38.2 f 3.2 35.6 f 1.6
CP-l a,c 46.7 f 1.3 29.5 f 2.4 25.9 *2.3
'Based on field observations: a, high yielding; b. submarginallandsanddroughtareas;
c, infilling and interplanting(Bezbaruah and Singh, 1988).
'Average of three separate experiments (h = SE).
'72 hr after inoculation.
much more susceptibleto diseases than the older leaves probably due to the
variations in the texture and physical structure of the leaves. The older
leaves havea very tough exterior,which probably makes penetration by the
pathogen difficult. The trichomes which are abundant in older leaves also
act as a barrier to certain fungi. Thisis found to be the case of the anthrac-
nose fungus Gloeosporium theae-sinensis,which infects the plant through
the trichome of young leaves. In many cases, the pathogen was inhibited
from gaining entranceby a callosity which was produced by swelling ofthe
trichome cell wall inward in sucha way that it enveloped and precededthe
invading hypha (Andoand Hamaya, 1986).
B. Biological Activities of leaf Diffusates and Exudates

Tea varieties, resistant and susceptible to each of the three pathogens (P.
theae, C. camelliae, and B. carbonum), wereselected for further study.
Leaf diffusates from the selected tea varieties were collected and assayed
ea of Defense 493
for their effecton spore germinationand germ tube growth of the respective
fungi. The results indicated that the diffusates from the resistant varieties
were more fungitoxic than those from the susceptible varieties (Chakra-
borty and Saha, 1989). Leaf exudates of tea also contained some fungitoxic
substances, which suggests the presence of antifungal compounds in the
healthy leaves. Presumably, tea leaves contain some diffusible preformed
antifungal compoundswhich play a role in their defense mechanism. Toda
et al. (1989) also reported that extracts of Japanese greentea leaves inhib-
ited the growth of various bacteria.
C. Detection and Identification of Antifungal Compounds in Tea leaves
The results of drop diffusate tests strongly suggested that tea leaves con-
tained both preformed and postinfectional antifungal compounds.To iso-
late these compoundsthe plants were inoculated withone of the pathogens,
B. carbonurn. After 72 hr of inoculation with the pathogen the leaves of
resistant (TV-26) and susceptible (TV-18) varieties were extracted in aque-
ous methanol, concentrated in vacuum, fractionated with ethyl acetate,
spotted on TLC plates,and developed in chloroform-methanol (9 : 1). On
spraying the plates with Folin-Ciocalteau reagent, several spots appeared
indicatingthepresenceofphenoliccompounds in the extracts of both
healthy and infected leaves of the two varieties. In addition, the extract
from healthy plants contained a brown-colored spot at R, 0.8, developed
when sprayed with vanillin-H2S0,.
These crude extracts were assayed for their antifungal activityby TLC
plate bioassay method. Extract from healthy leavesof both varieties (TV-l8
and TV-26) showed a prominent inhibition zone at Rf 0.8 in TLC plates
(Fig. 3, zone A). These inhibition zones of TV-l8 and TV-26 wereof 20 and
25 mm diameter, respectively. There was no evidence ofthis inhibition zone
after 72 hr of inoculation in TV-18, whereas in TV-26 the inhibition zone
was faintly visible. Inthe extracts from inoculated leavesof both suscepti-
ble and resistant varieties, prominent inhibition zones appeared at Rf 0.6
(Fig. 3, zone B). This antifungal compound in resistant variety (TV-26) was
approximately four times morethan that of the susceptible variety (TV-18).
In case of noninoculated control of both the varieties, muchless antifungal
activity was seen at the corresponding region.
The prominent inhibiting zoneat R, 0.8 was identified as catechinon
the basis of its color reaction and R, value (Kawamura and Takeo, 1989).
Results of the present investigation also indicate that the catechin is presum-
ably cleavedto some simpler phenols.The breakdown of this compound is
almost complete in the susceptible variety but incomplete in the resistant
variety as traces of catechin are located evenafter 72 hr of inoculation. The
TU -18 TV -26
B
I
H I H I
Figure 3 TLC bioassay of tea leaf extracts of susceptible (TV-18) and resistant
(TV-26) varieties. H, extracts from healthy leaves;I, extracts from leaves inoculated
with B. carbonurn (48 hr after inoculation).
inhibitory zone E of the extracts from inoculated leaves at Rf 0.6was

identified to be catechol. Accumulationof catechol in resistant variety (TV-
26) increased significantly(510 pg/g fresh wt tissue) in comparison to sus-
ceptible variety(187 pg/g fresh wt tissue) after 48 hr of inoculation withB.
carbonurn. In the susceptible variety, eventhough catechin is broken down
completely, accumulationof catechol is not greater than the resistant vari-
ety. It seems probable that increasedlevelofcatecholisrelated to the
resistance mechanism.
Since the results of the facilitated diffusion technique indicated the
involvement of catechin as a preformed fungitoxic substance as well as
accumulation of other antifungal phenols, mainly catechol, as a result of
inoculation, the quantitative changes in phenolcontents after 6, 12,24, 36,
and 48 hr of inoculation were determined in both varieties. At different
times after inoculation, the pattern of changes in total phenolic content
(ethyl acetate-soluble) ofboth the varieties differed significantly. The phe-
nolic content of resistant cultivar(TV-26) showed a significant continuous
increase from 24 hr of inoculation. In the uninoculated leaves of the same
Pathogens
Fungal
Defense of Tea
Against 495
variety, the phenolics showed a gradual decrease with time. Both healthy
and inoculated susceptible variety(TV-18)had less phenolicthan the resis-
tant variety, withthe inoculated leaves having morethan the healthy ones.
The level of water-soluble phenolic compound was low and did not differ
in resistant, susceptible,and noninoculated control plants (Fig.4).
All these resultstherefore suggest that the healthy leaves oftea plants
contain catechin, which is an antifungal polyphenol. Inoculation resulted
in the breakdown of catechin with the simultaneous appearance of other
antifungal phenols, mainly catechol. The concentration of this phenol kept
increasing with time thein resistant varietyup to 72 hr. Even though reports
are available on the quantitative differences in phenolcontents of resistant
and susceptible varieties, most of them are conflicting with each other.
Sridhar and Ou (1974)reported differences in total phenolic accumulation
during the interaction of Pyricularia oryzae with rice. However, no differ-
ences were found in the phenolic content in the interaction of Helmin-
thosporium maydis race T with N and T cytoplasms of a single maize
genotype (Macri et al., 1974). On the other hand, Hammerschmidt and
Nicholson (1977) demonstrated a clear difference between resistant and
susceptible interactions of maize to Colletotrichum graminicola based on
;Is5
- A B
ie51
e
WS
m
E o 1
6
1
12
1 1
24
1 1
36
1
48
1
6
1 1
12
1 1
24
1 1
36
1 t
48
HOURS AFTER INOCULATION

Figure 4 Changes in total water-soluble (A) and ethyl acetate-soluble (B) pheno-
lics in extracts of tea leaf tissue of susceptible (TV-18) and resistant (TV-26) varie-
ties. Healthy: TV-l8 (0-""0); TV-26 (0-0). Inoculated: m - 1 8 (o-"--.);
TV-26 ( 0 -0 ) .
accumulation of phenols. Apart from these conflicting results,

further com-
plications arise because phenols are not distributed uniformly in the in-
fected tissues. Some phenols increased, while
others decreased.
D. Elicitation of Antifungal Compound in Tea leaves by

Mycelial WallExtract of Bipolaris carbonurn
Having establishedthe fact the pathogenB. carbonum induces the accumu-
lation of phenols inthe tea leaves and more rapidlyin the resistant variety,
the next questionwas whether elicitors could play any rolethis ininduction.
It was therefore decidedto make mycelial cell wall extracts ofthe pathogen
and to determine its effect on the induction of diffusible antifungal com-
pounds in tea leaves. Cell wall extracts were prepared fromthe mycelia of
B. carbonum as described under “Materials and Methods.” To determine
the effect ofthesewallextractson the production of antifungal com-
pounds, fresh spores of B. carbonum were suspended in its own mycelial
wall extract and drops of suspension placed on leaf surfaces. After 48
hr, drops were collected, combined, and centrifuged. The supernatant was
assayed for fungitoxicity by spore germination bioassayand also analyzed
by TLC for detection of antifungal compound as described by Purkayastha
and Ghosh (1983). It was found that fresh mycelial wall extract was highly
stimulatory to growth of B. carbonum but inhibitory when collected from
leafsurfaces after 48 hr incubation. Spores suspended in mycelial wall
extract and incubated over leaf surfaces completely inhibitedthe germina-
tion of B. carbonum. The mycelial wall extract containingthe elicitor very
effectively inducedthe accumulation and subsequent diffusion of the anti-
fungal compound in the resistant variety(TV-26).
E. Elicitation of Antifungal Compound by Chemical Treatment

Other than the biotic phytoalexin inducers (elicitors), chemicals of widely
diverse natures are known to induce phytoalexin production in the plants.
Under certain circumstances these chemicals might induce host resistance
and could be usedto control plant diseases(Purkayastha, 1986).
In the present investigation 17 compounds belonging to four groups
(heavy metal salts, metabolic inhibitors, reducing agents,and amino acids)
were selected. These compounds were bioassayed for their toxicityby spore
germination tests. Results showed that the heavy metal salts were fungi-
toxic. At a concentration of 10-4 M, mercuric chloride, cupric chloride,
nickel chloride, and cadmium chloride inhibited germination of B. car-
bonum completely while 75% inhibition in germinationwas observed with
nickel nitrate. Cycloheximide and thioglycollic acid also inhibited germina-
tion of B. carbonum significantly. Other chemicals had no significant effect
a of Defense 497
on the spore germination. These chemicals were further tested on suscepti-

ble variety (TV-18)for their effect on disease development, as described by
Yanase and Takeda (1987). It was observed that nickel chloride and nickel
nitrate were very successful in reducing disease development. There was
no appearance of disease symptoms, even 2 weeks following inoculation.
Significant increase of antifungal compounds in treated inoculated plant
was observed which could be correlated with the reduction in disease symp-
toms. Keen et al. (1981) reported that sodium iodoacetate actsas an abiotic
elicitor of glyceollin in primary leaves ofCV. Harosoy soybeans and which
is associated with the expression of resistance. Glyceollin production also
increased significantly in soybean plants (CV. Soymax) after induction of
resistancebysodiumazidetreatment (Chakraborty and Purkayastha,
1987). Sinha and associates ina series of experiments, successfully demon-
strated chemical induction of disease resistance in rice infected with Hel-
minthosporium oryzae, which they correlated withan increased accumula-
tion of a phytoalexin-like substance in the treated plants (Sinha and Giri,
1979; Sinha and Hait, 1982; Giriand Sinha, 1983a,b). Ghosaland Purkay-
astha (1987) also observed a differential response of rice leaves to some
abiotic elicitors of phytoalexin (momilactone A). Gibberellic acid, sodium
azide, and penicillin induced production of momilactone A in rice. These
compounds also reduced the sheath rot diseaseofrice.Among the six
antibiotics used as foliar spray on a susceptible soybean cultivar (“Soy-
max”), cloxacillin induced maximum resistance against anthracnose (Pur-
kayastha and Banerjee, 1990).
W. EPILOG
The defense strategies tea of worked out in the course of the project pointed
out the involvement of phenolics in disease resistance, some of which were
preformed, and otherswere postinfectional. Thelatter could be defined as
phytoalexins (Chakraborty et al., 1989). It is well knownthat plants defend
themselves from microbial pathogens through different mechanisms includ-
ing the synthesis of antimicrobial phytoalexins,the hypersensitive response,
and cellwall modification(HalversonandStacey, 1986).Suchdefense
mechanisms may be induced by the expression of genes resulting from the
recognition of a particular microbe by a host. The exchange of molecular
signals between host and pathogen is considered to be one of the mecha-
nisms of specificity. The signal molecules involved in defense responses
need to be identified.
Preliminary work has been carriedout on both biotic and abiotic elici-
tors in the course of our project. It is now proposed to purify and identify
the fungal cell wall elicitors and establish their precise role in eliciting the
defense response. Detailed work on cell wall modification relating to dis-

ease resistance in tea is another proposed area of research.
Compatibility inthe host-parasite interaction in many instances is also
dependent on the presence of cross-reactive antigens (CRA) between them.
Mimicry of antigenic determinants in plants by various microbes is wide-
spread and is important in the compatibility of plant-parasite interactions
(Chakraborty, 1988). Except for a role in recognition phenomena (Dazzo
and Brill, 1977, 1979), a direct and active role of CRA is still speculative
but has been visualizedin a number of host-parasite systems(Purkayastha,
1989). The role thus visualized for certain common antigens is one that
provides both a stabilizing influence between interacting cells ofdifferent
organisms and a basis for continued compatibility.It is believed that CRA
forms a continuum between cells of hostand parasite that favors the pro-
gressive growthand establishment of the parasite (DeVay et al.,1981). The
role of.CRA is probably subjectto overriding effects of substances such as
phytoalexins or other inhibitory substances already present in host tissues
or induced by parasitic microorganisms. Alteration of specific host antigens
by suitable treatment may also induce disease resistance in plants (Chakra-
borty and Purkayastha, 1987; Ghosal and Purkayastha, 1987; Purkayastha
and Banerjee, 1990).
Keeping this in mind,it was decidedto determine the presence of cross-
reactive antigens betweentea and its pathogens. Isolationand purification
of CRA shared between tea plant and foliar pathogens as well as tissue
and cellular locations of CRA is being carried out. In addition, further
biochemical, physiological, and genetic research is in progress for a com-
plete understandingof the defense mechanism of tea plant againsta number
of foliar pathogens.
ACKNOWLEDGMENTS
This workwas supported inpart by the Council of Scientificand Industrial
Research, New Delhi. Financial assistance received from the University
Grants Commission byA. Saha is also acknowledged.
REFERENCES
surfaces and the biochemistry of pathogenesis. Annu. Rev. Plant Physiol. 2 6
31-52.
Anderson-Prouty, A. J., and Albersheim, P. (1975). Host pathogen interactions.
111. Isolation of a pathogen synthesized fraction rich in glucan that elicits a
defense response in the pathogen’shost. Plant Physiol.5 6 286-291.
tTea of Defense 499
Ando, Y.,and Hamaya, E. (1986). Defense reaction of tea plant against infection
of the tea anthracnose fungus.Study Tea69 35-43.
Bezbaruah, H. P., andSingh, I.D. (1988). Advances inthe use of planting material.
Field Management in Tea (S. K. Sarkar and H. P. Bezbaruah, eds.), Tocklai
Experimental Station, Jorhat, India.
Chakraborty, B. N. (1987). A new record of leaf disease of tea (Camelliasinensis
(L.)0 .Ktze). Curr. Sci.56(17): 900.
Chakraborty, B. N. (1988). Antigenic disparity.Experimental and Conceptual Plant
Pathology (R. S. Singh, U. S. Singh, W. M. Hess, and D. J. Weber, eds.),
Oxford and IBH Publishing Co. Pvt. Ltd.,New Delhi, pp. 477-484.
Chakraborty, B. N., and Saha, A. (1989). Biological activity of leaf diffusates of
tea in relation to resistance to Bipolaris carbonum. Environ. Ecol. 7(3): 717-
720.
Chakraborty, B. N., and Purkayastha, R. P. (1987). Alteration of glyceollin synthe-
sis and antigenic patterns after chemical induction of resistancein soybean to
Macrophomina phaseolina. Can.J. Microbiol. 33: 641-65 1.
Chakraborty, B. N., Chakraborty, U., and Saha, A. (1989). Isolation and charac-
terization of an antifungal compound from leaves challenged withPestalotiop-
sis theae. Indian Phytopathol. 42(2): 302.
Chakraborty, B. N., and Chakraborty, U. (1988). Biological control of foliar dis-
eases of tea. Proceedings of 5th International Congress of Plant Pathology,
Kyoto, p. 344.
Chessin, M., and Zipf, A. E. (1990). Alarm system in higher plants. Bot. Rev. 56:
193-235.
Dazzo, F. B., and Brill, W. J. (1977). Receptor site on clover and alfalfa roots for
Rhizobium. Appl. Environ. Microbiol.33: 132-136.
Dazzo, F. B., and Brill, W. J. (1979). Bacterial polysaccharide which binds Rhizo-
bium trifollito clover root hairs. J. Bacteriol. 137: 1362-1373.
DeVay, J. E., Wakeman, R. J., Kavanagh, J. A., and Charudattan, R. (1981). The
tissue and cellular location of a major cross-reactive antigen shared by cotton
and soil-borne fungal parasites.Physiol. Plant Pathol.18: 59-66.
Deverall, B. J., and Wood, R. K. S. (1961). Infection in bean plants (Viciafaba L.)
with Botrytis cinerea and B. fabae. Annals of Applied Biology.4 9 461-472.
Dickens, J. S. W., and Cook, R. T. A. (1989). Glomerella cingulata on Camellia.
Plant Pathol.39: 75-85.
Ghosal, A., and Purkayastha, R. P. (1987). Biochemical response of rice (Oryza
sativa L.) leaves to some abiotic elicitors of phytoalexin. Indian J. Exp. Biol.
25: 395-399.
Giri, D. N., and Sinha, A. K. (1983a). Effects of heavy metals on susceptibility of
rice seedlingsto brown spot disease. Ann. Appl. Biol. 103: 229-235.
Giri,D. N., andSinha,A. K. (1983b). Control ofbrown spot diseaseofrice
seedlings by treatment with a select group of chemicals. Z. Pfanzenkrunk.
Pfanzenschutz 90 479-487.
Halverson, L. J., and Stacey, G. (1986). Signal exchange in plant-microbe interac-
tions. Microbiol. Rev. 50(2): 193-225.
500 Chakraborty et ai.
Hamaya, E., Tsushida, T., Nagata, T., Nishino, C., Enoki, N., and Manabe, S.
(1984). Antifungal components of Camellia plants. Ann. Phytopathol. Soc.
Jpn. 50(5): 628-636.
Hammerschmidt, R., and Nicholson, R. L. (1977). Resistance of maizeto anthrac-
nose, Changes in host phenolsand pigments. Phytopathology 67: 251-258.
Harborne, J. B. (1973). Phytochemicalklethods. Chapman and Hall, London.
Hofmans, A. I., and Fuchs, A. (1970). Direct bioautography on thin-layer chroma-
tograms as a method for detecting fungitoxic substances. J. Chromatogr. 51:
327-329.
Kawamura, J., and Takeo, T. (1989). Antibacterial activity oftea catechin to Strep-
tococcus mutans. Nippon Shokuhin Kogyo Gakkaishi 36(6): 463-467.
Keen, N. T. (1978). Phytoalexins: efficient extraction from leaves by a facilitated
diffusion technique. Phytopathology 68 1237-1239.
Keen, N. T., Ersek, T., Long, M., Bruegger, B., and Holliday, M. (1981). Inhibi-
tion of the hypersensitive reaction of soybean leaves to incompatiblePseudom-
onas spp. by blasticidin S, streptomycin or elevated temperature. Physiol. Plant
Pathol. 1 8325-337.
Macri, F.,Dilenna, P., and Vianelli, A. (1974). Preliminary researchon peroxidase,
polyphenol oxidase activity and phenol content in healthy and infected corn
leaves, susceptible and resistant to Helminthosporium maydis race T. Rev.
Patologie Vegetale. 4 109-122.
Muller, K. 0.(1958). Studies on phytoalexins. I. The formation and the immuno-
logical significance of phytoalexin produced by Phaseolus vulgarisin response
to infections with Sclerotima fmcticola and Phytophthora infestans.Aust. J.
Biol. Sci. 11: 275-300.
Nagata, T.,Tsushida, T., Hamaya, E., Enoki, N., Manabe, S., and Nishino, C.
(1985). Camellidins, antifungal saponins isolated from CamelliaJaponica.
Agric. J. Biol. Chem. 49(4): 1181-1 186.
Nishino, C., Manabe, S., Enoki, N., Nagata, T., Tsushida, T., and Hamaya, E.
(1986). The structure of the tetrasaccharide unit of camellidinis,saponins pos-
sessing antifungal activity. J. Chem. Soc. Chem. Commun.720-723.
Purkayastha, R. P. (1986). Elicitors and elicitation of phytoalexins. Vistas in Plant
Pathology (A. Verma and J. P. Verma, eds.), Malhotra, New Delhi, pp. 25-
44.
Purkayastha, R. P. (1989). Specificity and disease resistance inplants. Presidential
address, Section of Botany, 76th Session of the Indian Science Congress, Ma-
durai, pp. 1-32.
Purkayastha, R. P., and Banerjee, R. (1990). Immunoserological studies on cloxa-
cillin-induced resistance of soybean against anthracnose. Z. Pfanzenkrank.
Pfanzenschutz 97(4): 349-359.
Purkayastha, R. P., and Ghosh, S. (1983). Elicitation and inhibition of phytoalexin
biosynthesis in Myrothecium-infected soybeanleaves. Indian J. Exp. Biol. 23:
216-218.
Sambandam, T., Sivaswamy, N., and Mahadevan, A. (1982). Microbial degrada-
tion of phenolic substances.Indian Rev. Life Sci. 2 1-18.
Pathogens
Fungal
Defense
Against
of Tea 501
Sarmah, K. C. (1960).Diseases of Tea and Associated Crops in North-East India.

Indian Tea Association, Memorandum No.26,pp. 33-40.
Schlosser, E. (1988). Preformed chemical barriers in host-parasite incompatibility.
Experimental and Conceptual and Plant Pathology (R.S. Singh, U. S. Singh,
W.M. Hess, and D. J. Weber, eds.), Oxford and IBH Publishing Co. Pvt.
Ltd., New Delhi, pp. 465-476.
Schlosser, E. W. (1980). Preformed internal chemical defenses. Plant Disease and
Advanced Treatise:How Plants Defend Themselves, Vol.5 (J. G. Horsfall and
E. B. Cowling, eds.), Academic Press,New York, pp. 161-177.
Schonbeck, F.,and Schlosser, E.(1976).Preformed substances as potential protect-
ants. Encyclopedia of Plant Physiology, Vol. 4 (R.Heitfuss and P. H. Wil-
liams, eds.), Springer-Verlag, Berlin, pp.651-678.
Sinha, A. K., and Giri, D. N. (1979). An approach to control brown spot of rice
with chemicals knownas phytoalexin inducers.Curr. Sci.4 8 782-784.
Sinha, A. K.,and Hait, G. N. (1982).Host sensitization as a factor in the induction
of resistance in rice plants against Dreschlera infection by seed treatment with
phytoalexin inducers. Trans. Br. Mycol.Soc. 79(2): 213-219.
Sridhar, R., and Ou, S. H. (1974). Phenolic compounds detectedinriceblast
disease. Biologia Plantarum1 6 67-70.
Stahl, E.(1967).Diinnschicht - Chromatographie.Springer-Verlag, Berlin.
Toda, M., Okubo, S., Chrishi, R., and Shimamura, T. (1989). Antibacterial and
bactericidal activitiesof Japanese green tea. Jpn. J. Bacteriol.44(4): 669-672.
Trivedi, N., and Sinha, A. K. (1976). Resistanceinducedinriceplantsagainst
Helminthosporium infection by treatment with variousfungal fluids. Phytopa-
thol. Z. 86: 335-331.
Van Etten, H. D., Mathews, D. E., and Mathews, P. S. (1989).Phytoalexin detoxi-
fication: Importance for pathogenicity and practical implications. Ann. Rev.
PhytOpathOl. 27: 143-164.
Van Sumere, C. F., Wolf, G., Teuchy, H., and Kint, J. (1965). A new thin-layer
method for phenolic substancesand coumarins. J. Chromatogr.22: 48-60.
Wang, Z. N. (1991). Chinese famous teas and their characteristic constituents bio-
formation. World Tea: International Symposium on Tea Science, Shizuoka,
Japan, pp. 23-33.
Ward, E. W. B. (1986). Biochemical mechanisms involved in resistance of plants to
fungi. Biology and Molecular Biologyof Plant-Pathogen Interactions(J. Bai-
ley, ed.), Springer Verlag, Berlin. pp.107-131.
Werder, J., and Kern, H. (1985). Resistance of maize to Helminthosporium car-
bonum: changes in host phenolics and their antifungal activity. J. Plant Dis.
Prot. 92(5): 477-484.
Yamanishi, T. (1991). Flavor characteristics of various teas. World Tea: Interna-
tional Symposium on Tea Science,Shizuoka, Japan, pp. 1-1 l.
Yanase, Y., and Takeda, Y. (1987). Method for testing the resistance of tea gray
blight caused by Pestalotia longiseta Spegazzini in tea breeding. Bull. Natl.
Res. Inst. VegetableOrnamental Plants and Tea. (Kanaya) I : 1-9.
22
Effects of Age-Related Resistance and
Metalaxyl on Capsidiol Production in
Pepper Plants Infected with
Phytophthora capsici
Byung Kook Hwang
of Korea
Korea University, Seoul, Republtc
1. INTRODUCTION
A. AgeRelated Resistance and Metalaxyl for Control
of fhyfophthora Blight
Phytophthora blight of pepper (Capsicum annuum L.) plants caused by
Phytophthora capsici Leonian is a serious soil-borne disease in the major
pepper-growing areas of the world (Leonian, 1922; Weber, 1932; Barksdale
et al., 1984). It has been reported that P. capsici attacks plants such as
pepper, eggplants, cucumber, honeydew melon, pumpkin, squash, tomato,
and watermelon (Polach and Webster, 1972). Phytophthora rots of the
lower stem of plants favored by high temperature, high soil moisture,and
atmospheric humidity are common and most severe in low-lying, poorly
drained areas. In pepper plants, a brown girdlingrot of the stems becomes
enlarged, the lower leavesdrop, and eventually the whole plant wilts.
Phytophthora blight of pepper plants has been controlled crop by rota-
tion, the use of resistant cultivars, and the application of systemic fungi-
cides such as metalaxyl. Althoughthis disease cannot be readily controlled
by any one measure, growing resistant cultivars may be very effective in
reducing damagefrom P . capsici. Recently, Kim et al. (1989) demonstrated
an age-related resistance of pepper cultivars, which is distinctly expressed
as pepper plants mature. Since the Phytophthora disease causes severe dam-
age in pepper plants onlyat later growth stages inthe field, the age-related
resistance must be considered in breeding resistant pepper cultivars.
Our earlier studies demonstrated the expression of age-related resis-
tance in pepper plants infected with P. capsici under controlled environ-
mental conditions(Kim et al., 1989). We also suggestedthat the appearance
503
504 Hwang
of age-related resistancein the pepper cultivars may be due to the morpho-

logical and nutritional changes in tissues of pepper stems during aging (i.e.,
the drastic increase in amount of dry matter; the significant decrease in
amounts of mineral nutrients such as nitrogen, phosphorus, potassium,
calcium, and magnesium; and the lower contents of fructose, glucose,and
sucrose in the stem tissues) (Jeun and Hwang, 1991). Such morphological
and physiological changes in pepper stems of different ages associated with
age-related resistance to P. capsici may be caused by the production of
some proteins or enzymes considered to be the result of differential gene
action (Hwang et al., 1991).
Metalaxyl [methylN-(2-methyoxyacetyl)-N-(2,6-xylyl)-~,~-alaninate] is
a systemic fungicide which is specifically active against members of the
Oomycetes such as Phytophthora spp. and downy mildews. Recently, this
fungicide has been intensively applied in pepper-growing countries for con-
trolling Phytophthora blightcaused by P. capsici.Metalaxylhasbeen
shown to be highly effective againstPhytophthora blight of pepper plants
by inhibition of mycelial growth, germination of zoospores, zoospore re-
lease from sporangia, and sporangium formation, even at low concentra-
tions (Sung and Hwang, 1988). When applied to pepper plants, either as a
soil drench or as a foliar spray, metalaxyl exhibits both protective and
curative effects againstPhytophthora blight (Sungand Hwang, 1988).
B. Phytoalexin Production Associated with Age-Related

Resistance and Metalaxyl Treatment
Such differences in the degree of resistance of pepper plants may also be

due to preformed inhibitory substances present in plants and/or antimicro-
bial compounds, phytoalexins, accumulating after exposure to P. capsici.
Recently, Hwang and Heitefuss (1982) demonstrated that preformed anti-
fungal compounds in leaf extracts of barley, though present in trace quanti-
ties, may contribute to the expression ofa gradual, decreased susceptibility
or quantitative resistanceto powdery mildew with increasing age of plants.
In the soybean-Phytophthora megasperma f.sp. glycinea (Pmg) interac-
tion, Paxton and Chamberlain (1969) reported that soybean plants became
increasingly resistantto Pmg as theymatured, despite greatly reduced phy-
toalexin (glyceollin) production. In contrast, Keen (1971) concluded that
the highest glyceollin concentrations accumulated in infected older plants
of soybeans. Lazarovits et al. (1981) also demonstrated that increased gly-
ceollin production was associated with age-related resistance of soybean
hypocotyls to the compatible race, whereas glyceollin production decreased
with tissue age inthe incompatible race interactions despite increased resis-
tance.
Phytophfhora Infection in Pepper 505
Phytoalexins are produced by plants not only in response to interac-

tions with fungi, bacteria, viruses, nematodes,and other living organisms,
but also followingtreatment with many chemicals, irradiation by ultraviolet
light, and exposureto the products of microbial metabolism (Bailey,1982).
In addition, phytoalexin production can also be induced by fungicides.
Reilly and Klarman(1972) demonstrated that several fungicides induced the
phytoalexin glyceollin in soybean [Glycine max (L.)Merr.] hypocotyl tis-
sues. Cartwright et al.(1977) also reportedthat dichlorocyclopropanes may
exert their systemic fungicidal activity against the rice blast disease by caus-
ing riceto synthesize the phytoalexins monilactonesA and B.
Several years ago, it was demonstrated that control of Phytophthora
rot of soybean hypocotyls by the systemic fungicide metalaxyl caused a
hypersensitive response accompanied by glyceollin production (Ward et al.,
1980; Borner et al., 1983). Subsequently, it was found that in plants only
partially protectedby metalaxyl glyceollin concentrations exceeded the EC,
in vitro in all lesions. However, spreadof the pathogen was not restricted
(Lazarovits and Ward, 1982), suggesting that glyceollin may not play a
significant role in inhibiting spread of the fungus in metalaxyl-treated seed-
lings.
Stossl et al. (1972) first reported that pepper fruits produced the anti-
fungal sesquiterpenoid phytoalexin capsidiol in response to infection by
several pathogenicand nonpathogenic fungi. Other fungi and bacteria also
elicit accumulation of capsidiol (Stossl et al., 1972; Ward et al., 1973). A
series of extensive studies the of rate and magnitude of capsidiol accumula-
tion coupled withultrastructural studies to support a role for the compound
in the restriction of fungal growth and development in infected pepper
fruits and stems (Jones et al., 1974; Hwang et al., 1990). It also accumu-
lated around sitesoflocalizedinfectioninleaves (Ward, 1970, but its
accumulation or the sensitivity of fungi to capsidiol cannot explain the
resistance or susceptibility of pepper to all fungi. The significance of capsid-
io1 accumulation in compatibleand incompatible interactions between rip-
ening fruits and fungi has been assessed (Ward and Stossl, 1972; Ward et
al., 1973; Stossletal., 1973,1977; Jones et al., 1975a, b). The relation
between capsidiol concentration and speed of fungal invasion in stemsof
pepper cultivars susceptibleor resistant to P. capsici has been assessed by
Molot et al. (1981). However, they could not demonstrate a convincing
connection between capsidiol induction at the infection front and inhibition
of lesion development,or a'clear difference inthe reaction of the resistant
cultivar.
The role of capsidiol in the age-related resistance of pepper plants to P.
capsici has not been studied extensively.In the present studies,the involve-
ment of postinfectionally formed capsidioltheindevelopment ofP.capsici
506 Hwang
in stemsand roots of pepper plantsat different growth stages was examined

in relation to age-related resistance of pepper plants. It has also not been
demonstrated that applications of the systemic fungicide metalaxyl for con-
trol of Phytophthora blight of pepper plants affect capsidiol production
in infected pepper plants. In the experiments reported here, we examined
capsidiol production in stemsof pepper plants treated with metalaxyl and
inoculated with P . capsici. We also compared capsidiol production in the
susceptible and resistant cultivars treated with metalaxyl. The majority of
results presented in this chapter have been reported elsewhere (Hwang and
Sung, 1989; Hwang and Kim, 1990).

A. Plant, Fungus, and Fungicide
The pepper (Capsicum annuum L.) cultivars used were Hanbyul and King-
to Phytophthora blight. Sixor
kun, differing quantitatively in susceptibility
ten plants per pot were grown in plastic pots (5 x 15 x 10 cm) containing
a mixture of steam-sterilized loam soil,sand, and peat (3 :5 :2 v/v). Fertil-
izer was applied at the rate of 0.27-0.13 g of actual N-P-K per pot at 3
weeks after planting. Plants were maintained at 25 f 2OC during a 16-hr
*
light period from0600 to 2200 hr of about 5000 lx and at 16 2OC during
darkness.
An isolate of P. capsici was cultured on V-8 agar for 7 days at 25 *
1OC. Zoospores were produced as described previously (Kim et al., 1989)
and a zoospore suspensionwas usedas inoculum. Zoospores in suspensions
were counted with a haemocytometer, and the concentration was adjusted
to 1 X 105/ml with steriletap water.
The fungicide used in this study was metalaxyl, formulated as a 25%
a.i. wettable powder. All concentrations are given as active ingredient(Ri-
domil, Seoul Agricultural Chemical Co.). Metalaxyl suspensions were pre-
pared in sterile tap water to give appropriate concentrations. Metalaxyl
treatments were performed by pouring the 20 ml of metalaxyl suspensions
per pot uniformly over the surface a day before stem inoculation with
P. capsici, except in the experiment on the effect of time of metalaxyl
application.
B. Inoculation Procedure and Disease Assessment

A longitudinal wound, about 1 cm long and 1 mm deep, was made with a
razor blade on each stem of pepper plants, approximately1 cm (bottom) or
9 cm (middle) from the soil surface and at the top ofstems. A small
quantity of sterile cotton soaked in zoospore suspension(1 x 10S/ml)for
Phyfophfhora Infection in Pepper 507
30 min was placed on the wounded sites of stems. The inoculation siteswere
then covered with plastic tape to maintain the moist condition. Metalaxyl
suspensions at various concentrations were poured into the soil before or
after stem inoculation withP. capsici.
For root samples, plants at the six-leaf and first branched-flowering
stages were inoculated by uniformly pouring 20m1 of a suspension of
motile zoospores(1 x 105/ml) perpot over the surface of soil. Immediately
after soil drenching,pots were placed in a large plastic tray filled with tap
water.
Disease development, as measured by lesion length, in pepper plants
was rated every day after inoculation. The data are the means of three
replicates from the disease ratings of15 infected plants.
C. Extraction and Estimation of Capsidiol
Fifteen stems of pepper plants per treatment replicate were harvested at

various intervalsafter inoculation. Three replicatesof each treatment were
used. The infected stems were sliced transversely in 3-mm-thick sectionsfor
4 cm upward (bottom and middle) or downward (top) from the inoculated
sites. Comparable areasof healthy, metalaxyl-treated stems also were simi-
larly sliced.
At different intervals following inoculation, soil
was removed from the
roots of 10 plants by using running tap water. The roots were carefully
washed, blotted with filter paper,and sampled. The sliced stem sectionsor
roots were placed in 250-m1 Erlenmeyer flasks with 15 ml/g fresh wt of
40% aqueous ethanol and were then vacuum-infiltrated. The flasks were
stoppered and placed on a reciprocal shaker (110 strokedmin) for 7 hr.
After agitation by shaking, the stem or root tissues were removed byfiltra-
tion andthendriedovernight at 95OC. The filtrate was vacuum-
concentrated at 4OoC to approximately one-half volume.The concentrate
was partitioned twice with ethyl acetate. The pooled ethyl acetate layerwas
dehydrated with MgSO, and evaporated to dryness at 4OOC. Following
transfer of the residue to vials with peroxide-free diethyl ether
and evapora-
tion of ether under nitrogen, the dry residue was dissolved in 0.5 m1 ethyl
acetate and then storedat -2OoC until usedfor gas chromatographic anal-
ysis of capsidiol.
Allcapsidiolanalysesweremadebygaschromatographyfollowing
Ward et al. (1973). The ethyl acetate sample was evaporated under nitro-
gen. The final residues were dissolved in a solution of methyl myristate (4
mM) in ethanol, with the ester serving as an internal standard. Aliquots
(2-3 pl) from each ethanol solution were injected into a Packard model 419
gas-liquid chromatograph fitted with a glass column (182 cm long, 2 mm
S08 Hwang
i.d.) containing Gas Chrom Q (80-120 mesh) coated with 3% SE30. The
column was kept at 162OC, the injector at 192OC, and the flame ionization
detector at 230OC. The carrier gas was nitrogen at a 40 ml/min flow rate
with hydrogen and air at 30 and 300 ml/min, respectively. Retention times
were 7.5 and 14.5 min for methyl myristate and capsidiol, respectively.
Capsidiol in the sampleswas identified by cochromatography withauthen-
tic capsidiol, which was obtained from Dr. A. Stossl, AgricultureCanada,
Research Center,London, Ontario, Canada.
111. RESULTS
A. Effect of Plant Age and Stem Part on Disease Development
About a day after inoculation withP.capsici, the inoculation sitesof stems
of pepper plants became greyishto brown, as previously described(Kim et
al., 1989). Slight differences between susceptibleCV. Hanbyul and resistant
CV. Kingkun in lesion length were observed at the six-leaf stage (Fig.l), and
the differences were more apparent at the first branched-flowering stage
(Fig. 2). When inoculated on three stem sites, lesions at the top of stems
progressed more rapidlythan those of the middle or bottom (Figs.2 and 3).
0
U
1 2 3 C 5 6 7 0
. f2 i 3t
1
. . . .
C S 6 7
In
Oays after inoculation
Figure 1 Time courses of (A) capsidiol accumulation and(B) disease development
in the stemsof pepper cultivars Hanbyul (susceptible) and Kingkun (resistant)theat
six-leaf stage inoculated with zoospore suspensions of P. capsici at the bottom of
stems. Vertical bars represent standard deviations.
Phytophthora Infection in Pepper 509
Figure 2 Time courses of (A) capsidiol accumulation and (B) disease development
in the stemsof pepper cultivarsHanbyul (susceptible) and Kingkun (resistant) at the
of P. capsici at
first branched-flowering stage inoculated with zoospore suspensions
the middle andbottom of stems. Vertical bars representstandard deviations.
The ranks of lesion length were top > middle > bottom of stems at the
first branched-flowering stage. The lesion development in Kingkun was
much slower than that in Hanbyul, as the plants reached the first branched-
flowering stage that is considered the approach of maturity.
B. E f f e c t of Plant Age and Stem Part on Capsidiol Production in Stems

The concentrations of capsidiol were determined in the stems of the CV.
Hanbyul and the CV. Kingkun infected by P. capsici at the six-leaf stage
(Fig. 1). Capsidiol was not detected in stems from healthy plants. The
amount of capsidiol in the stems could be measured only until 7 days after
inoculation because all infected plants died. The lesion development in the
510 Hwang
Days after inoculation

Figure 3 Time courses of (A) capsidiol accumulation and(B) disease development
in the stems of pepper cultivars Hanbyul (susceptible) Kingkun
and (resistant)at the
six-leaf and first branched-flowering stages inoculatedwith zoospore suspensionsof
P. capsici at the top of stems. Vertical bars represent standard deviations.
stems of the CV. Hanbyul was more rapid than that of the CV. Kingkun. In
contrast, the concentrations of capsidiol in the two cultivars were similar,
with a maximum at the third day after inoculation. The resistant CV. King-
kun always contained more capsidiolthan the susceptible CV. Hanbyul. In
particular, a secondary accumulation of capsidiol at 7 days after inocula-
tion occurred inthe resistant CV. Kingkun. Atthe six-leaf stage, the level of
capsidiol at the top of stems2 days after inoculation was higher in Hanbyul
than in Kingkun, but declined thereafter (Fig. 3).
The lesion developmentand capsidiol concentrationat the top,middle,
and bottom of stems of the susceptible CV. Hanbyul and the resistant CV.
Kingkun at the first branched-flowering stage are compared in Figs.2 and
3. The lesion length at the top,middle, and bottom of stems increased more
rapidly in the susceptible CV. Hanbyul than in the resistant CV. Kingkun.
l The differencesbetweenthetwocultivarsbecamemoreconspicuous at the
l lowerpartsofstems.Capsidiolwasfirstdetected at allparts of stems a
day after inoculation, thereafter accumulating more in Kingkun than in

Hanbyul. At 7 days after inoculation, the levels of capsidiolat the middle
and bottom of stems of the CV. Kingkun were about 2-3.5 times as highas
those in the susceptible CV. Hanbyul. A high quantity of capsidiol was
produced at the top of stems 2 days after inoculation, but thereafter it
declined slowly. Significant differences between the two cultivars in capsid-
io1 accumulation were observedat the bottom and middle of stemsfrom 5
and 7 days 'after inoculation, respectively. More capsidiolwas produced at
the upper parts than at the bottom of stems.
C. Effect of Plant Age on Capsidiol Production in Roots

The concentrations of capsidiol which accumulated inroots of the suscepti-
ble CV. Hanbyul and the resistant CV. Kingkun at various intervals after
inoculation withP. capsici are presented in Fig.4. Capsidiol was not detect-
able in roots from healthy plants.At the six-leaf stage, the concentrations
of capsidiol in roots of the two cultivars during the disease progresswere
similar to each other, except for a higher amount in the resistant CV. King-
kun at 6 days after inoculation. At the first branched-flowering stage, the
accumdlation of capsidiol inthe roots gradually increased with the maxima
Daysafter soil drench

Figure 4 Timecourses of capsidiolaccumulationin roots of peppercultivars
Hanbyul (susceptible) and Kingkun (resistant) at the six-leaf and first branched-
flowering stages drenched with zoospore suspensionsof P. capsici in the soil. Verti-
cal bars represent standarddeviations.
512 Hwang
at 10 and 12 days after inoculation in the cvs. Hanbyul and Kingkun,

respectively, and declined afterward. In particular, higherquantities of
capsidiol accumulated in Kingkunthan those in Hanbyul until6 days after
inoculation. DuringPhytophthora infection inroots of pepper plants, cap-
sidiol greatly accumulatedat the first branched-flowering stage when com-
pared to the six-leaf stage.
D. Effect of Concentration of Metalaxyl onCapsidiol Production

When metalaxyl was applied to the soil a day before stem inoculation of
zoospore suspensionof P. capsici, lesion developmentfrom the inoculation
site of the stem decreased in the two cvs. Hanbyul and Kingkun, with
increasing amounts of metalaxyl (Fig. 5A). Four days after stem inocula-
tion at the eight-leaf stage, susceptible Hanbyul
was more severely diseased
than resistant Kingkunat 1 and 3 pg/ml. A s the concentration of metalaxyl
increased, disease progress gradually decreased, with the decline being more
METALAXYLbtg rn1-l)
Figure 5 Influence of concentration of a metalaxyl drench in soil on (A) disease

development and (B) capsidiol production in the stems of pepper cultivarsHanbyul
(susceptible) and Kingkun (resistant). Plants were inoculated at the eight-leaf stage
with zoospore suspensions of P. capsici at the bottom of the stem. Data were
obtained on day 4 after stem inoculation. Vertical bars represent standard devia-
l tions.
fhytophfhora Infection in Pepper 513
pronounced in Hanbyulthan in Kingkun.In particular, at the drench treat-

ment of 11 pg/ml of metalaxyl, lesion development was strikingly inhibited
to l - c m lesion length inthe two cultivars.
In contrast to the disease development inboth cultivars, the production
of capsidiol in pepper stems increased with increasing amounts of metalaxyl
(Fig. 5B). In comparable inoculated, drenched plants, resistant Kingkun
always contained more capsidiol in stems than susceptible Hanbyul, irre-
spective of metalaxyl concentration. However, capsidiol accumulationby
metalaxyl was more noticeable in Hanbyul than in Kingkun. At 11 pg/ml
of metalaxyl, the two cultivars had similar levels of capsidiol. Metalaxyl
treatment did not induce capsidiol in healthy, uninoculated plants of both
cultivars.
E. Effect of the Time of Metalaxyl Application on Capsidiol Production

When metalaxylwas applied to soil at 7 pg/ml3 or 7 days beforeand at the
time of stem inoculation on pepper plants at the 10-leaf stage, Phytophth-
ora blight was inhibited in resistant Kingkunand greatly restricted in sus-
ceptible Hanbyul (Fig. 6). Small quantities of capsidiol accumulated simi-
larly in the two pepper cultivars. When the soil drench of metalaxyl was
delayed after inoculation, lesion development gradually progressedboth in
pepper cultivarsand capsidiol accumulated in large amounts. In particular,
the accumulation of capsidiol in metalaxyl-treated plants was much more
marked in susceptible Hanbyulthan in resistant Kingkun 1 or 2 days after
inoculation.
F. Time Course of Capsidiol Production by Metalaxyl

A s previously described,the retardation of lesion developmentwas directly
proportional to the amount of metalaxyl applied (Fig.7). The accumulation
of capsidiol in stems ofthe inoculated, undrenchedcontrols increased to a
maximum at 3 days after inoculation in the two cultivars and declined
afterward. At 1 or 5 pg/ml of metalaxyl, accumulation of capsidiol in-
creased morethan in the inoculated, undrenched control, with a maximum
at 2 days after inoculation in both cultivars.
In these treatments, no signifi-
cant differences between the two cultivars in capsidiol accumulation by
metalaxyl were observed. Capsidiol production in both cultivars was about
three times higher at 5 pg/ml of metalaxyl than at 1 pg/ml of metalaxyl at
2 days after inoculation. The concentration of capsidiol in stems decreased
after maximum accumulation following2 days of stem inoculation, regard-
less of metalaxyl treatmentand pepper cultivar.
514 Hwang
0 Hanbyul(susceptible1
-5 R Kingkun (resistant)
-
3
3 100
E!
0
50
U
U
TIME OF METALAXYLAPPLICATION
DAYS BEFORE AN0 AFTER INOCULATION
Figure 6 Influence of time of soil drench of metalaxyl relative to the time of

inoculation on (A) disease developmentand (B) capsidiol production in the stems of
pepper cultivars Hanbyul (susceptible) and Kingkun (resistant). Plants were inocu-
lated at the 10-leaf stage with zoospore suspensions of P. capsici at the bottom of
the stem. Data were obtained on day 4 after stem inoculation. Vertical bars repre-
sent standard deviations.
IV. DISCUSSION
A. Effect of Plant Ageon Capsidiol Production
Infection systems used here allowed a precise determination both
of disease
development and capsidiol concentrations inP. capsici-infected tissues of
pepper plants of the susceptible CV. Hanbyul and the resistant CV. Kingkun
METALAXYL 1119 m(-' METALAXYL SHg ml"
1 2 3 4 5
DAYS AFTER STEM INOCULATION
Figure 7 Time courses of (A) disease progress and(B) capsidiol production in the
stems of pepper cultivars Hanbyul (susceptible) and Kingkun (resistant). Plants were
inoculated at the eight-leaf stage with zoospore suspensions of P. capsici at the
bottom of thestemafter soil treatmentwithmetalaxyl.Verticalbarsrepresent
standard deviations.
at different development stages.The resistance of pepper plantsto P. cap-

sici increased as theymatured, irrespective of cultivar. The differences be-
tween susceptible and resistant responses to P. capsici were quantitative
rather than qualitative because symptoms slowly developed even on resis-
tant plants. Whatever the basis for age-related resistance in the different
tissues of pepper plants, it appears that P. capsici has a preference for
516 Hwang
young tissues. In stem infections, for instance, the pathogen spread much
more slowlyat the bottom than in the middle of stems (Fig. 2). In contrast,
more capsidiol was produced inthe middle than at the bottom of the stem
following wound inoculation. These findings indicatingthat capsidiol pro-
duction decreased in older tissues of stems despite increased resistancedo
not support the concept ofa general inverse relation between the amount of
phytoalexin accumulating inside the host tissue and the rate ofdisease
development. Accordingly, the increases in disease resistance in older tis-
sues of stems may be derived from other mechanisms including possible
morphological barriers and the production of inhibitory compoundsother
than capsidiol. Hahn et al. (1985) demonstrated that most of the phyto-
alexinaccumulatedin the epidermisin P . megasperma f. sp. glycines-
soybean system. Because there is more dry matter in the bottom tissue of
pepper stems, the number of epidermal cells that can synthesize capsidiol
would be less per unit dry weight at the bottom than in the middle of stems.
On these assumptions, it should be considered that the higher amount of
capsidiol may be produced per epidermal cell at the bottom than in the
middle of stems.
Our data demonstrate that, in general, capsidiol concentrations in in-
fected organs of pepper plants were correlated with the degree of resistance
expressed to P. capsici. The resistant CV. Kingkun always contained more
capsidiol in all infected organs such as stem and root than the susceptible
CV. Hanbyul, suggestingthat the resistant cultivar synthesizes more capsid-
io1 in responseto infection. Even the susceptible cultivar accumulated con-
siderable capsidiol, especially when older plants were inoculated. As pre-
viously observed by Molot et al. (1981), however, capsidiol accumulation
may not bethe main factor restricting P . capsici growth in resistant pepper
plants, but other antifungal compounds may be induced, together with
capsidiol by the infection of P . capsici. It would appear that susceptible
cultivars also have the ability to produce capsidiol, possibly being differ-
ently expressed as plants mature. Other mechanisms, such as the preferen-
tial degradation of capsidiol in susceptible plants, should not be excluded
(Ward and Stossl, 1972; Stossl et al., 1977).
The basis for the increased resistance ofmature pepper plants to Phy-
tophthora blight seemsto be distinguishedfrom that of older tissues ofthe
same stems. Phytophthora infection was reduced and capsidiol accumu-
lated more in stemsand roots at the first branched-flowering stagethan at
the six-leaf stage. These results suggest that capsidiol accumulation in in-
fected stems and roots of mature pepper plants may contribute to the in-
crease in resistanceto P . capsici, being especially affectedby host genotype
and plant age. However, possible morphological barriersand other inhibi-
tory compounds inmature pepper plants may also playimportant roles in
Phytophthora Infection in Pepper 517
retarding growthof P. capsici, as previously described in other host-patho-

gen combinations (Hwang and Heitefuss, 1982).
There were wide differences in the concentrations of capsidiol deter-
mined in different organs. Lower concentrations of capsidiol were found in
roots than in stems. The production of low amounts of capsidiol in roots
appear to be dueto different biochemical interactionsof host and pathogen
in these organs. Tissues of the stems were uniformly necrotic and slightly
sunken, but tissues of the roots were disintegrated ina watery rot and died.
Small quantitiesof capsidiol (below 30 pg/g dry wt) in the young roots at
the six-leaf stage may not contribute to the expression of resistance to P.
capsici. In contrast, old roots of pepper plants at first branched-flowering
stage responded to inoculation with zoosporesof P. capsici by accumulat-
ing considerable quantitiesof capsidiol, indicatingthat capsidiol may con-
tribute to the resistance expression of pepper plantsat late growth stages.
Similarities in lesion developmentand time courses of capsidiol accu-
mulation in root and stem of pepper plants at late growth stages would
suggest that similar mechanisms govern the increased resistance of pepper
plants to P. capsici with increasing age.
B. Effect of Metalaxyl on Capsidiol Production

In addition to the efficacy of metalaxylfor control of P. capsici on pepper
plants (Sung and Hwang, 1988), metalaxyl treatment stimulated capsidiol
production in pepper stems infected with the fungus. Whereas some fungi-
cides are themselves capable of inducing phytoalexin production in plants
(Reilly and Klarman, 1972), it seems likely that metalaxyl itself does not
elicit capsidiol production in pepper stemsbut rather increasesthe capacity
of pepper plants to synthesize capsidiol in responseto infection by P. cap-
sici. This observation is supported by the absence of capsidiol production
in uninoculated, metalaxyl-treated plants inthis study. Applicationof met-
alaxyl in pepper plants infected with P. capsici retarded disease develop-
ment but did not change the rapidly growing stem lesions into brownish
necrotic ones of the hypersensitive reaction. In contrast to the findings of
Ward et al. (1980) that control by metalaxyl of P. megasperma f.sp. gly-
cines in soybean hypocotyls resulted ina hypersensitive resistant response
accompanied by glyceollin production, our results suggest that metalaxyl
can also stimulate phytoalexin production even in a susceptible responseto
Phytophthora infection in plants, even though there is no hypersensitive
resistant response. Our previous electron microscopic observations that in
metalaxyl-treated stems an electron-dense material was apposed in those
sites of the host cell wall in most intimate contact with the fungal cell
wall (Hwang et al., 1989, 1990) also provide the possibility that metalaxyl
518 Hwang
treatment may induce the plant defense reaction.In the control of rice blast
by dichlorocyclopropanes, Cartwrightet al. (1977) also suggested that the
compound acts by sensitizing the plant such that subsequent inoculation
with P. oryzae Br. and Cav. results ina resistant reaction.
Withincreasingmetalaxylconcentrations,capsidiolproductionwas
stimulated in infected pepper stems. The increase followed similar patterns
in the susceptible and resistant cultivars (Fig. 5). Inhibition of disease devel-
opment aswell as stimulationof capsidiol production by metalaxylappear
to be greatly influencedby the degree of resistance of pepper cultivars to P.
capsici. Therefore, resistant pepper cultivars may possess a genetic basisfor
producing more capsidiol than susceptible ones, although the difference
between the susceptible and resistant responses of pepper plantsto P. cap-
sici is quantitative rather than qualitative (Kim et al., 1989). Whereas large
quantities ofglyceollinproducedinsoybeanby an incompatiblehost-
pathogen interaction remainedconstant over a range of metalaxyl concen-
trations (Ward et al., 1980), it isof interest that capsidiolamountsin
pepper increased significantly in infected stems of resistant cultivars with
increasing doses of metalaxyl.
When compared with inoculated, metalaxyl-untreated controls, metal-
axyl application before inoculation prevented disease development but did
not stimulate capsidiol production. However, larger quantities of capsidiol,
especially in susceptible Hanbyul, were produced in infected stems of pep-
per plants treated with metalaxyl at intervals during disease development
(Fig. 6). These results suggest that metalaxyl may cause pepper stems to
accelerate capsidiol production by affecting the host-pathogen interaction
after the pathogen spreadsinto stem tissue. Our earlier finding, that metal-
axyl is very effective for inhibition of mycelial growth of P. capsici (Sung
and Hwang, 1988), raises the possibility, although unproven, that fungal
cell wall polymers, which elicit capsidiol accumulation, may also be released
in infected stem tissue from mycelium of P. capsici by metalaxyl treatment,
as observed by Yoshikawa et al. (1981) in the P.m.f. sp. glycinea-soybean
system.
The stimulationof capsidiol productionby metalaxyl reachedits maxi-
mum at 2 days after inoculation, when typical symptoms beganto appear
in wound-inoculated stems (Fig. 7). These results provide evidence that
metalaxyl plays a critical role in inducing capsidiol production in the early
stages of host-pathogen interaction. In contrast, capsidiol production in
the inoculated, metalaxyl-untreatedcontrols increased, with a maximum at
3 days after inoculation when considerable disease progress occurred,thus
suggesting that Phytophthora infection may be important for stimulation
of capsidiol production in the later stages of the host-pathogen interaction.
The amount of capsidiol decreased to base levels after the maximum
accumulation. In particular, the effectiveness of metalaxyl treatment was

conspicuous in reducing the amount of capsidiol in infected stems. The
disappearance of capsidiol with time may not be due to degradation by P.
capsici (Ward and Stossl, 1972; Stossl et al., 1973) but, rather, due to the
metabolism of it bypeppertissue(Stossletal., 1977) or possibly to a
reaction with metalaxyl. Although unproven, it is also possible that some
unknown factors at a late stage of the host-pathogen interaction may de-
grade or otherwise modify capsidiol in infected, metalaxyl-treated stems.
v. EPILOG:
Yield losses of pepper plants caused by Phytophthora blight can be best
minimized bya combination of practices including crop rotation, host resis-
tance, and chemicalapplication(Bruin and Edgington, 1981; Davidse,
1981; Shew, 1985; Sung and Hwang, 1988; Kim et al., 1989). However, due
not only to economic burdenbut to the limit of cultivatedland, the growers
mainly rely on the use of resistant pepper cultivars and chemical control
against the Phytophthora disease.
P. capsici severely infects pepper plants only at later growth stages in
the fields. Therefore, a stable and durable type of resistance such as age-
related resistance would be valuable for use in breeding programs of pepper
cultivars. The expression of age-related resistance of pepper plants at later
developmental stages may be due to the physiological and morphological
changesinpepperstemsduringplantdevelopment (Jeun and Hwang,
1991). However, preformed antifungal substances, such as phenolics, may
not contribute to the expression of age-related resistance of pepper plants,
but rather phytoalexin “capsidiol,”which accumulates after exposure to P.
capsici, may playan important role inthis resistance.
In the present studies, it has been demonstrated that capsidiol concen-
trations in infected stems and roots of pepper plants were correlated with
the degree of resistance expressed to P. capsici. In comparable infected
organs, the resistant CV. Kingkun always contained more capsidiol than the
susceptible CV. Hanbyul. The stem and root of the two cultivars accumu-
lated more capsidiol and became increasingly resistant as plants matured.
These results suggestthat capsidiol production hasa role in increasingthe
.resistance of pepper plants with aging.
A combination of direct fungitoxicand indirect effects involving acti-
vation of natural defense mechanisms has been demonstrated in pepper
plants treated with the systemic fungicide metalaxyl (Borner et al., 1983;
Hwang et al., 1989). Metalaxyl treatment not only produced directeffects
in the fine structure of the fungal cellbut also effectedthe defense reaction
of pepper plants (Hwang et al., 1991). Due ‘to the intimate nature of the
520 Hwang
interaction between P. capsici and pepper plants, fungicide interference

with metabolism ofthe pathogen must inevitably lead to alteration of host
physiology which resultsfrom disturbances at the pathogen-host interface.
Enhanced capsidiol accumulation has also been reported in response to
inoculation withP.capsici in pepper plantstreated with metalaxyl.
In the present study, capsidiol production, associated with control of
Phytophthora blight of pepper plants by metalaxyl, was determined in the
infected stems of pepper cultivars resistant and susceptible to P. capsici.
Increasing concentrationsof metalaxyl in soil treatments a day before stem
inoculation with P. capsici gradually retarded the lesion development on
the stems of pepper plants but stimulated capsidiol production in the in-
fected stems. Metalaxyl treatments did not change the rapidly growing stem
lesions into the brownish necrotic ones of the hypersensitive reaction. In
particular, the accumulation of capsidiol by metalaxyltreatment was more
pronounced in the resistant CV. Kingkun than in the susceptible CV. Han-
byul. At metalaxyl concentrationsof 1 and 5 pg/ml, lesions appearedon the
stems 2 days after inoculation, witha maximum productionof capsidiol. As
the stem lesions developed and enlarged, the production of capsidiol in
metalaxyl-treated, infected stems declined to a final base level similar to
that in the infected control stems 5 days after inoculation. The metalaxyl
treatments after 1 and 2 days of inoculation produced more capsidiol the in
susceptible seedlingsthan before inoculation.
In conclusion, age-related resistance and metalaxyl treatment might
make an important contribution to the effective prevention ofPhytophth-
ora blight in pepper plants by directly affectingthe pathogen itself or indi-
rectly enhancing capsidiol accumulation in infected tissue. However,there
is a need for critical investigation of the natural sequence of events when
age-related resistance is expressed in pepper plants in relationto capsidiol
production. Further detailed researchon the relative rates of fungal expan-
sion and capsidiol production and degradation at the metalaxyl-treated
infection siteis also needed.
REFERENCES
Bailey, J. A. (1982). Mechanisms of phytoalexin accumulation.Phytoalexins (J. A.
Bailey andJ. W. Mansfield, eds.), Blackie, London, pp. 289-318.
Barksdale, T. H., Papavizas, G. S., and Johnston, S. A. (1984). Resistanceto foliar
blight and crown rotof pepper caused by Phytophthora capsici. PlantDis. 6 8
506-509.
Borner, H., Schatz, G., and Grisebach, H. (1983). Influence of the systemic fungi-
cide metalaxylon glyceollin accumulation in soybean infected withPhytophth-
ora megaspermaf.sp. glycinea. Physiol. Plant Pathol.23: 145-152.
Phyfophthora Infection in Pepper 521
Bruin, G. C. A., and Edgington, L. V. (1981). Adaptive resistance to Peronospor-

ales to metalaxyl. Can. J. Plant Pathol.3: 201-206.
Cartwright, D., Langcake, P., Pryce, J., and Leworthy, D. P. (1977). Chemical
activation of host defence mechanisms as a basis for crop protection. Nature
267: 511-513.
Davidse, L. C. (1981). Mechanism of action of metalaxyl in Phytophthora meg-
asperma f.sp. medicaginis. NetherlandsJ. Plant Pathol.87: 254-255.
Hahn, M.G., Bonhoff, A., and Grisebach, H. (1985). Quantitative location of
fected with Phytophthora megasperma f.sp.glycinea. Plant Physiol. 78 591-
601.
Hwang, B. K., and Heitefuss, R. (1982). Antifungal compounds in leaves of spring
barley in association withadult plant resistance to powdery mildew (Erysiphe
graminis fsp. hordei). Phytopathol. Z. 104: 272-278.
Hwang, B. K., and Kim, Y. J. (1990). Capsidiol production in pepper plants associ-
ated with age-related resistance to Phytophthora capsici. KoreanJ. Plant Pa-
thol. 6: 193-200.
Hwang, B. K., and Sung, N. K. (1989). Effect of metalaxylon capsidiol production
in stems of pepper plants infected with Phytophthora capsici. Plant Dis. 73:
748-75 1.
Hwang, B. K., Kim, W. B., and Kim,W.K. (1989). Ultrastructure at the host-
parasite interface of Phytophthora capsici in roots and stems of Capsicum
annuum. J. Phytopathol. 127: 305-315.
Hwang, B. K., Ebrahim-Nesbat, F., Ibenthal, W. D., and Heitefuss, R. (1990). An
ultrastructural study of the effect of metalaxyl on Phytophthora capsici in-
fected stems ofCapsicum annuum. Pest. Sci.29 151-162.
Hwang, B. K., Yoon, J. Y., Ibenthal, W. D., and Heitefuss, R. (1991). Soluble
proteins, esterases and superoxide dismutasein stem tissueof pepper plants in
relation to age-related resistanceto Phytophthora capsici.J. Phytopathol. 132:
129-138.
Jeun, Y. C., and Hwang, B. K. (1991). Carbohydrate, amino acid, phenolic and
mineral nutrient contents of pepper plants in relation to age-related resistance
to Phytophthora capsici.J. Phytopathol. 131: 40-52.
Jones, D. R.,Graham, W. G., and Ward, E. W. B. (1974). Ultrastructural changes
in pepper cells ina compatible interaction with Phytophthora capsici. Phytopa-
thology 64: 1084-1090.
Jones, D. R., Unwin,C. H., and Ward, E. W. B. (1975a). The significance of
capsidiol induction in pepper fruit during an incompatible interaction with
Phytophthora infestans. Phytopathology65: 1286-1288.
Jones, D. R., Unwin, C. H., and Ward, E. W. B. (1975b). Capsidiol induction in
pepper fruit during interactions with Phytophthora capsici and Monilia fmcti-
cola. Phytopathology 65:1417-1421.
Keen, N. T. (1971). Hydroxyphaseollin production by soybeans resistant and sus-
ceptible to Phytophthora megasperma var. sojae. Physiol. Plant Pathol. I :
265-275.
522 Hwang
Kim, Y. J., Hwang, B. K., and Park, K. W. (1989). Expression of age-related

resistance in pepper plants infected with Phytophthora capsici. PlantDis. 73:
745-747.
Lazarovits, G., and Ward,E. W. B. (1982). Relationship between localized glyceol-
lin accumulationand metalaxyl treatment in the control of Phytophthora rot in
soybean hypocotyls. Phytopathology 72:1217-1221.
Lazarovits, G., Stossl, and Ward, E. W. B. (1981). Age-related changes in specific-
ity and glyceollin production in hypocotyl reaction of soybeansto Phytophth-
ora megaspermavar. sojae. Phytopathology 71:94-97.
Leonian, L. H. (1922). Stem and fruit blight of peppers caused by Phytophthora
capsici sp. nov.Phytopathology 12: 401-408.
Molot, P. M., Mas, P., Conus, M., and Ferriere, H. (1981). Relations between
capsidiol concentration, speed of fungal invasion and level of induced resis-
tance in cultivars of pepper (Capsicum annuum) susceptible or resistant to
Phytophthora capsici. Physiol. Plant Pathol. 18: 379-389.
Paxton, J. D., and Chamberlain, D. W. (1969). Phytoalexin production and disease
resistance in soybeanas affected by age. Phytopathology 5 9 775-777.
Polach, F. J., and Webster, R. K. (1972). Identification of stems and inheritance of
pathogenicity inPhytophthora capsici. Phytopathology 6 2 20-26.
Reilly, J. J., and Klarman, W. L. (1972). The soybean phytoalexin, hydroxypha-
seollin, induced by fungicides. Phytopathology 62:1113-1 115.
Shew, H. D. (1985). Response of Phytophthoraparasitica var. nicotiana to metal-
axyl exposure.Plant Dis. 6 9 559-562.
Stossl, A., Unwin, C. H., and Ward, E. W.B. (1972). Postinfectional inhibitors
from plants. 1. Capsidiol, an antifungal compound from Capsicumfmtescens.
Phytopathol. 2. 74: 141-152.
Stossl, A., Unwin, C. H., and Ward, E. W.B. (1973). Postinfectional inhibitors
from plants: fungal oxidation of capsidiol in pepper fruit. Phytopathology 63:
1225-123 1.
Stossl, A., Robinson, J. R., Rock, G. L., and Ward, E. W. B. (1977). Metabolism
of capsidiolby sweet pepper tissue: some possible implicationsfor phytoalexin
studies. Phytopathology 67:64-66.
Sung, N. K., and Hwang, B. K. (1988). Comparative efficacy and in vitro activity
of metalaxyl and metalaxyl-copper oxychloride mixture for control of Phy-
tophthora blight of pepper plants.Korean J. Plant Pathol.4: 185-196.
Ward, E. W.B. (1976). Capsidiol production in pepperleavesinincompatible
interaction with fungi.Phytopathology 6 6 175-176.
Ward, E. W. B., and Stossl, A. (1972). Postinfectional inhibitors from plants. 111.
Detoxification of capsidiol, an antifungal compound from peppers. Phytopa-
thology 6 2 1186-1 187.
Ward, E. W.B., Unwin, C. H., and Stossl, A. (1973). Postinfectional inhibitors
from plants. VII. Tolerance of capsidiol by fungal pathogens of pepperfruit.
Cm. J. Bot. 51: 2327-2332.
Ward, E. W. B., Lazarovits, G., Stossl, P., Barrie, S. D., and Unwin, C.H. (1980).
Glyceollin production associated withcontrol of Phytophthora rot of soybeans
by the systemic fungicide, metalaxyl.Phytopathology 7 0 738-740.
Pbytopbfbora Infection in Pepper 523
Weber, G.F. (1932). Blight of peppers in Florida caused byPhytophthora capsici.

Phytopathology 22: 775-780.
Yoshikawa, M., Matama, M., and Masago, H.(1981). Release of a soluble phyto-
alexin elicitor from mycelialwalls of Phytophthora megasperma var. sojae by
soybean tissues.Plant Physiol.67: 1032-1035.
23
Induced Chemical Resistance in
Sesamum indicum Against
Alternaria sesami
R. K. S. Chauhan
Jiwaji University, Gwalior, Madhya Pradesh, India
B. M. Kulshrestha
K. R. G. College, Gwalior, Madhya Pradesh, India
1. INTRODUCTION
This project was undertaken to investigate whether the fruits of Sesamum
indicum L. could produce any phytoalexin in response to infection by Al-
ternaria sesami (Kawamura) Mohanty and Beraha. S. indicum (Pedalia-
ceae), commonly known as til (sesame), is one ofimportant
the oil-yielding
plants. The plant is susceptibleto various diseases such asAlternaria blight
caused by A . sesami, Cercospora leaf spot caused by C.sesamicola, Phy-
tophthora blight caused by P . parasitica var. sesami, and root rot caused
by Macrophomina phaseolina.However, noinformation is availableso far
regarding postinfectionalantifungal compounds in sesame plant and their
role in disease resistance.
II. METHODOLOGY
The seeds of S. indicum were procured from National Seed Corp. Alfer-
naria sesami (Kawamura) Mohantyand Beraha was isolated from infected
parts of S. indicum. Test organisms, Aspergillus niger (Van.) Tieghem,
Cladosporiumcladosporioides @res.)DeVries, Colletotrichum capsici
(Syd.) Butler and Bisby, Curvularia lunata (Wakker) Boedijn, Fusarium
solani (Mart.)Sacc. Helminthosporiumtetramera McKinney, Mycos-
phaerella rabiei Kovachevsky, and Rhizopus stolonifer (Ehrenb. Ex Fr.)
Lind were obtained from the Indian Type Culture Collection (IARI, Delhi).
Phytoalexin was detected fromfruits following the drop diffusate tech-
nique. Thoroughly washed fruits were split open longitudinally and seeds
were removed. Thesefruit halves were surface-sterilized with sodium hypo-
525
Kulshrestha
526 and Chauhan
chloride solution and washed twice with sterilized distilled water. Fruits
wereplaced in a humid chamber, inoculated aseptically with the spore
suspension of A . sesami (1 x lo5spores/ml), and incubated at 28 2OC.*
The drops were collected centrifugedand the supernatant is collected.
The chemicals used for induction of phytoalexin, i.e., aureofungin,
bavistine, 2,4-dinitrophenol, dithan M-45, mercuric chloride, sulfex, and
thiobendezole are applied to fruit cavities as solutions of different ppm
(200, 600, 1OOO). After incubation for 48 hr diffusates were collected, cen-
trifuged, and the supernatant was tested for phytoalexins (Kulshresthaand
Chauhan, 1988).
111. SESAME PHYTOALEXINS

A.
Bioassay
Spore germination, germ tube growth, and radial mycelial growth bioassays
wereused in the present investigation (Kulshrestha, 1982; Chauhan and
Kulshrestha, 1984). In different bioassays, Alternaria sesami, Aspergillus
niger,Cladosporiumcladosporioides,Colletotrichumcapsici, Curvularia
lunata, Fusarium solani, Helminthosporium tetramera, and Mycosphaer-
ella rabiei were used as the test organisms. In spore germination bioassay,
the effect of active principle from the host plant on spore germination of
test organisms is observed and is done by estimating the inhibition of spore
germination by slide germination method.To avoid influence by nutrients
present in the spores during bioassay, germ tube growth bioassay is pre-
ferred (Muller, 1958). For this purpose, spores of different test organisms
were kept in sterilized distilled water for the time period just enough to
initiate germination. Water then removed by centrifugation and germinated
spores were taken in the diffusate in a cavity slide and the effect on germ
tube growth was observed. Radial mycelial growth bioassay consisted of
measuring the diameters of colonies of test organisms grown on phyto-
alexin-amended media and their comparison with those of control ones
(Langcake and Pryce, 1976). All the test organisms werefound sensitive to
varying degrees (Table 1).
It is clearthat phytoalexins from sesame, besides reducing spore germi-
nation, frequently retard germ tube growth and further the growth of the
mycelium, and thus make them less effective in causing infection. There-
fore, phytoalexins appear to discourage infection by reducing germination
of spores and subsequent growthof germ tubes.
B. Effect of Incubation Time on the Production
of Sesame Phytoalexin
To determine the effect of incubation time on phytoalexin productionS. in
indicum, diffusates were collected after 4, 8, 16, 24, 32, 48, and 72 hr of
S. indicum Resistance to A. sesami 527
Table 1 Effect of Phytoalexin from Fruits of S. indicum on Spore Germination,

Germ Tube Growth, and Radial Mycelial Growth of Fungi
Vo Inhibition over control

Radial
mycelial
tube
Germ
Spore
Fungus growth'
germination'
growthb
Alternaria sesami 37 63 48
Aspergillus niger 69 69 93
Cladosporium cladosporioides 41 50 70
78
Colletotrichum capsici 88 37
Curvularia lunata 77 69 70
Fusarium solani 65 66 56
Helminthosporium tetramera 100 88 88
Mycosphaerella rabiei 78 50 67
'Mean of 100 observations taken after8 hr of incubation at28 2OC.

bMeanof 30 germ tubes observed after 24 hr.
8 days of incubation.
'Mean of three replicates observed after
inoculation. Phytoalexin activitywas detected in the droplets only after 8

hr and concentration increased gradually up to 32 hr, reaching a maximum
at 48 hr. There was no further increase or decrease in concentrationof the
phytoalexin up to 72 hr (Table2).
C. Phytoalexin Production by Different Plant Parts

Fruit, leaf, and stem of sesame plant were taken to investigate the produc-
tion of phytoalexin in theseparts after inoculation with fungi. Phytoalexin
Table 2 Time Course of Phytoalexin

Production
Vo Inhibition in the
spore germinationof
Incubation (hr) A . sesami
4 0
8 0
16 20
24 30
32 45
48 60
72
Kulshrestha
528 and Chauhan
was detected infruits, leaves, and stem of sesameplant although concentra-

tion was variable. Leaves and fruits produce more phytoalexin than stem
(Table 3).
D. Studies on the Chemical Nature of Phytoalexin

Three solvent systems were employedfor the separation of sesame phyto-
alexin on thin layer chromatograms, namely, chloroform-methanol (100 :
2), benzene-acetic acid-water (125 :72 :3), and butanol-acetic acid-water
(4 : 1 :5). Fluorescence under UV light (254 nm), Rf values, reaction with
1% ferric chloride and diazotized sulfanilic acid werethe criteria used for
chemical characterization of the phytoalexin (Table 4). Fluorescing spot
was also testedfor antifungal activity. The phytoalexin of sesame plantwas
found to be phenolic in nature. It completely inhibited spore germination
of A. sesami.
E. Chemical Induction of Phytoalexin

Aureofungin, bavistin, 2,4”dinitrophenol,dithan “ 4 5 , mercuric chloride,
sulfex, and thiobendazole were the chemicalstested for their ability to
induce phytoalexin production in fruits of S. indicum. 2,CDinitrophenol
and mercuric chloride induced phytoalexin production in higher concentra-
tions (Table5).
IV. EPILOG
The phytoalexin inS. indicum is found to be phenolic innature. It is highly
inhibitory to the spore germination, germ
tube growth, and mycelial growth
Table 3 Phytoalexin production in fruits, leaves and stem

of S. indicum in
response to inoculation withA. sesami
Vo Inhibition inthe spore germination over

control
Fungus
Alternaria sesami 60 90 30
Colletotrichum capsici 88 80 20
Curvularia lunata 60 69 35
Fusarium solani 65 84 38
Helminthosporium tetramera 100 13 40
Mycosphaerella rabiei l8 61 30
529
530 Chauhan and Kulshrestha
Table 5 Chemical Induction of Phytoalexin in Fruits of S. indicum

~ ~~~ ~ ~~ ~
Phytoalexin production
RI = 0.16
Concentration
butanol-acetic
acid-water
Chemical @Pm) (4 : 1 :5 )
Aureo fungin 200

600
lo00
Bavistine 200
600
lo00
2,CDiNtrophenol 200
600
lo00
Dithan M45 200
600
lo00
Mercuric chloride 200
600
lo00
Sulfex 200
600
lo00
Thiobendazole 200
600
loo0
Sterilized distilled water -
(control)
- ,Absence; + ,presence in traces; + f ,presence in moderate concentrations;+ + + ,pres-
ence in higher concentrations.
of some fungi tested. Phytoalexin was detected in the drops only after8 hr
of inoculation and the concentration increased up to 48 hr. A. sesami
induced phytoalexin in fruits, leaves, and stem of sesame plant. Certain
chemicals also could induce the phytoalexin. Further characterization of
this compound, elucidation of its biosynthetic pathway, mechanism of ac-
tion, and development of analogs for use in the crop protection are the
future prospects.
S. indicum Resistance to A. sesami 531
ACKNOWLEDGMENTS
The authors thank Dr. S. Chauhan, Professor, School of Studies in Botany,
Jiwaji University, Gwalior for her keen interest in the preparation of this
manuscript.
REFERENCES
Chauhan, R. K. S., and Kulshrestha, B. M. (1984). Production of phytoalexin in
Sesamum indicum against Alternaria sesami.Indian Phytopathol. 37: 482-485.
Kulshrestha, B. M. (1982). Studies on induced chemical resistancefactors in certain
plant diseases. PhD Thesis, Jiwaji University, Gwalior,India, pp. 255.
Kulshrestha, B. M., and Chauhan, R. K. S. (1988). Physicochemical nature of the
phytoalexin producedby fruits of Sesamum indicum in response to inoculation
with Alternaria sesami and its chemicalinduction. Indian Phytopathol. 41: 92-
95.
Langcake, P., and Pryce, R. J. (1916). The production of resveraterol by Vitis
vinifera and other members ofthe Vitaceae as a response to infection or injury.
Muller, K.0.(1958). Studies on phytoalexins. 111. The formation and the immuno-
logical significance of phytoalexin producedby Phaseolus vulgarisin response
to infection with Sclerotinia ficticola and Phytophthora infestans. Aust. J.
Biol. Sci. 11: 215-300.
24
Phytoalexins andOther Postinfectional
Compounds of Some Economically
Important Plantsof India
M. Daniel
The Maharaja Sayajirao Universityof Baroda, Vadodara, Gujarat, India
1. INTRODUCTION
Studies on phytoalexins have taken rapid strides thein recent past, but still
the available data are grossly inadequate to draw any valid conclusion on
the distribution, metabolism, and role of these compounds in higher plants.
Out of the 250,000 or more angiosperms, hardly a couple of hundred spe-
cies have been analyzedfor their phytoalexins. Exceptfor the Fabaceae, no
other family has been systematically screened for their defense reactions.
Almost all the plants studied so far are the cultivated species. Nothing is
known onthe phytoalexins ofwild plants, weeds, or tree species.
Study of phytoalexins in isolation, away from other disease resistance
factors, is another reason for the ambiguity existing on the concept of
phytoalexins. All the known phytoalexins and the preformed toxic sub-
stances (inhibitins) are secondary metabolites. Though the secondary me-
tabolites in plants once were considered as waste products/excretory sub-
stances, the overwhelming evidence accumulated so far ascertains that the
plants owe their survival to these compounds.It is these compounds which
act as UV screens, deter herbivores, repel insects,and ward off pathogens.
Some of them are responsible for effective pollination, dispersalof seeds,
and dormancy. It is explicitthat during evolution plants experimented with
different defensive chemicalsand therefore the various taxa elaborated dis-
tinct classes of compounds. Concurrent with the modifications to perfect
the defense mechanisms were the attempts of the pathogens to detoxify
these compounds. The struggle between plants and microbes/herbivores to
outwit each other is the reason for the plethora of secondary metabolites
found in nature. Most plants also have to face the threats of animals and
therefore are compelled to produce the taste modifiers (bitter substances)
along with antimicrobials. In this context a compound which possessesthe
533
S34 Daniel
ability to alter taste as well as inhibit fungal invasion would have been the
compounds of choice for the plant. A number of secondary metabolites
such as phenolics, tannins, flavonoids, sesquiterpenes, saponins, and iri-
doids fit well into this prescription.
Phytoalexins are produced as a result of triggering the biosynthetic
pathways by the microbial invasion. It is therefore explicit that the cell
should already possess the needed machinery for doing so. The switching
on of a defense reaction meansthat the expression ofthe gene concerned is
kept suppressed under normal conditions and this repression is removed
once the signal from the pathogen is received. This gene would have formed
de novo in evolutionor would have beena modified form of a gene inher-
ited from the ancestors. This also indicates that the precursors of phyto-
alexin would be present in the plant at any giveninstant and in this context
the study of preformed substances assumes greater significance.
Our interest in phytoalexinswas purely academic, i.e., to find out the
defense reactionsof plants. Therefore we based our studies on tree species.
Trees are different from herbs in that they store more compounds for a
comparatively longer time. The life span of leaves also is longer in trees
than in herbs. The selection of plants was random. Later on some of the
herbaceous perennials also were included. The physiological similarity of
Cuscuta to fungal pathogens promptedus to include this angiospermpara-
site in our study. The studies on different plants are at various stages of
progress. All the results available with us are presented here, which collec-
tively may shed some light the on defense reactionsof plants.
II. TREES AND SHRUBS

A. Tectona grandis L.
T. grandis (teak; Verbenaceae) is one the of most important timber treesof
India. The wood is used in high-class furniture, boat and ship building,
houses, bridges, railway sleepers, etc. The bark of the tree is medicinally
important as an astringent and the wood is used as an anthelmintic. The
flowers and seeds are known diuretics.
Powdery mildew caused by Uncinula tectonae Salm. (Karadge et al.,
1980), rust due to Chaconia tectonae Ramakr. K. and Ramakr. T.S. (Ra-
makrishnan and Ramakrishnan, 1949), and leaf spots due to Cercospora
tectonae Stevens (Tirumalachar and Chupp, 1948) and Sphaceloma tecto-
nae Wani and Tirum. (Wani and Tirumalachar, 1969) are the leaf diseases
reported from Tectona. Wood rot caused by Ganoderma applanatumPat.
and Daedaleaflavida Lev. are the diseases seen in the wood.
In Gujarat, leaf spot diseases appear during the months of July to
Phytoalexins of Economically
Plants
Important 535
September. The brownish spotslater enlarge to blackish brown patches.In

extreme cases the whole leaf turns brownish black.Curvularia clavataJain,
the fungus isolated from the infected areas, was found to be the causative
organism of this disease.
Chemical analysis of both healthy and infected leaves was carried out
by standard procedures. The identities of flavonoids were confirmed by
spectral analysis and cochromatography with authentic samples. Phenolic
acids were identified on the basis of their spectral
data, color reactionswith
diazotized p-nitraniline and sulfanilic acid, chromatographic properties,
and cochromatography withstandard compounds.
The healthy leaves of Tectona contained two flavones: 4' -0Me api-
genin and luteolin. The phenolic acids present were syringic, sinapic, vanil-
lic, melilotic, and gentisic acids. The other constituents of the leaves were
quinones (lepachol and tectoquinone), proanthocyanidins, iridoids, alka-
loids, and tannins. The infected leaves did not contain any flavone but a
flavonol 3 ',4 '-dimethoxyquercetin insteadand phenolic acids such as feru.
lic, vanillic, melilotic,and gentisic acids. They contained the same quinones
as of the healthy leaves as well as proanthocyanidins, iridoids, alkaloids,
and tannins.
There was no significant chemical difference between the diffusates of
control and treated leaves when the healthy leaves were treated with the
spore suspension of C. clavata. But when the leaves were treated with a
nonpathogen, Fusarium solani, the diffusate contained p-hydroxybenzoic
acid.
Mycelial growth assay ofF. solani with p-hydroxybenzoic acid showed
that this fungus is strongly inhibitedat all concentrations. Maximum inhibi-
tion was noted at lo00 ppm. Spore germinationand germ tube elongation
of F. solani were also inhibited byp-hydroxybenzoic acidat 300, 500, and
lo00 ppm concentrations. Maximums of85% inhibition of spore germina-
tion and 64% inhibition of germtube elongation were noted at 1000 ppm.
Similarly, the spore germination, germ tube elongation, and mycelial
growth of C.clavata were inhibited byp-hydroxybenzoic acid at all concen-
trations. Maximum inhibition of mycelial growth was noted at loo0 ppm.
At this concentration the inhibition of spore germination was 58% and
inhibition ofgerm tube elongation63% (Abraham, 1989).
The replacement of flavones by the flavonol 3 ',4 '-diOMe quercetin in
diseased leaves appears to be highly significant. This involves a change in
the biosynthetic pathway in which flavonols with more potent phenolic
hydroxy groups are produced. Production of ferulic acid in placeof syrin-
gic and sinapic acids alsois noteworthy. Ferulic acid is known for its activ-
ity against Poria weirii (Li et al., 1972) and Fusicoccum amygdali (Borys
and Childers, 1964). It is found to inactivate tungrovirus invivo (Sridhar et
536 Daniel
al., 1979) and impart resistance in wheat varieties (Chigrin and Romru,
1969).
The production of p-hydroxybenzoic acid as a response to the non-
pathogen is interesting. This supports the view expressed by Debnam and
Smith (1976) that the non-pathogen-infested tissue produces more phyto-
alexins than the tissue infected by pathogen. The strong inhibitionof myce-
lial growth, spore germination, and germ tube growth of F. solani and
C. clavata proves the antifungal nature ofp-hydroxybenzoicacid.The
immediate production of this compoundand its marked antifungal activity
leads to the conclusion that p-hydroxybenzoic acid is indeed a phytoalexin.
This phenolic acid had been reported as a phytoalexin in Nectria-infected
apples (Swinburne, 1975).
Though p-hydroxybenzoic acid or its derivatives occur in plants as a
component of lignin, its role in imparting resistanceto the plant is proved
by the studies on carrot slices using Botrytis cinerea. Here the infection
leads to the production of inhibitors such as 6-methoxymellein, p-hydroxyben-
zoic acid, and falcarinol (Harding and Heale, 1981). The increase in p-
hydroxybenzylgroupsalsoleads to the increased production of lignin,
which either directly or indirectly acts as a potential barrier to infection
(Henderson and Friend, 1979).
B. Cassia fistula L.
C. fistula (Caesalpiniaceae) isa small to medium-sized deciduoustree com-
monly recommended in social forestry programs. The timber is used for
making carts and agricultural implements. The leaves find application an as
emollient for dropsy and rheumatic pains. The pulp from the podsis offi-
cially included inthe British Pharmacopeia asan ingredient of sennaand is
used as a mild, pleasant,and safe purgativeeven for children and expectant
mothers. The flowersof C. fistula are sometimes used asa food by natives.
Its root is a tonic, febrifuge,and a strong purgative.
The common disease observed in C. fistula is powdery mildew caused
by Oidium sp. (Salam and Rao, 1958) or Phyllactinia corylea (Pers.) Karst
(Yadav, 1963). Otherfungireported from thisplant are Pestalotiopsis
adusta (Ell. Ev.) Steyart (Kanujia and Singh, 1978) and Phoma comlanata
Tode ex. Fr. (Rajak and Rai, 1982).
The leaf spot disease of C. fistula is found in the months of January
and February after which period the leaves are shed. The spots are small,
irregular, and black. Aspergillus niger Van Tiegh. was the organism iso-
lated from diseased areas. During pathogenicity tests the lesions begin to
appear 6-8 days after inoculation withthe pathogen.
The healthy leaves from the uninfected trees were found to contain
Plants
Important 537
threemethoxyflavonols:4’-OMequercetin,4’-OMekaempferol, and
7,3 ’4”triOMe quercetin; phenolic acids such as syringic and p-coumaric
acids; proanthocyanidins; tannins and alkaloids. The diseased leaves con-
tained the hydroxyflavonolsquercetin and kaempferol,syringic and o-
coumaric acids, alkaloids, proanthocyanidins, and tannins. The amountof.
flavonols in infected leaveswere much higher than those of healthy leaves
(Abraham and Daniel, 1988a).
The healthy leaves when exposed to the pathogen A. niger produced
quercetin and kaempferol whilethe control contained the methylated deriv-
atives of these compounds.The nonpathogen F. solani also evoked similar
response when inoculatedon healthy leaves.
Quercetin, oneof the flavonols induced by the fungus, was assayed for
its antifungal activity. This compound inhibited the spore germination of
A. niger by 49% and germ tube elongation by 35% at 500 ppm. Mycelial
growth also was inhibited at 100 and 500 ppm concentrations of quercetin
(Abraham, 1989).
Evidently the process triggered bythe infection of C.fistula leaves by
A. niger is the demethylation of 4’-OMe kaempferol, 4’-OMe quercetin,
and 7,3 ’,4’-triOMe quercetin to theirhydroxylatedparentcompounds
quercetin and kaempferol. The hydroxylated compounds are more toxic
than their relatively unreactive methoxy derivatives.
The antifungal properties of quercetin and kaempferol were reported
earlier by Sporoston (1957). Quercetin inhibited the growth of fungi such as
Daedalea quercina and Fomes annosus (Alcubilla-Martin, 1970; Walchii
and Scheck, 1976). Both quercetin and kaempferol were found to be absent
from Populus maximowiczii, P. laurifolia, and other hybrids which were
susceptible to Dothichiza populea, while they were located inthe resistant
varieties of P. nigra var. italica and its hybrids.
The induction of o-coumaric acid in place of p-coumaric acid by the
pathogen and nonpathogen is interesting. Though p-coumaric acidis also
known to possess antifungal activity (Trappe et al., 1973), the formation of
o-coumaric acid had an added significance in that this phenolic acid pro-
duces coumarinby lactonization. Coumarinis a more potent antimicrobial
agent(Berkenkemp,1971).o-Coumaricacidhasbeenreported to bea
systemic fungicide (Gangulee and Kar, 1985).
C. Morinda tornentosa Heyne

M. tomentosa (Rubiaceae) is a small tree commonly found in the deciduous
forests of Gujarat and other central andsouthern parts of India. The heart-
wood of this tree is durable, used for furnitures. A red dye “Suranj” is
obtained from the roots of M. tomentosa. The leaves are used as a tonic
538 Daniel
and febrifuge. Its fruit is a deobstruent and is used against dysentery and
asthma.
The diseases reportedon M . tomentosa leaves are leaf spots caused by
Botryodiplodiatheobromae Pat. (Shreemali and Bilgrami,1968), Cer-
cospora morindae Syd, Gleosporum morindae Payak & Tirum. (Payak,
1953), and Macrophoma morindae Ramakr. & Sund. (Ramakrishnan and
Sundaram, 1954).
The leaf spot diseaseof Morinda is noted inthe months of October to
December. The symptoms are blackish brown irregular patches on both
the surfaces and at times the whole apical portion of the leaf turns dark.
Colletotrichium gleosporoides Penzig and Sac. is the fungus isolated and
found to be pathogenic.
The results obtained from M . tomentosa were similar to those of C.
fistula. The healthy leaves contained the two methoxyflavonols 4’-OMe
kaempferol and 3 ’,4’-diOMe quercetin,and the four phenolic acids vanil-
lic, syringic, gentisic, and ferulic. The infected leaves contained the hy-
droxyflavonols kaempferol and quercetin along with four phenolic acids
found in healthy leaves. The diffusates of both the pathogen- and non-
pathogen-treated leaves contained quercetin and kaempferol (Abraham and
Daniel, 1988a).
Quercetin at 500 ppm inhibited the spore germination of C.gleospor-
oides to 32% and germ tube elongation up to 53%. The mycelial growth of
the fungus was inhibited at 100 and 500 ppm (Abraham, 1989).
D. Eucalyptus globulus Labill.

E. globulus (Myrtaceae) is a large tree extensively planted under aforesta-
tion programs throughout India. Its wood is useful in a variety of ways.
The bark contains tannins and the leaves yield a volatile oil rich in 1,8-
cineole, which is used asa mosquito repellantand local antiseptic.
A number of antifungal compounds are reported from various species
of Eucalyptus. DaCosta and Rudman (1958) found that the outer heart-
wood of E. microcorys to be extremely resistant to decay-causing organisms
such as Coniophora carebella, Coriolus versicolor, and Fomes d u m . But
methanol-extracted outer heartwood was promptly decayedby these fungi,
and when the methanol extractwas added to the heartwood of E. regnans
susceptible to thefungus it conferred protection against decay. Inhibition
of wood-rotting fungiby stilbenes and other polyphenols ofE. sideroxylon
has also been reported (Hartand Willis, 1974). Antimicrobial compounds
are reported from E. globulus (Osborne and Thrower, 1964) and E. trifora
(Egawa et al., 1977) also.
The common diseases found in E. globulus are leaf spot caused by
Plants
Important 539
Pestalotiopsis funerea (Desm.) Stey. (Bilgrami, 1963) and root rot caused
by Ganoderma lucidumKarst. (Bakshi,1974).
E. globulus grown in Baroda develops leaf spot disease inthe months
of December and January. This disease is characterized by small, round,
black spots scattered on the upper surface of the leaf. Alternaria alternata
(Fr.) Keissler is isolated from the lesions and its pathogenicityis proved in
the laboratory.
On chemical analysis,both the diseasedand the noninfected leaves were
found to contain the same compounds, i.e., 3’-OMe quercetin, 4’-OMe
kaempferol, and vanillic, syringic,and p-hydroxybenzoic acids.The proan-
thocyanidins, iridoids,and alkaloids were also the same inboth healthy and
infected leaves.
When inoculated with the spores of the pathogen, the healthy leaves
produced an additional compound fluorescing blue in UV light (long wave-
length) and having an absorption maxima of 276, 285, 296, 330, and 341
nm in methanol. The chromatographic properties and the spectral data
suggested this compound to be a coumarin. However, the nonpathogen
used (F.solani) failed to induce any response in the healthy leaves. The
diffusate from the treated leaves inhibitedthe mycelial growth ofthe patho-
gen (up to 54%) at 10 m1 dilution. The inhibitionof spore germinationand
germ tube elongation of A . alternata were 65% and 71070, respectively, at
10 m1 dilution of the diffusate. This provesthe blue fluorescing compound
induced by A . alternata to be a phytoalexin (Abraham, 1989).
E. Syzygium cumini (L.) Skeels

S. cumini (Myrtaceae) isan evergreen tree found widely cultivatedthrough-
out India for its edible fruits. The fruit jambolan (Java plum) is used as
fresh fruit as well as for jellies, wines,and cordials. The wood of S. cumini
is hard and durable, used for agricultural implementsand fuel. The bark of
the tree is astringent dueto tannins and is used against bronchitis,asthma,
ulcers, and dysentery. Thefruit juice and seed extract are hypoglycemic.
Tannins from S. cumini are found to inhibit the growth of fungi such
as Colletotrichium falcatum and Pyricularia oryzae (Janardhanan et al.,
1963).
The diseases reported in S. cumini are leaf spot caused by Elsinoe
kamatii Tewari (Tewari, 1968),Tripospermum juglandis(Thumen) Hughes
(Nath and Bhargawa, 1976), and Phyllosticta eugeniae Thirum. (Kamat
and Kalani, 1964).
The leaf spot disease ofSyzygium was found to occur in November and
December. The diseased areas appear as small yellowish spots toward the
base and tip of the leaf. The spots then grow in size and cover a large
540 Daniel
portion of the leaf blade, which becomes greyish yellow. The lesions are
produced on both the surfaces of leaves but are more prominent on the
upper surfaces. Aspergillus niger van Tieghum, the fungus isolated from
the diseased portions ofthe leaf, was found to be the causative organism.
The flavonoid chemistry of the diseased leaves remained the same as
that of the healthy leaves in containing 3 '-OMe quercetin and myricetin.
The healthy leaves contained vanillic, gallic, p-hydroxybenzoic, syringic,
and p-coumaric acids, while in infected leaves, the p-coumaric acid is re-
placed by gentisic acid.
When phytoalexin was induced in the leaves of S. cumini with spores
of the pathogen, the leachates contained a visibly colored pinkish brown
compound (brown in UV, turning to pink with sodiumcarbonate spray and
&,/MeOH 274, 318, 332, and 400 nm) was located in the diffusates. The
color reactions and spectra denote this compound to be a benzoquinone.
Studies onthe characterization of this compoundis in progress. There were
no qualitative or quantitative differences in the diffusates when the leaves
were treated with sporesof the nonpathogen F. solani. Facilitated diffusion
also showedthe same result.
The diffusate from the pathogen-treated leaves containing the phyto-
alexin inhibitedthe mycelial growth ofthe pathogen at l and 5 m1 dilution.
A maximum of 35% inhibition of the mycelial growthwas observed at 5 m1
dilution of the diffusate. Spore germination and germ tube elongation of
A. niger also was inhibited (87% and 48070, respectively) at 5 m1 dilution
(Abraham, 1989).
F. Mangifera indica L.
M. indica (Anacardiaceae) is the source of mango fruits. It is a tall ever-
green tree cultivated not only for fruits but also for its timberand medicinal
values. The wood is used for plywood and its bark for tanning. The ripe
fruit is considered diuretic and laxative. The seed kernel is a medicine for
asthma and diarrhea. Baked and sugared pulp of unripe fruits is recom-
mended for cholera and plague patients.
Different parts of mango tree are known to suffer from a number of
diseases. Apart from the fruit the leaves and inflorescences ofthe tree also
are attacked by a number of pathogens. The important diseases are 1.) leaf
spot caused by species ofAsterofibertina, Aureobasidium, Cercospora, Cil-
iochoreffa, Coffetotrichium, Curvufaria, Diplodia, etc. (Mukherji and Bhasin,
1986), 2.) anthracnose by Chaetomium mangiferae Batista and Lima, and
3). powderymildewbyErysiphecichoracearumDC(Uppal and Patel,
1945).
The leaf spot disease found in Baroda is taken up for the study of
Plants
Important 541
phytoalexins. This disease occurs during January to March. The lesions are
small in size, yellow-brown in color, and are seen scattered in the upper
surface only.
Aspergillus niger van Tieghum was the fungus isolated from the dis-
eased areas of leaves. The pathogenicity of this fungus is proved by tests.
The diseased leaves contained the same components, i.e., saponins,
flavonols such as quercetin and quercetagetin, the xanthone mangiferin,
and phenolic acids, vanillic, syringic, and p-hydroxybenzoic acids. There
was no qualitative change between the diffusates of control and leaves
treated with the pathogen. However, when the nonpathogenF. solani was
used to examine the phytoalexin response, mangiferin was seen leaching
out into the diffusate. This indicates the possible role of mangiferin as a
phytoalexin. The antifungal activity of mangiferin against Fusarium has
already been proved (Ghosal et al., 1977). It is found that the walls of F.
oxysporum hyphae suspended in mangiferin were lysed within 72 hr. The
mycelium turned black and the protoplasts contracted, got detached from
the cell wall and were seen collected at one comer or the center of the
cell. The seeds of safflower treated with mangiferin was resistant to F.
oxysporum. In Mangifera, the content of mangiferin in flowers or shoots
were very high when infected F. by rnoniliformae whereas the healthy flow-
ers and shoots containedthis compound in traces.
The increased productionof mangiferin in response to F. solanimay be
the reason why this fungus is unable to infect the mango tree. It is also
evident that mangiferin is less toxicto A . niger (Abraham, 1989).
G. Anogeissus hfifolia Wall.

Anogeissus latifolia(Combretaceae) isa tall, magnificent tree found in the
deciduous forests of western and southern India. This is one of the most
important timber trees of Gujarat. It yields a gum, “gum ghatti,” which is
usedas an adhesive, in calicoprinting, and in medicine. The leaves and
twigs ofAnogeissus, being tanniniferous, are used in the leather industry.
The fungus causing leaf spot A in. latifolia is reported to be Pestaloti-
opsis versicolor (speg.) Steyart (Agarwal and Ganguli, 1959). The disease,
seen in Baroda, involves the appearanceof small brownish spots in leaves,
gradually spreadingand ultimately dryingup a large portion of the lamina.
Sometimesthe entire leafis dried up.Alternaria alternata(Fr.) Kiessler was
found to cause this disease.
The flavonols 3’4’-diOMe quercetin and 7,3’4’-triOMe quercetin were
found in both infected and healthy leaves of Anogeissus. But the infected
leaves also contained vitexin, a glycoflavone. The concentration of flavo-
noids was much higher in diseased leaves. The other phytochemicals, i.e.,
542 Daniel
proanthocyanidins, tannins, and phenolic acids (vanillicand ferulic acids),

also were common to both noninfected and infected leaves (Abraham and
Daniel, 1990).
The diffusates from the leaves treated with spores ofthe pathogen and
nonpathogen (F. solani) showed no qualitative differences from those of
the control.
H. Madhuca indica Cmel.

M . indica (Sapotaceae), the source of mowra butter, is a large deciduous
tree abundant in the forests of Gujarat and adjoining areas. Its wood is
used for furnitures and construction work,and the fleshy sweet flowersare
edible. A local liquor, “mowdi,” is brewed from the flowers. The seed fat,
mowra fat, is edible and used extensively for the making of margarines,
chocolates, and soaps. This oil has medicinal importance due to its emol-
lient properties and is used against skin diseases, rheumatism, piles, and
hemorrhoids.
M. indica is infected by Cylindrocladium scoparium Morg (leaf spot;
Nirwan and Singh, 1967), Scopella echinulata (Niessl.) Mains (leaf rust;
Patil and Tirumalachar, 1971), and Sphacelomamadhucae Wani and
Tirum. (leaf spot, Wani and Tirumalachar,1971).
The spots observed here are small blackish ‘brown patches growingto
bigger patches coveringa major portion of the leaf blade.The pathogenic-
ity of Colletotrichium gleosporoides Penziz and Sacc., isolated from the
lesions, was proved in the laboratory.
The healthy leaves of Madhuca contained a single flavonol (4‘-OMe
myricetin), phenolic acids such as p-hydroxybenzoic and vanillic acids, iri-
doids, and saponins. The infected leaves contained all these phytochemicals
except for 4’-OMe myricetin which is replaced by myricetin. The diffusates
from the leaves treated with spores of the pathogen and nonpathogen (F.
solani) were alike in chemical constitution. However, when the pathogen-
inoculated leaves were subjected to facilitated diffusion, the extract con-
tained myricetin (Abraham,1989).
1. Heterophragmaadenophyllum Seem.
H. adenophyllum (Bignoniaceae) isa tall tree cultivated for its timber.The
important diseases ofthis tree are 1.)powdery mildew caused by Acrospor-
ium sp. (Patwardhan, 1966), 2.) leaf rust byPhragmidiella heterophragmae
Tirum. and Mund. (Ling, 1951) and Phyllospella stakmanii Sathe (Sathe,
1965), and 3). leaf spot by Sphaceloma heterophragmae Wani and Tirum.
(Wani and Tirumalachar, 1969).
Plants
Important 543
The leafspot disease prevalent in Baroda appears as brown spots which

spread to the entire lamina of leaflet whichin turn becomes cream-colored.
The culture of diseased areas of leaf yielded Botryodiploidia theobromae
Pat. The tests proved this organism to be the pathogen.
On analysis, the healthy leaves yielded 4’-OMe apigenin, benzoqui-
nones, procyanidin, and syringic acid. The diseased leaves contained api-
genin (in place of its methoxy derivative), the same quinones, procyanidin,
and syringic acid. Drop diffusates with the pathogen contained apigenin
(Chitra, unpublished). Further details are being investigated.
J. Spafhodea cornpanulafa Beauv.
S. companulata (Bignoniaceae) is a tall, handsome tree cultivated for its

beautiful red flowers.It yields a valuable red timber and the various parts
of the plant are said to be useful in a number of ailments. Glomerella
cingulata Spauldand Schrenk is reported to cause leaf spot disease in this
plant (Ponnappa, 1967). But the black oval diseased lesions in leaf yielded
Curvulariaprasadii R.L. Mathur and B.L. Mathur, which is proved to be
the pathogen.
The healthy leaves were found to contain 3’,4’-dimethoxyquercetin,
vanillic acid, p-hydroxybenzoic acid, organic bases, and proanthocyanid-
ins.The diseased leavescontainedquercetin,vanillicacid,p-hydroxyben-
zoic acid, syringicand gentisic acids, bases, and proanthocyanidins. Quer-
cetin is found to be produced whenthe healthy leaves were inoculated with
the pathogen. Further studies are in progress(Chitra, unpublished).
K. Zizyphus oenoplia Mill.

Z. oenoplia (Rhamnaceae) is a straggling shrub, often semiscandent, pro-
ducing edible drupes. The leaves are used againsta wide variety of ailments
such as tuberculosis, persistent fever, etc.,and the seeds possess rejuvenat-
ing properties.Leaf rust due to Catenulopsora zizyphi Ramakr. & Subram.
(Ramaskrishnan and Subramanian, 1946) is reported from this plant. The
brownish diseased areason the leaves ofthe plants in Barodais found to be
caused byAlternaria alternata (Fr.) Kiessler.
The healthy leaves of Zizyphus contained quercetin, 3’,4’-diOMe quer-
cetin, vanillic acid, ferulic acid, p-coumaric acid, proanthocyanidins, alka-
loids, and saponins. The diseasedleavescontainedquercetininlarge
amounts (but no methoxy derivatives)and the other phytochemicals of the
healthyleaves(Abraham and Daniel,1988b). The healthyleaves,when
inoculated withthe pathogen, produced quercetin in greater amounts.
544 Daniel
1. Carvia callosa (Nees) Bremek.

C. callosa (Strobilanthes callosus, Acanthaceae) is a tall shrub growing
profusely as an undergrowth in many deciduous moist forests in western
India. The stem is used for the fibers it contains and the leaves exhibit
expectorant properties.Leaf rust caused byAecidium carviae Sathe (Sathe,
1966) isreported from Poona. The red-tinged diseased patches seenon the
plants growing on Pavgadh hills near Baroda is found to be caused by
Aspergillus aculeatusLizuka.
The healthy leaves ofCarvia yielded 7-OMe apigenin, 7-OMe luteolin,
coumarins, vanillic acid, syringic acid, and proanthocyanidins. The dis-
eased leaves contained 4’-OMe apigenin, 4’-OMe vitexin, 7-OMe luteolin,
coumarins, vanillic acid, syringic acid, ferulic acid,
and proanthocyanidins
(Abraham and Daniel, 1988b). During pathogenicity trialsit was also found
that the diseased leaves produced 4’-OMe vitexin and ferulic acid.
4‘-OMe vitexin is a glycoflavone in which 4‘-OMe apigenin istolinked
glucose by C-C bonds. This compound is water-solubleand is resistant to
alkaline and acidic hydrolysis. Preliminary studies indicate that 4’-OMe
vitexin is toxicto many fungi (Daniel, unpublished).
111. MEDICINAL PLANTS

A. Adhatala vasica Nees
A. vasica (Acanthaceae) is a well-acclaimed medicinal plantof India. The
leaves of this tall shrub yield two quinazoline alkaloids, vasicine
and vasici-
none, which are effective expectorants and bronchodilators. The whole
plant is recommendedfor treatment of bronchitis, asthma, fever, and rheu-
matism.
A number of fungi are known to cause diseases in Adhatoda. Impor-
tant pathogens reportedare Aecidium adhatodae Syd. (Sydow et al., 1906)
and Chnoopsora butleri Diet. and Syd. (Dietal, 1906) causing leaf rust;
Collelotrichiumdematium Grove(Roy,1976)causinganthracnose; and
Cercosporaadhatodae Chowdhury(Chowdhury,1948), Colletotrichium
capsici Butler and Bisby, Drechslera speciferurnNicot. (Roy, 1976),Phoma
vasicae Shreemali (Shreemali, 1972), and Corynespora cassicola Wei (Mun-
jal and Gill, 1962) causing leafspot.
The brown spot disease prevalent in plants in and around Baroda is
caused by Colletotrichium dematium Groove. These spots enlarge as the
infection proceedsand at times more than half of the leaf turns brownish
black in color.
Both the healthy and infected leaves contained the flavonols kaemp-
ferol and quercetin, glycoflavones vitexinand isovitexin, and phenolic acids
Plants
Important 545
such as p-hydroxybenzoic, syringic, and p-coumaric acids. But the diseased

leaves contain a flavone 7-OMe apigenin inaddition. The amount of flavo-
noids and phenolic acids in the diseasedleavesismuchhigher than in
the noninfectedones.Between the twoprincipalalkaloidsvasicine and
vasicinone, the latter was absent inthe diseased leaves (Darshikaand Dan-
iel, 1992).
The production of 7-methoxyapigenin and the absence of vasicinone as
a result of infection is proved when the healthy leaves were inoculated
with the pathogen. But Aspergillus niger,the nonpathogen tested, failedto
induce any chemical changethe inleaves.
B. Triantbemaportulacastrum L.
T. portulacastrum (Aizoaceae) isa succulent annual weed. The whole plant
is used as a pot herb. The leaves are diuretic and used in edema and dropsy
due to various causes. Theroots possess cathartic and irritant properties.
Cercospora trianthemae Chiddarwar is reported to cause leaf spot in
Trianthema (Chiddarwar,1962). But the leaf spot disease occurring in Bar-
oda is found to be due to Fusarium sp.
The healthyleavesofTrianthemacontained6,7-dimethoxy-3,5,4’-
trihydroxyflavone, vanillic acid, p-hydroxybenzoic acid, and phytoecdy-
sones. The diseased leaves, inaddition to these compounds, contained quer-
cetin and ferulic acid.By using drop diffusate technique it isfound that the
pathogen induces the formation of quercetin and -ferulic acid. The non-
pathogen Aspergillus niger failed to evoke any response in the healthy
leaves (Darshikaand Daniel, 1992).
Quercetin is found to inhibit spore germination of Fusarium at all
concentrations (33% at 200 ppm, 40% at 400 ppm, and 70% at IO00 ppm).
The germ tube growth also was found to decrease with increase in concen-
tration of quercetin.
C. Withania somnifera Dunal

W. somnifera (Solanaceae) is a woody herbaceous perennial extensively
used in different systems of medicine. It contains a number of alkaloids
belonging to tropane, pyrrazole, or piperidine groups. Withaferins, a group
of steroidal lactones,contribute to the bacteriostatic and antitumor proper-
ties of this plant. The whole plant is used aasrejuvenating tonic, sedative,
and antiinflammatory drug. The roots are recommended for female disor-
ders, rheumatism, dropsy,and in senile debility.
Leaf rust in Withania is caused by Aecidium withaniae Thuem. (Sy-
dow, 1912) and leaf spot is by Cercospora withaniaeH. & P. and Colleto-
trichium capsici Butlerand Brisby (Pavgi and Singh, 1970). The black spot
546 Daniel
disease inthe leaves observed in Baroda found

is to be caused by Chaetom-
ium globosum Kunze ex Fr.
Both the healthy and diseased leaves contained the same flavonoids
(4’-OMe kaempferol and 3’4”diOMe quercetin), phenolic acids, (vanillic,
p-hydroxybenzoic, and ferulic acids), alkaloids,and withaferins. The con-
centration of flavonols and phenolic acids was very high in the diseased
leaves. Both the pathogen and nonpathogen failed to yield any new com-
pound when inoculated in healthy leaves (Darshika, unpublished).
D. Tylophora asthmatica W. & A.

T. asthmatica (T. indica, Asclepiadaceae) is a twiner exhibiting marked
medicinal properties.The leaves of this plant contain a number of phenan-
throindolizidine alkaloids, with tylophorine, tylocebrine, and tylophorinine
being the principal ones among them. These alkaloids aswell as the crude
extract of the leaves possess powerful vesicant propertiesand therefore are
used effectively in fighting asthma. Tylophorine has a paralyzing action on
the heart muscle but a stimulating action on the muscles of blood vessels.
All three principal alkaloids possess anticancer properties.
Cercospora tylophorina Rav. is reported from the diseased leaves of
Tylophora (Rao, 1962). But the brownish black diseased spots on leaf are
found to be caused by Colletotrichium sp.
The infected leaves contained quercetin addition
in to kaempferol, gen-
tisic, p-hydroxybenzoic, and caffeic acids present theinhealthy leaves. The
analysis of alkaloids inthe diseased leaves indicatedthat they are devoid of
the major alkaloid tylophorine. The leaves inoculated with pathogen also
showed similar results. However,the nonpathogen Aspergillus niger failed
to evoke any metabolic change in the leaves (Darshika, unpublished).
W. CUSCUTA AND ITS HOSTS

Cuscuta is a leafless yellow twiner found parasitizing a wide variety of
hosts. Out of the 150 species of this genus, only three are available in this
part of the country, of which C. reflexa Roxb. and C. chinensis Lam. have
been taken up for the present study. The hosts selected are Cordia myxa L.
(Boraginaceae), Streblus asper Lour.(Urticaceae),Clerodendroninerme
Gaertn. (Verbenaceae), and Calotropis gigantea Br. (Asclepiadaceae). The
first three plantsare hosts to C. reflexa and the fourth plant to C. chinensis.
Both the leaves and bark of the hosts were analyzedfor their phytochemi-
c a l ~and fungal pathogens. This study, though preliminary in nature, pro-
vides a number of interesting observations which would give insights to the
defense reactionsof various plants in question.
Phytoalexins of Economically Important Plants 547
A. Cordia myxa L.
Two trees growing side by side, one infested with Cuscuta and the other
noninfected, were taken up for chemical studies. The tree infected with
Cuscuta possessed dark green leaves but seldom bore flowers or fruits,
while the tree free of Cuscuta possessed pale green leaves and abundant
flowers and fruits. Even the diseases of the two plants were different. The
leaves of the Cuscuta-infested tree had small brown diseased spots caused
by Gleosporium sp. whilethe other tree contained grey blisters on the leaf
surface caused by Botryodiplodia sp.
On analysis,the leaves of noninfectedCordia yielded flavonols such as
quercetin, 7,3 '-diOMe quercetin, and 7,3 ',4'-triOMe quercetinand pheno-
lic acids like vanillic, p-hydroxybenzoic, melilotic, gentisic,
and p-coumaric
acids. They contained saponins and tannins also. The leaves the
of tree with
Cuscuta contained methoxyquercetins (7,3 '-diOMe and 7,3 '4"triOMe)
and syringic acid, ferulic acid, saponins,and tannins. The steroids of these
two plants (whichare being characterized) alsoare different. The study on
the barks also yielded similar results (Julie, unpublished).The phytoalexin
response of these plants is being investigated.
B. Streblus asper lour.

In StrebIus free of Cuscuta, the diseased areas appeared near the tip and
margins of the leaves and were colored brown with a grey center. The
culture of thesespotsyieldedCladosporiumsp. The tree infectedwith
Cuscuta hadthe diseased areas ofthe leaf colored brownishyellow on and
around midrib, caused by Chaetomium sp. Flavonoids in both the leaves
were in traces and could not be identified.The phenolic acids were also the
same inboth trees. There were differences in steroids betweenthe two trees.
The constituents of bark of the respective trees also were similar
to those of
leaves. The leaves of tree with Cuscuta,on inoculation with Cladosporium,
produced quercetinand its derivatives in detectable amounts (Julie, unpub-
lished).
C. CIerodendroninerme Caertn
The plants with Cuscuta had white elliptic lesionsin the lamina caused by
AIternaria sp. whereasthe Clerodendron resistantto this parasite haddark
brown spots near the margin and near midrib causedby Cladosporium sp.
The leaves of the former plant contained very low amounts of 4'-OMe
apigeniii and 7,4'-OMe apigenin while the leaves of latter plant produced
large amountsof 4'-OMe apigenin and 7,4'-diOMe apigenin. The phenolic
acids, steroids, and alkaloids were same in both the trees. The bark of
548 Daniel
Cuscuta-infectedplantproducedscutellarein(a6-hydroxyflavone),api-
genin, and 4’-OMe apigenin, whereas the bark of the plant free of the
angiosperm parasite possessed apigenin and 4’-OMe apigenin in traces.
The
steroids of both the plants were different (Julie, unpublished).
D. Ca/otmpis gigantea Br.

The Cuscuta-infected plant possessed brown diseased spots on the leaves
caused by Gleosporium sp. The plant free of Cuscuta was having brown
patches witha pink centeron the leaves caused byAlternaria sp. Theleaves
of the former plant produced methoxyquercetinsand vanillic acid while the
leaves ofthe latter plant contained quercetin, vanillic acid, ferulic acid,and
p-hydroxybenzoic acid. The steroids, cardiac glycosides, and alkaloids were
similar in both plants. The barks of the respective plants also were similar
in chemical constitution (Julie, unpublished).
On chemical analysis, C. refexa was found to possess chlorophylls (a
and b) in traces, abundant carotenoids, 3-5% of waxes, saponins, alka-
loids, and very little flavonoids. Onthe other hand, C. chinensis contained
high amounts of flavonoids (quercetin and its methoxy derivatives), alka-
loids, and negligible amountsof carotenoids, waxes, and saponins. Chloro-
phyll was absentfrom this species.
The results of studies conducted on all four hosts indicate that the
plants resistantto Cuscuta are different chemically fromthe plants infested
by the parasite. The major differences are inthe flavonoids, phenolics,and
triterpenoids. The resistant plants contained higher amounts of flavonoids
always. Another interestingfeature noted inthe present study is the possible
role of Cuscuta in deciding the resistance of the host part to certain fungal
pathogens. The plants with Cuscuta are resistant to the fungal pathogensof
the Cuscuta-resistant hosts. It is also observed that the leaves of Cuscuta-
infected plant, when kept separate from Cuscuta and inoculated with the
pathogen of the resistant plant, developeddiseasesymptoms.Cuscuta
plants, though very tender and possessing very little mechanical tissue,are
resistant to fungal diseases. The disease-resistant factors may be higherwax
content (inC. refexa) or very high flavonoid content (inC. chinensis).
V. DISCUSSION
The reactions of the 16 plants to the fungal invasion can be grouped into
three categories: 1.) demethylation of existing polyphenols, 2.) increase in
concentration of phenolics, and 3). production of new compounds. In al-
most all casesthe compounds converted/producedare phenolics, in which
flavonoids formthe major group followed by phenolic acids.
Plants
Important 549
Demethylation of methoxyflavonoids is seen in seven plants: Cassia,

Morinda,Zizyphus,Madhuca,Trianthema,Spathodea, and Hetero-
phragma. Except for Heterophragma where flavones are located, all others
produced flavonols, quercetin being the most common one.The preference
to flavonoids may be explained by the fact that they are multipurpose
compounds useful to the plant in a variety of ways such as UV screens,
pigments, antimetabolites, and antimicrobials. It is known that the primi-
tive angiosperms (Magnolidae) elaborated flavonols in leaves and during
the course of evolution methoxylation is introduced, which isan advanced
step seen in moderately advancedtaxa (e.g., Rosidae). Reduction in flavo-
nols and introduction of flavones, further advancements in flavonoid evo-
lution, are seen in many of the Asteridae. Phenolics being antimetabolites,
their increased concentration, especially flavonoids, in the cell sap may
exert a harmful effect on the metabolism of plants. Therefore the mecha-
nism in plants would have been to methylate these compounds to make
them more lipophilic and store them in outer lipid layers of waxes and
cutins. Such a placement of flavonoids does not interfere with their func-
tion as UV screens, antifeedants, or protective compounds. It has an added
advantage that these compounds canbe converted to more potent hydroxy
derivatives and transport to the site of infection. This corroborates the
observation that the leaves ofCitrus cultivars resistantto Malesecco disease
contain larger quantitiesof methylated flavonoids than the leaves of suscep-
tible ones(Harborne, 1983).
The increase in the concentration of phenolics in diseased leaves is a
well-known phenomenonand is seen inalmost allthe plants screened inthe
present project. InMangifera and Withania the greater amountsof pheno-
lics constitute the only detectable reaction the plant exhibits to invading
fungi. These compounds are abundant on the tissues bordering diseased
areas. Most of the diseased spots exhibita red border indicating the accu-
mulation of anthocyanins.
Production ofnew compounds is the mechanism of resistanceex-
pressed in Tectona, Carvia, Anogeissus, Eucalyptus, Syzygium, Adhatoda,
and Tylophora. It is observed that each plant produces the typical corn-
pounds characteristic of the particular family to which it belongs. Carvia
and Anogeissus produced glycoflavones which are otherwise Seen in the
primitive members ofthe Acanthaceae (Danieland Sabnis, 1987) and Corn-
bretaceae. The Myrtaceae are known for their coumarinsand quinones, and
therefore Eucalyptus produced a coumarin andSyzygium a benzoquinone.
Flavones are more common in the Acanthaceae and therefore Adhatoda
could produce a flavone, while Tylophora belonging to the flavonol-rich
Asclepiadaceae synthesizeda flavonol. Though the Verbenaceae are known
for their flavones and quinones, Tectona produced p-hydroxybemoic acid,
550 Daniel
a phenolic acid widely distributed inthe plant kingdom, but implicated in

disease resistance. It is worth mentioning here that during drop diffusate
studies,theleachates from theleaveskeptas control containedsimple
phenol. This may indicate that during pathogenesis phenol would have been
converted to p-hydroxybenzoic acid.
The studies on Cuscuta also emphasize the reliability of flavonoids as
the resistant factors. The plants rich in flavones/flavonols are resistant to
the attack of this parasite.The vulnerabilityof the hosts to the parasite may
depend on the steroid constitutionof the host also since these compounds
are found to be different in the plants free of parasite and plants with
parasite.
VI. EPILOG
The role of phenolics in the disease resistance of plants is proved beyond
doubt. Some of these compounds act as phytoalexins and others work
synergistically with the phytoalexins. The common flavonoids which are
widely distributed in plant kingdomare definitely involved in disease resis-
tance. Detailed studieson this aspect wouldbe conducted inour laboratory.
More and more plants are to be screened before drawing conclusions
on the distribution, properties, and metabolism of phytoalexins. The ten-
dency to describe phytoalexins asthe sole disease resistancefactors is to be
met with caution.It is definitely true that they form one of the very impor-
tant resistant factors of plants. They have a major defensiverolein a
number of angiosperms but in many others they may have a supportive
role. All these facts point to the highly complexnature of defense mecha-
nism of plants.
Our future work on Cuscuta would beto find out whether this plant or
any other angiosperm parasite induces phytoalexins on their hosts or not.
The induction of resistant factors in the hosts of Cuscuta also is to be
studied in detail. Another interesting feature to be probed is the nature of
relationship between Cuscuta and its hosts. It may be a case of symbiosis
also.
ACKNOWLEDGMENTS
I thank my research students (past and present) Joy Abraham, Darshika
Parikh, Chitra Arya, and Julie Sebastianfor permitting me to incorporate
their unpublished results in
this chapter.
REFERENCES
Abraham, K.J. (1989). Phytoalexins and other post-infectional compoundsof some
economically important treesof Gujarat. PhD thesis, M.S.University, Baroda.
Plants
Important 551
Abraham, K. J., and Daniel, M. (1988a). Phytoalexins of Cassia fistula L. and

Morinda tomentosaHeyne. Natl. Acad. Sci. Lett. ll(4): 101-102.
Abraham, K. J., and Daniel, M. (1988b). Phytoalexins and related post-infectional
compounds of forest crops of Gujarat. Adv. For. Res. IndiaI: 191-197.
Abraham, K. J., and Daniel, M. (1990). Chemical changes in the fungal infected
leaves of Ziz,yphus oenoplia Mill and Anogeissus latifolia Wall. Natl. Acad.
Sci. Lett. 13(7): 267-268.
Agarwal, G. P., and Ganguli, S. (1959). A leaf spot disease of Anogeissus lati-
folia Wall due to Pestalotiopsis versicolor(Speg) Steyaert. Cum. Sci. 2 8 295-
296.
Alcubilla-Martin, M. (1970). Fungus inhibitors insprucebark. Landwrit For-
schung. 25: 96-101.
Bakshi, B. K. (1974). Control of root disease in plantations in reforested stands.
Indian Forester Z O O 77-78.
Berkenkemp, B. (1971). Effect of coumarin on some pathogens of sweet clover.
Can. J. Plant Sci. 51: 299-303.
Bilgrami, K. S. (1963). Leaf spot diseases of some common ornamental plants.
Proc. Natl. Acad. Sci. USA 33(B): 429-452.
Borys, M. W., and Childers, N. F. (1964). The growth of Fusicoccum amygdali in
vitro, as influenced by pH, phenolic compounds and unpurified extracts from
peach twigs.Acta Microbiologia Polond13: 21 1-220.
Chiddarwar, P. P. (1962). Contribution to our knowledge of the Cercosporae of
Bombay state 111. Mycopathol. Mycol.Appl. 17:71-78.
Chigrin, V. V., andRozuru,L. V. (1969). Changesinphenolicmetabolismof
spring wheat infected with stem rust. Soc. Plant Physiol.1 6 269-273.
Chowdhury, S. (1948). Some fungi from Assam111. Indian J. Agric. Sci,18: 177-184.
DaCosta, E. W. B., and Rudman, F. (1958). The causes of natural durability in
timber. I. The role of toxic extractives in
the resistance of tallow wood (Euca-
lyptus microcorysF. Muell) to decay. Aust. J. Biol. Sci.I I : 45-57.
Daniel, M., and Sabnis, S. D. (1987). Chemosystematics of the Acanthaceae. Proc.
Indian Acad. Sci. (Plant Sci.)97(4): 315-323.
Darshika, P., andDaniel, M.(1992). Changes in the chemical content of Adhafoda
and Trianthemadue to fungal diseases.Indian J. Pharm. 54(2): 73-75.
Debnam, J. R., and Smith, I. M. (1976). Changes in the isoflavones and pterocar-
pans of red cloveron infection with Sclerotinia trifoliarumand Botrytis cineria.
Dietal, P. (1906). Uber Chnoopsora eine neue Uridineen-Gattung Ann. Mycol. 4:
424.
Egawa, H., Tsutsui, D., Tatsuyama, K., and Hatta, T. (1977). Antifungal sub-
stances found in leaves ofEucalyptus species. Experientia33: 889-890.
Gangulee, H. C.,andKar,A. K. (1985). CollegeBotany. New Central Book
Agency, Calcutta.
Ghosal, S., Biswas, K., Chakraborti, D. K., and Basuchaudhary,K. C. (1977). Control
of Fusarium wilt of safflower by mangiferin.Phytopathology 67: 548-550.
Harborne, J. B. (1983). The flavonoids of the Rutales. Chemistry and Chemical
Taxonomy of the Rutales (P. G. Waterman and M. F. Grundon, eds.), Aca-
demic Press,London, p. 149.
552 Daniel
Harding, V. K., and Heale, J. B. (1981). The accumulation ofinhibitor compounds

in the induced resistance response of carrot root slices to Botrytis cinerea.
Physiol. Plant Pathol.1 8 7-15.
Hart, J. H., and Willis, W. E. (1974). Inhibition of wood-rotting fungi by stilbenes
and other polyphenols inEucalyptussideroxylon. Phytopathology64: 939-948.
Henderson, S. J., and Friend, J. (1979). Increase in PAL and lignin likecompounds
asrace specific responses of potato tubers to Phytophthora infestans. Phytopa-
thol. Z. 94 323-334.
Janardhanan, K. K., Ganguly, D., Baruah, J. B., and Rao, P.P. (1963). Fungitox-
icity of extracts from tannin-bearing plants. Curr. Sci. 32: 226-227.
Kamat, M. N., and Kalani, I. K. (1964). Pathogenic fungi on Syvgium cumini
Skeel. J. Univ. Poona28 122-124.
Kanujia, R. S., and Singh, C. S. (1978). Some new host records of parasitic fungi I.
Indian Phytopathol. 31:225-236.
Karadge, B. A., Chavan, P. D., and Thite, A.N. (1980). Changes in phenolic
compounds of teak leaves inducedby powdery mildewinfection. Indian Phyto-
pathol. 33: 114-1 16.
Li, C. Y ., Lu, K. C., Trappe, J. M., and Bolle, W. B. (1972). Poria weiriiinhibiting
and other phenolic compoundsin red alter and Doughlas fir. Microbios 5: 65-
68.
Ling, L. (1951). Taxonomic notes on Asiatic smuts 111. Sydowia 5: 40-48.
Mukherji, K. G.,and Bhasin, J. (1986). Plant Diseases of India: A Source Book.
Tata-McGraw-Hill, New Delhi.
Munjal, R. L., and Gill, H. S. (1962). Some dematiaceous Hyphomycetes from
India. Indian Phytopathol. 1 6 62-68.
Nath, V., and Bhargava, K. S. (1976). Some leafspot fungi recorded inGoraphpur.
Indian Phytopathol. 2 9 461-462.
Nirwan, R. S., and Singh, B.B. (1967). Leaf spot of Madhuca indica caused by
Cylindrocladium scoparium. Indian Phytopathol.2 0 169-170.
Osborne, L. D., and Thrower, L. B. (1964). Thiamine requirement of some wood-
rotting fungi and its relation to natural durability of timber. Trans. Br. Mycol.
SOC.47: 601-611.
Patil, B. V., and Tirumalachar, M. J. (1971). Studies on EIsinoe and Sphaceloma
diseases ofplants in Maharashtra State (India). VII. Sydowia 25:47-50.
Patwardhan, P. G. (1966). Some new records of powderymildew fungi. Plant Dis.
Rep. 5 0 709-710.
Pavgi, M. S., and Singh, U. P. (1970). Parasitic fungi from North India.IX.
Sydowia 24:113-1 19.
Payak, M. M.(1953). Some new records of fungifrom Bombay State. Sci. Cult. 1 8
342-343.
Ponnappa, K. M. (1967). Someinterestingfungi. I. Miscellaneousfungi. Proc.
Indian Acad. Sci. 66B: 266-272.
Rajak, R. C., and Rai, M. K. (1982). Speciesof Phoma from legumes. Indian
Ramakrishnan, T. S., and Ramakrishnan, K. (1949). Chaconia tectonaeRamakrish-
nan T.S. & K.Sp. Nov. on teak. Indian Phytopathol. 2(1): 17-19.
Plants
Important 553
Ramakrishnan, T. S., and Sundaram,N. V. (1954). Additions to fungi of Madras.

XVI. Proc. Indian Acad. Sci. 40B: 17-23.
Ramakrishnan, T. S., and Subramanian, C. L. (1946). On Catenulopsora zizyphi
on Zizyphys oenopliaMill. Curr. Sci. 15:261-262.
Rao, V. G . (1962). Some new records of fungi imperfecti from India. Sydowia 1 6
41-45.
Roy, A.K. (1976). Some new records of fungi on medicinal plants. Curr. Sci. 45:
464-465.
Salam, M. A., and Rao, P. N. (1958). Fungi from Hyderabad (India) I11 (Oidium
sp.). Indian Phytopathol. 11: 126-129.
Sathe, A. V. (1965). Some new or revised species of Physopella (Uredinales) from
India. Sydowia 19:138-142.
Sathe, A. V. (1966). Some additions to rust fungi of India. Mycopathol. Mycol.
A&. 2 9 333-338.
Shreemali, J. L. (1972). Two new pathogenicfungicausingdiseasesonIndian
medicinal plants. Indian J. Mycol. Plant Pathol.3: 84-85.
Shreemali, J. L., and Bilgrami, K. S. (1968). Range of variation in morphology of
different isolates of Botryodiplodia theobromaePat. Indian Phytopathol. 21:
351-360.
Sporoston, T., Jr. (1957). Studies on the disease resistanceto Impatiens Balsamina.
Sridhar, R., Mohanty, S. K., and Anjaneyulu, A. (1979). In vivo inactivation of
rice tungro virus by ferulic acid.Phytopathol. Z. 94 279-281.
Swinburne, T. R. (1975). Microbial proteases as elicitors of benzoic acid accumula-
tion in apples.Phytopathol. 2.82: 152-162.
Sydow, H. (1912). Fungi Indiae OrientalisPars. IV. Ann. Mycol. I O 243-280.
Sydow, H., Sydow, P., and Butler, E. J. (1906). Fungi Indiae Orientalis Pars. I.
Ann. Mycol. 4: 424-445.
Tewari, I. (1968). On the morphology and taxonomy of Elsinoe kamati Sp.Nov. on
Syzygium cuminifrom India. Nova Hedwigia 15:213-219.
Tirumalachar, M. J., and Chupp, C. (1948). Notes on some Cercosporae ofIndia.
Mycologia 4 0 352-362.
Trappe, J. M., Li, C. Y., Lu, Y. C., and Bollen, W. B. (1973). Differential response
of Poria weirii to phenolic acids from Douglas fir and red alder roots. Forest
Sci. 1 9 191-196.
Uppal, B. N.,and Patel, M. K. (1945). Powdery mildewofMango. J. Univ.
Bombay 9 12-16.
Walchii, O., and Scheck, E. (1976). Natural resistance of Castanea sativawood to
fungal attack and its cause.Beih Mat. Organismen 3: 77-89.
Wani, D. D., and Tirumalachar, M. J. (1969). Studies on Elsinoe and Sphaceloma
diseases of plants inMaharashtra (India). VI. Sydowia 23:257-260.
Wani, D. D., and Tirumalachar, M. J. (1971). Studies on Elsinoe and Sphaceloma
diseases of plants in Maharashtra (India). VII. Sydowia 25:47-50.
Yadav, A. S. (1963). Additions to the microfungi of Bihar 1. Erysiphaceae. Indian
Phytopathol. 1 6 164-166.
25
Possible Role of Phytoalexin Inducer
Chemicals in Plant Disease Control
Asoke Kumar Sinha
Bidhan Chandra Krishi Viswavidyalaya, Kalyani, West Bengal, India
1. INTRODUCTION
The main objective of plant pathology is to control an increasing number
of diseases and to do so with maximal effectiveness. Since suitable resistant
varieties are neitheralwaysavailable nor availableagainstallkindsof
pathogens, the grower often is forced to resort to chemical control based
on the use of chemicals with direct toxic action on the pathogen. Such
control is mainly available against fungi, which constitute the most impor-
tant group amongthe plant pathogens. However, the lack of suitable fungi-
cides against someof the major fungal pathogensand particularly the eco-
logical hazards to which their regular and large-scale use may lead have
worried concerned scientists, prompting them to look for some suitable and
safer alternative approaches to plant diseasecontrol.
Mostplantshave a versatile,multicomponentdefense,adequately
equipped to provide them protection against most of their potential patho-
gens; only a few of them can overcomethis defense and cause disease. The
general assumptionis that varieties withina host speciesare resistant when
they possess oneor more resistantgenes and susceptible when they lack any
such gene. Plant defense primarily depends on some need-based dynamic
responses to attemptedinfection,mostly an induciblephenomenon,its
qualitative and quantitative aspects being regulated by, signals from the
invading pathogen. Even a susceptible host variety is neither completely
defenseless nor totally lacking in any genetic information for resistance.
Even sucha plant has itslatent defense potential, which finds expression at
certain growth stages and under certain stress conditions created during
cropping. This naturally led to the feeling that by manipulation of cropping
conditions or by creating certain stresses,it may be possibleto activate this
latent potential and put it into operation. In the background of great suc-
cess achieved with active immunization, based on specific antigen-antibody
reaction, in controlling some of the most serious human and animal dis-
555
556 Sinha
eases, this wouldnot seem impossible, particularly because much similarity

existsbetween the basic life processesinplantsandanimals. It isnow
well established that something analogous to active immunization of the
vertebrates also occurs in plant systems, though its components and regula-
tion may differ from those in animals(KuC, 1982, 1987;Sequeira, 1983).
Evidence is now available to conclude that effective resistance can be
developedinsusceptibleplantsagainsttheirfungal,bacterial,andviral
pathogens by the use of both biotic and abiotic agents, i.e., by 1.) prior
inoculation withthe same, related, or even unrelated pathogens of homolo-
gous or heterologous kind; 2.) prior treatment with cell-free microbial flu-
ids containing their metabolites and/or cell constituents; or 3.) treatment
with many synthetic compounds (Price,1964; Matta, 1971,1980; Wain and
Carter, 1972; Goodman, 1980; Langcake, 1981; Lazarovits, 1988; Sinha,
1990,1992). Resistance acquired by susceptible plant on treatment with
various biotic agents was, in many instances, systemic in nature, lasting in
effect and also broad spectrum in action, i.e., effective against more than
one pathogen, may be of different kinds (KuC, 1982, 1987; Mucharromah
and KuC, 1991). Particularly impressive had been the findings on tobacco
(Matta, 1980; KuC and Tuzun, 1983), cucumber (Jenns et al., 1979), and
melons (Caruso and KuC, 1977) where field level, long-term, systemic pro-
tection could be induced by and were effective against a broad range of
fungi, some bacteria,and many viruses (Goodman, 1980; Hamilton, 1980).
However, this approach had little acceptance from the growers probably
becauseofheavycostsinvolvedin the laboratory production of biotic
inducer agentsand also the logistic difficulties faced in delivering the agent
to the crop, particularly when the cropped area is large.
The phytoalexin concept of Muller and Borger (1940), further devel-
oped withthe incorporation of new findings overthe next 50 years (Cruick-
shank, 1963, 1980; Dixon, 1986; Van Etten et al., 1989), probably provides
the most rational explanation and also a basis for the dynamic defense of
many plant species, primarily against their fungal and only occasionally
against bacteria and nematode pathogens as well (Mansfield, 1982). Rapid
postinfection production of phytoalexin-type antipathogenic principle in
the affected tissue and its rapid accumulation to a toxic level results in
resistant reaction. Such high-level accumulation of phytoalexin is linked
with incompatible host-pathogen interaction. Though not universal in oc-
currence, phytoalexin accumulation is now accepted as one of the major
dynamic defense mechanisms of the plant. Besides the pathogen itself and
other infectious entities, 1.) its metabolites and cell constituents, 2.) some
plant products, 3.) some synthetic compounds including even some fungi-
cides and toxins, and also 4.) some stress factors can induce phytoalexin
synthesis. Those in 1.) are termed as biotic “elicitor.” While their presence
Phytoalexin
Disease
Control
Chemicals
Inducer
and 557
and involvement in phytoalexin production have been reported in many

cases, only a few could be isolatedso far from the pathogen and character-
ized. More significant appears to be the fact that many synthetic com-
pounds having no link either with the pathogen or host also share this
ability. Many prefer the term abiotic elicitor for this group of compounds
(Yoshikawa, 1978; Darvill and Albersheim, 1984), but for the purpose of
this chapter the terminducer will be usedto distinguish themfrom elicitors
of biotic origin. Feelings have often been expressed that phytoalexins or
their biotic elicitors can be directly used on host plants for disease control,
particularlyagainstthose of fungalorigin,buttheexploratorystudies
yielded no significant information. However, during the 50 lastyears, many
synthetic compounds of diverse nature had been tested on various crops,
and with some of them considerable amelioration of disease symptoms was
noticed, often linked with significant changes in host metabolism (Van-
Andel, 1966; Langcake, 1981; Lazarovits, 1988; Sinha, 1992). At Bidhan
Chandra Krishi Viswavidyalaya, Kalyani, we have continued exploratory
work for more than a decade with a large group of synthetic compounds
that are known phytoalexin inducers in the hope that these may trigger
some active defense responses the in host plant including phytoalexin accu-
mulation. Results in general have indicated great promise and may suggest
an alternative approach to conventional chemical control. These findings
are presented here in the background of what could so far be achieved in
this direction with phytoalexins, their biotic elicitors,and synthetic chemi-
cals with little involvement thein natural defense responsesof plants.
II. PLANTDISEASECONTROL BY NONCONVENTIONAL

CHEMICAL AGENTS
Besides phytoalexins, the known resistance factor produced in plants in
response to attemptedinfectionbypotentialpathogens,mostlyfungi,
pathogen metabolites, or cell constituents that trigger host responses lead-
ing to phytoalexin production, i.e., elicitors, and miscellaneous synthetic
compounds withor without a link to biosynthesis of phytoalexins in plants
have been testedon many plant species witha view to exploring whetheror
not these biotic or abiotic agents may protect them from their pathogens.
These findings are presented in the following pages and their implications,
if any, discussed.
Phytoalexins
In spite of some anomalous results, there is now enough evidence
to accept
phytoalexins as a primaryfactorinplantdefense,mostlyagainsttheir
fungal pathogens. Theyare neither universal in occurrence nor responsible
558 Sinha
for host defense inall cases. Rapid production of phytoalexinand its early
accumulation to a toxic levelat the site of infection is now known to be the
basis of a major dynamic defense mechanism in many plants. About 300
phytoalexins have been characterized so far, mostly from dicotyledonous
plants, and many more, for which good evidence is available, still remain
to be characterized. Phytoalexins show great diversity in form, e.g., isofla-
vonoids, sesquiterpenoids, furanoterpenes, diterpenes, polyacetylenes, stil-
bene, etc. Some plant species produce more than one phytoalexin when
infected. In a particular host-pathogen interaction, generally one phyto-
alexin plays a dominant role. If the phytoalexin is removed from the af-
fected tissue, the normal resistance reaction is changed to a susceptible
reaction(KlarmanandGerdemann, 1963). Similarly,ifphytoalexinis
added to the site of infection, there is a change from susceptibleto resistant
host reaction (Chamberlain and Paxton, 1968). Because of their broad-
spectrumfungitoxicaction and diverseforms,alogicalconsequenceof
these studies was to explore if phytoalexins could be directly utilized to
protect plants from disease. However, few attempts were made in this direc-
tion and these did not indicate much promise in this respect. Weyrone could
provide only some protection to French bean from rust(Uromyces sp.) and
broad bean from chocolatespot (Botrytis fabae) (Fawcettet al., 1969)
diseases. Ward et al. (1975) reported good control of late blight to tomato
(Phytophthorainfestam) with capsidiol butthe protective effect lasted only
up to 8 days. Viniferin also protected vines from downy mildew (Plasmo-
para viticola) for only 7 days (Langcake, 1981). A direct comparison be-
tween seven isoflavonoid phytoalexins and two widely used fungicides, be-
nomyl and macozeb,ledRathmell and Smith (1980) to conclude that
phytoalexins have little commercial potentialfor direct use in plant disease
control like fungicides. This does not seem to be unlikely in view of their
mild toxicity, unstable nature, and lack of movement in the plant tissue.
Further, their complex ring structure does not make them suitable for easy
isolation fromthe reacting host tissueand low-cost synthetic processes.
Phytoalexin Elicitors
Elicitors are molecules of biotic origin
that signal plantsto begin the process
of phytoalexin biosynthesis. The fact that, besides the live inoculum of
pathogen, its cell-free mycelium extract, spore germination fluid, culture
filtrate, and so forth also induce phytoalexin synthesis makes this obvious.
Since this is an area of much activity, many reviews are available (Keen,
1975; Yoshikawa, 1978; Bailey, 1982; Darvill and Albersheim,1984; Dixon,
1986). Though the occurrence and functioning of elicitor-active compounds
Phytoalexin
Disease
Control
Chemicals
Inducer
and 559
have been indicated for many plant-pathogen systems, only in a small

number of cases could elicitors be isolated and characterized. It has been
found that glucans, mostly those with branched structure, and glycopro-
teins constitutethe most common types of elicitors. Besides these, polypep-
tides, lipids, lipopolysaccharides, lipoglycoprotein, fatty acids, and chito-
san, the deacetylated derivative of chitin, are also elicitor-active. Some of
these are extremely active at very low (10"' M) doses. Though the occur-
rence of race-specific elicitors have been claimed for Rhizopus stolonifer
(Stekoll and West, 1978) and Phytophthoramegasperma f.sp. gljrinea
(Keen and Legrand, 1980), the evidence in support is not conclusive. Some
enzymes of fungal origin, e.g., endopolygalacturonase from R . stolonifer
and endopolygalacturonate lyase from Erwinia carotovora, and others of
plant origin, e.g., chitinase and 0-endoglucanase, also participate in the
elicitation process, though not directly.
At the initial stage of elicitor-based research,it was felt that there might
be some scope for their use in plant disease control. Though a hepta- and
octasaccharide obtained from mycelial wall of P . megasperma f.sp. gly-
cines with elicitor activity could be synthesized (Sharp al.,et1984), because
of the large molecules and often branched structures involved, elicitor syn-
thesis cannot beeasy and must be very expensive. Further, when foliar
spray with some of the elicitors was tested as a possible method for plant
disease control, nopromise was indicated. This discouraged further at-
tempts inthis direction.
Miscellaneous Synthetic Compounds
Concurrent with studies to unravel the mysteries of host defense against

plant pathogens have been attempts to study diverse, mostly nontoxic or
mildly toxic synthetic compounds with a view to their potential applications
in plant diseasecontrol. Although initial studies inthis direction date back
to the early part of this century, most serious studies were made during the
last three decades (Dimond, 1963; VanAndel, 1966; Wain and Carter, 1972;
Langcake, 1981; Lazarovits, 1988; Sinha, 1990, 1992). A variety of com-
pounds with knownor no effect on plant systems, particularly metal salts,
amino acids, plant growth regulators, and miscellaneous compounds, had
been tested for this purpose, mostly on major crop plants against their
important fungal pathogens, and impressiveresultswererecordedwith
some of them. The main idea was to look for nonhazardous but effective
compounds that can possibly replacethe conventional toxic agents in plant
disease control. A general idea is given below, by group, of the protective
effects achieved with such compounds.
560 Sinha
Metal Salts
Among the metallic compounds, copper, zinc,and lithium saltsand boron
compounds have shown particularly good effect against many important
crop diseases, e.g., lithium salts on cereals against powdery mildews (sys-
temic) and rusts; copper salts against wheatrusts, brown spot of rice, and
of wheat and rust of
Verticilliumwilt of cotton; zinc salts against leaf rusts
linseed (Wain and Carter, 1972). Cupric chloride-induced resistance in cot-
ton against Verticillium wilt was associated with increased phytoalexin pro-
duction (Bell and Presley, 1969). Subramaniam (1963) achieved complete
control ofpigeonpeawilt. (Fusarium udum) byseed treatment, foliage
spray, or soil drench with100 ppm manganese sulfate. Additionof micro-
nutrients haveoften led to suppression of soil-borne pathogens, particularly
those causing vascular wilts.
Amino Acids
Among the amino acids screened for their possible suppressive effect on
plant diseases, some showed distinct promise, e.g., phenylalanine against
apple scab (Venturia inaequalis); serine and threonine against cucumber
scab (Cladosporium cucumerinum); serine against chocolate spot of bean
(Botrytis fabae);and some sulfur-containing amino acids against pea root
rot (Aphanomyces euteiches ) (VanAndel, 1966; see references). Effective
control of Fusarium wilt of tomato was achieved with root application of
methionine (Jones and Woltz, 1969) and phenylalanine (Carrasco et al.,
1978). In the later treatment, plants developed much increased phenol level.
Promising results were recorded with some amino acid derivatives also.
Presowing seed treatment with dodecyl-DL-alaninate provided rice plants
long-lasting systemic protection from blast disease (Arimoto et al., 1976).
Treatment with esters of N-allylglycine and N-allylsarcosine led tomato
plants to acquire strong resistance to Fusarium wilt; they responded to
infection with increased levels of phenolicsand stimulated peroxidase activ-
ity (Kirino et al., 1980).
Plant Growth Regulators
Plant growth regulators have been particularly effective against vascular
wilts, e.g., indoleacetic acid (IAA) 2,4-dichlorophenoxyacetic acid (2,4-D),
naphthaleneaceticacids (NAA), and 2,3,5-triiodobenzoicacid(TIBA)
against Fusarium wilt of tomato; I A A and naphthalene acetamide (NAM)
against Verticillium wiltof the same plant; 2,3,6-trichlorophenoxyacetic
acid (2,3,6-T), and halogenated benzoic acids against Dutch elm disease
(Ceratostomella ulmi);and benzoic acids, IAA, and TIBA against oak wilt
(Ceratocystis fagacearum) .(Wain and.Carter, 1972). In all such effective
treatments, plants showed inhibited growth. It has been suggestedthat these
Phytoalexin
Disease
Control
Chemicals
Inducer
and 561
treatments ledto a transformation of cell wall pectic compounds from their

methylated state to calcium/magnesium pectates that are more rigid and
less amenableto pectic enzyme action. Various growth retardants like cyco-
cel (chlorocholine chloride), Phosphon-D, N,N-dimethylpyrrolidinium io-
dide, and N,N-didimethylpiperidiniumiodide also inhibitthe development
of Fusarium or Verticillium wilt of tomato or both (Sinha and Wood, 1964;
Buchnauer, 1971; Buchnauer and Erwin, 1973; Erwin, 1978). Resistance
induced by the above iodides was associated with increasedformation of a
phytoalexin-typecompoundin Verticillium wilt-affected cotton plants.
Some growth regulators also act against diseases other than wilt, such as
chocolate spot of broad bean (Botrytisfabae), wheat rusts, and cucumber
scab (Wain and Carter, 1972; see references).
Miscellaneous Organic Compounds
Some ofthe many synthetic compounds tested on various crops have shown
promising results against their diseases. Primarily those cases where
there is
evidence or at least the suggestion that the induced protective effect in
plants is mediatedthrough changes in host metabolism merit consideration,
and the details are presented here. Phenylthiourea effectively protects cu-
cumber from scab (C. cucumerinum), and the protected plants respondto
infection with increasedPO activity and enhanced lignification (Sijpesteijn,
1969). With probenazole treatment, plants achieve substantial protection
from both blast (Watanabe etal., 1979) and bacterial leaf blight (Xantho-
monas oryzae) (Tomiku et al., 1979). When challenged with the pathogen,
plants responded with higher accumulation of fungitoxic substances (Shi-
mura et al., 1981) and increased PO activity and lignification (Iwataet al.,
1980) as comparedto the untreated plants. Thereis also the report that rice
plants treated with WL28325, a prophylactic fungicide, get protected from
blast, and thisisassociatedwithincreasedproductionofphytoalexins
(momilactones)inresponse to infection (Cartwright et al., 1977,1980).
Ghosal and Purkayastha (1974) reported that treatment with gibberellic
acid was effective in providing protection to rice plants which, in its pro-
tected state, developed much higher levels of PO as compared to the un-
treated plants.
Ethylene itself (Pegg, 1976) or its precursor, ethephon (Retig, 1974)
effectivelychecks Fusarium wiltof tomato. Such protected plants also
showedmuchincreasein PP0 and PO activities. Fosetyl-Al (TEPA =
aluminum tris-O-ethyl phosphonate) has been found to be very effective
against different diseases, particularly those caused by fungal pathogens,
such as downy mildew of grapes(Plasmopara viticola(Lafon et al., 1977);
Phytophthora infection of tomato, tobacco, pepper, and citrus (Vo-Thi-
Hai et al., 1979; Bompeix et al., 1980; Gutter, 1983); and Phomopsis infec-
562 Sinha
tion of grapes (Bertrand et al., 1977). No phytoalexin accumulated in the

susceptible plants inthe presence of Fosetyl-Al alone or when infected in its
absence, but more than the normal levels of phytoalexin accumulate inthe
treated plants. Phosphorous acid, a breakdown product of fosetyl-Al, is
more effective inthis respect. Fenn and Coffey(1985) reported that fosetyl-
induced protection coincided in time with the appearance of necrotic symp-
toms, accumulationof phosphorous acid,and increased PO activity. Appli-
cation to the treatedplants of a-aminooxyaceticacid, an inhibitor of
aromatic biosynthesis and phenylpropanoid pathway, partially neutralizes
the induced protective effect. Sodium azide treatment has been found to
protect soybean plants fromMacrophomina phaseolinainfection (Chakra-
borty and Purkayastha, 1987). Similarly, cloxacillin and penicillin protect
the same plantfrom Colletotrichurn dematium infection (Purkayastha and
Banerjee, 1990). In both sodium azide and cloxacillin treatment, but not
with penicillin, the effect was associated with a change in the antigenic
pattern of the host towarda reduced commonness between the host and the
virulent pathogen isolate. In sodium azide treatment plants also responded
to inoculation with M. phaseolina with increased phytoalexin production.
A recent report states that a spray with oxalate or dibasidtribasic potas-
sium phosphate inducesstrong resistance in cucumber plants to four fun-
gal, two bacterial,and two viral pathogens (Mucharromahand KuC, 1991).
However, the chemically induced protective effect was not as strong as that
induced by prior inoculation withColletotrichum lagenarium.
The above experimental findings show that many synthetic compounds
without any evident toxicity to the pathogen can induce strong resistance in
a host plantnot only to one but sometimes to more than one pathogen, may
be of different kinds, and may also be effective on more than one plant
species. Such an induced protective effect is systemic in nature and has
often been shownto be long-lasting ineffect.
Synthetic PhytoalexinInducer Compounds
Results presented so far, though often very significant and also suggestive
in nature from the perspective of plant disease control, mostly involved
one, rarely a few chemicals appliedto a plant host. There was little justifica-
tion in most casesfor selecting and using a particular chemical ona particu-
lar host against a disease. In that context, the studies conducted in our
laboratory are an exception. Studies conducted here for more than a decade
(and still continuing) have involved a large group of synthetic compounds,
tested on some important crop plants against some of their major fungal
pathogens with different levels of parasitic specialization,are more broad-
based in nature, and provide comprehensive information in the area of
Phytoalexin
Disease
Control
Chemicals
Inducer
and 563
induced resistance in plants as a possible disease control measure. This is

probably the first time that the test chemicals were used for some of their
known properties, i.e.,the ability to induce the production of phytoalexin-
type resistant factor in plants (Uritani et al., 1960; Uehara, 1963; Condon
etal., 1963; Perrin and Cruickshank, 1965;Schwochau andHadwiger,
1968; Hadwiger, 1972). These include heavy metal salts, amino acids, plant
growth regulators, metabolic inhibitors, reducing agents, antibiotics, drugs,
and so forth. Since phytoalexins are now accepted as resistance factors
having a major role in host defense in many plant-fungus interactions, it
was felt that their induced production in (susceptible) plants by treatment
with inducer chemicals might also trigger a series of metabolic responses
that do not favor the pathogen. Such induced resistancemay then become
useful in plant diseasecontrol in the same way that immunization is effec-
tive against humanand animal diseases.
The idea of exploringthe possibility came from early studies in BCKV
on biological induction of resistance in rich plants to brown spot disease
incited by Helminthosporium oryzae. It was noted that optimum protection
from H. oryzae was achieved when3-to 4-week-old plants were inoculated
with the mildly virulent isolate of this pathogen 2 days before challenge
inoculation (Sinhaand Trivedi, 1969; Sinha and Das, 1972). A mild isolate
of H. oryzae had the strongest effect as inducer agent as compared to four
other speciesof Helminthosporium or twominorpathogens ofrice,
namely, Cercospora oryzae and Nigrospora oryzae (Trivedi and Sinha,
1976a). Of the various kindsof fungal fluids, e.g., culturefiltrate, mycelial
extract, spore germination fluid, etc., that can also be used as a foliage
spray to induce resistance in rice plants; maximum effect was achieved with
spore germination fluid of the mild isolate, obtained from spore suspension
with 3 x lo6conidia/ml, applied 2 days before challenge inoculation (Tri-
vedi and Sinha, 1976b; Mukhopadhyayand Sinha, 1980). This was as effec-
tive as the optimum concentration of live inoculum (- 5 x 10’ conidia/
ml). It was strikingly evident from these studies that, with minor excep-
tions, the magnitude of resistance induced by different treatments had good
correlation withthe level of fungitoxicity postinfectionally developedthe in
treated plants after challenge withthe virulent isolate.The toxicity initially
developed following the treatment wasless significant in terms of both
concentration and the final effect. Such observations suggested that the
initial treatment withthe live inoculum ofa biotic agentor fluid containing
its metabolitesnot only inducedthe accumulation of phytoalexin-type resis-
tance factor(s) in the host tissue that might function as a passive defense
causing some initial limitation to the challenger pathogen, but must have
also activated the defense potential that usually remainslatent in a suscepti-
ble host. In this background, testingof phytoalexin inducer compoundson
h
S Y
2-
+++++ +
+++ ++++++ +
Y E= + + + l + + + ++ +
+ I +I +++ +
c
a
l +
c
7
e Y
+ + +
0 "En 1 + + 1 + 1 + + + I + + ++
%:E
w 3 + +++
bl
e
Y
0
+ ++
2% +I= ;+I++ ++ ++ ++
2
a
8
%
z
0 -5
553 + ++
- 3 I I. + +
+
do
a .S + +
+ ++ ++
zQa g s ++
Q
I l l
+1 + 1
+ +I ++
-5
+ ++ ++ + ++ ++ ++ ++ + +I++ +
+
$%"- Q) + +++++++++ + ++ + I+I + +
+ +++ + ++
I +++ + ++
I + + + + ++ I
+ +++ + ++
+ + + +
+
+
+
+
I +++++++++ + + I
+ +
+
+
+ +++ + +++
++ + +++ + +++
++ + I I + + + + +++
+ +++ + +++
+
+++
:+: I +
+ '==+
+ + + + I+
:I++ +
+ + + + I + + +
++++ I:+
++ +
+
++++ +++ +++ ++
I:++
++++
+++
++
+++
:++
I I ++
++
I+++ + + +++ + + +
+++ + l + ++++ + + +
:+++ + + + + + +I++
566 Sinha
crop plants against their fungal pathogens seemed worth exploring. De-
tailed observations are given disease-wise. Table l gives an approximate
idea of the magnitude of protective effect achieved against different crop
diseases with different compounds and Table 2 summarizes the nature of
changes in symptom expression achieved with strongly effective compounds
and the associated changes in host metabolism.
Brown Spot of Rice (Helminthosporium oryzae)
More than 50 compounds, mostly known phytoalexin inducersand a few
related compounds, had been testedon susceptible rice plants in different
pot experiments for their possible effectiveness against artificial inoculation
with the brown spot pathogen. In the initial experiments, plants were spray-
treated with aqueoussolutions of differentchemicals,most at M, and
a few at lower doses
or
M). Theconcentrationsinitially
tested
were closeto their optimumfor phytoalexin induction in plants. Such treat-
ment, administered 2 days before challenge inoculation at the age of 3
weeks by spraying conidial suspension to the leaves till dripping, resulted in
appreciable reductions in leaf symptoms in most cases. Rarely, a compound
totally failed. Only those compounds which could reduce symptoms moder-
ately to considerably werefurther tested at a range of three concentrations,
most at and lo-’ M, and cycloheximide and growth regulators
at lo-’, and M.
Of the three
methods of application initially
tested-i.e., presowing seed treatmentby soaking seeds in chemical solution
for 24 hr, 24-hr rootdip of seedlings in chemical solution before transplant-
ing to field plots, and foliage spray 2 days prior to artificial inoculation-
wet seed treatment appearedto provide a more significant as well as a more
persistent protective effect. In all subsequent studies seed treatment was
used as the preferred modeof application.
In seed treatment, compounds such as lithiumsulfate, sodium selenite,
thioglycollic acid, L-cysteine, p-chloromercuricbenzoic acid (p-cmb), and
cycloheximideshowedsubstantialprotectiveactionagainstbrown spot.
Thesereducedleafsymptomsbymore than 70%. Compounds which
showedsomewhatlesspronounced but stillverysignificanteffects and
reduced symptoms by morethan 50% and up to 70% included ferric chlo-
ride, nickelnitrate, cadmium chloride, barium chloride, sodium molybdate,
sodium iodoacetate, IAA, 2,4-D, DL-methionine, and DL-nor-leucine (Giri
and Sinha, 1983a, b). With many of the above compounds, their induced
protective action appeared to be concentration-independent and showed
little relation to their in vitro fungitoxic action, if any. These treatments
adversely affectedboth the components of symptom expression witha sub-
stantially reduced lesion number and reduced lesion size. Treated plants
developed comparatively fewer lesions of large size groupsand many more
G
5 F
OD
52
W
567
is
2
W
cd
2
5
c
vj
6
569
570 Sinha
of smaller size groups. Moderateto high levels of fungitoxicity that devel-

oped in the young rice seedlings in these treatments gradually declined and
disappeared totally between 3 and 4 weeks. Challenge inoculation at this
stage led the treated plants to freshly develop high to very high levels of
fungitoxicity in vivothat showed good correlationto the level ofprotection
achieved (Sinha and Hait, 1982). Such active response by the treated plants
was believed to be due to a conditioning of their tissue, basedon sensitiza-
tion, under the influence of chemical treatment, so that they respond to
challenge inoculation almost like resistant plants. Later, Hait and Sinha
(1987) provided strong biochemical evidenceto support this contention. In
their experiment both untreated susceptible and resistant plants as well as
susceptible plants receiving treatment with two highly effective compounds,
namely, sodium seleniteand L-cysteine, were inoculatedat the age of 2, 4,
6, and 8 weeks. Leaf materials from both untreated and treated plants 3
days after each inoculation were used for analysis. It was observed that
treated susceptible plants showed appreciably reduced total protein content
and increased P P 0 activity as compared to the untreated plants. Such
changes brought them closer to the conditions existing in resistant plants.
Treated susceptible plants responded to challenge inoculation with signifi-
cantly greater increase inboth total phenol and protein contents aswell as
in PP0 activities as compared to the control plants. Treated susceptible
plants showed almostthe same trends as those recorded for resistant plants.
The induced effectwas most pronounced in4-week-old plants but declined
thereafter. However,withsodiumselenitetreatment that yields a more
persistent effect, induced effects remained appreciable up to 6 weeks in
most respects, although not with L-cysteine. Further evidence of tissue con-
ditioning came fromgel and immune electrophoretic studies with plants in
the above treatments (Hait, 1982). In both L-cysteine and sodium selenite
treatments, plants appeared to have lost the three antigenic proteins that
they shared with the virulent isolate of H. otyzae. This was noted only in
selenite treatment but not with L-cysteineat 2 weeks; at 4 weeks the loss of
proteins was evident in both treatments, but not thereafter. A point of
further interest was that plants in both treatments showed twonew protein
bands corresponding to the two bands already present in resistant plants.
These observationsfit in well with the common antigen hypothesis (Doubly
et al., 1960; DeVayand Adler, 1976).
Later studies by Das (1993) recorded strong protective action on rice
plants with cupric chloride, sodium selenate, 2,4,5-T, and cycocel, and very
strong action with chitosan, a 0-1,I-linked glucosamine polymer derived
from chitin by deacetylation. Kar and Sinha (1988) reported good protec-
tion from brown spot also with propantheline bromide, an anticholinergic
drug, known as an effective inducerof phytoalexin (Hadwiger, 1972a).
Phytoalexin
Disease
Control
Chemicals
Inducer
and 571
Blast of Rice (Pyricularia oryzae)

Many of the compounds withstrong protective actionon rice plants against
brown spot disease were also tested against blast disease in multilocation
field trials at hot-spot locations inthe states of Tripura, West Bengal, and
Bihar. In most of these trials, infectionwas first detected after 5-7 weeks,
but in one case infection came quite late, i.e., after 13 weeks. To prepare
field plots for these experiments, nitrogen was added at a high dose, 100-
120 kg/ha. Test compounds that provided very strong protection from blast
in different trials included ferric chloride, cupric chloride, p-cmb, sodium
malonate, cycloheximide, 2,4-D, and chitosan (Sinha and Sengupta, 1986;
Das, 1993). Those with strong but less significant action were cadmium
chloride, lithiumsulfate, thioglycollic acid, L-cysteine, sodium selenite,so-
dium selenate, sodium molybdate, sodium iodoacetate, sodium fluoride,
DL-methionine, DL-phenylalanine,and IAA. Plants in effective treatments
showed many fewer blast lesions of reduced mean size, and lesions rarely
coalesced.Wheninfectionoccurredrather late, after 13 weeksin a trial,
symptoms were assessed2 weeks later, cupric chloride, p-cmb, and cyclohexi-
midereducedleafsymptoms in the range of 85-89%. In treatments with
sodium selenite, L-cysteine, and sodium malonate as well, the induced effect
was quite strong and60-71070 reductions in leaf symptoms were noted. Seed-
soaking treatmentfor 24 hr gave a better result than48-hr treatment.
Biochemical studies with leaf samples from untreated and variously
treated rice plants from the infected crop showed that plants treated with
highly effective cupric chloride and sodium selenate and fairly effective
IAA contained higher phenol and protein levels and more PO activity than
the untreated plants. With minor exceptions, correlation existed between
such increases and the magnitude of protection achieved with different
chemicals.
Sheath Blightof Rice (Rhizoctonia solani)
In limited pot trials of exploratory nature, three compounds with strong
effect against both brown spot and blast diseases of rice were also tested
against sheath blight disease that has gained much importance in recent
years. When 5-week-old pot-grown plants of susceptiblenature were artifi-
cially inoculated, those in all three treatments developed significantly re-
duced symptoms as compared to fairly heavy symptoms developed in the
untreated plants (Sarkar and Sinha, 1991). While cycloheximide showed
fairly strong protective action against sheath blight, the other two had only
moderate effects.Further studies are in progress.
Leaf Spotof Wheat (Helminthosporium sativum)
A large number of phytoalexin inducer chemicals were tested inpot trials
against leaf spot disease of wheat incited by Helminthosporium sativum.
572 Sinha
Out of about 40 compounds tested in wetseed treatment, the majority

could provide 3- to 5-week-old wheat seedlings with strong to very strong
systemic protection from artificial inoculation with H. sativum (Hait and
Sinha, 1986; Chakraborty and Sinha, 1989). Those reducing leaf symptoms
by 70% or more included the chlorides and sulfates of copper, zinc, mer-
cury, and barium; nitrates of copper, zinc, nickel, and iron; and nickel
chloride,sodiumsulfite,sodiummalonate,sodiumfluoride, and IAA.
Compounds such as lithium sulfate, L-cysteine, thioglycollic acid, sodium
azide, sodium iodoacetate, sodium molybdate, p-cmb, cycloheximide, DL-
phenylalanine, and 2,4-D also showed protective action that was strong
but slightly less significantthan that in the first group. These compounds
primarily prevented the successful establishment of infection, as reflected
in the very limited number of lesions developed, but many of them re-
stricted lesion enlargement also. At the age of2 weeks, diffusates from leaf,
in most of the above treatments, showed high levelsof fungitoxicity. This
toxicity didnot persist, declined gradually,and finally disappeared between
3 and 4 weeks. Inoculation at these stages ledto the development of strong
fungitoxicity in both treated and untreated plants, with the difference be-
tween their levelsnot sufficient to explain the difference in their leaf symp-
toms. The induced effect declined with time.
Damping Off of Chili (Pythium sp.)
Out of 15 compounds screened in small-scale nursery trials against damping
off of chili, most showed significant effects in protecting chili seedlings
from damage and mortality. The trials continued for 3 successive years.
Compounds found to be highly effective in presowing wet seed treatment
against the seedling disease include barium chloride, boric acid, sodium
selenite, thioglycollic acid, sodium sulfite, sodium molybdate, cyclohexi-
mide, and 2,4-D (Sahana, unpublished information). In one or more experi-
ments,thesecompoundsreducedseedlingmortalityby 60% or more.
Somewhat less significant effects were recorded with cadmium sulfate, L-
cysteine, sodium malonate,p-cmb, and IAA.
Fusarium Wilt of Pigeon Pea (Fusarium udum)
Only a few of the 14 chemicals tested in wet seed treatment against Fu-
sarium wilt of pigeon pea (i.e., chitosan, mercuric sulfate, cupric sulfate,
and sodium sulfite) could significantlycheck symptom expression(3748%)
and reduce plant mortality (41-49%) (Maity, 1991). Root exudates and root
extracts from susceptible plants that underwent such treatments showed
appreciable fungitoxicity, measured against germination of both conidium
and chlamydospore. Such toxicity in root extracts increased further in re-
sponse to artificial inoculation withF. udum. Upon challenge,treated sus-
Phytoalexin
Disease
Control
Chemicals
Inducer
and 573
ceptible plants also recorded significant increases in flavonol content and

peroxidase activity,and the increased levels came close to those of resistant
plants. Susceptible plants, in treatments with chitosan and mercuric sulfate
(the two most effective compounds), showedthe loss of two of five anti-
genic proteins they share with the virulent isolate of F. udum but have
acquired at the same time two new proteins correspondingto two already
present in resistant plants. This is in conformity withthe common antigen
hypothesis.
Fusarium Wilt of Chickpea (Fusarium oxysporumf. sp. ciceri)
In pot experiments, successfulcontrol of vascular wilt of chickpea has been
achieved by presowing wet seed treatment with some of the 40 chemicals
tested. Highly effective among them are mercuric sulfate, cycloheximide,
DL-phenylalanine, DL-methionine, IAA,2,4-D, 2,4,5-T, cycocel, and chito-
San (Sahana, 1991). When symptoms were finally assessed 5 weeks after
inoculation, the above treatments reduced disease incidence by 55-68%,
disease indexby 79-89% against severe wilting incontrol plants, and plant
mortality by 71-86%, and also limited both upward and lateral spread of
the fungus in the host vascular system. Root exudates from such treated
plants show considerable fungitoxicity and adversely affect both germina-
tion and germ tube growth of chlamydospores and conidia. The inhibitory
effect was more pronounced on germ tube growth. The toxicity was also
evident in the root extract and stem extract with further increases when
plants become infected. Metal salts such as cupric chloride, zinc chloride,
and nickel chloride also have good protective action against chickpea wilt,
though less significantthan the above compounds. Working with the highly
effective compounds, Chowdhury (1992) reported that susceptible chickpea
plants in these treatments respond to infection with significantly greater
increases in total phenols, O-dihydroxyphenol, and total protein contents,
as well as in P P 0 and PO activities. In these respects, responsesof treated
susceptible plants closely follow those of resistant plants.
Collar Rot of Chickpea (Sclerotium rolfsii)
The heavy mortalitythat characterizesS. rorfsiiinfection of chickpea plants
in pot experiments is substantially checked bywet seed treatment with the
sulfates and chlorides of copperand zinc, mercuricnitrate, L-cysteine, and
thioglycollicacid(Sahana, 1991). Thesereducedplantmortality by 55-
78%. In the treated plants, pectolytic enzyme (polygalacturonases) activity
of the pathogen in the rot-affected tissue was found to be significantly
lower than in the untreated plants. Compounds such as ferric chloride,
cupric nitrate, zinc nitrate, barium sulfate, and barium nitrate also are
fairly activeon chickpea plants in limiting collar
rot infection.
574 Sinha
Root Rot of Sugarbeet (Sclerotium rolfsii)

Results froma replicated fieldtrial with a mixed group of nine phytoalexin
inducers used for seed treatment before sowing in S. rolfsii-infected soil
showed that all except IAA could limit root rot of beet resulting from
natural infection and contribute in some measureto improved plant health
and better productivity(Das, personal communication). All-around strong
effects were recorded with 2,4,5-T, chitosan, zinc chloride, and 2,4-D in
that order. They reduced disease incidence by61-72%, plant mortality by
25-72%, and increasedroot yield (ton/ha) by 32-61% and sugar yield(ton/
ha) by 29-63%. Cycloheximide and cycocel effectively reducedboth disease
incidence and mortality by 55-65'70 and 46-50%, respectively, but caused
only moderate increases root
in yield and sugar yield.
Stem Rot of Groundnut (Sclerotium rolfsii)
In different series of experiments with pot-grown plants raised from both
untreated and variously treated groundnut seeds, barium chloride, barium
nitrate, zinc chloride, zinc sulfate, and lithium sulfate (Acharya, unpub-
lished), IAA, 2,4-D, 2,4,5-T, cycocel, cycloheximide, chitosan, L-cysteine,
and DL-phenylalanine substantially checkedthe progress of rotting resulting
from artificial soil inoculation with S. rorfsii. Plant mortality was mostly
markedly reduced(72-95%) in such effective treatments. Pectolytic enzyme
activity on which the pathogenic activityof this pathogenis based was also
significantlysuppressedinthesetreatments.Intreatmentswithorganic
compounds, affected tissue responds to challenge inoculation with accumu-
lation of calcium and magnesium, and also enhanced lignification. It has
further been shownthat in these treatments plants respond to infection also
with significantly greater increases total
in phenol, o-dihydroxyphenol, and
total protein contents, and with increased P P 0 and PO activity (Chowd-
hury, 1992).
Early Leaf Spot of Groundnut (Cercospora arachidicola)
In a series of fieldtrials conducted at two locations, susceptiblegroundnut
plants raised from seeds treated with different metal salts and boric acid
exhibited moderateto strong resistance to natural infection withC. arachid-
icola. In comparison, fairly heavy infection and symptoms developed in
plants raised from untreated seeds. Salts such as barium sulfate, barium
nitrate, mercuric chloride, and lithium sulfate exercised strong protective
effect and reduced leaf symptoms by 60% or more (Acharya, unpublished).
In another series of field experiments, seed treatment with growth regula-
tors, such as IAA, 2,4-D, 2,4,5-T, cycocel, as well as cycloheximide, ledto
substantial (62-67%) reductions in leaf symptoms (Chowdhuryand Sinha,
1989). Later studies also achieved very strong reductions (73-87'70) with
Phytoalexin
Inducer
Chemicals
and Disease Control 575
chitosan, L-cysteine,and p-Cmb and a somewhat lesser effect with sodium

azide (Chowdhury and Sinha, 1990). In all of the above cases, induced
protectiveeffect was systemic and long-lasting. In the treatmentswith
growth regulators, podyield was appreciably increased. Studyof inoculum
production (conidia) ona leaf at fixed position (sixthfrom the base) overa
period of 10 days revealed that in groundnut plants receiving varioustreat-
ments spore production appeared to be much limited as compared to that
in the untreated plants,and the differencewas quite significant. Generally,
the drop in inoculum production has good correlation with the induced
protectionachievedwith a treatment.Biochemicalstudiesshowed that
groundnut plants in different effective treatments responded to natural
infectionwithmarkedincreasesin total phenol and o-dihydroxyphenol
contents, and less pronounced but still significant increases in
total protein
content and both PP0 and PO activities (Acharya, unpublished; Chowd-
hury, 1992).
111. DISCUSSION ,
Results from both pot and field experiments using a large and diverse group
of phytoalexin inducer chemicals, mostly in wet seed treatment, against
some important crop diseases of fungal origin establish beyonddoubt that
such treatments can provide a rational and safe alternativeto conventional
chemical control using toxic compounds like fungicides that act directly on
the pathogen. Based on these observations, the following generalizations
seem possible:
1. While most of the phytoalexin inducer chemicals tested have the
potential for providing some degree of protection to different plant species
from their fungal pathogens,when used at rather low concentrations
to M), many of them show substantial suppressive effects on symptom
expression and plant morality.
2. Resistance induced in a particular plant species by such treatment
appears to be broad spectrum innature and may be active againsta variety
of fungal pathogens with different modes of pathogenesis as well as differ-
ent levels of parasitic specialization. Many of the inducer compounds,dif-
ferent both in their chemicalnature and their biological action, if any, can
induce apparently similar resistance in a particular plant species against its
one or more pathogens. The same compoundmay also be active on more
than one plant species against their different pathogens.
3. For most of the effective compounds, the induced protectiveeffect
has little relation eitherto their in vitro toxicity(if any) or to their concen-
tration gradient. Some show their maximum effect at the lowest concentra-
tion tested, often as low as M.
576 Sinha
4. The induced resistance developed in a plant species and the conse-

quent protective effect against its pathogen@) is systemic and mostly long-
lastingin nature. The inducedprotection mayremainactivealmost
throughout crop life.
5 . The effective compounds mostlyappear to act on the pathogen not
through any direct toxic action, any,if but through dynamic host-mediated
responses that create in vivoan environment unfavorablefor growth and/
or activity of the fungus involved. Supporting evidence comes from bio-
chemical studies.
6. In a highly effective treatment, susceptible host plant may lose all
or some of its antigenic proteins that it has in common with the virulent
isolate of the pathogen and/or acquire one or more new proteins corre-
sponding to protein@)the resistant plant already has.In terms of common
antigen hypothesis, such changes in the host indicate its shift to reduced
compatibility with the pathogen and/or increased resistance to it. With
respect to the parameters commonly associated witha plant host’s normal
defenseresponses,suchasphenolics,phytoalexin-typeantifungalsub-
stance, lignin, and protein contents, as well as phenylalanine ammonia-
lyase (PAL) PPO, and PO activities, susceptible plants in effective treat-
ments initially recorded small to moderate increases, which declined with
time. However, if and when challenged with pathogen, the same plants
recorded significantly greater increasesthan the control plants. Their post-
infection responses mostly show the same trends as characterize the resis-
tant plant.
7. Effectivecompounds appear to conditionsusceptiblehostplant
through activation of its latent defense potential and sensitization of its
tissue, so that it becomes competent and also ready to interact with the
challenger pathogen, if and when it would infect, with dynamicand vigor-
ous responses likethe resistant plant.
The above findings clearly establishthe fact that even susceptible plants
are not totally defenseless against a pathogen. Even in the absence of a
specific gene for resistance, they also have some genetic information for
defense reactions. This meansthat all plants have the genetic potentialfor
disease resistancelatent in them,and this can be activated and made opera-
tive by treatment with many different bioticor chemical agents. It is inter-
esting to note that much similarity exists between the resistances induced by
these agents. This has been experimentally demonstrated on wheat plants
against Drechslera sorokiniunu (Chakraborty and Sinha, 1984) and on cu-
cumber against four fungal, two bacterial,.and two viral pathogens (Mu-
charromah and KuC, 1991). With both kinds of agents, induced resistance
is time-dependent, systemic, and long-lasting. It is also nonspecific in re-
spect to both the inducer agentsand the pathogens againstwhich it is active.
Phytoalexin
Disease
Control
Chemicals
Inducer
and 577
In both cases, the inducedresistanceappears to havedevelopedin the

host through a processof conditioning based on tissue sensitization. Such
conditioned susceptible host tissue responds to challenge inoculation or
infection withthe pathogen with physiological changes that mostly indicate
the same characteristic trends of the natural defense responses of the host
species. These responses mostly involve stimulated production of phenolics,
phytoalexin,lignin,protein,hydroxyproline-richglycoprotein, and in-
creased activity of PAL, chitinase, /3-1,3-glucanase, PO, and PP0 among
others. Not all but someof them get activated and combineto provide the
treated susceptible plant with a multicomponent and multilayered protec-
tive system in which the host tissue, now sensitized and made competent
and ready to react dynamically, responds with changes in different compo-
nents as may be requiredto contain the particular pathogen. The versatility
of such induced resistance, i.e., systemic acquired resistance (SAR) in the
treated plant, can only be explainedby such coordinately regulated multi-
componentsystems.Resultsfromourstudieswithphytoalexininducer
chemicals and also from other studies basedon compounds withno known
recordofphytoalexininductioninplants(Langcake,1981;Lazarovits,
1988) support this view. Various observations suggest that while many of
the above components may be activated as a part of the tissue sensitization
process following treatment and record small to moderate increases, these
may constitute onlya broad-spectrum protective effect of mildnature that
is effective against morethan one pathogen. When challenged with differ-
ent pathogens, only a few of thee components will be moderately or vigor-
ously stimulatedand these will havea primary role in containing the patho-
gen.Themagnitudeofresistanceultimatelyexpressedagainstdifferent
pathogens may then dependon thequality of the signal, i.e., nature of the
elicitor, and its intensity, i.e., concentration of the elicitor, received from
the pathogen. While the quality, i.e., the chemical nature of the elicitor
from pathogen, will probably determine the defense componentsto be fur-
ther activated, its strength will regulate the speed of response with respect
to such components,and this will determinethe level of protection achieved
against a particular pathogen. The multicomponent nature of SAR in a
plantmayhaveanotheradvantage,i.e.,thedifferentcomponentsmay
adversely affect the pathogen at different stages* its interaction with the
host, as reported for chickpea wilt (Chowdhury, 1992).
Beside the large group of phytoalexin inducers tested by us, significant
levels of protection of susceptible plants against their fungal pathogens
have also been reportedto be achieved with compounds like sodium azide,
probenazole, WL28325, TEPA, gibberillic acid, cloxacillin, and penicillin
among others (Langcake, 1981; Lazarovits, 1988). While sodium azide was
the only known phytoalexin inducer among them, some others like piperi-
578 Sinha
dinium iodide, pyrrolidinium iodide, probenzaole,and trifluralin-likeher-

bicides also appear to induce the production of a phytoalexin-type fungi-
toxic substance in healthy plants, though at low levels. However, plants in
most of these treatments, except with cloxacillin and penicillin, responded
to challenge with the pathogen with significantly high levelsof such toxic
compounds. It is felt, however, on the basis of aboveresults that our
search for sensitizer-type compounds need not be restricted exclusively to
phytoalexininducersandtheirrelatedcompounds.Onthebasis of his
extensive studies on blast of rice with more than 150 chemicals, Rathmell
(1982) concluded that any compound withthe ability to cause tissue necro-
sis in plants would be ableto protect rice plantsfrom blast disease. Almost
simultaneously, KuC (1982) also concluded that both biotic and chemical
agents that effectively induce resistance in plants have the common prop-
erty of creating, through a mild metabolic perturbation, a common stress
of persistent nature in the host. Only rarely has any adverse effect on seed
germination or seedling health been recorded with a phytoalexin inducer
chemical tested by us. Visible localized necrosis may occur when a live
inoculum is used as the inducer agent but not generally with a microbial
fluid or a dilute chemical solution used or seed treatment. In the case of
chemical agents, the induced metabolic perturbation may not be such high
magnitude as to cause cell necrosis but maybe adequate to result in a
certain byproduct(s)that functions asan endogenous elicitorand triggers a
mild resistance response locally (i.e., in the neighboring cells). Either that
byproduct or a secondary endogenous elicitor is translocated as a signal
carrying information to distant parts of the host plant where it conditions
the cells making them competent and also alert, through the process of
sensitization. Such conditioned cells respond dynamically to infection like
the cells of resistant plant. In the case of biotically induced resistance in
plants, there is evidence of transmission of an immunity signal in cucumber
(Jenns and KuC, 1979) and tobacco (KuC and Tuzun, 1983). In the latter,
movement of such a signal through phloem tissue has also been reported
(Jenns and KuC, 1977). In no caseof chemically induced resistance has such
evidence been available until now. Compounds postulatedto have a role as
transmissible immunity signal include ethylene (VanLoon, 1982), salicylic
acid (Ward et al., 1991), and oligogalacturoronide fractions of plant cell
wall (Uknes et al., 1992). Both ethylene and salicylic acid can themselves
induce many defense-related responses in plants and make them acquire
SAR. However, no compelling evidence is available in support of such a
role for them. Were such a role possible, these could probably be directly
used for seed treatment or foliage sprayfor the purpose of diseasecontrol.
While screening a large number of synthetic organic compounds to
identify phytoalexin inducers, Hadwiger and associates notedthat the effec-
Phytoalexin
Inducer
Chemicals
and Disease Control 5 79
tive ones concomitantly induced new synthesis of RNA and protein and
increased PAL activity, and these were correlated with phytoalexin synthe-
sis (Hadwiger, 1971, 1972a, b; Hadwiger and Schwochau, 1970; Hess and
Hadwiger, 1971). It was concluded that such a property of the effective
compounds was related to their ability to bring about specific changes in
the conformation of host cell DNA by intercalating into it or complexing
with it. Heavy metals can also do the same job by complexing the host
DNA. However, most of the phytoalexin inducers we have used in our
attempts to induce resistancedo not have any abilityto complex with DNA
(Hadwiger, personal communication), except for chitosan which showed
very strong effect against all the diseases tested.It has a strong affinity with
cell DNA. Chitosan protects pea pod tissue from its pathogen, Fusarium
solani f.sp. pisi, as strongly as does prior inoculation withF. solani f.sp.
phaseoli, a nonpathogen.
Within 15 min of such treatment or inoculation withthe latter, chitosan
can be detected within host cell cytoplasm and nucleus. Chitosan induces in
pea tissue, besides phytoalexin production, increased protein synthesis and
PAL activity (Hadwigerand Beckman, 1980) as well as enhanced lignifica-
tion (Pearce and Ride, 1982)-all componentsof the host’s natural defense
response that are also induced by infection with F. solani fsp. phaseoli.
This shows that both biotically and chemically induced resistance function
almost similarly. In our extensive studies with phytoalexin inducer chemi-
cals for induction of host resistance in plants, it may seem quite natural to
place the major emphasis on phytoalexin as the causal factor. However,
most of them also coordinately induce the increased synthesis of phenolics,
particularly o-dihydroxyphenols, proteins, and lignin, and enhanced activ-
ity of PAL and PO (all componentsof both induced and natural resistance
in plants), thus broadening its scope of action. Some organic compounds
like probenazole, TEPA, and WL28325, notknownyetasphytoalexin
inducers, induce strong resistance in some crop plants, and the protected
plants respond to challenger pathogen with increased leveland/or activity
of some of the above defense components including phytoalexin synthesis
(Vo-Thi-Hai et al., 1979; Iwata et al., 1980; Cartwright et al., 1980; Shi-
mura et al., 1981). This does not happen in all cases. Phytoalexin inducer
compounds may have certain advantages over others as the effective in-
ducer of resistance because they stimulate early protein synthesis and PAL
activity, a key enzyme in aromatic metabolism that regulates not only syn-
thesis of some phytoalexinsbut also that of many toxic phenolsand lignin.
Increased protein synthesis constitutes one of the responses of many
plants with SAR to infection with their pathogen. This may involve in-
creased synthesis of existing protein or de novo synthesis of some new
protein including pathogenesis-related protein(s) (PR protein). Such pro-
580 Sinha
teins have been reported from tobacco and some other plants acquiring
SAR (mostly against viral pathogens) by treatment with acetylsalicylic acid,
aspirin, benzoic acid,2,4-D, IAA, polyacrylic acid, etc. (Bozarthand Ford,
1988 and references therein). Some of them belong to know class of pro-
teins, such as chitinase, p-l,3-glucanase, etc., with suggested involvement
in induced resistance,but not with any clear evidence. Further evidence of
possible involvement of proteins in induction of resistance in plants came
from studies on the antigenic relationship between host cultivars and patho-
gen, commonness in antigenic protein standing for compatibility, and dis-
parity for incompatibility between them (Doubly et al., 1960; DeVay and
Adler, 1976). Investigations on chemically induced resistance in rice-Hel-
minthosporium oryzae (Hait, 1982), pigeon pea-Fusarium udum (Maity,
1991), soybean-Macrophomina phaseolina (Chakrabortyand Purkayastha,
1987)/Colletotrichum dematium (Purkayastha and Bannerjee, 1990)
showed that a susceptible plant acquiring resistance lost all or some of the
antigenic proteinsit had in common with the pathogen, indicatinga shift to
reduced compatibility, but also developed (in rice and pigeon pea) one or
more new proteins correspondingto some constitutively present in resistant
plant, indicating a shift to greater resistance. Since alteration of specific
antigen@) ina susceptible host plant by chemical treatment appears to be
linked with induced resistance, their immunochemical identification and
regulation may provide a basis for their use in plant disease control. Pur-
kayastha (1973) suggested a close relation between antigens, phytoalexins,
and disease resistance.
Recognition of the potential pathogen by the host at the pathogen
elicitor-host receptor level as “nonself” is deemed as the primary eventthat
triggers a cascade of host responses leadingto the expression of resistance
(Callow, 1977; Yoshikawa, 1983). In the case of SAR in tomato to Phy-
tophthora sp. (Vo-Thi-Hai et al., 1979) and in rice to Pyricularia oryzae
(Sekizawa and Mase, 1980) developed,respectively, by treatmentwith
TEPA and probenazole,ithasbeensuggested that specific structural
changes at the host-receptor surface following chemical treatment make
such recognition of the pathogen by the susceptible plant possible. No
clear evidence is available in support; however, the fact that many of the
compounds can induce effective resistance in one or more plant species
against oneor more of their pathogens makes specific structural changes at
the receptor level that would allow such recognition seem unlikely.
It is evident from the various observationsthat even susceptible plants
having no gene for resistance specifically active againsta particular patho-
gen have some genetic information for a generalized resistance against most
of its pathogen. Such defense normally remains latent but can be activated
by prior restricted inoculation withan infectious agentor treatment with a
Phytoalexin
Disease
Control
Chemicals
Inducer
and 581
sensitizer-type chemical. According to KuC (1987), resistance in plant is

determined by the speed and magnitude with which genes for resistance
mechanism are expressed and the activity of gene productsrather than the
presence or absence of such a gene. In spite of the undoubted promise of
induced resistance, there has been no genetic analysis of the immunized
plants. Weknow nothing about how the genes are activated. Recently,
Ward et al. (1991) reported that the onset of SAR in tobacco to tobacco
mosaic virusby restricted inoculation with the same virusor treatment with
salicylic acid or 2,6-dichloroisonicotinic acid (INA), an immunizing agent,
is correlated with coordinate gene expression and induction of nine classes
of mRNA. Later, Uknes et al. (1992) made similar observations with re-
spect to SAR in Arabidopsis sp. to Pseudomonas syringaepv. tomato and
Peronospora parasitica by treatment with salicylic acidand INA or infec-
tion with the pathogen. In both types of treatment, onset ofSARwas
associated with high-level accumulation of three proteins and that genes
corresponding to these proteins could also be induced by the same treat-
ments. On pea pod tissue, chitosan,a natural elicitor of fungal origin, has
been shownto induce many defense-related responses including the synthe-
sis of 20 major proteins as can also be achieved by infection withFusarium
solani f.sp. phaseoli, a nonpathogen (Hadwiger and Waggoner, 1983). Dis-
ease resistance responses at the protein or mRNA level indicate that the
number of genes associated with the disease resistance response is quite
large. It has been shown that chitosan authenticallymimics F. solani f.sp.
phaseoli in enhancingthejn vivo and in-vitro synthesis of all these proteins
associated with host resistance in pea. So Hadwiger and Wagoner (1983)
proposed that it would be useful to identify those genes whose activation
most closely corresponds with resistance reaction and to investigate their
regulatory properties. Onlyafter it has been shown that the specific func-
tion of a gene contributes to or regulates host resistance or this has been
genetically verified, can beit accepted as a disease resistance gene.
IV. EPILOG
Immunity providesthe best form of defensefor a plant. Substantial experi-
mental findings leave littledoubt about the fact that strong systemic resis-
tance of a lasting nature can be induced in a susceptible plant cultivar
against its pathogen by prior restricted inoculation with infectious agents
of both homologous and heterologous nature, prior treatment with fluids
containing their metabolites or cell constituents, or even treatment with
synthetic chemicals. With its systemic and persistent nature, time depen-
dence for optimization of the effect, and dynamic response to challenge
with the pathogen, if and when that would occur, such biologically or
582 Sinha
chemically induced resistance outwardly resembles immunization of ani-

mals. However, induced resistance plantsin with its nonspecificand broad-
spectrum effect, differs from immunization of animals, which characteristi-
cally results in strong, specific action against a particular pathogen based
on highly specific antigen-antibody reaction.
Great similarity exists betweenthe biologically and chemically induced
resistance in plants, at least in the expressive phase, though some differ-
ences may be there in the initial and/or determinative phase. Chemical
agents may enjoy certain advantages over biotic agents in respect of their
early and better absorption into the plant system,rapid distribution into its
different parts, and also longer persistence in host cells. Extensive work by
KuC and associates (KuC, 1987; Mucharromah and KuC, 1991) on tobacco,
cucumber, and melon illustrates both the characteristic featuresand effec-
tivenessofbiologicallyinducedresistance.However, the useof biotic
agents suffers from certain limitations. Costs involved in maintaining a
laboratory with adequate technicalhands for the production of large
amounts of inoculumand logistical difficulties faced in properly delivering
suchinoculum to the corp, particularly under conditions oflarge-scale
cultivation, make itnot only cumbersomebut uneconomic and noncompet-
itive. Also, such treatments do not fit in with the modern concept of agri-
culture. The above disadvantages may be overcome or bypassed by using
synthetic,nontoxicchemicalswhichcanmimic the signals from biotic
agents in triggeringa series of defense-related responses. Cultivators accus-
tomed to the useoftoxicchemicals for conventionalchemical control
would not find any difficulty in using the nonconventional chemicals. Their
nonhazardous nature, simple use in seed treatment at micro doses, and
broad-spectrum protective effect against a variety of plant pathogens both
of different kindsand with different levels of parasitic specialization mark
them as having great promise in plant disease control. These compounds
act on the pathogen not directly but through host-mediated responses based
on sensitization of susceptible host tissue and activation of its latent defense
potential. Susceptible plants with SAR, when challenged the withpathogen,
respond dynamically like the resistant plants, limiting symptom expression
as a consequence. The conditioned and protected plants do not as such
show any great change in any one of the defense-related responses but
respond dynamically and vigorously to infection with significant increases
in few or more componentsof defense responses, i.e., their responses occur
at the right time and the right location where needed. Thus, any wasteful
expenditure of energy is prevented, i.e., the induced resistance is also en-
ergy-efficient. This would help in the cultivation of commercially desirable
but defense-susceptiblecrop varieties in preferenceto resistant ones,which
are mostly not very good yielders.
It has been admitted that neither phytoalexins as such or their biotic
elicitors canbe of any use in plant diseasecontrol. Use of biotic agents also
has many snags. Insteadof wasting time on such bioticproducts, it would
be sensible to look for synthetic chemicals which can mimic the natural
signals from the biotic agentsand trigger most, if not all, of the multicom-
ponent defense responses of the host. Among the chemicals we tested,
cupric chloride, lithium sulfate, L-cysteine, thioglycollic acid, barium SUI-
fate, zinc chloride, sodium malonate, cycloheximide,p-Cmb, I U , 2,4-D,
2,4,5-T, cycocel, and DL-phenylalanineshowed strong inhibitory effect
against more than one disease on different crops. Chitosan had strong
effect against all the diseases tested. All of them are known phytoalexin
inducers. However, search for sensitizer chemicals need not be kept re-
stricted to phytoalexin inducers only; others may also have similar effect
(Langcake, 1981; Lazarovits, 1988). However, phytoalexin inducer chemi-
cals may enjoy some initial advantages over other chemicals in this respect.
Despite the many advantages, the use of sensitizer chemicals has yet to
be accepted as a safe and effective alternative to conventional chemical
control. This isa time for plant pathologists to realize their immense poten-
tial as a tool for plant disease control. With such synthetic chemicals, they
can possibly plana subtle, natural, and selective strategyfor this purpose.
Though mainly tested and found to be effective against fungal pathogens,
indications are that they may also be effective against bacterial and viral
pathogens for which no good control measure is yet known. Lazarovits’s
(1988) assertion that chemically induced resistance doesnot have the same
broad base of protection as that induced by biotic agents doesnot appear
to be validin the light of more recent observations (Mucharromah and KuC,
1991).
ASthe conventional chemicalcontrol constitutes an integral component
of modern agriculture, its large-scale replacement with the use of noncon-
ventional, sensitizer-type chemicals outis of the question, but their increas-
ing use overthe years may moreand more reduceour dependence on toxic
compounds. That would not be a small gain.
Immunization has openedan uncharted area of molecular biology re-
lated to the role of stress and infection on a plant’s genome and its expres-
sion. Besides its practical utility, exploratory studies may also shed Some
light on its scientific significance.What are the basic eventsat a molecular
level that trigger a series of responsesthat regulate and shape the induction
Of resistance in susceptible plants? Characterization of the immunity sig-
nal, factors that regulate its production,its cell-to-cell and long-range corn-
munication, and the hormonal mechanism that makes the plant respond
to stress are points that need investigationand elucidation at the molecular
level.
584 Sinha
REFERENCES
Arimoto, Y., Homma, Y., Ohtsu, N., and Misato, T.(1976). Studies on chemically
induced resistance of plants to disease. I. The effect of soaking rice seeds in
dodecyl-DL-alanine hydrochloride on seedling infection byPyricularia oryzae.
Ann. Phytopathol. Soc. Jpn. 42: 397-400.
Bailey, J. (1982). Physiological and biochemical events associated with the expres-
sion of resistance to disease. Active Defence Mechanisms in Plants (Wood,
R. K.S. ed.), Plenum Press, New York, pp. 39-65.
Bell, A. A., andPresley, J. T. (1969). Heat-inhibited or heat-killedconidia of
Verticillium albo-atruminduce disease resistance and phytoalexin synthesis in
cotton. Phytopathology 5 9 1147-1151.
Bertrand, A., Ducret, J., Debourge, J. C., and Homere, D. (1977). Etude des
propriets d’une novelle famillede fongicides: less monoethyl phosphites metal-
liques. Characteristiques physiochimiques et propiites biologiques. Phytiatr.
Phytopharm. 26: 3-18.
Bompeix, G., Ravise, A., Raynal, G., Fettouche, F., and Durand, M. C. (1980).
Modalites de l’obtention des necroses-bioquantes sur feuillies detachees de to-
mate par l’action du tris-0-ethylphosphonated’aluminium (phosethyl d’alumi-
num), hypotheses sur son mode d’action in vivo. Ann. Phytopathol. 12: 337-
351.
Bozarth, R. F., and Ford, R. E.(1988). Viral interactions: induced resistance (cross
protection) and viral interference among plant viruses. Experimental and Con-
ceptual Plant Pathology (R. S. Singh, U. S. Singh, W. M. Hess, and D. J.
Weber, eds.), Oxford-IBH, New Delhi, pp. 551-567.
Buchenauer, H., and Erwin, D. C. (1973). Reduction of the severity of Verticillium
wilt of cotton by the growth retardants, N,N-dimethyl pyrrolidinium iodide
associated with the production of antifungal compounds. 2. Pflanzenkrank-
heit. Pfanzenschutz8 0 576-586.
Buchenauer, H., and Erwin,D.C. (1976). Effect of plant growth regulator pydanon
on Verticilliumwilt of cotton and tomato.Phytopathology 66 1140-1 143.
Callow, J. A. (1977). Recognition, resistance and the role of plant lectins in host-
parasite interactions. Ad. Bot. Res. 4 1-49.
Carrasco, A., Boudet, A. M.,andMarugo, G. (1978). Enhancedresistanceof
tomato plants to Fusarium by controlled stimulation of their natural phenolic
production. Physiol. Plant Pathol. 12: 225-232.
Cartwright,D.,Langcake, P., Pryce, R. J., and Ride, J. P. (1977). Chemical
activation of host defence mechanism as a basis for crop protection. Nature
267: 511-513.
Cartwright, D., Langcake, P., and Ride, J. P. (1980). Phytoalexin production in
rice and its enhancement by a dichloropropane fungicide. Physiol. Plant Pa-
thol. 47: 259-267.
Caruso, F. L., and KuC, J. (1977). Field protection of cucumber, watermelon,and
muskmelon against Colletotrichurn lagenarium by Colletotrichum Iagenarium
Phytoalexin
Disease
Control
Chemicals
Inducer
and 585
Cassels, A. C., and Flynn, T.(1978). Studies on polyacrylic acid induced resistance
to viral and non-viral plant pathogens.Pest. Sci.4 365-371.
Chakraborty, B. N., and Purkayastha, R. P.(1987). Alteration in glyceollin synthe-
sis and antigenic pattern after chemical induction of resistance in soybean to
Macrophomina phaseollina. Can.J. Microbiol. 33:835-840.
Chakraborty, D., and Sinha, A. K. (1984). Similarity between the chemically and
biologically induced resistance in wheat seedlingsto Drechalera sorokiniana.2.
Pfanzenkrankheit. Pfanzenschutz91: 59-64.
Chakraborty, D., and Sinha, A. K. (1989). Differential effects of anionic forms of
selected heavy metal salts in seed treatment on Helminthosporium infection of
wheat seedlings. Indian Phytopathol. 42: 157-160.
Chamberlain, J. W., and Paxton, J. D. (1968). Protection of soybean plants by
phytoalexin. Phytopathology 5 8 1349-1350.
Chowdhury, A. K. (1992). Management of selected fungal diseases of groundnut
and chickpea by the useof unconventionalchemicals. PhD thesis, Bidhan
Chandra Krishi Viswavidyalaya, Kalyani.
Chowdhury, A.K., and Sinha, A.K. (1989). Control of early leafspot of groundnut
by treatment with a select group of chemicals.Znt. Arachis Newslett.6: 16-18.
Chowdhury, A.K., and Sinha, A. K. (1990). Control of foliar diseases of groundnut
by seed treatment with non-conventional chemicals. Znt. Arachis Newslett. 8:
16-17.
Cruickshank, I. A. M. (1963). Phytoalexins. Ann. Rev. Phytopathol. 1: 351-374.
Cruickshank, I. A. M. (1980). Defences triggeredby the invader: chemical defence.
Plant Disease: An Advanced Treatise, Vol. 5 (J. G . Horsfall and E. B. Cowl-
ing, eds.), Academic Press,New York, pp. 247-267.
Darvill, A. G., and Albersheim, P. (1984). Phytoalexins and their elicitors. Ann.
Rev. Plant Physiol. 35:243-275.
Das, A. R. (1993). Resistance induced in rice plantto brown spot andblast diseases
by seedtreatment with chemicals andthe associated host responses.PhD thesis,
Bidhan Chandra Krishi Viswavidyalaya, Kalyani.
Davis, D., and Dimond, A. E. (1953). Inducing disease resistance with plant growth
regulators. Phytopathology, 43: 137-140.
DeVay, J. E., and Adler, H.E. (1976). Antigens common to host and parasite.
Ann. Rev. Microbiol. 3 0 147-163.
Dimond, A. E. (1963). The modes of action of chemotherapeutic agents in plants.
Perspectives of Biochemical Plant Pathology (S. Rich, ed.), Connecticut Ag-
ricultural ExperimentalStation Bulletin 663, pp. 62-77.
Dixon, R. (1986). The phytoalexin responses: elicitation, signalling and control of
host gene expression.Biol. Rev. Cambridge Phil. Soc.61: 239-292.
Doubly, J. A.,Flor, H. H., andClagett, C. D. (1960). Relationofantigen of
Melampsora Iini and Linum usitatissimum to resistance and susceptibility. sei-
ence 131: 229.
Fawcett, C. H.,Spencer, D. M., and Wain, R. L. (1969). The isolation and proper-
ties of a fungicidal component present in seedlings ofViciafaba. Netherlands
J. Plant Pathol. 75:72-81.
586 Sinha
action of fosetyl-Aland phosphorus acid. Phytopathology 75: 1064-1068.
Ghosal, A., and Purkayastha, R. P. (1984). Elicitation of momilactoneby gibberil-
lin in rice. Curr. Sci.53: 506-507.
Giri, D. N., and Sinha, A. K. (1983a). Effects of heavy metalsalts on susceptibility
of rice seedlingsto brown spot disease. Ann. A& Biol. 103: 229-235.
G i , D. N., and Sinha, A. K. (1983b). Control of brown spot disease of rice
seedlings bytreatment with agroup of chemicals.Pfanzenkrankheit. Pflanzen-
schutz 9 0 479-487.
Goodman, R. N. (1980). Defences triggered by previous invaders: bacteria. Plant
Disease: An Advanced Treatbe, Vol. 5 (J. G. Horsfall and E. B. Cowling,
eds.), Academic Press, New York, pp. 305-317.
Grinstein, A. N., Lisker, N.,Katan, J., and Eshel Y. (1976). Effect of dinitroaniline
herbicides on plant resistance to soil-borne pathogen.Phytopathology 66:517-
522.
Grinstein, A. N., Lisker,N., Katan, J., and Eshel, Y. (1981). “Trifluralin,” a
sensitizer for Fusarium resistance in tomatoes and cotton. Phytoparasitica 9
235-236.
Guest, D. I. (1984). Modification of defence responses in tobacco and capsicum
following treatment with Fosetyl-Al [aluminum tris (0-ethyl phosphonate)].
Physiol. Plant Pathol.25: 125-134.
Gutter, Y. (1983). Supplementary anti-mold activity of phosethyl-Al,a new brown
rot fungicide for citrus fruits. Phytopathol. Z. 107: 301-308.
Hadwiger, L. A., and Schwochau, M. E. (1971). Specificity of deoxyribonucleic
acid intercalating compounds in the control of phenylalanine ammonia lyase
and pisatin levels. Plant Physiol.47: 346-351.
Hadwiger, L. A. (1972a). Increased levels of pisatin and phenylalanine ammonia
lyase inPisum sativumtreated with antihistaminic, antiviral, antimalarial, tran-
quilizing or other drugs. Biochem. Biophys. Res.Commun. 46: 71-79.
Hadwiger, L. A. (1972b). Induction of phenylalanine ammonia lyase and pisatin by
photosensitive psoralen compounds.Plant Physiol.49: 779-782.
Hadwiger, L. A., and Beckman, J. A. (1980). Chitosan as a component of pea-
Fusarium solaniinteractions. Plant Physiol.66 205-21 l.
Hadwiger, L. A., and Schwochau, M. E. (1970). Induction of phenylalanine ammo-
nia lyaseand pisatin in pea pod by poly-lysine, spermidine,or histone fractions.
Biochem. Biophys. Res. Commun, 38: 683-691.
Hadwiger, L. A., and Wagoner, W. (1983). Electrophoretic patterns of pea and
Fusarium solani protein synthesized in-vitro or in-vivo which characterize the
compatible and incompatible interactions. Physiol. Plant Pathol.15: 211-218.
Hait, G . N. (1982). Nature of see&borne infection in selected crop diseases and
S,
control of seedling damage by see treatment with non-toxic chemicals. PhD.
Thesis, BidhanChandra Krishi Visw idyalaya, Kalyani.
Hait, G. N., and Sinha, A. K. (1986). Protkction of wheat seedlings from Helmin-
thosporium infection by seed treatment‘ with chemicals.J. Phytopathol. 115:
97-107.
Phytoalexin
Disease
Control
Chemicals
Inducer
and 587
Hait, G. N., and Sinha, A. K. (1987). Biochemical changes associated with induc-
tion of resistance in rice seedlingsto Helminthosporium oryzae by seed treat-
ment with chemicals.Z.Pflanzenkrankheit. Pflanzenschutz 9 4 360-368.
Hamilton, R. 1. (1980). Defence triggeredby previous invaders: viruses.Plant Dis-
eases:An Advanced Treatise, Vol. 5 (J. G. Horsfall and E. B. Cowling, eds.),
Hess, S. L., and Hadwiger, L. A. (1971). The induction of phenylalanine ammonia
lyase and phaseollin by 9-amino acridine and other deoxyribonucleic acidinter-
calating compounds. Plant Physiol.4 8 197-201.
Iwata, M., Suzuki, Y., Watanabe, T., Mase, S., and Sezikawa, Y. (1980). Effect of
probenazole on the activities ofenzymes related to the resistant reaction in rice
plants. Ann. Phytopathol. Soc. Jpn.4 6 297-306.
Jenns, A. E. and KuC, J. (1977). Localized infection with tobacco necrosis virus
protects cucumber against Colletotrichum lagenarium. Physiol. Plant Pathol.
11: 207-212.
Jenns, A. E., and KuC, J. (1979). Graft transmission of systemic resistance of.
cucumber to anthracnose induced by Colletotrichum Iagenarium and tobacco
necrosis virus. Phytopathology, 69 753-756.
Jenns, A. E., Caruso, F. L., and KuC, J. (1979). Non-specific resistanceto pathogen
induced systemically by local infection of cucumber with tobacco necrosis vi-
rus, Colletotrichum lagenariumor Pseudomonas lachrymans. Phytopathologia
Mediterranea 18: 129-134.
Kar, S. C., and Sinha, A. K. (1988). Drugs suppress brown spot infection of rice.
Science and Culture,5 4 276-277.
Keen, N. T. (1975). Specific elicitorsof plant phytoalexin production: determinants
of race specificity inthe pathogen? Science, 187: 74-75.
Keen, N. T.(1986). Phytoalexins and their involvement in plant disease resistance.
Iowa StateJ. Res. 6 0 477-.
Keen,N. T., and Legrand, M. (1980). Surface glycoproteins: evidence that they
may function as race-specific elicitors of Phytophthora megasperma f. sp.
glycinea. Physiol. Plant Pathol.17: 175-192.
Kirino, O., Oshita, H., Oishi, T., and Kato, T. (1980). Preventive activity of N-
allylamino acids against Fusarium diseases and their mode of action. Agric.
Biol. Chem. 4 4 35-40.
Klarman, W. L., and Gerdemann, J. W. (1963). Induced susceptibility in soybean
plants genetically resistant to Phytophthora sojae. Phytopathology, 53: 863-
864.
KuC, J. (1982). Plant immunization mechanismsand practical implications. Active
Defence Mechanisms in Plants (R.K. S. Wood, ed.), Plenum Press,New York,
pp. 275-298.
KuC, J. (1987). Plant Immunization and its applicability to disease control. Innova-
tive Approachesto Plant Disease Control(I. Chet, ed.), Wiley, New York, pp.
225-274.
KuC, J. (1990). A case for self defence in plants against diseases. Phytoparasitica
1 8 3-8.
588 Sinha
KuC, J., and Tuzun, S. (1983). Immunization for disease resistance in tobacco.Rec.
Adv. Tobacco Sci.P 179-213.
Lafon, R., Bugaret, Y.,and Bulit, J. (1977). Nouvelles perspectives de lutte contre
le mildiou de la Vigne [Plasmopara viticola(B.C.) Berlese et de Toni] avec un
fongicide systemique, l’ethylphosphite d’aluminium. Phytiatr. Phytopharm.2 6
19-40.
Langcake, P. (1981). Alternativechemicalagents for controllingplantdiseases.
Trans. R. Soc. London B295 83-101.
Lazarovits, G. (1988). Induced resistance: xenobiotics. Experimental and Concep-
tual Plant Pathology(R.S. Singh, U. S. Singh, W. M. Hess, and D. J. Weber,
eds.), Oxford-IBH, New Delhi, pp. 575-592.
Maity, B. R. (1991). Studies on Fusarium resistance in pigeon peaand its induction
by chemical treatment. PhD thesis, Bidhan Chandra Krishi Viswavidyalaya,
Kalyani.
Mansfield, J. W. (1982). The role of phytoalexin in disease resistance.Phytoalexins
(J. W. Mansfield and J. A. Bailey, eds.), Blackie, London, pp. 319-322.
Matta, A. (1971). Microbial penetration and immunization of congenial host plants.
Ann. Rev. Phytopathol. 9 387-410.
Matta,, A. (1980). Defence triggered by previous diverse invaders. Plant Disease:
An Advanced Treatise, Vol. 5 (J. G . Horsfall and E. B. Cowling, eds.), Aca-
demic Press, New York, pp. 345-361.
Mucharromah, E., and KuC, J. (1991). Oxalate and phosphateinducesystemic
resistance against diseases caused by fungi, bacteria and viruses in cucumber.
Crop Prot. 10 265-270.
Mukhopadhyay, S., and Sinha, A. K. (1980). Spore germination fluid as inducer of
resistance in rice plants against brownspot disease. Trans. Br. Mycol.Soc. 74:
69-72.
Muller, K. O., and Borger, H. (1940). Experimentelle Untersuchungen uber die
Phytophthora resistenz der kartoffel. Arbeiten. Biol. Reichsanst. Land. Fors-
twirtsch. 23:189-23 1.
Ossowski, O., Piloth, A.,Garreg, P., and Lindberg,B. (1984). Synthesisof a
glucoheptose that elicit phytoalexin accumulation in soybean. J. Biol. Chem.
259 113-137.
Pearce, R. B., and Ride, J. P. (1982). Chitin and related compounds as elicitorsof
lignification response in wounded leaves.Physiol. Plant Pathol.2 0 119-123.
Pegg, G. F. (1976). The response of ethylene treated tomato plants to infection by
Verticillium albo-atrum. Physiol. Plant Pathol. 215-226.
9
Perrin, D., and Cruickshank, I. A. M. (1965). Studies on phytoalexins. VII. Chemi-
cal stimulation of pisatin formation in Pisum sativum 1. Aust. J. Biol. Sci. 18:
803-810.
Price, W. C. (1964). Strains, mutations, acquired immunityand interference. Plant
Virology (M. K. Corbett and H. D. Sisler, eds.), University Press, Gainesville,
FL, pp. 93-1 17.
Purkayastha, R. P. (1973). Phytoalexins:plantantigens and diseaseresistance.
Science and Culture, 39:528-535.
Purkayastha, R. P., and Bannerjee, R. (1990). Immunoserological studies on cloxa-
Phytoalexin
Disease
Chemicals
Inducer
and Control 589
cillin-induced resistanceto soybean againstanthracnose. J. Plant Dis. Prot. 97:

349-359.
Rathmell, W. G.(1982). Active defence mechanisms of plants in relation to Protec-
tion of crops from pathogens with chemicals. Active Defence Mechanisms in
Plants (R.K. S. Wood, ed.), Plenum Press,New York, PP. 347-348.
Rathmell, W. G.,and Smith, D. A. (1980). Lack of activity of selected isoflavonoid
phytoalexins asprotectant fungicides. Pest. Sci. 11: 568-572.
Retig, N. (1974). Changes in peroxidase and polyphenoloxidase associated with
natural and induced resistance of tomato to Fusarium wilt. Physiol. Plant
Pathol. 4: 145-150.
Sahana, D. (1991). Control of chickpea diseases by seed treatment with non-toxic
chemicals. PhD thesis, BidhanChandra Krishi Viswavidyalaya, Kalyani.
Sarkar, M. L., and Sinha, A. K. (1991). Control of sheath blight of riceby uncon-
ventional chemicals.Indian Phytopathol.4 4 379-382.
Schwochau, M. E., and Hadwiger, L. A. (1968). Stimulation of pisatin production
in Pisum sativum by Actinomycin D and other compounds. Arch. Biochem.
Biophys. 126 731-732.
Sekizawa, Y., and Mase, S. (1980). Recent progress in studies on nonfungicidal
controlling agent probenazole with referenceto the induced resistance mecha-
nism of riceplant. Rev. Plant Prot. Res. 13: 114-121.
Sequeira, L. (1983). Mechanisms of induced resistance inplants. Ann. Rev. Micro-
biol. 37: 51-59.
Sharp, J., Albersheim, P., Ossowski, O., Pilotti, A., Garreg, P., and Lindberg, B.
(1984). Comparison of the structure and elicitor activities of a synthetic and
mycelial wall derivedhexa-8-D-glucopyranosy1)-D-glucitol.J. Biol. Chem.259
113-141.
Shimura, M., Iwata, M.,Tashiro, N.,Sekizawa, Y., Suzuki, Y., Mase, S., and
Watanabe, T. (1981). Anti-conidial germinationfactor induced inthe presence
of probenazole in infected host leaves.Isolation
I. and properties of four active
substances. Agric. Biol. Chem.45: 1431.
Sijpesteijn, A.K. (1969). Mode of action of phenylthiourea, a therapeutic agent for
cucumber scab.J. Sci. Food Agric. 2 0 403-405.
Sinha, A. K. (1976). Induced resistance asa factor in plant disease control. Indian
Biologist, 9 1-16.
Sinha, A. K. (1984). A new concept in plant disease control. Science and Culture,
5 0 181-186.
Sinha, A. K. (1990). Basic research on induced resistance for crop disease manage-
ment. Basic Research for Crop Disease Management (P. Vidyasekaran, ed.),
Daya, New Delhi, pp. 87-100.
Sinha, A. K. (1992). An alternative approach to chemical control of plant diseases.
Plant Science in the Nineties(R. D. Bannerjee, S. P. Sen, K. R. Samaddar, U.
Sen, A. K. Sulcar, and A. K. Biswas, eds.), University of Kalyani, Kalyani,
pp. 97-113.
Sinha,A. K., and Das, N. C. (1972). Inducedresistanceinrice plants against
Helminthosporium infection. Physiol. Plant Pathol.8: 401-410.
Sinha, A. K., and Hait, G.N. (1982). Host sensitization asa factor in the induction
590 Sinha
of resistance in rice against Drechslera byseed treatment with phytoalexin

inducers. Trans. Br. Mycol. Soc. 7 9 213-219.
Sinha, A. K., and Sengupta, T. K. (1986). Use of unconventional chemicals in the
control of blast of rice. Proceedings of the 2nd International Conference on
Plant Protectionin the Tropics,Genting Highlands, Malaysia, pp. 19-21.
Sinha, A. K., and Trivedi, N. (1969). Immunization of rice plants against Helmin-
thosporium infection. Nature, 223:963-964.
Sinha, A. K., and Wood, R. K. S. (1964a). Control of Verticilliumwilt of tomato
plants with “Cycocel” [(2-~hloroethyl)trimethylammonium chloride]. Nature,
202: 824.
Sinha, A. K., and Wood, R. K. S. (1964b). The effect of growth substances on
Verticillium wilt of tomato plants. Ann. Appl. Biol. 6 0 117-128.
Stekoll, M. S., and West, C. A. (1978). Purification and properties of an elicitor of
castor bean phytoalexinfrom culture filtrate of the fungus Rhizopus stolonifer.
Plant Physiol. 61: 38-45.
Subramaniam, S. (1963). Fusarium wilt of pigeonpea.111. Manganese nutrition and
disease resistance.Proc. Indian Acad. Sci. (Plant Sci.),57: 259-274.
Tomiku, T., Fuzikawa, T., Sato, S., and Ando, S. (1979). Occurrence of Xantho-
monas oryzaeinfection of rice and time and number of application of probena-
zole granules.Nogyo OyobiEngei 54: 339-340.
Trivedi, N., and Sinha, A. K. (1976a). Factors affecting the induction of resistance
in rice plants to Helminthosporium oryzae.J. Soc. Exp. Agric.I : 20-24.
Trivedi, N., and Sinha, A. K. (1976b). Resistance induced in rice plants against
Helminthosporium infection by treatment with various fungal fluids.Phytopa-
thol. Z. 8 6 335-344.
Trivedi, N., and Sinha, A. K. (1980). Effect of pre-inoculation treatments with
some heavy metal salts and amino acids on brown spot disease in rice seedlings.
Proc. Indian Acad. Sci. (Plant Sci.)8 9 283-289.
Uehara, K. (1963). On the production of phytoalexinby metallic compounds.Bull.
Hiroshima Agric.Coll. 2: 41-42.
Uknes, S., Mauch-Mani, Mayer, M., Potter, S., William, S., Drencher, S., Chan-
dler, D., Susarenko, A., Ward, E., and Ryals, J. (1992). Acquired inArabidop-
sis. Plant Cell, 4: 645-656.
Uritani, I., Uritani, M., and Yamada, H. (1960). Similarmetabolic alterations
induced insweet potato by poisonous chemicalsand by Ceratostomellajimbri-
Uta. PhytOpathOlO~,5 0 31-34.
VanAndel, 0 . M. (1966). Amino acids and plant diseases. Ann. Rev. Phytopathol.
4: 349-368.
Van Etten, H. D., Mathews, D. E., and Mathews, P. S. (1989). Phytoalexin detoxi-
fication: importance for pathogenicity and practical implications. Ann. Rev.
Van Loon, L. C. (1983). The induction of pathogenesis-related proteins by patho-
gens and specific chemicals.Netherlands J. Plant Pathol.8 9 265-273.
Vo-Thi-Hai, G., Bompeix, G., and Ravise, A. (1979). Role dutris-0-ethyl phospho-
nate d’aluminium dan la stimulation des reactions du tomate contre le Phy-
tophthora capsici.C. R. Hebd. Seances Acad. Sci. C 288: 1171-1 174.
Phytoalexin
Disease
Control
Chemicals
Inducer
and 591
Wain, R. L., and Carter, G. L. (1972). Historical aspects. SystemicFungicides

(R. W. Marsh, ed.), Longmans, London, pp. 6-33.
Ward, E. W. B., Unwin, C. H., and Stoessl,A. (1975). Experimental control of late
blight of tomatoes with capsidiol, the phytoalexin from pepper. Phyfopathol-
Ogy 65: 168-169.
Ward, E. R., Uknes, S. J., Williams, S. C., Dincher, S. S., Wiederhold, D. L.,
Alexander, D. C., Ahl-Goy, P., Metraux, J. P., and Ryals, J. A. (1991). Plant
Cell, 3: 1085-1094.
Watanabe, T.,Sekizawa, Y., Shimura, M., Suzuki, Y., Matsumoto, K., Iwata, M.,
and Mase, S. (1979). Effects of probenazole (Oryzemate) on rice plants with
reference to controlling rice blast.J. Pest. Sci. 4: 53-59.
elicitors. Nature, 275: 546-547.
Yoshikawa, M. (1983). Macromolecules, recognition and the triggering of resis-
tance. Biochemical Plant Pathology(J. Callow, ed.), Wiley,London, pp.267-
298.
Abelmoschus esculentus,25 Aecidium
Abscisic acid, 10, 16 adhatodae, 544
Acanthaceae, 544,549 carviae, 544
Acer saccharum,171 withaniae, 545
Acetonitrile, 202,204,240-242,410, Aflatoxin(s), 161,218,219
411,413 B,, 161,218
Acetophenone, 353 B,, 218
4-alkoxy-, 417 B,, 218
4-benzyloxy-2-hydroxy-,415 G,, 218
4-butoxy-, 417 Agarospirol, 461
2-hydroxy-4,6-dimethoxy-,263 Agglutination, conidia, 147, 148
Acetyl CoA, 216,407,468,480 Agrobacterium tumefaciens,24, 64
Acetyl salicyclic acid,580 Aizoaceae, 545
Achylaflagellata, 12 Ajmalicine, 12
Acrocylindrium, 25 Alarms, in defense systems,486
oryzae, 21,26 Alanine, 144
Acrosporium, 542 Alfalfa, 48,96,222,223,391,392,
Actinomycin D, 8,233,245,246 405 (see also Medicagosativa)
Adenosine monophosphate (AMP), Alkaloids, 535,537,539, 543,545-548
429 Alkanals, 178
cyclic-, 429,430,431,438 Alkanones, 178
dibutyryl cyclic-,430,431 Alkenals, 175, 178
S-Adenosyl-L-methionine:bergaptol, C,&,-, 175
72 Alkyl bis-phenylether, 216,217
S-Adenosyl-L-methionine:xanthotoxol Alkyl bis-phenyls,209,211,214,215,
0-methyl transferase, 72 220,221
Adenyl cyclase,429,430,43 1 Allylic alcohol, 450
Adhatoda, 544,549 N-Allyl amino acids, 560
vasica, 544 N-Allyl glycine,560
593
I
594 Index
N-Allyl sarcosine,560 Apple, 536

Alternaria, 547, 548 scab, 560
alternata, 50,539, 541,543 Arabidopsis, 581
blight, 525 thaliana, 237,256
brassicae, 238 Arabinose, 147
sesami, 525-531 Arachidins, 203,208,211,212,215,
Aluminum (M), 144,384 217,219
Aluminium trk-0-ethyl phosphonate Arachidonic acid, 12,13,432
(TEPA), 375,561 Arachis, 223,361
(see also Fosetyl-Al) hypogea, 24,50,199,287,309
Amino acidsin phytoalexin elicita- (see also Groundnut)
tion, 560,563 Arginine, 144
y-Aminobutyric acid, 144 Aromatic hydrocarbon fungicide
Aminocyclopropane carboxylic acid (AHF), 328,329
(ACC), 106,107 Aromatic substituent constant, 328
synthase, 107 Arthrobactor luteus, 62
Aminoethoxy vinylglycine(AVG), 2-Arylbenzofuran, 377
383 Asclepiadaceae, 546,549
L-a-Aminohydrazino-B-phenyl propi- Ascochyta
onic acid (AHPP), 383 imperfecta, 96
a-Aminooxy acetic acid(AOA), 266, pki, 17
275,281,379,382,383,398, rabiei, 15,21,397
399,562 Asparagine, 144
L-a-Aminooxy-&phenyl propionic Aspartic acid, 144
acid (AOPP), 380,400 Aspergillus,218,219,222
Anacardiaceae, 540 aculeatus, 544
Anogeissus, 541,549 flaws, 161-181,200,217-220,343,
latifolia, 541 361
Anthocyanin(s), 124 fumigatus, 361,365
deoxy-, 193 japonicus, 102
Anthacnose, in niger, 335,343,358,359,361,525-
Adhatoda, 544 528,536,537,540,541,545,
bean, 46 546
Dioscorea, 336 parasiticus, 161,200,218
soybean, 27 Aspirin, 580
Antibiotics, 2,136,288,333,497, Asteridae, 549
563 Asterolibertina,540
Antigens, 5,20-27,570 ATPases, 51-53, 75
Aphanomyces euteiches,12,358, Aureobasidium, 540
560 Aureofungin, 526,528,530
Apigenin, 543,548 Aureol, 341,355,359,360
4’-OMe-, 535,543,544,547,548 Autofluorescence, 191-193
7-OMe-, 544,545 Auxins, 110
7,4‘-diOMe-, 547 Avenalbumin, 14
Apoplast, 187, 189,401 Avena sativa, 14,24
Index 595
Bacillus cereus,343 Boron, 560

Bacteria, 117-119, 125,153,169,183- Botryodiplodia, 547
198,202,242,336,344,361, natalensis, 268
505 theobromae, 19,343,348,350,356,
cultures, 118, 344 538,543
Bacterial pustule disease, soybean, Botrytis, 191,486
118 cinerea, 3, 10, 14, 19,20,281,288,
Barium chloride, 564,567-569 290,291,293,294,299,300,
Barium sulfate, 564, 568,583 304,308,309,318-326,329,
Barley, 42-43, 51,52,54 357,536
Batatasin(s), 336,342-344,348-351, fabae, 3, 19,558,560,561
356,361-363,365,366 Brassenin, 8-methoxy, 236
demethyl-, 342,344,348-350,356, Brassica, 229-261
361-363,365 adpressa, 247
Bavistin, 526, 528, 530 campestris, 229,259
B chromosomes, 201 .
var pekinensis, 259
Beloruskii rannii,12 carinata, 229-232,247-250,252
Benomyl, 558 insularis, 231
Benzimidazole, 377,382 juncea, 229-233,235,241,243-246,
Benzoic acid, 560, 580 248-252
halogenated, 560 naponigra, 260
p-hydroxy-, 535,536,539-545,547- napus, 229-232,235,244,246-253,
549 255
2,3,5-triiodo-(TIBA), 560 var. oleifea, 229
Benzoquinone, 540,543,549 nigra, 229-232,248-253
Benzyl amino purine, 10 oleracea, 229,230,232,237,247,
Benzyl penicillin, 11 248,250
BergaptokS-adenosyl methionine, 72 rapa, 229,230,231,247,248,250,
Bibenzyl 251
2'-hydroxy-3,5-dimethoxy-,349 Brassicanal
3-hydroxy-S-methoxy-, 336,348, A, 236
349 B, 236
2,4'-dihydroxy-3,5-dimethoxy-, C , 236
342,349 Brassilexin, 234-240,244-255
Bignoniaceae, 543 Brassinin, 235-238,240,244, 245,
Biochanin A, 15,90, 120 247-250,254
Biosol-2, 141 c ~ c ~ o234-240,242,244-250,253,
-,
Bipolaris carbonum,486,489492, 254
494 Cmethoxy-, 235,236,238,240,
Blackleg disease, crucifers,230, 231, 242,244-250
254 spiro-, 235,245
Blight, bacterial, in cotton, 183- Brassitin, 236
198 methoxy-, 236
Boraginaceae, 546 Bremia lectucae,384
Boric acid, 564,568 Broadbean, 3,191,558,561
596 Index
Brussels sprouts, 405 Capsidiol, 9, 15, 18,381,384,503-

Buckwheat, 402 523,558
2-Buten-1-01, 176 effect of metaloxyl, 512-514, 517-
2-Butoxy alcohol, 176 519
plant age, 509-512, 514-517
Cadalene, 168,169, 178, 195 time course,513-514
biosynthesis, 193-194 Carbon monoxide, 476,477
2,7-dihydroxy-(DHC), 164-166, Carotenoids, 548
168-173,178, 184-194 Carrot, 430 (see also Daucas carota)
2-hydroxy-7-methoxy- Carvia, 544,549
(HMC,DHMC), 164-166, callosa, 544
169-173, 178, 185-194 Caryophyllene, 177
6-Cadinene, 185, 194 Casbene, 9, 14
Cadmium chloride, 12,27,496,571- synthetase, 14
575 Cassia, 549
Cadmium sulfate, 478,479 fistula, 536
Caesalpiniaceae, 536 Catalase(s), 125, 169
Caffeic acid, 487, 546 Catechin, 486-489
0-methyl transferase, 124 2,3-dioxygenase, 487
Cajanol, 338, 352 Catechol, 487,488,494
Calcium (Ca,Ca++), 13,54,74, 138, Catenulosporazizyphi, 543
144,429,431-433,435,438, Cauliflower mosaic virus,184,
475 188
chloride, 238 Cell aggregates in culture,120, 121
Callose, 72 Cellular responses,85-1 15
Calmidazolium, 43 1 competency factors, 102-106
Calmodulin (CaM), 43 1,432, conditioning, 102-106
438 mediating, 107-1 10
Calotropis gigantea,546, 548 modulating, 107-1 10
Calphostin C, 437 Ceratocystis
Camalexin(s), 236 hfagacearum, 560 .
methoxy-I-, 236 fimbriata, 4,24,467-483
Camelina sativa, 237,251 Ceratostomella ulmi, 560
Camellia Cercospora, 210,540
japonica, 489 adhatodae, 544
sinensis, 485-501 arachidicola, 200,208,209,215-
var. assamica, 485 2 17,220-222
var. sinensis, 485 beticola, 2 1
Camellidin I & 11,489 morindae, 538
Camphene, 177 oryzae, 563
Canavalia ensiformis,370,388 sesamicola, 525
Capsella bursa-pastoris,252 tectonae, 534
Capsenone, 18 trianthemae, 545
Capsicum tylophorina,546
annuum, 15,503,506 withaniae, 545
frutescens, 522 Chaconia tectonae, 534
Index 597
Chaetomium, 547 limon, 263

cupreum, 487 macrophylla, 264
globosum, 14,546 paradisi, 264
mangiferae, 540 reticulata, 264
Chalcone(s), 93, 354,415 sinensis, 263,264
isomerase, (CHI), 5,55,86,354, Cladosporium, 207,217,218,234,
407 235,547
synthase, (CHS), 5,55, 86,354, cladosporoides,343,525-528
357,407 cucumerinum,203,335,343,358-
Charge transfer complexes (CTC), 360,560,561
318,325,328 fulvum, 376,401
Chickpea, 16,397,400,564,568,573 herbarum, 203
(seealso Cicerarietinum) Clerodendron inerme,546,547
Chili, 572 Cloxacillin, 11,27,497,562,577,578
Chitinase, 107,401, 559, 577,580 Cochliobolus
Chitosan, 10,429,558,565,567,568, carbonum, 401,489
570-575, 579,581,583 miyabeanus, 44,489
Chloramphenicol, l5 Coffea arabica,24
Chlorocholine chloride (Cycocel), Collar rot, chickpea, 564,573
561,565 Colletotrichum, 191, 546
Chlorogenic acid,488 camelliae, 486,489,491,492
p-Chloromercuric benzoic acid(p- capsici, 525-528
Cmb),567,569,571-575, corchori, 24
583 dematium, 544,562,580
Chnoopsora butleri,544 var. truncata, 11,24,26,27
Chocolate spot, broad bean, 558,560, falcatum, 539
561 gloeosporoides,268,538,542
Cholera toxin, 430 graminicola, 193,495
Chromosomes, B,201 lagenaricum, 562
Chrysanthemum,52 lindemuthianum,9, 10, 17,55, 378,
CHS gene, 54 400,425
Cicer arietinum, 15,21 (seealso Combretaceae, 356,541,549
Chickpea) Competency factors, 102
Ciliochorella,540 Compositae, 4
Cinnamate-4-hydroxylase,124,216, Coniophora carebella,538
407,476,479 Convolvulaceae, 4
Cinnamic acid, 169,216, 280,281, Copper(Cu), 144,560
287,356,423425 chloride, 233,241-243,247,248,
trans-, 407 252 (seealso Cupric chloride
Cinnamoyl-CoA ligase, 334,354 and Cuprous chloride)
Citronerol, 472 sulfate, 15, 568
Citrullus vulgaris,24 Corchorus capsularis,24
Citrus, 263-286,384,549,561 (see Cordia, 547
also Orange) myxa, 546,547
aurantium, 263,264 Cordycepin, 422,432
jambhiri, 264 Coriolus versicolor,538
598 Index
Corn, 73 (see also Zeamays) Cupric chloride, 9,496,560,564,567,

Corticum 570-575,583 (see also Copper
invisum, 486 chloride)
theae, 486 Cupric sulfate, 564, 567
Corynebacterium insidiosum,25 Cuprous chloride, 449 (see also Cop-
Corynespora cassicola,544 per chloride)
Cotton, 129-198,560 (see also Gos- Culture darkening, 117-128
sypium spp.) Curvularia, 540
boll, 162,163, 169, 170 clavata, 535,536
seed, 161-181 lunata, 525-528
Cotyledon(s), 94-98, 189-191,219 prasadii, 543
bioassay, 102-106, 186,201,252, spicata, 210
253 Cuscuta, 534,546-548,550
4-Coumarate:CoA ligase,5, 216,407 chinensis, 546,548
o-Coumaric acid,537 reflexa, 546,548
4-methoxy-, 281 Cyanogenic glycosides, 486
p-Courmaic acid, 281,318,356,407, Cyclic AMP, 429,430,431,432
488,537,540,543-547 2‘,3”, 431
m-Coumaric acid, dihydro-,356 3‘,5‘-, 431,432
p-Coumaroyl CoA, 216,287,334, Cyclic nucleotide diphosphodiester-
356,406,407 ase, 43 1,432
Coumarin(s), 275,537, 539,544 Cyclobrassinin, 234-240,242,245-
6,7-dimethoxy-(see Scoparone) 250,253,254
6,7-dimethyl pyrano-(see Xanthy- sulfoxide, 235,236,247,249,250,
letin) 254
7,gdimethyl pyrano- (see Seselin) Cyclodehydroisolubimin, 18
7-hydroxy- (see Umbelliferone) Cyclohexamone, 177
7-methoxy-, 281 Cycloheximide, 233,245,246,496,
6-methoxy-7-hydroxy- (see Scopo- 565,567-569,571-575,583
letin) Cyclokievitone, 338,339,347,355,
Coumestans, 337,341,351,355,359, 359,360
360 1‘‘,2‘ “dehydro-, 359
Coumestrol, 118-120,341,346 hydrate, 339,347
Cow pea, 51,52,123,376-379,385 3‘,4’,5’-trimethyl-, 359
(see also Vigna Unguiculata) Cyclopolyether derivatives,141
Cristacarpin, 340,348 Cyclopropane carboxylic acid,2,2-
Cross-protection phenomenon, 149- dichloro3,3’-dimethyl-, 14,
153 375
Cross-reactive antigens(CRA), 27, Cycocel (see Chlorocholinechloride)
497 Cylindrocladium scoparium,542
Crown-6, 141 Cysteine, 144,564,567, 569-575,
Crucifers, 187,229-261 583
Cucumber, 222,385,503,556,562, Cystine, 144
576 Cytochrome c, 476
scab, 560,561 reductase, 328, 329
Index 599
Cytochromep-450,476,479,480 8,2"dihydroxy-, 337,346, 352,355

isoenzymes, 16 8,2"dihydroxy dihydro-, 337,346,
monooxygenase, 16 352,355
Cytokinin(s), 109, 110, 124 2'-hydroxy-, 337,346,352,354,
355,359
Daedalia 2"hydroxy dihydro-, 337,346,352,
flavida, 534 355
quercina, 537 Diaporthe citri, 264,269,489
Dalbergioidin, 337,347,348,352, Diazotised p-nitraniline (DPN), 203,
355,359-361 212-214,490,535
5,2'-dimethoxy-, 337,347,352,355 Diazotised sulphanilic acid(DSA),
Daucas carota, 124 (see also Carrot) 203,212-214,535
Damping off, chili, 564 Dibenzo-18, 141
Datura stromonium,465 Dibutylphthalate (DBP), 410,411,
1-Decanol, 176 414,416,417
N-Decyl aldehyde, 176 Dichlorocyclopropanes, 505,s 18
Defense genes, 107, 112,291,436 Dichloromethoxyphenol (DCMP),
expression, 107 328
regulation, 107, 108 2,CDichlorophenoxy acetic acid (2,4-
transcription, 291 D), 123,124,560,565,567-
Defense reaction, 41,54,64-65, 129, 569,571-575,580,583
136, 141 Dienols, 216-218,220,221
elicitation, 53 Diethyl amine,451
protein kinase in, 53-54 Di-2-ethyl hexyl phthalate, 137
strategies of, 161-181,485-501 Diethylmalonate (DME), 451
suppression, 12,41,49-53, 142 Digitonin, 72
Dendrolasin, 4 Dihomo-y-linolenic acid, 13
6-hydroxy-, 469,474 Dihydroazulene, 450
~-oxo-,469,474 Dihydrostilbene(s), 334,337,342,
Dendrolasinoide, 6-oxo-, 474 344,348-351,356,357,361-
Desoxy hemigossypol, 167 365
Detoxification, 5 , 16-19, 155,355, antimicrobial properties, 361
356,391-403 biosynthesis, 356,357
glycosides, 391-403 chemotaxonomy, 358
pterostilbene, 309 growth inhibitory properties, 361-
resveratol, 309 365
E-viniferin, 309 Chydroxy-3,5-dimethoxy-, 342
Diacyl glycerol(DAG), 429,435-438 3-hydroxy-5-methoxy-, 342
Diadzein, 86,90,93,94,96-98, 118, o-Dihydroxyphenol, 573-575,579
119, 121-123,209,211,213, 2,4-Dinitrophenol, 526,528,530
215,216,337,346,352,355, Dioscorea, 333-335,343,348,351,
359,360,400 356,358
biosynthesis, 93,216, 355 alata, 335, 336,348,356
conjugates, 90,91,95-97, 106, 109, botatas, 348,351,358,361
110,400 bulbifera,336,349,358,366
600 Index
[Dioscorea] Elicitors, 5,6, 12, 15,27,53,54,61-

cayanensis, 334 67,70-73,77,94,117-122,
chemotaxonomy, 358 124, 148, 149, 155, 162, 166,
decipiens, 336 . 241-243,409,411,416,418-
dumentorum, 335,336,349,358, 427,436,437,467,480,556,
366 557,558
hispida, 336 abiotic, 6,8,233,235,241,243,
mangenotiana, 336,349,358,366 244,254,497,557
opposita, 348.349 biotic, 86,233,242, 556
rotundata, 335,336,349,365,366 CO-, 106
Dioscoreaceae, 348 fatty acid, 167
Diplodia, 540 primary, 99, 101
gossypina, 168 multiple, 100
zeae, 70 pmg, 62
Dipogon, lignosus, 348 receptor complex,53
Dithan “45,526,528,530 releasing factor, 62
cDNA, 55,195 specific, 6
chalcone isomerase,55,65 synergists, 99, 101, 102, 105, 167
chalcone synthase,55,65 VI-,418-427
elicitor-releasingfactor, 64 Vz-,418427,436,437
molecular cloning,64 volatile, 171-178
NADPH: isoflavone oxidoreduc- Ellagic acid,488
tase, 403 Elsinoe kamatii,539
phenyl alanine ammonia lyase, 55, 8-1,3-Endoglucanase, 62,63,66,72,
65 100,108,559
pine stilbene synthase,357 Endopolygalacturonatelyase, 559
soybean-8-1,3-endoglucanase, Endopolygalacturonidase,102
64 Epicatechin, 487
DNAase I, 184 gallate, 487
DNA plasmids, 184 Epigallocatechin, 487
Dodecyl aldehyde, 176 gallate, 487
Dodecyl-DL-alaninate,560 Epioxylubimin, 452
Dolichin (A&B), 340,348 Erysiphe
Dolichos biforus, 348 graminis, 7,42,44,47,48
Dothichizapopulea, 537 f.sp. cichoraceamm,540
Downy mildew, 504 f.sp. hordei, 42-45
tobacco, 41 f.sp. tritici, 45,48
vitis, 558, 561 pisi, 44-47, 54
Drechslera Erythrina, 357
sorokiniana, 576 Erwinia
speciferurn, 544 caratovora, 102,243,559
ssp. atroseptica, 232
EGTA, 54,431,432,434 chrysanthemi, 243
Eicosapentenoic acid, 12,13 pv. dahliae, 232
Electron transport inhibition, 75 Bcherichia coli, 343
Index 601
Ethephon, 561 Formarin, 451

Ethylene, 9,72, 106, 107,246,247, Formononetin, 15,90, 120,209,211,
383,561,578 213,215,216,220,407
N-Ethyl maleimide, 375 7-0-glycoside-6’ ‘-0-malonate
Eucalyptus, 538,549 (FGM), 400
globulus, 538, 539 Fosetyl-Al, 264,266,277-280,282,
microcorys, 538 290,292,293,305-307,375,
regnans, 538 376,383-384,561,562
siderovlon,538 Free radicals, 78, 187, 188,328
trifrora, 538 French bean,9,52,558(see also
Evans blue, 185 Phaseolus vulgaris)
Exobasidium vexans,486 Furanocoumarins, 101
Exopolysaccharides, 11 Furanosesquiterpenoids,4,467-483
biosynthesis, 468,475
Fabaceae, 199,253,334,345,533(see enzyme conversion,467-483
also Leguminoseae) Furanoterpenoids, 445,558
Falcarinol, 536 Fusarium, 108,541, 545,560,561
Famesal, 470 moniliformae, 14,541
Farnesoic acid, 9-hydroxy-, 474 oxysporum, 189,541
Farnesol, 468-470,472,474,475,480 f.sp. ciceri, 568,573
dehydrogenase, 470,472,473,480 f.sp. lycopersici, 10
9-hydroxy-, 469,474,480 f.sp.pisi, 17
~-oxo-, 469,480 f.sp. vasinfectum, 19,24,25
Farnesyl pyrophosphate, trans, trans roseum, 393,397
(FPP), 185,193-195,468,480 sambucinum, 17
Fatty acids, 209,210,214-216,558 semitectum, 24
Ferulic acid, 97, 168,169,535,538, solani, 7,18,24,525-528,535-537,
542-548 539-542
Femc chloride, 564,567,571-575 f.sp. cucurbitae, 358
Flavone(s), 353,535, 550 f.sp. phaseoli, 17,356,579,581
6,7-dimethoxy-3,5,4’-trihydroxy-, fsp. pisi, 579
545 sulphureum, 19
6-hydroxy flavone, 548 udum. 25,560,572,573,580
Flavonoids, 208,210,213,215,216, Fusarium wilt, 561
317,324,534,541,545,546, chickpea, 564,573
548,549,550 pigeonpea, 537,572
demethylation, 17,537,549 tomato, 560,561
methoxylation, 549 Fusicoccum amygdali, 535
Flavonol(s), 90,91,535,537,541,
542,546,549,550,573 Galactose, 138,144,146-148,489
Fluoresein, 184 a-l,CGalacturonides, 86
Fomes Gallic acid,488, 540
annosus, 537 Gallocatechin, 487
durus, 538 Gamma irradiation, 264,265,271-
Forage legumes, 391-403 277,282
602 Index
Ganoderma Glucanase(s), 62, 107,401,577,580

applanatum, 534 Glucans, 95
lucidum, 539 wall-, 103
Gaseous phytoalexins,171-178 yeast wall-, 103
Genes, 136 B(1 3)-, 63,86, 100, 101
+
antisense, 222 B(1 + 6)-, 63,86, 100, 101

avirulence, 20, 183 Glucobrassicin, 236,253
bacterial blight resistance,cotton, 4-methoxy-, 236
186 neo-, 236
CHS, 54-56,428 Glucosamine, 144
glucanase, 64 Glucose, 138,144,146-148,489
glyceollin, 65 2-desoxy-, 293
PAL, 56,428 Glucose-6-phosphate dehydrogenase,
phosphorylated, 54,57 187
phytoalexin biosynthesis,54 B-Glucosidase, 393,397
Pda, 16,201 Glucosinolates, 245,253,254
pisatin demethylase,385 Glucuronic acid, 489
resistance, 20 Glutaryl CoA, 3-hydroxy-3-methyl
Rps, 70,85, 89,92 (HMG CoA),468
silencer regions, 54 Glutathione, 106, 155
stilbene synthase,309 S-transferase, 155, 156
suppressor, 22 Glycanase, 62,77
Genistein, 90-96, 119, 121-123,337, released elicitors,62,63
345,347,348,352,355,359, Glyceollin, 8,26,50,61,62,71-73,
360 85,92-94,96,98, 101-106,
biosynthesis, 355 117-119, 121-125, 193,201,
conjugates, 90,96, 106 354,400,497,504,505
8,2'-dihydroxy-, 337,347,352,355 biosynthesis, 65,86, 365
8,2'-dihydroxy dihydro-, 337,347, elicitors, 71, 72, 86
352,355 isomers, 73,85,120
2'-hydroxy-, 337,347,348,352, Glycetine, 90
355,359-361 Glycine, 73,334
2'-methoxy-, 337,347, 352,355, max, 24,50,61,69,117,505 (see
359,361 also soybean)
Gentisic acid,535,538,540,543,546, sojae, 92
547 wightii, 70
Geraniol, 472 Glycinol, 340,346,348
dehydrogenase, 472 2,lO-( dimethyl allyl-)-,340
Geranyl pyrophosphate, 468 Glycolate oxidase,329
Gibberellapulicaris, 17, 18 Glycolic acid, 329
Gibberellic acid, 11, 14,26,497, 561, Glycopeptides, 49
577 Glycoprotein(s), 70,71, 107, 138,
Gloeosporium, 547, 548 143,144,418
morindae, 538 Glycosidases, 75
theae-sinensis, 489,492 Glyoxylic acid, 329
Glomerella cingulata,19,543 Glyphosate, 397-400
Index 603
Heavy metals,6,210 HPLC profiling of phytoalexins,97,

Helminthosporium, 563 205
carbonum, 13,208,210,217,288, crucifers, 234,235,240,254
401,489 groundnut, 205,212-214
maydis, 495 Medicago, 410,412-414,416,417
oryzae, 14,210,497,563,567,570, soybean, 119,120
580 Vitis, 292
sativum, 406,567,571 Hydrogen peroxide, 124,125,329
tetramera, 525-528 Hydroperoxide(s), 125
turcicum, 70 organic, 125
victoriae, 14 Hydrophobicity, 325,328
Helianthus annuus, 24 Hydroxycinnamic acids,97-99
Heliocide Hz,174 3-Hydroxy-3-methyl glutaryl CoA
Hemigossypol, 133, 167 (HMG-COA),468
6-dioxy-, 133 reductase, 469,478,480
6-methoxy-, 133 9-Hydroxy octadecadien-10,12-
Hemigossypolone, 174 methyloate, 209,211,215,220
Hemileia vestatrix,24,29 13-Hydroxyoctadecadien3,lO-
Hendersonula tontloidea, 263,268 methyloate, 209,211,215,220
Heptaglucosides, 15 Hydroxystilbenes, 289,317-331
Hepta-&glucoside alditols, 15 effect on cellular structure, 320-323
Heptanal, 176 effect on conidia, 320
Heptanes, 146, 147 effect on fungal cells, 318-323
Hepta-P-D-glucopyranoside (G7), effect on respiration, 318-320
62 Hypersensitive resistance,249,251,
l-Heptanol, 176 252
2-Heptanone, 177 Hypersensitive response(HR), 117,
3-Heptanone, 177 125,130, 183-189,231,251
trans-2-Heptenal, 176 Hyphantria cunea, 171
3-Hepten-1-01, 176
Heterodera glycines, 201 IAA oxidase, 169
Heterophragma, 549 Immunodepressors, 148
adenophyllum, 542 Immunosystem, 154, 155
Hexachloroacetone, 45 1 Indole, 253
2,4Hexadienal, 176 acetic acid(M), 560,565,567-
n-Hexanal, 176 569,571-575,580
cis-3-, 176 3-aminoethyl-, 237
trans-2-, 175, 176 phytoalexins, 233,240,244,249,253
Hex-2-enal, diethylacetal, 176 3-Indole carbaldehyde,237
cis,2-Hexen-l-ol, 176 Indole-3-carboxaldehyde,238
cis,3-Hexen-l-o1, 176 Inducers, 70, 136, 137, 141,293,555-
Hinesol, 44547,457 591
Histidine, 144 Infection inhibitor, 46
Homeostasis, 129, 155 Inositol
Hordeum aestivum, 25 -1,4,5-triphosphate( IP3), 435-437
Host-parasite specificity, 41-60 -1,3,4,54etrakisphosphate,(IP4), 436
604 Index
Invertase, 72,73 Isohemigossypol, 132-134, 139, 142,

Ipomoea batatas,24,467,481 146,149-151,154,155
(see also sweet potato) Isoleucine, 144
Ipomoeamarone, 4, 19,467-470,472, Isoliquiritigenin, 407
474,475,477,478,480 Isolubimin, 18,457,458
dehydro-, 469,475,480 Isonicotinic acid(INA), 375
-12-hydroxylase, 472,476,478 2,6-dichloro-, 375,581
Ipomeamaronol, 467-469 Isopentenyl pyrophosphate, 468
Ipomeamaronolide, 474 Isoprunetin, 337,347, 352,355,
Iron(Fe), 145 359
Iridoids, 534,535,539,542 2'-hydroxy-, 337,347,352,355,
Isobutyric acid, 185 359
Isofemerin, 337,347,348,352,355, Isorhamnetin, 90-92
359,361 Isosojagol, 341,346
Isoflavan(s), 337,351,357-359,392, Isosophoranone, 338,352
397
Isoflavanol, 407 Jasmonic acid, 105, 106
2',7-hydroxy,-4'-methoxy-, 407 Juglone, 78
Isoflavanone, 16,337,351,352,358,
360 Kaempferol, 90-92,537,538,546
7,4'-dimethoxy-;!'-hydroxy-, 209, 4'-OMe-, 537-539,546
211,213,215,217,221 Kievitol, 337,338,347,355,359,
7,2'-dihydroxy-4'-methoxy-,209, 360
211,213,215 2,3-dehydro-, 337,347,355,359,
5-hydroxy-, 353 360
Isoflavone(s), 16,90,110, 118,337, 5-deoxy-, 338,346,355
351-353,358,359 Kievitone, 8, 17, 18,337,338,344,
biosynthesis, 216,353, 354 345-348,352,355,358-360,
conjugates, 90, 101, 110,393 378,380
cycloderivatives,338 CyClO-, 338,339,347, 355,359,360
S-deoxy-, 94 2,3-dehydro-, 337,347,352,355,
2',7-dihydroxy,4'-methoxy-, 407 359
5-hydroxy-, 353 2",2"-dehydrocyclo-, 339,347
7-hydroxy,4'-methoxy-, 407 5-deoxy-, 337,346,352,355,359,
reductase, 222,223 360
Isoflavonoids, 4, 183, 187,317,334, 3 '-(y,y-dimethylallyl)-, 338,347,
344,351,353,354,358,365, 352,359,360
445,558 hydrate, 17, 18,337,338,347,355,
antimicrobial properties, 358-365 359,360
biosynthesis, 353-356 hydratase, 17,356
in chemotaxonomy,357 4'-methoxy-, 338,347,352
detoxyfication, 356 2',4',5,74etramethyl-, 338, 359,
UV spectra, 351-353 360
Isoflavonone, 337,359 trimethoxy-, 338,339,359,360
Isogenic isolines, soybean, 89 Kinetin, 109
Index 605
La3+,432,434 Lubiminol, 446,457,458

Lablab, 357 Lucerne, 405-443(see also Medicago
niger, 348 sativa)
Lacinilene, 168, 178, 195 callus culture,409
Lacinilene C (LAC,LC), 163-167, Lupin, 384
169-173, 178, 184, 186-194 Luteolin, 535
Lacinilene C7-methylether(LAC- 7-OMe-, 544
ME,LCME), 163-167,169- Luteone, 337,347,352,358
173, 178, 186, 187-194 Lycopersicum esculenturn,25,409,
Lactonisation, 169 418 (see also Tomato)
Lactose, 144 Lysine, 144
Laminarin, 72 POly-L, 14
Laxifloran, 339,348
Lectins, 144-148, 156 Maackiain, 8, 15, 16,21,317,334,
Leguminoseae, 4, 199,334 (see also 341,39141
Fabaceae) dihydro-(DHMaa), 393,396
Lemna minor, 166,167 Macozeb, 558
Lepachol, 535 Macrophoma morindae,538
Leptosphaeria maculans,230-235, Macrophomina phaseolina,21,24,26,
238,239,241-243,249,251, 27,525,562,580
252,254-256 Macroptilium
Lespedeza buergeri, 50 atropurpureum, 345,348
Lettuce, 384 bracteatum, 348
Leucine, 144 lathyroides,348
Lignin, 72, 383,384,407,536,579 marti, 345
Limonone, 177 Macrotyloma axillare,348
Linoleic acid, 178 Madhuca, 542,549
Linolenic acid, 178 indica, 542
Linum usitatksimum, 22,U Magnesium (Mg), 14, 145
Lipid-suberin conjugates, 406 Magnolidae, 549
Lipoglycoproteins, 559 Malacosoma californiapolviale,171
Lipopolysaccharide, 11,559 Malate dehydrogenase, 184
Liquiritigenin, 407 Malonated conjugates, 91
Lithium salts, 560, 564, 567, 569, 571- Malonyl CoA,216,287,334,356,407
575,583 Malonyl diadzein, 9,193
Lotus Malonyl genistein,91,93
bicolor, 50 Manganese (Mn), 145
coniculatus, 50 sulfate, 560
Lubimin, 9,12, 13, 17,18,432,445, Mangifera, 541,549
452,461 indica, 540
15-dihydro-, 18 Mangiferin, 541
10-epioxy-, 457,458 Mannitol, 190
iso-, 18,457,458 Mannose, 138,144,146-148
OXY-, 445,447,450,452,455,458, Medicago sativa, 24,50,405-443 (see
46 1 also Alfalfa and Lucerne)
606 Index
Medicarpin, 8,15, 16,21,48,208, Monoiodoacetamide, 475

211,213,215-217,220- Monooxygenases, 328;378
222,334,340,346,358,359, Morinda, 538,549
361,391-399,401,406,407, tomentosa, 537,538
411-413,415-417,419-422, Mucic acid, 293
430 Mucor, 328
biosynthesis, 216 Mustard oils,486
demethyl-, 209,211,213,215,217, Mycelial wall extract, 137, 162, 163,
218,220,221 233
demethylase, 222 Mycolaminarin, 64, 101, 103
iso-, 340,348,358,359,361 Mycosphaerella
Medicinal plants, 544-546 ligulicola, 7,50, 52, 53
Melampsora h i , 22,24 melonis, 7, 50
Melilotic acid, 535, 547 pinoides, 7,46,47,49-53,55
Mellein, 6-OMe, 430,431, 536 rabiei, 525-528
Mercaptoethanol, 184 Mycotoxin, 467
Mercuric acetate, 13 Myo-inositol, 436
Mercuric chloride, 8,10,242,342, 6-Myoporol, 474
350,362,467,478,496,526, Myoporone
528,530,564,567 6-dihydro-7-hydroxy-, 474
Metalaxyl, 11,503-523 4-hydroxy-, 474,480
in phytoalexin production,512- 4-hydroxy dehydro-,474,475,480
520 Myrcene, 174, 177
Metal salts,560, 563 Myricetin, 540, 542
Methionine, 144,560,565,568,571- 4'-OMe-, 542
575 Myrosinase, 253
3-Methyl-l-butanol, 176 Myrothecium roridum, 10,24,26
3-Methyl-2-butano1, 176 Myrtaceae, 538,539,549
Methyl( 3-indolylmethy1)-
dithiocarbamate S-oxide,238 NADP, 470,472
Methyl linolenate, 209,211,214-217, NADPH, 471,475
220,221 -cytochrome c reductase, 479
Methyl myristate,507, 508 cytochrome ~-450,478
0-Methyl transferases, 72 -isoflavone oxidoreductase,397
Mevalonate (MVA), 468 -oxidoreductase, 397
phosphate, 468 -reductase, 354
pyrophosphate, 468 Naphthalene
Mevalonolactone, 193 acetamide (NAM), 560
(5-3H), 185, 194 acetic acid(NAA), 560
Mevinolin, 382,383 Nectria, 400, 536
Microsomal fractions, 148 haematococca, 16,17,201,216,
Milletia japonica,50 385
Momilactone(s), 9, 14,26,497,505, Necrosis, 189-193,267
561 Neodunal, 341,348
Moniliniafructicola, 10, 19,288 Neorautanea mitis, 348
Monilicolin A, 10 Nerol, 472
Index 607
Nickel Parsley, 101, 107

chloride, 496,497,565 Pda gene, 16,201
nitrate, 496,497,565, 567 Pea, 7, 16,49,51,52,56, 108, 187,
Nicotiana tabaccum,376,462 20 1, 5 18(see also Pisumsat-
(see also Tobacco) ivum)
Nigrospora oryzae, 563 root rot, 560
Nitric oxide, 78 root rust, 560
2-Nonanone, 177 Pectate lyase, 102, 105, 106
l-Nonanol, 176 Pectinolytic enzymes, 14,573
trans-2-Nonenal, 176 Pedaliaceae, 525
Nonyl aldehyde, 176 Penicillin, 497,562, 577, 578
Nonyl phenol,209,211,214,215,220 Penicillium
Norcaradiene derivatives, 450 crysogenum, 329
digitatum, 264,268,271
Oat, 191 (see also A vena sativa) italicum, 268
Obligate parasitism,41-44 CPentanoic acid, 177
Ocimene, 177 2-Pentanone, 177
Octanol, 176 3-Pentanone, 177
3-Octanone, 177 1-Pentenol, 176
Octasaccharide, 559 4-Penten-1-01, 176
trans-2-Octenal, 176 Pepper, 503-523,561 (see also Capsi-
Oidium, 536 cum annuum)
l-Oleoyl-2-acetylglycerol,437 Peppermint, 405
Oligogalacturonide elicitors,11,100, Peronospora
578 parasitica, 581
0-Methyl transferase tabacina, 57, 169
caffeic acid, 124 Peroxidase, 97, 120, 123-125, 169,
S-adenosine-L-methionine: 560,573
xanthotoxol, 72 assay, 120
5-hydroxyferulic acid,124 isoenzymes, 97
Ophiobolus, 24 Pestalotia longiseta,489
Orange, 42 (see also Citrussinensis) Pestalotiopsis
Orchinol, 4 adusta, 536
Orchis militaris,4 funerea, 539
Orthovanadate, 52 theae, 486,489,491,492
Oryza sativa, 24 (see also Rice) versicolor, 541
Oryzalexin, 14 Phaeoisariopsis personata,200,
Oxadiazone, 10 220
Oxalic acid, 106 Phaseol, 341,346
Oxidative phosphorylation, uncou- Phaseollidin, 8, 17,344, 346, 348,
plers, 317 355,359,360,378,386
Oxindoles, 237,253 2(y,y-dimethyl allyl)-, 340,346
Oxygenase, 473,474 4(y,y-dimethyl allyl)-, 340, 346
Oxylubimin, 445447,450,452,455, isoflavan, 339
458,461 l-methoxyl-, 340, 348
Oxyrhynchus volubilis,348 Phaseolinae, 334,345
608 Index
Phaseollin, 3,8-10, 17,42,48, 187, Phenoxy acetic acid

317,341,344-346,348,359, 2,4-dichloro-( 2,4-D), 560
360,392 2,3,6-trichloro-(2,3,6-T),560
biosynthesis, 355 Phenyl alanine, 144,216,264,266,
hydroxy-, 17,21,42 280,281,287,351,354,356,
-isoflavan, 17,334,339,346,357, 406,407,411,425,560,565,
359 568,569,571-575,583
Phaseolus, 333-335,343,346,351, Phenyl alanine ammonia lyase (PAL),
354,356,357 5, 10,55,73,86, 107, 124,216,
aureas, 335,336,344-346,354,357, 275,281,344,348,350,354,
360,365 356,379,380,383,398,401,
chemotaxonomy, 357 407,409-411,421-425,430,
coccineus, 335,336,345,346,348, 432-434,437,576,577,579
355,357,365 assay, 410-412
lunatus, 335,336,345,346,354- extraction, 41 1
356,359,365 isoforms, 425
mungo, 335,336,344-346,354- Phenyl isocyanide,478,479
356,365 Phenyloxy alkyl carbonic acid,293
vigna complex,334,335,357 Phenylpropanoids, 85-115
vulgaris, 9, 10, 17, 166,335,336, genetics, 92
344-346,354,357,365(see pathway, 85-115,169,406,407,421
also French bean) regulation, 90-92
Phaseoluteone, 337,347,352, 359, Phenyl-tert-butyl nitrone, 188
360 Phenyl sepharose,431
Phases of phytoalexinproduction, 44- Phenyl thiourea, 561
49 Phloroglucinol, 487,488
Phenanthrene(s), 336,348,349,356 reactive compounds, 174
biosynthesis, 356 Phoma
dihydro-, 337,343,356 arachidicola, 200,208,215,220,
2,7dihydroxy4,60xy- 222
9,10-dihydro-, 336,348, comlanata, 536
349 lingam, 230
6,7-dihydroxy-2,4ethoxy- vasicae, 544
9,10-dihydro-, 336,348, Phomopsis, 561
349 Phorbol-l2-myristate-l3-acetate,
2,7-dihydrOxy-1,3,5-trimethOxy-, 437
9,10-dihydro-, 336 Phosphatase, 54
Phenanthro-indolizidines, 546 Phosphatidyl choline,436
Phenol(s), 275,488,489,550,560, Phosphatidyl inositol(PI), 436
570,574,575 Phosphatidyl inositol4,S-
p - [242-halogenomethyl- bisphosphate (PIP2), 435-
1,3-dioxalan-2-yl)ethyl], 437
449 Phosphatidyl inositol4phosphate
Phenolic acids,97,487-489,535-550 (PIP), 436,437
Phenolics, 72,97,98,305,376,544, Phosphalidyl serine,436
549,550,577,579 Phosphite, 305,306
Index 609
PhosphoinositidaseC, 435,436 &Pinene, 177, 178

Phosphoinositide signal system, 429, Pinosylvin, 342,356,357,361
435-437 dihydro-, 336,342,345,348-351,
Phospholipase C,437 356,358,362,365
Phosphonate, 305,375-390 synthase, 356
Phosphon-D, 561 Pinus, 356,361
Phosphonic acid, 305,379 sylvestris, 356, 357
tritiated, 379 Piperazine, 437
Phosphorous acid, 264,266,277,280,. Piperidinium iodide, 577,578
282,305 Pisatin, 3,7,8, 16,44,46,49,54, 108,
Phototoxic phytoalexin, 188, 189 109,201,216,317
Phragmidiella heterophragmae, biosynthesis, 54
542 demethylase, 16,74,201, 385
Phyllactinia corylea,536 synthesis, 49
Phyllospella stakmanii,542 Pisum sativum,50 (see also Pea)
Phyllosticta eugeniae,539 Plant moisture stress (PMS), 171
Phytoecdysones, 545 Plasmalemma fractions, 148
Phytoimmunity, 141 Plasmids
Phytophthora, 3,75,76, 101, 375- DNA, 184
390,580 R-l, 202
blight, 503,525 Plasmopara viticola,288,290,291,
capsici, 19, 376, 379, 503-523 293,300-309,558,561
cinnamomi, 384 Polyacetylenes, 4,445, 558
citrophthora, 263-286, 384 Polyacrylamide gel electrophoresis
cryptogea, 376-380 (PAGE), 184
fragarieae, 384 Polyacrylic acid,580
infestans, 2, 12, 13,23,25, 119, Polyamines, 108, 109
121, 193,288,450,558 Polygalacturonase, 406,573
megasperma, 360 endo-, 558
var. sojae, 10, 12,21,85-116, Polygenes, cotton, 186
193 Polyphenol oxidase(PPO), 124,344,
f.sp. glycinea (Pmg), 10,13,21, 351,561,570,573-577
50,61,63,66,201,504, 516, Polyphenols, 123, 124, 130,486
517,559 Populus
f.sp. medicaginia, 21 euroamericana, 171
nicotianae laurifolia, 537
var. nicotianae, 381 maximowiczii, 537
parasitica 12, 263, 384 nigra var. italica, 537
var. sesami, 525 Poria weirii, 535
root rot, 69,70,74,503,505 Post-infectional compounds, 533-
sojae, 69-72 553
vignae, 21 Potassium phosphate,562
Phytuberin, 381,446 Potato, 2,3, 12,23,48,55,56, 164,
Phytuberol, 381 167, 169,193,431,445,447,
Pigeon pea, 572,580 450 (see also Solanum tuber-
a-Pinene, 177, 178 osum)
610 Index
Powdery mildew,542,560 3-hydroxy-22-thia(3-OHTP)-, 393,

barley, 42-44 394
Cassia, 536 3-methoxy-l l-thia-(3-OMeTP),
pea, 42,44 393,394
Vitk, 308 :NAPPH oxidoreductase, 397
Preformed glycosides, 393,397- oxidoreductase, 392
400 phytoalexins, 15, 401
Primary recognition, microbes,42 1l-thia-(TP), 393-396
Proanthocyanidins, 535,537,539, Pterostilbene, 288,289,293,309,318,
542-544 . 320-323,326,328,329
Probenazole, 375,561,577-580 Puccinia, 191
Production potential, stilbenes, 294 arachidis, 200,208,209,215,217,
Proline, 144 218,220,221
Propanoic acid, 1,2,3-( 2-acetyloxy)tri- graminis, 25,41
Carboxylic-, 209,211,214,215, Putrescine, 108, 109
220 Pyricularia, 191
Propylene oxide(POX), 382,383 oryzae, 14,489,495,518,539,567,
Propantheline bromide, 570 571,580
Protease(s), 107 Pyrrolidinium iodide, 578
inhibitors, 107 N,N-dimethyl-, 561
Protein-phenol complexes, 318 Pythium, 568,572
Protein kinase(s), 53,54,435,437, aphanidermatum, 12
438
Protein-lipopolysaccharide complex, Quercetagetin, 541
137,418 Quercetin, 90,91,537,538, 541,543,
Protocatechuic acid, 487,488 545-548
Pseudomonas 3’,4’-dimethoxy-, 535,538,541,
aeruginosa, 343 543,546
cichorrii, 348 7,3’-dimethoxy-, 457
syringae 3’-methoxy-, 539,540
pv. glycinea (Psg), 117, 187 4’-methoxy-, 537
pv. pisi, 187 7,3’,4’-trimethoxy-, 537, 541, 547
pv. tomato, 581 Quinazoline alkaloids, 544
Psophocarpus tetragonolobus, Quinic acid, 488
348 Quinones, 155
Psoralidin, 341,346
Pterocarpan(s), 85,337,340,351, Radioimmunoassay, 201,202,210,
353,354,358-361,406 430
biosynthesis, 354,355 Raffinose, 144
fUanO-, 337,341 Rajania, 335
&hydroxy-, 16 Raphanussativus, 237,247,251
3-hydroxy-9-methoxy- (see Medic- Receptor(s), 64,66,136, 142,428,
arpin) 429
3-hydroxy-8.9-methylene dioxy- sites, 63,428,429
(see Maackiain) substances, 144,156,429
Index 611
Resistance Rotunol, anyhdro-B-, 446

age-related, 503-523 Rps gene, 70,85,89,92
insect, 75,77 Rubiaceae, 537
reactions, 130, 144 Rust, 560
Resveratrol, 5,20,210-212,215,217, groundnut, 200
219,287-289,294-297,300, linseed, 560
303-310,318,320,322,325, wheat, 560,561
329,356
biosynthesis, 356 Saccharomyces cervisiae,11
dihydro-, 342,349,356,361 Salicylic acid, 106,222,578, 581
trans-, 342, 361 Salk, 171
Reverse-phase HPLC, 184,210,240, Saponins, 486,489,534,541-543,
241,292 547,548
Rhamnaceae, 543 Sapotoceae, 542
Rhamnose, 144 Saptoria nodorum, 17
Rheum rhaponticum, 356 Sarocladium oryzae, 25,26
Rhizobium trifollii, 25 Sativan, 339,358,406,407,411,412,
Rhizoctonia 414417,419-422,430,433
bataticola,2 15 Sclerotiniafructigena, 323
repens, 4 Sclerotium rolfsii, 568,569,573,574
solani, 17,70, 153, 154,571 Scoparone, 263-286,384
Rhizopus biosynthesis, 280,281
nigricans, 174 production
stolonifer, 14,525,559 effect of fosetyl-Al,277-280
Rice, 14,497,560,561,563,567,570, effect ofphosphorous acid, 277-
571,580 (see also Oryzasativa) 280
blast, 14,505, 518,560,564,567, radiation, 271-277
571,578 temperature, 270,271
brown spot, 560,564 Scopella echinulata,542
sheath blight, 571 Scopoletin, 163, 164, 168-173, 178
Rishitin, 8, 12, 13,21,48, 193,317, Scopolin, 169
381,445,446,447,450,461 Scutellarein, 548
Rishitinol, 446 Second messenger,246,429
mRNA, 6,27,55, 107, 195,354,422, Serine, 560
432,581 Sesamum indicum,525-531
chalcone isomerase,55 Seselin, 263
chalcone synthase,428 Sesquiterpene cyclase,185, 195
Ccoumarate: CoA ligase, 72 Sesquiterpenoid phytoalexins,12, 18,
phenyl alanine ammonia lyase,55, 132, 164,183-198,381,383,
56,72,428,434 445,558
pinosylvin synthase, 357 antibacterial activity, 187-189
Root rot Shikimate dehydrogenase,124
cotton, 153 Shikimate pathway, 398
pea, 560 Shikimic-polymalonic acid pathway,
Rosidae, 549 216,287
612 Index
Signal(s), 98-1 10 Spiroannulation techniques, 447

molecules, 55, 98-1 10 Spirobrassinin, 236,245
perception, 99 Spirodienone, 449,450
systems, 429,432,435437,438 Spirovetivane, 445
transduction, 53,54,99,427-438 phytoalexins, 445-466
Silver nitrate, 233,235,241-243 sesquiterpenes, 445
Sinapic acid,535 Staphylococcus aureus,343
Sinapis alba,247,251 Stemphylium
Sirodesmin PL,233,242,245,246 botryosum, 17,392-397
Sodium sarcinaeformae, 7,47
azide, 8, 14,26,497,562,565,572, Stenophylus
577 angustifolia, 348
diethyl malonate anion, 450 stenocarpa, 348
dodecyl sulfate, 11, 184 Stereoselective synthesis, spiroveti-
fluoride, 565,567,571-575 vanes, 445-466
iodoacetate, 8,497, 565,568,571- Steroids, 547,548
575 Stilbene(s), 208,210,212,215,219,
malonate, 565,567,571-575,583 287-315,317-331,333-373,
molybdate, 565,568,571-575 558
nitrate, 15 biosynthesis, 215,216,287,290,
selenite, 8,564,567, 568,570-575 292-299,303,310,356,357
sulfite, 564, 567, 568 3-chloro-4'-hydroxy-, 328
Solanaceae, 4,334,445,446,545 detoxification, 309
Solanescone, 457,458 dihydro-, 334,337, 342, 344,348,
Solanum tuberosum, 25,462(see also 351,356,357,361-365
Potato) 3,5-dimethoxy-4'-hydroxy-,26,318
Solavetivone, 445447,450,455,457, (see also Pterostilbene)
458,461 effect of fosetyl-Al, 305-307
3-hydroxy-, 457,458 3-hydroxy-5-methoxy dihydro-,342
synthesis, 452 mode ofaction, 309,317-331
Sophoraisoflavone,338,352 production potential, 294,296, 300,
Sorghum, 191 301,309
bicolor, 336,345, 361,363,364 synthase, 5,20,287,309,356, 357
Soybean, 26,27,42, 51,52,65,66, 4,3',5'-trihydroxy-, 208,212,215,
69-115, 117-128, 187, 193, 256,287,318(see also Resvera-
201,385,393,400,562,580 trol)
(see also Glycine maw) Stophostyles helvola,348
Sphathodea, 549 Strawberry, 384
campanulata, 543 Streblus, 547
Spermidine, 108, 109 asper, 546,547
Spermine, 108,109 Strobilanthes callosus(see Carvia cal-
Sphaceloma losa)
heterophragmae, 542 Strontium, 13
madhucae, 542 Sugar beet, 21,368
tectonae, 534 root rot, 574
Spinacia oleracea,3 18 Sugar maple (seeAcer saccharum)
Index 613
Sulphex, 526,528,530 Transaminase, 380

Superoxide ion, 78 Transaminic acid, 137
Suppressor(s), 41,50-52,55,56,66, Transgenic plants, 5,20,64, 222, 309
142-144,401 tobacco, 5,64,309
gene, 22 Traumatic acid, 106
Suspension cell cultures, 54,55,88, Trianthema, 545, 549
118, 124,411,421,422,425, portulacastrum, 545
430,431 2,3,5-Trichlorophenoxyaceticacid
Sweet potato, 4,429,467-483(see (2,3,5-T), 560,565,568,569,
also Ipomoeabatatas) 583
Synthesis inhibitors, 245,246 Tricophyton mentagyrophytes,343
Syringic acid, 535,537-541,544,545 Tricyclazole, 375
Systemic acquired resistance(SAR), Trifluoroperazine, 431
222,310,577,582 Trifluralin, 578
Syzygium, 539,549 Trifolieae, 334
cumini, 539,540 Trifolirhizin, 393,397-399
Trifolium
Tagetes, 189 prattens, 50
Tamus, 335 repens, 25,50,358
Tannins, 534,535,537,542,547 Trigonella, 334
Tea (see Camellia sinensis) 2,3,5-Triiodobenzoic acid(TIBA),
Tectona, 535,536,549 560
grandis, 534 Tripospermum juglandis,539
Tectoquinone, 535 Triticum aestivum,25
Terpene reductase,472,479 Trypsin, 421
Terpenoid phytoalexins,429 Tryptophan, 144
Tetrachloroethane, 413 Tungrovirus, 535
Thielaviopsis basicola,25 Tylocebrine, 546
Thiobendazole, 526,528, 530 Tylophora, 546,549
Thioglycollic acid,496,564, 567,568, asthmetica, 549
571-575,583 indica, 549
Thiophenes, 189 Tylophorine, 546
dihydrobenzo- (DHBTP), 393,394 Tylophorinine, 546
Threonine, 144,560 Tyloses, 135, 136, 140
Tobacco, 2,20,23,64, 169,222,309, Tyrosine, 144,287,328
376,381-384,445,447,556, . Tyrosine ammonia lyase(TAL), 344,
561,580,581(see also Nicoti- 348,350
ana tabaccum)
isogenic plants, 385 Ultraviolet light(UV),6,8,210,233,
mosaic virus,42,378 241,243,245,254,282,287,
transgenic plants, 5,64,309 288,293,295-297,299,301,
a-Tocopherol acetate, 328 308,445
Tomato, 13,21,23, 191,376,401, effect on scoparone production,
408,418,503,558,560,580 271-277
(see also Lycopersicum escu- effect on Vitis phytoalexin produc-
lentum) tion, 293,308
614 Index
Umbelliferone, 264 Viniferins, 288,293

Uncinula a-viniferin, 288,289,293,309
necator, 308 winiferin, 288,289,293-297,303,
tectonae, 534 304,309,310
Uromyces, 558 Vitaceae, 317,318,320,329,356
appendiculatus, 42,57 Vitexin, 541
Urticaceae, 546 4’-OMe-, 544
Ustilago maydis,25 VitiS,287-3 16
acerifolia,297,301
Valine, 144 andersoni, 296,297,299,301
Vasicine, 544, 545 argentifolia, 296,297,299,301
Vasicinone, 544,545 champini, 296,297,299,301
van der Waalsvolumes, 325,328 cinerea, 296,299,301
Vanillic acid, 535-548 doaniana, 294,295,297,301
Venturia inaequalis,70,288,560 Iabrusca, 297
Verapamil, 54,432,434 longii, 296,299
Verbenaceae, 534,546 riparia, 294-297,299,301
Verticillium, 405,425 rupestris, 294,297,299,301,302,301
albo-utrum, 13,21, 174,405- treleasi, 296,297,299,301
443 vinifera, 287,294-297,299,301,
dahliae, 131,134, 135,137-139, 302, 307 (see also Grapevine)
142,143,145,146,148,149, Voandozeia subterranea,335
152, 154, 156,268,405,408 &Vulgarin, 21
wilt, 129-160,560,561
Vestitol, 334,339,358,406,407,411, Water cress, 239
412,414417,419-421 Wheat, 285,560,561,564,567,571,
demethyl, 339,346,348,359 576
iso-, 339,348 Wine analysis,292
a-Vetispirene, 447 Withaferins, 545,546
Vetiver oil, 445 Withania, 545,549
8-Vetivone, 445,446,457 somnifera, 545
Viciafaba, 3,50 Wound response, 167, 168
Victorin, 14 Wyerol, 19
Vigna, 333-335,346,357 Wyerone, 19
angularis, 335,345,346 acid, 19
chemataxonomy, 357 epoxide, 19
mungo, 336
radiata, 335,336 Xanthomonas
sinensis, 50 campestris, 125, 126,243
subterranea, 335,345,346,348 pv. campestris, 187,232
umbellata, 335, 345,346 pv. glycines, 118, 120-123
unguiculata, 123,335,346,354,376 pv. malvacearum, 24, 153, 154,
(see also Cowpea) 168,183-198
Vignafuran, 342,347,348,355,377, oryzae, 561
378 Xanthone, 541
Index 615
Xanthotoxol, 72 Zea mays, 25,58

Xanthoxylin, 263, 369 Zeatin, 109
Xanthyletin, 263,269 Zinc, 56
Xenobiotics, 154-155 Salts, 564,567-569,583
Xylem, 132,135,138,140-143, 151, Ziwphus, 543,549
152 oenoplia, 543
Xylose, 147

Handbook of Phytoalexin Metabolism and Ac

Uploaded by

Copyright:

Available Formats

Handbook of Phytoalexin Metabolism and Ac

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Handbook of Phytoalexin Metabolism and Ac

Uploaded by

Copyright:

Available Formats

Handbook of

Soil Biochemistry, Volume l , edited by A. D. McLaren and G. H. Peterson

Organic Chemicalsin theSoil Environment, Volumes l and 2, edited by C.

Additional Volumes in Preparation

Seed Development and Germination,edited by Jaime Kigel andGad Galili

Nitrogen Fertilization in the Environment, edited by Peter Edward Bacon

Phytohormonesin Soils: Microbial Production andFunction, W. T. Franken-

Marcel Dekker, Inc. New York. Basel Hang Kong

Handbook of phytoalexin metabolism and action / edited by M. Daniel,

This book is printed on acid-free paper.

Copyright 0 1995-by MARCEL DEKKER, INC. All Rights Reserved.

MARCEL DEKKER, INC.

Current printing (last digit):

PRINTED IN THE UNITED STATES OF AMERICA

Ever since Mullerand Borger proposedthe concept of phytoalexins in 1940,

Phaseolus, Vigna, Dioscorea, tobacco, alfalfa, redclover, potato, sweet

the Past 50 Years

2. Phytoalexins and Host Specificity in Plant Diseases 41

3. Elicitor and the Molecular Basesof Phytoalexin Elicitation 61

4. Soybean Phytoalexins:Nature, Elicitation, Modeof Action,

5. Cellular Biochemistryof Phenylpropanoid Responsesof

6. Culture Darkening, Cell Aggregate Size,

7. Phytoalexins as a Factor in the Wilt Resistanceof Cotton 129

8. Cotton (Gossypium hirsutum)Strategies of Defense Expression:

10. Chemistry, Biology,and Role of Groundnut Phytoalexins

12. Scoparone (6,7-Dimethoxycoumarin),a Citrus Phytoalexin

13. Stilbene Phytoalexinsand Disease Resistance inVitis 287

14. Mode of Toxic Actionof Vitaceae Stilbeneson Fungal Cells 317

15. Inducible Compounds in

16. Involvement of Phytoalexins inthe Response of

18. Induction of Phytoalexin Synthesis Medicago

19. Stereoselective Synthesis of Spirovetivane-Type Phytoalexins 445

20. Enzymic Conversionof Furanosesquiterpene in Ceratocystis

21. Defense Strategies of

22. Effects of Age-Related Resistanceand Metalaxyl on

23. Induced Chemical Resistance in Sesamum indicum

24. Phytoalexins and Other Postinfectional Compounds

25. Possible Roleof Phytoalexin Inducer Chemicals in

S. A.Adesanya,Ph.D. Department of Pharmacognosy,Faculty of

M. Daniel, Ph.D. Reader, Department of Botany, Faculty of Science, The

C.J.Luczka-Bayles,M.S. AssistantDirector,FlowCytometry and

Asoke Kumar Sinha, PhD. Professor, Department of Plant Pathology,

Christopher J. Smith, PhD. BiochemistryResearch Group, Schoolof

Richard N. Strange, PhD. Department ofBiology,UniversityCollege

P. V. SubbaRao, PhD.* Centre de CoopQation Internationale en

Abraham Sztejnberg, Ph.D. Associate Professor of Plant Pathology and

YoshijiTakemoto, PhD. Faculty of PharmaceuticalSciences,Osaka

IkuzoUritani, PhD.? Faculty ofDomesticScience,NagoyaWomen’s

J. Michael Williams,PhD.. Department of Chemistry, University of Wales,

Masaaki Yoshikawa, PhD. Department of Biology, Faculty of Science,

Robert M. Zacharius, PhD. Principal Consultant, R. M. Zacharius and

S. S. Zeltser, PhD. Institute of Experimental Biology ofPlants, Academy

H. J. Zeringue, Jr. Chemist, Commodity Safety ResearchUnit, Southern

+Currentq f f i l i o n : Department of Biology, University College Lnndon, London, England

A. Some Relevant Research Prior to Phytoalexin Theory Proposed

pointed out that symptoms of recovery in tobacco plants infected with

B. Experiments lead to Phytoalexin Theory and

antimicrobial, low molecular weight, secondary metabolite formed de novo

C. Early Work on Phytoalexins

Muller and Borger (1940) first detected an antifungal substance known as