Glutenworkshop

Proceedings
11th International Gluten Workshop

Beijing, China
August 12–15, 2012
Zhonghu He and Daowen Wang (eds.)

Organizers
Chinese Academy of Agricultural Sciences/Institute of Crop Science/National Wheat
Improvement Center
Chinese Academy of Sciences/Institute of Genetics and Developmental Biology/State Key
Laboratory of Plant Cell and Chromosome Engineering
International Maize and Wheat Improvement Center (CIMMYT)
Sponsors
Ministry of Science and Technology (MOST), National Basic Research Program on Genetic
Studies of Wheat Agronomic Traits and Varietal Improvement (2009CB118300)
Ministry of Agriculture (MOA), International Collaboration Program (2011-G3)
National Natural Science Foundation of China (NSFC)
Conference Exhibitors
Brabender
Chopin Technologies
Perten
Headquartered in Mexico, the International Maize and Wheat Improvement Center (known by its
Spanish acronym, CIMMYT) is a not-for-profit agriculture research and training organization. The
center works to reduce poverty and hunger by sustainably increasing the productivity of maize
and wheat in the developing world. CIMMYT maintains the world’s largest maize and wheat
seed bank and is best known for initiating the Green Revolution, which saved millions of lives
across Asia and for which CIMMYT’s Dr. Norman Borlaug was awarded the Nobel Peace Prize.
CIMMYT is a member of the CGIAR Consortium and receives support from national governments,
foundations, development banks, and other public and private agencies.
© International Maize and Wheat Improvement Center (CIMMYT) 2013. All rights reserved.
The designations employed in the presentation of materials in this publication do not imply the
expression of any opinion whatsoever on the part of CIMMYT or its contributory organizations
concerning the legal status of any country, territory, city, or area, or of its authorities, or concerning
the delimitation of its frontiers or boundaries. The opinions expressed are those of the author(s),
and are not necessarily those of CIMMYT or our partners. CIMMYT encourages fair use of this
material. Proper citation is requested.
Correct citation: Zhonghu He and Daowen Wang (eds.). 2013. Proceedings: 11th International
Gluten Workshop, Beijing, China, August 12–15, 2012. Mexico, D.F.: International Maize and Wheat
Improvement Center (CIMMYT).
ISBN: 978-607-8263-30-1
AGROVOC Descriptors: Gluten; Wheat; Seed quality; Protein quality; Protein content
AGRIS Category Codes: Q04 Food Composition

Q01 Food Science and Technology
Dewey decimal classification: 572.651 HE
Printed in Mexico.
ii
Contents
v Zhensheng Li: Welcome Speech

vi Tom Lumpkin: Letter to the Participants of the 11th International Gluten Workshop
viii Shumin Wang: Welcome Speech
ix Acknowledgements
1 Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics

1 Wheat A genome sequencing and its application for quality modification
2 Wheat proteogenomics: mapping proteins and grain phenotype to the genome
7 Creating a detailed map of the wheat flour proteome: a critical step in understanding the
effects of environment on flour quality and immunogenic potential
13 Monitoring of storage proteins: accumulation and impact of high temperature in wheat kernels
using proteomics
18 Proteomic analysis of mature kernels of Fusarium graminearum-infected transgenic bread
wheat expressing PGIP
22 Genomic analysis of the expressed /-gliadin gene region in hexaploid wheat
27 Profiling gliadins from common wheat using two-dimensional gel electrophoresis
31 Session 2: Biosynthesis, structure, and functional analysis of storage protein
31 An asparagine residue affects LMW-GS maturation processing
34 Intracellular trafficking and tissue distribution of gluten proteins: the epitope-tagging approach.
38 The trafficking of wheat prolamins in rice endosperm
43 Gluten protein structures: variation in wheat grain and for various applications
47 A model for the transcriptional regulation of low molecular weight glutenin genes in wheat
52 Variation and classification of toxic epitopes related to celiac disease among α-gliadin genes
from four Aegilops genomes
62 The interactive effects of wheat storage proteins in a gluten free model system
67 Genetic variation of dough properties independent of high and low molecular weight glutenin
subunit genes in wheat
71 What causes extreme gluten quality variation in winter wheat in Norway?
73 Changes in protein components and gluten content in noodle processing and cooking
78 In silico analysis of transcriptional regulation of HMW glutenin genes in wheat
83 Session 3: Improvement of end use qualities by genetic and alternative approaches
83 Transferring research findings into industry application
85 Wheat quality improvement in China: progress and prospects
91 Transformation of common wheat (Triticum aestivum L.) with an avenin-like b gene improves
flour mixing properties
96 Update on low-molecular-weight glutenin subunit identification
102 Gliadin composition of 137 doubled haploid wheat lines derived from a cross between
Kariega and Avocet
105 Comparison of glutenin subunit composition among Australian and North American
wheat classes
107 Understanding the effects of HMW-GS and LMW-GS on processing quality using near-isogenic
lines and development of functional markers for LMW-GS in wheat
114 Estimating dough properties and end-product quality from flour composition
119 Reliability of gluten-related small scale tests to estimate dough viscoelasticity and bread
loaf volume
iii
123 New possibilities in micro-scale wheat quality characterisation: micro-gluten determination
and starch isolation
128 Characteristics and evaluation parameters associated with cooking quality of Chinese
fresh noodles
137 Optimization and application of capillary electrophoresis for rapid separation and
characterization of water-soluble proteins in wheat grains
143 The effects of bug (Eurygaster spp. and Aelia spp.) damaged flours on quality characteristics
of cakes, cookies and breads
148 Molecular cloning and characterization of four i-type LMW glutenin subunit genes from
T. zhukovskyi and T. compactum
153 Effects of interactions between high molecular weight glutenin subunits and waxy alleles on
dough-mixing properties in common wheat
158 Application of molecular markers for HMW-GS in wheat cultivar improvement
162 Complex quality characterisation of Hungarian wheat cultivars
166 Bread-making qualities of a blend of protein-removedrice flour and extra-strong wheat flour
169 Effect of cultivar and roller milling on the levels of phenolic compounds in Italian durum wheat
174 The effect of bug (Eurygaster spp. and Aelia spp.) damaged flours on formation of acrylamide
and hydroxmethylfurfural (HMF) in cakes, cookies and breads
179 Session 4: Starch and health attributes of the wheat grain
179 Starch biosynthesis in rice grains: natural variation and genetic improvement
180 Bio-fortification for high micronutrients in wheat breeding programs in China
185 Allergen potential of non-prolamin seed proteins in Brachypodium distachyon
190 Blood pressure lowering peptides from wheat gluten
195 How reliable are the measurements of residual gluten in gluten-free foods?
200 Effects of HMW- & LMW-glutenins and grain hardness on size of gluten polymers
206 Structural and mechanical properties of compression-molded wheat gluten, gliadin and
glutenin enriched films
209 DNA sequence variations of four Wx alleles generating polymorphic Wx-A1 protein for
decreased amylose content in wheat
iv Proceedings 11th InternaƟonal Gluten Workshop

Zhensheng Li: Welcome Speech at 11th International
Gluten Workshop, Beijing, China, August 12–15, 2012
Ladies and gentlemen, dear colleagues:

On behalf of all wheat scientists in China, I warmly welcome you to Beijing and to take part in the 11th
International Gluten Workshop.
Wheat is the most widely cultivated food crop in the world, and provides vital energy and protein for
over 60% of the population on earth. As is well known, the most important characteristic of wheat,
which no other crop can replace, is the presence of its flour gluten proteins that allow the processing
of a multitude of delicious foods such as breads and noodles. In this sense, research focused on
gluten is very important, and the International Gluten Workshop, dedicated to gluten and related
research topics, is a great venue for wheat researchers who study and improve wheat quality. I believe
this workshop will yield fresh insights into the genetic, biochemical and molecular bases of gluten
functionality, which will help to make wheat an even better food crop for all people. This is a great
challenge, and it is also our sacred goal and duty.
As a wheat researcher for over 60 years, I am very happy and excited to see that many young scientists,
who work on wheat gluten and quality, are present at this workshop. With the efforts of so many
talented people, I am sure that the above goals can be reached in the not too distant future.
Finally, I hope all of you enjoy the workshop and have a pleasant stay in Beijing.
Thank you.
Zhengsheng Li: Welcome Speech 11th InternaƟonal Gluten Workshop v

Tom Lumpkin: Letter to the Participants of the 11th
International Gluten Workshop
Dear colleagues and friends:

Warm greetings from Mexico!
I am very sorry not to be able to join you for the 11th International Gluten Workshop, due to my prior
commitments. By way of this letter, it gives me pleasure to express the full support of CIMMYT for
this important workshop and our strong commitment to collaboration with China in wheat research
and development.
In May 2012 CIMMYT and CAAS celebrated 30 years of successful collaboration. I know that
CIMMYT’s contributions in germplasm, knowledge sharing, networking, and training are highly
appreciated in China. More than 260 Chinese wheat varieties are derived from CIMMYT germplasm;
40 of those have been sown on a combined area of more than 45 million hectares. The 5+10 glutenin
subunit in quality-improved Chinese commercial wheat was derived from CIMMYT germplasm.
More than 200 Chinese scientists have participated in training courses or visiting scientist programs
organized by our center, and more than 60 postgraduates have received training.
CIMMYT has the largest investment in China among CGIAR centers. Five collaborative research
programs led by CIMMYT scientists posted in China have been established at CAAS and in Yunnan
and Sichuan. This has created a new model for CGIAR-China collaboration and increased CIMMYT’s
impact in the country.
An excellent example of this impact is the CIMMYT-CAAS wheat quality laboratory. In the early
1990s, wheat quality was a new area of research for China. When CAAS requested additional
collaboration from CIMMYT on the topic, cereal chemist Dr Roberto Javier Peña and CIMMYT
liaison officer in China, Dr Zhonghu He, worked together to organize training activities in Mexico
and China. Led by those two scientists, partnerships were established with Australia, USDA-ARS,
Japan, France, UK, and research institutes in other countries.
As a result, there has been tremendous progress in standardizing quality testing systems and in
developing and perfecting molecular markers. Forty functional wheat quality markers are now
used in various wheat breeding programs in China. Three advanced lines developed using those
markers are expected to be released in the next few years, and the markers have been widely used to
characterize Chinese and CIMMYT germplasm. The Chinese wheat quality project led by Zhonghu
He received the first class science award from State Council in 2008, and Dr. He was recently
promoted to the category of CIMMYT Distinguished Scientist—only the seventh researcher to receive
this distinction in the history of our center.
Speaking of history, humanity currently stands at a crossroads in the evolution and development of
one of its oldest and most important domesticated crops. Wheat accounts for one-fifth of the world’s
food. It is second to rice as a source of calories in the diets of developing country consumers, but it is
first as a source of protein. Demand for wheat in the developing world is projected to increase 60%
by 2050. At the same time, climate-change-induced temperature increases are likely to reduce wheat
production in developing countries by 20–30%.
vi Proceedings 11th InternaƟonal Gluten Workshop

As wheat scientists, we know that new technology will help us to provide solutions for these
problems. We also know that wheat is a global crop that faces global challenges; the challenges can be
met only through strong partnerships worldwide. Events like this workshop represent an invaluable
opportunity to share experiences, establish linkages, and commit to joint work on the issues facing the
wheat crop. Given the long time frame for outputs from our research, we know that what we do now
will bear fruits for a new generation of farmers and consumers a decade hence.
As an organization, CIMMYT focuses on applied science and technology, as well as networking and
sharing knowledge. But we can achieve nothing without your collaboration and support. On behalf
of CIMMYT, I would like to express my sincere gratitude for our partnerships with all of you to date.
Through initiatives like the CGIAR Research Program WHEAT, we are moving swiftly and actively to
strengthen and expand those relationships.
I encourage you to contact Roberto Javier Peña and Zhonghu He, or write to me directly, with ideas,
suggestions, or criticism, and above all I wish you a successful workshop and a nice stay in Beijing.
Sincerely yours,
Thomas A. Lumpkin
Director General
The International Maize and Wheat Improvement Center
Tom Lumpkin: LeƩer to the ParƟcipants of the 11th InternaƟonal Gluten Workshop vii
Shumin Wang: Welcome Speech at 11th International
Gluten Workshop, Beijing, China, August 12–15, 2012
Dear colleagues and friends:

On behalf of the Institute of Crop Sciences, Chinese Academy of Agricultural Science, I am very
pleased to welcome all participants of the 11th International Gluten Workshop, and particularly,
the international participants, who have travelled long distances to come to this workshop. We are
honored to be joint organizers of this workshop with our long term collaborators in the Genetic and
Biology Development Institute of the Chinese Academy of Science, and the International Maize and
Wheat Improvement Center. These three institutes, together with others, successfully organized
the International Wheat Quality Conference in 2004. We are also grateful to the sponsors of this
workshop, i.e. National Natural Science Foundation of China, International Collaboration Project
from MOA, National Basic Research Project from MOST, and three private companies involved with
quality testing equipment.
Wheat is the third leading cereal in China, with a production of around 110 million tonnes,
representing 23% of cereals grown in China. China is the largest wheat producer and consumer;
therefore, Chinese wheat production is crucially important both to domestic food security and
international markets. Before 1990, improvements of yield and disease resistance were the major
objectives due to China’s large population size and limited arable land. Wheat quality research,
mainly processing quality, was initiated in the mid-1980s to meet an increasing demand for better
quality and food diversity due to significant increases in living standards. Very recently, consumers
have become concerned about nutritional quality and food safety; therefore, both processing and
nutritional quality are currently important subjects. In terms of food type, quality of traditional
products such as noodles and steamed bread needs further improvement, but pan bread and soft
wheat products have much higher commercial values, largely due to the demand for food diversity.
Chinese wheat has acceptable protein content, but very weak gluten quality, and is therefore not
suitable for producing pan bread, and even not suitable for production of noodles and steamed bread
under mechanized or semi-mechanized conditions. Improvement of gluten quality is now a key
objective for meeting these challenges. Clearly, it is an excellent choice to have the 11th International
Gluten Workshop in China. During the last 20 years, our institute has made significant progress in
developing methodologies for testing Chinese products and development of molecular markers using
the comparative genomics approach. High quality varieties have been released and promoted by
provincial programs. This success is largely due to our domestic and international collaborations. Our
institute has established close collaborations with CIMMYT, Australia, USDA-ARS, Japan, and France,
and many other countries; indeed, your collaborations and contributions are highly appreciated.
However, we have to recognize that wheat quality research in China is still new and not much
research was done on many traditional products; consequently, current wheat quality still fails to
meet the needs of the milling and food industries. We are facing great challenges; for example, how
to combine yield and quality together, how to reduce quality variation under climate change, and
how to transfer the new technology into practical breeding programs? I believe that collaboration is
the key way to deal with these challenges. Therefore, I hope this workshop will provide an excellent
opportunity to exchange the latest developments in gluten research, and particularly in the area of
integration of new and conventional technologies, and in development of new tools and new methods
for breeding programs. I strongly believe that this workshop will provide valuable information and
ideas for improving wheat quality in China.
In ending, I wish you a successful workshop; have a nice stay in Beijing, and thank you for your attention.
viii Proceedings 11th InternaƟonal Gluten Workshop

Acknowledgements
The 11th International Gluten Workshop would not have been possible without the valuable
contributions of the Organizing Committee. I would like to sincerely acknowledge the advice and
help of my committee colleagues: Xu Liu, CAAS, China; Rudi Appels, Murdoch University, Australia;
Roberto Javier Pena and Hans-Joachim Braun, CIMMYT, Mexico; Gerard Branlard, INRA, France;
Craig Morris and Susan Altenbach, USDA-ARS, USA; T.M. Ikeda, NARO, Japan; W. John Rogers,
Universidad Nacional del Centro de la Provincia de Buenos Aires, Argentina; Wujun Ma, Western
Australia Department of Agriculture and Food, Australia; Domenico Lafiandra, University of Tuscia,
Italy; Perry K. W. Ng, Michigan State University, USA; Shunhe Cheng, Yangzhou Agricultural
Research Institute, China; Aimin Zhang, CAS, China; Xianchun Xia, CAAS, China; Yueming Yan,
Capital Normal University, China; and Daowen Wang, CAS, China.
Many committee members also chaired sessions along with scientists listed below: Angela Juhasz,
Hungarian Academy of Sciences, Hungary; Guangyuan He, Huazhong University of Science and
Technology, China; Frank Bekes, FBFD PTY LTD, Australia; Yimin Wei, CAAS, China; and Tuula
Songtag-Strohm, University of Helsinki, Finland. I am also very grateful to Aimin Zhang and his staff
at the Institute of Genetics and Developmental Biology, CAS, and Na Zhang, Simin Li, Yonggui Xiao,
Yong Zhang, and Yan Zhang at the Institute of Crop Science, CAAS, for organizing the visit to the
wheat labs in these two institutes and for other logistical support.
Financial support was generously provided by the National Natural Science Foundation of China
(NSFC, 31210303039), National Basic Research Program on Genetic Studies of Wheat Agronomic
Traits and Varietal Improvement (2009Cb11830) from the Ministry of Science and Technology, and the
International Collaboration Program (2011-G3) from the Ministry of Agriculture. I also particularly
wish to acknowledge the participation, exhibition and financial contributions of Brabender, Chopin
Technologies, and Perten in this workshop.
This workshop would not have been possible without administrative support from the Institute of
Crop Science, CAAS, the Institute of Genetics and Developmental Biology, CAS, and CIMMYT.
Finally, I am sure all authors and the organizing committee join with me in giving special thanks
to Professor Robert McIntosh for editing the manuscripts included in these proceedings, and to the
CIMMYT editorial team for design and printing.
Z. H. He
Chair of the Organizing Committee
Acknowledgements ix
Session 1: Capturing discoveries from genomics,
proteomics, and transcriptomics
Wheat A genome sequencing and its application for

quality modification
H.Q. Ling, A. Zhang, D. Wang, D. Liu, J.Y. Wang, H. Sun, H.J. Fan, Y.W. Li, Z.S. Li, Y. Zhao, D.W. Wang,
K.P. Zhang, Y.S. Yang, J.J. Wang, and L. Dong
The State Key Lab of Plant Cell and Chromosome Engineering, Institute of Genetics and Developmental Biology,
Chinese Academy of Sciences, Beijing 100101, China
Wheat is one of the most important food crops in the world, and is more widely cultivated and
consumed than other major cereals. Because of its immense importance for human subsistence, major
international efforts are being devoted to study and improve the agronomic traits of wheat, such
as yield, quality, nutrient use efficiency, and resistance to biotic and abiotic stresses. Wheat quality
includes multiple aspects (i.e., processing, milling, flour color, starch, functional nutrient, etc.), each
involving the action of multiple genes and being influenced by environmental conditions. Although
conventional genetic and breeding studies have contributed substantially to understand and improve
wheat quality traits, they usually focus on a small number of genes, and may not be highly efficient
in revealing the functional complexities underlying the different types of quality traits. Functional
genomics studies, conducted at genome wide level, offer high prospects for dissecting, understanding,
and modifying complex agronomic traits. However, these kinds of studies are currently not possible
for wheat owing to the lack of complete genome sequence information. The huge and complex
polyploid genomes in common and durum wheats have made complete genome sequencing difficult
in these species. Thus, alternative and complementary strategies are needed for obtaining genome
sequencing information for wheat. Our State Key Laboratory has chosen to sequence the A genome
of Triticum urartu (AA, 2n = 2x = 14), which is the progenitor of the A genomes in many polyploid
wheats (including common and durum wheats). Using next generation sequencing facility, coupled
with transcriptome sequencing of multiple organs, a draft genome sequence of Triticum urartu
has been generated. In this presentation, the salient features of Triticum urartu A genome will be
briefly described. The major emphasis of the talk will be placed on several lines of ongoing research
concerning the application of Triticum urartu genome sequencing information and resources in studying
and improving wheat quality traits.
Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 1

Wheat proteogenomics: Mapping proteins and grain
phenotype to the genome
B. Chapman1, P. Moolhuijzen1, W. Ma1, M. Bellgard1, Y. Yan2, S. Wang2, A. Juhasz3, F. Bekes4,
and R. Appels1
1Center for Comparative Genomics, Murdoch University, Perth, WA 6150, Australia; 2Capital Normal University,
Beijing, China; 3Agricultural Institute, Center for Agricultural Research of the Hungarian Academy of Sciences,
Martonvásár 2462, Hungary; 4FBFD Pty Ltd, Beecroft, NSW 2119, Australia
Abstract
The unique transcripts from a range of wheat gene transcripts datasets and coding sequences predicted
from the wheat genome survey sequences (International Wheat Genome Sequencing Consortium; IWGSC)
were resources used to establish a dataset of predicted protein entities with their respective molecular
weights estimated. This dataset was then cross-referenced to available data for the molecular weights
of proteins reported in the literature as well as from our own analyses of grain protein in samples from
structured mapping populations. The chromosomal locations from a mapping population of 225 doubled
haploid lines using high throughput MALDI-TOF analyses could be compared to predicted locations of
proteins determined from the chromosome arm-based analysis of the wheat genome survey sequences. A
pilot study on the proteogenomic analysis of wheat utilized published tandem mass spectrometry spectra
from proteins separated on 2D gels, and this study validated annotations defined by AUGUSTUS gene
predictions (using the maize model). We found 170 peptides in total, mapped to 76 known genes and 5
novel genes. A more complete analysis using gene models predicted from FL-cDNAEST and RNA-seq
evidence and mass spectra from shotgun mass spectrometry will provide a higher proteome coverage
and ultimately provide a more comprehensive insight into the gene space within the wheat genome. The
study highlighted some technical challenges in resolving proteins close to each other in molecular weight.
This can now be addressed with newer mass spectroscopy technologies. The available dataset of predicted
proteins was also analyzed for epitopes of potential interest for identifying grain proteins that may cause
health issues in consumers of wheat-based products.
Introduction in dryland/rain-fed environments. Understanding

Wheat is a major food crop and although the specific requirements for producing Chinese
Australia produces only about 24 million tonnes, food products in the Australian environment will
it is a major exporter at a value of approximately anticipate many environmental conditions that
$5 billion per year (ABS 2013). Australian wheat may develop in China as the effects of climate
performs poorly in China, mainly due to the fact change and loss of arable land are felt in future
that the Australian wheat evaluation system is years. Defining the entire protein complement of
based on western style food products. China has wheat is now feasible (Juhász et al. 2012) to aid
unique food products made from wheat, which in calibrating the grain from a selected cultivar
include a wide range of steam-bread and noodle grown in a particular environment, to wheat
types across the provinces that require different grain in particular areas of China for specific food
flour processing characteristics. For example, products.
steam-bread in North China requires chewing
resistance and stomach filling attributes, whereas
in South China it needs to be soft in texture. It Materials and methods
is widely accepted that Chinese wheat cultivars Proteins analyzed
possess better attributes for making Chinese
food products, hence the low marketability of Gluten protein fractions from 225 doubled haploid
Australian wheat in China. However, Australia lines from a Westonia × Kauz cross (Zhang et al.
has a long history of modern wheat production submitted) were analyzed on a standard Voyage
2 Proceedings 11th InternaƟonal Gluten Workshop

MALDI-TOF to identify peaks that segregated The gene annotations from the proteogenomic
in the population, in order to locate them to a analysis were deduced from matches to the
genetic position in the chromosomes. 6-frame translations of the genome sequences
in the International Wheat Genome Sequencing
Mass spectra from wheat cv. Butte 86 (Dupont
Consortium (IWGSC) survey sequence database
et al. 2011) grain endosperm were kindly
(wheat-urgi.versailes.inra.fr/Seq-Repository),
supplied by S. Altenbach and W. Vensel (USDA,
which were cross-referenced with AUGUSTUS-
California). The mass spectra were generated
predicted genes (using the maize gene model).
using a QSTAR Pulsar i quadrupole time-of-
Any peptides matching outside the predictions
flight mass spectrometer (Applied Biosystems/
were classified as novel annotations and any
MDS Sciex, Toronto, Canada), derived from
peptides matching within the gene predictions
protein spots separated by 2D gel electrophoresis
were classified as known annotations. A
and digested with chymotrypsin, trypsin and
valuable feature of the IWGSC database is that
thermolysin. The spectra from the trypsin digest
the genome sequences are effectively binned
were used in this analysis.
into chromosome arm locations because of the
Bioinformatics chromosome arm flow-sorting that was carried
out to produce the DNA for sequencing (Feuillet
Wheat proteome mapping was carried out et al. 2012; Doležel et al. 2012).
using the proteogenomic pipeline developed
in Arabidopsis (Castellana et al. 2008), using
the latest revision of the pipeline, which also Results and discussion
includes a new rescoring step based on MSGF
(Kim et al. 2008). The assignment of annotations Assigning MALDI-TOF peaks in the low
at the final stage of the pipeline uses an Event molecular glutenin group of proteins to a
Probability, which is calculated based on the genetic location
number of locations to which the peptide was The assignment of high and low molecular
mapped in the genome and the best Local False weight glutenin subunits observed in MALDI-
Discovery Rate (lFDR) for the peptide (lFDR TOF mass spectroscopy to respective subunits
for the best spectrum representative for that assigned to genetic loci followed Liu et al. (2010)
peptide). The software, InsPecT (Castellana and Gao et al. (2010). Presence/absence variation
et al. 2008), was run setting the protease to between the two parents (Westonia and Kauz)
Trypsin, the instrument to QTOF, with fixed was used to generate presence/absence scores
modifications to carbamidomethylation, variable in a double haploid genetic population derived
modifications to oxidation of methionine, from a cross between the two parent lines and
N-terminal carbamylation, and allowing for two thus to assign genetic locations as described by
post translational modifications per peptide. Chen et al. (2007). Even though the presence/
The database was set up using a 6-frame absence scoring of the protein peaks meant
translation (stop-to-stop open reading frame) that the genetic locations did not have high
with a minimum open reading frame size of 40 confidence scores, the HMW and LMW glutenin
amino acids and a contaminants database, which subunits mapped to the long and short arms of
consisted of known peptide sequences derived the group 1 chromosomes in a genetic map with
from trypsin and keratin. A decoy database approximately 2,000 markers (mainly SNPs),
was also included which consisted of reversed compiled using 225 doubled haploid lines
sequences from the target database. from the Westonia × Kauz cross. In addition to
Peptides were mapped to the genome and these expected assignments to the genetic map,
clustered. Each peptide cluster required at some LMW glutenin subunit peaks (molecular
least one peptide, which was unique within the weights 31,965 and 32,831) mapped to the long
genome. If one peptide was found in the cluster arm of chromosome 7A, consistent with an early
and it was unique, then the respective cluster was report by Salcedo et al. (1980) who reported low
considered in the proteogenomic analysis. The molecular weight gliadins on chromosomes 4B
maximum distance between each cluster was set and 7A. A search of gene predictions within the
to 7,400 bp, which was chosen based on 95% of IWGSC survey sequence data for gliadin-related
gene sizes in rice. genes revealed hits on 4AL and 7DS for Cys-rich

gliadin N-terminal domain-containing genes to the available wheat genome sequences in the
and we are currently investigating whether these IWGSC survey sequence database. The example
genes may be homologs to the LMW glutenin reported here is based on peptides from trypsin
subunits mapped to the long arm of 7AL in the digests, derived from the Dupont et al. (2011)
Westonia × Kauz map. data, and indicated that a total of 164 non-
redundant peptides were mapped to known
The newer mass spectroscopy technology
locations (based on the IWGSC AUGUSTUS gene
(described below) to analyze the amino acid
predictions), whereas 6 were mapped to novel
sequence of specific peptides will provide a more
locations. The 170 peptides in total, mapped to
accurate approach for assaying polymorphisms
76 known genes and 5 novel gene annotations
in proteins for genetic mapping.
(which included 1 novel gene mapping outside
Mapping proteome data to the developing predicted genes). This analysis did not utilize
wheat genome sequence a splice-graph (Castellana et al. 2008), due to
technical hurdles at the time as a result of the
The proteogenomic pipeline developed in large size of the wheat genome, and as a result
Arabidopsis by Castellana et al. (2008) was used novel splicing events may have been missed. An
in a pilot study to assign peptide mass spectra example of the analysis is shown in Table 1.
Table 1. Sample alignments of peptides to International Wheat Genome Sequencing Consortium

(IWGSC) survey gene predictions.
IWGSC_6DS_v1_2096398 11638 11680 + Peptide=LQCVGSQVPEAVLR

IWGSC_6DS_v1_2096398 11680 11725 + Peptide=DCCQQLADINNEWCR
IWGSC_6DS_v1_2096398 11824 11857 + Peptide=LTAASVPEVCK
IWGSC_6DS_v1_2096398 11824 11887 + Peptide=LTAASVPEVCKVPIPNPSGDR
IWGSC_7BS_v1_3131167 3297 3329 - Peptide=SIGNGVQFLNR
IWGSC_5BS_v1_2275541 8501 8561 + Peptide=EGVVYGAGIGPGVYDIHSPR
IWGSC_7AS_v1_4256526 1537 1582 + Peptide=WIADSDVITQVLEEK
IWGSC_2DS_v1_5030111 1 17 - Peptide=LVLANAIYFK
IWGSC_2DS_v1_5030111 364 376 - Peptide=LVLANAIYFK
IWGSC_5BL_v1_4998394 55 102 - Peptide=DLLGDTDKLTNVALGR
IWGSC_3B_v1_10519302 310 358 + Peptide=TPNVFDNKYYVNLVNR
IWGSC_3B_v1_10414304 1405 1452 - Peptide=LPIVIDASGDGAYVCK
IWGSC_3B_v1_10414304 1501 1554 - Peptide=EHGVQEGQAGTGAFPSCR
IWGSC_3B_v1_10414304 1555 1593 - Peptide=CGALYSMLDSMYK
IWGSC_3B_v1_10414304 1636 1677 - Peptide=LQCNGSQVPEAVLR
IWGSC_3B_v1_10414304 1453 1485 - Peptide=LTAASITAVCK
IWGSC_7BL_v1_6269081 128 184 - Peptide=KLASVADLYVNDAFGTAHR
IWGSC_4BL_v1_7039291 559 622 + Peptide=IGNEDCTPWMSTLITPLPSCR
IWGSC_4BL_v1_7039291 682 727 + Peptide=QQCCGELANIPQQCR
IWGSC_4BL_v1_7039291 760 808 + Peptide=SRPDQSGLMELPGCPR
IWGSC_1BL_v1_3794722 27 74 - Peptide=DVSPGCRPITVSPGTR
IWGSC_1DS_v1_1894798 364 369 - Peptide=NVFNDLLSPGQLLIIPQNYVVLK
IWGSC_1DS_v1_1894798 1296 1358 - Peptide=NVFNDLLSPGQLLIIPQNYVVLK
IWGSC_1DS_v1_1894798 364 369 - Peptide=NVFNDLLSPGQLLIIPQNYVVLKK
IWGSC_1DS_v1_1894798 1293 1358 - Peptide=NVFNDLLSPGQLLIIPQNYVVLKK
IWGSC_1DS_v1_1894798 1158 1217 - Peptide=NSILGALPVDVIANAYGISR
IWGSC_1DS_v1_1894798 1098 1136 - Peptide=FSREEELGVFAPK
IWGSC_1DS_v1_1894798 1218 1256 - Peptide=TNANSMVSHIAGK
IWGSC_1DS_v1_1894798 1098 1145 - Peptide=SLKFSREEELGVFAPK

The peptides mapping to a gene prediction on Assigning phenotypes to specific proteins
1DS provide a typical example of the analysis
A phenotype of grain proteins as discussed
being undertaken in that they identified the
in detail by Osorio et al. (2012) and Juhász
triticin coding gene (four non-overlapping
et al. (2012) utilizes wheat proteome data for
peptides assigned the region 472-560 of the
predicting the allergen attributes of grain
triticin protein (Fig. 1), consistent with the
proteins. The high throughput analysis for
studies of Singh et al. (1993). Gene predictions
peptides that are unique identifiers for proteins
using AUGUSTUS are currently being refined
carrying epitopes associated wheat allergies
using the extensive transcriptome data sets (EST
or toxicities resulting in celiac disease will be
and assembled RNA-seq) becoming available
feasible for characterizing cultivars. It is expected
for wheat, instead of using the maize gene
that protein signatures for grain proteins can
model and will provide a more wheat-specific
include information identifying epitopes as well
proteogenomic analysis that is generated from
as a range of other features that relate to the
defined wheat tissues.
predictive values of flour from the grain to be
used in the food industry.
triticin [Triticum aestivum]

Sequence ID: gb|ACB41346.1|Length: 577Number of Matches: 1
Related Information
Range 1: 472 to 494GenPeptGraphics Next Match Previous Match
Alignment statistics for match #1

Score Expect Identities Positives Gaps
77.0 bits(174) 7e-15 23/23(100%) 23/23(100%) 0/23(0%)
Query 1 NVFNDLLSPGQLLIIPQNYVVLK 23
NVFNDLLSPGQLLIIPQNYVVLK
Sbjct 472 NVFNDLLSPGQLLIIPQNYVVLK 494

43.5 bits(95) 4e-04 13/13(100%) 13/13(100%) 0/13(0%)
Query 1 TNANSMVSHIAGK 13
TNANSMVSHIAGK
Sbjct 508 TNANSMVSHIAGK 520

64.3 bits(144) 1e-10 20/20(100%) 20/20(100%) 0/20(0%)
Query 1 NSILGALPVDVIANAYGISR 20
NSILGALPVDVIANAYGISR
Sbjct 521 NSILGALPVDVIANAYGISR 540

52.8 bits(117) 5e-07 16/16(100%) 16/16(100%) 0/16(0%)
Query 1 SLKFSREEELGVFAPK 16
SLKFSREEELGVFAPK
Sbjct 545 SLKFSREEELGVFAPK 560
Fig. 1 The alignment of trypsin peptides from the predicted gene (“query”), triticin, on chromosome
1DS to the respective peptides from the triticin gene in GenBank (“Sbjct”). The perfectly matched
peptides were located in the region reference triticin amino acid sequence from position 472 to 560.

References Gao L, Ma W, Chen J, Wang K, Li J, Wang S, Bekes
F, Appels R, Yan Y (2010) Characterization and
Australian Bureau of Statistics (2013) Available at: www.
comparative analysis of wheat high molecular weight
abs.gov.au
glutenin subunits by SDS-PAGE, RP-HPLC, HPCE,
Castellana NE, Payne SH, Shen Z, Stanke M, Bafna V, and MALDI-TOF-MS. J Agric Food Chem 58:2777-2786
Briggs SP (2008) Discovery and revision of Arabidopsis
Juhász A, Gell G, Békés F, Balázs E (2012) The epitopes
genes by proteogenomics. Proc Natl Acad Sci USA
in wheat proteins for defining toxic units relevant to
105:21034-21038
human health. Funct Integr Genomics 12:585-598
Chen J, Lan P, Tarr A, Yan YM, Francki M, Appels R, Ma
Kim S, Gupta N, Pevzner PA (2008) Spectral probabilities
W (2007) MALDI-TOF based wheat gliadin protein
and generating functions of tandem mass spectra: a
peaks are useful molecular markers for wheat genetic
strike against decoy databases. J Proteome Res 7:3354-
study. Rapid Comm Mass Spectrometry 21:2913-2917
3363
Doležel J, Vrána J, Safář J, Bartoš J, Kubaláková M,
Liu L, Ikeda TM, Branlard G, Peña RJ, Rogers WJ, Lerner
Simková H (2012) Chromosomes in the flow to
SE, Kolman MA, Xia X. (2010) Comparison of low
simplify genome analysis. Funct Integr Genomics
molecular weight glutenin subunits identified by SDS-
12:397-416
PAGE, 2-DE, MALDI-TOF-MS and PCR in common
Dupont FM, Vensel WH, Tanaka CK, Hurkman WJ, wheat. BMC Plant Biol 10:124
Altenbach SB (2011) Deciphering the complexities of
Osorio C, Wen N, Gemini R, Zemetra R, von Wettstein
the wheat flour proteome using quantitative two-
D, Rustgi S (2012) Targeted modification of wheat
dimensional electrophoresis, three proteases and
grain protein to reduce the content of celiac causing
tandem mass spectrometry. Proteome Sci 9:10
epitopes. Funct Integr Genomics 2012 12:417-438
Feuillet C, Stein N, Rossini L, Praud S, Mayer K, Schulman
Salcedo G, Prada J, Sanchez-Monge R, Aragoncillo C
A, Eversole K, Appels R (2012) Integrating cereal
(1980) Aneuploid analysis of low molecular weight
genomics to support innovation in the Triticeae. Funct
gliadins from wheat. Theor Appl Genet 56:65-69
Integr Genomics 12:573-583
Singh NK, Donovan GR, Carpenter HC, Skerritt JH,
Langridge P (1993) Isolation and characterization of
wheat triticin cDNA revealing a unique lysine-rich
repetitive domain. Plant Mol Biol 22:227-237

Creating a detailed map of the wheat flour proteome: A
critical step in understanding the effects of environment
on flour quality and immunogenic potential
S.B. Altenbach, F.M. Dupont, W.H. Vensel, C.K. Tanaka, P.V. Allen, and W.J. Hurkman
USDA-ARS Western Regional Research Center, Albany, CA 94710, U.S.A.
Abstract
Proteomic strategies combining high-resolution 2-dimensional gel electrophoresis (2-DE) and tandem
mass spectrometry (MS/MS) offer considerable potential for understanding the effects of environment
on flour quality and immunogenic potential. However, it is challenging to create a detailed map of
the wheat flour proteome, largely because of the complex nature of the major gluten proteins. Using
improved methods, a comprehensive proteomic map of wheat flour from the bread wheat Butte 86 was
developed in which 230 proteins were associated with specific gene sequences. The proteomic map
was used to identify proteins that are differentially expressed in developing Butte 86 grain in response
to post-anthesis fertilizer. These results coupled with recent success in the genetic transformation of
Butte 86 facilitate targeted efforts to alter flour composition in the US bread wheat and to more clearly
define the roles of individual proteins in flour quality and in the response of the grain to the growth
environment.
Introduction individual environmental factors affect the

The protein composition of wheat flour is very accumulation of proteins in the grain. Central
complex. The gluten proteins, consisting of to this work is the development of a detailed
gliadins and glutenins, comprise 70-80% of the map of the wheat flour proteome in which
flour protein and are the main determinants individual proteins were separated by 2-DE and
of flour quality. Both gliadins and glutenins identified by MS/MS (Fig. 1). The proteomic
contain repetitive sequences and an abundance map can be used in conjunction with controlled
of glutamine and proline. Each group contains growth experiments to determine the effects
many closely related proteins that differ of environmental treatments on the proteome.
considerably among cultivars. Gliadins are Additionally, because individual proteins
generally present as monomers and are separated in the map are linked to gene sequences, the
into alpha-, gamma- and omega-gliadin proteomic map facilitates efforts to manipulate
subgroups, each with distinct primary protein the expression of genes encoding proteins of
sequences. The glutenins consist of two types interest in transgenic plants. As shown in Fig. 1,
of proteins; the high molecular weight-glutenin these transgenic plants can be used to provide
subunits (HMW-GS) and the low molecular information about the relationship between
weight-glutenin subunits (LMW-GS), linked by specific genes, proteins and flour quality, and to
disulfide bonds into large polymers. In addition, better define the roles of certain proteins in the
some proteins with sequences similar to alpha-, response of the plant to the environment.
gamma- and omega-gliadins contain an extra In this paper, we bring together several recent
cysteine residue and are linked into the gluten studies using the US wheat cv. Butte 86 that
polymer, presumably functioning as chain- begin to unravel the response of the wheat
terminating glutenin subunits. flour proteome to the growth environment and
Because growth conditions of the wheat plant establish relationships between specific proteins
during grain development influence the quality and flour quality.
of the flour, it is important to understand how

Temperature Drought Fertilizer
Quality Proteome Transgenic Plants
Genes
Fig. 1 Research strategy to decipher the effects of environment on wheat flour quality. Plants are grown
in greenhouses under defined environmental regimens. Effects of the environmental treatments on flour
quality are assessed and quantitative 2-DE is used to identify individual flour proteins that change in
abundance. Expression of genes encoding differentially accumulated proteins is modified in transgenic
plants. Transgenic plants are used to relate specific changes in the proteome to flour quality and to better
define the roles of specific proteins in the response of the grain to environment.
Materials and methods Results and discussion

The US bread wheat (Triticum aestivum) cv. Butte
Proteomic analysis of wheat flour
86 was used for all studies. Plants were grown
under controlled conditions in greenhouses as In many proteomic studies, identification
outlined in Altenbach et al. (2003). Methods for of gluten proteins is based on relatively few
protein extraction and 2-DE analysis are reported peptides, making it impossible to distinguish
in Hurkman and Tanaka (2004). Improved methods individual proteins within the major groups
for MS/MS identification of gluten proteins are (Cunsolo et al. 2012). This is in part because the
detailed in Vensel et al. (2011) and a detailed wheat gluten proteins are not readily cleaved
map of the Butte 86 flour proteome is described with trypsin, the enzyme used in most MS/
in Dupont et al. (2011). A comparative proteomic MS studies. Additionally, the identification of
analysis of flour from Butte 86 plants grown with proteins by MS/MS is based on the best fit of
and without post-anthesis fertilizer is presented spectral data to databases of protein sequences
in Altenbach et al. (2011). Biolistic transformation and most sequence databases do not reflect the
methods for Butte 86 and an RNA interference considerable heterogeneity of gluten proteins
construct targeting the omega5-gliadins are in different cultivars. In a recent study, MS/MS
described in Altenbach and Allen (2011). methods were optimized for Butte 86 (Vensel et

al. 2011). This included digesting each protein with trypsin. Thermolysin proved particularly
with chymotrypsin and thermolysin in addition useful for MS/MS identification of alpha- and
to trypsin. Additionally, the sequences of Butte omega-gliadins, generating 69% and 72% of the
86 gluten proteins were deduced from assembled peptides, respectively.
expressed sequence tags (ESTs) and incorporated
In the Dupont et al. (2011) study, maximum
into sequence databases used for analysis of
sequence coverage was 89% and average
spectral data (Altenbach and Kothari 2007;
sequence coverage was 48% for the LMW-GS,
Altenbach et al. 2010a,b; Dupont et al. 2011).
80% and 54% for the alpha-gliadins and 63%
Using these methods, a comprehensive 2-DE
and 44% for the gamma-gliadins, respectively.
map of Butte 86 flour proteins was generated
The high sequence coverage made it possible to
in which individual 2-DE spots were linked to
distinguish very similar proteins. For example,
specific gene sequences (Dupont et al. 2011).
gamma-gliadins containing eight cysteines
In the analysis, nearly 4,500 peptides were
were distinguished from those containing nine
identified. These corresponded to 168 distinct
cysteines, alpha-gliadins containing six cysteines
protein sequences. Of these, 5 HMW-GS, 22
were distinguished from those containing
LMW-GS, 4 omega-gliadins, 13 gamma-gliadins
seven cysteines and very similar alpha gliadins
and 23 alpha-gliadins were distinguished. In
containing epitopes known to be involved in
addition to the gluten proteins, three farinins,
celiac disease (CD) were distinguished from
three purinins, three triticins, eight globulins,
those that do not contain these epitopes.
three grain softness-related proteins, 16 alpha-
amylase/protease inhibitors, nine serpins, The 2-DE analysis also demonstrated that the
three beta-amylases, three tritins, one xylanase flour proteome has multiple layers of complexity.
inhibitor, and 33 distinct enzymes were identified Because of the large numbers of peptides
in the flour. Of the peptides identified, 62% identified, it was clear that many 2-DE spots
were from gluten proteins and 38% were from contain more than one protein, particularly
non-gluten proteins (Table 1). Only 22% of the in regions of the gel where LMW-GS overlap
peptides identified from the gluten proteins with alpha- and gamma-gliadins. Additionally,
were generated with trypsin, while 26% were multiple 2-DE spots were sometimes identified
generated with chymotrypsin and 52% with as the same protein. In some cases, this was likely
thermolysin. In comparison, 81% of the peptides due to post-translational modifications such as
from the non-gluten proteins were generated glycosylation, observed for some alpha-amylase
with trypsin and only 14% and 5% were inhibitors, or proteolytic processing, observed
obtained with chymotrypsin and thermolysin, for certain globulins and farinins. In other cases,
respectively. Digestion of proteins with multiple spots probably represented charge
chymotrypsin and thermolysin was particularly trains due to sample extraction or 2-DE. For this
important for the alpha- and omega-gliadins, reason it is important to consider the sum of all
since only 5% and 1% of the peptides identified spots with the same identification in comparative
for these proteins, respectively, were generated proteomic experiments.
Table 1. Numbers of peptides identified by MS/MS with different proteases. Data are summarized from
Dupont et al. (2011).
Total % peptides % peptides % peptides

# peptides trypsin chymotrypsin thermolysin
Gluten proteins 2,785 22.0 26.1 51.9
HMW-GS 745 41.7 23.5 34.8
LMW-GS 814 20.0 25.6 54.4
alpha-gliadins 691 5.1 26.5 68.5
gamma-gliadins 405 25.7 30.9 43.5
omega-gliadins 130 0.8 26.9 72.3
Non-gluten proteins 1,698 80.9 14.0 5.1

Changes in the flour proteome that result from elongation factor EF1A and puroindoline. It is
plant growth conditions notable that several of these proteins are either
confirmed or suspected food allergens.
The nutritional status of the plant influences
how the grain responds to environmental factors
such as high temperatures and drought. Thus,
Table 2. Effects of post-anthesis fertilizer on
comparative proteomics was first used to analyze
the levels of gluten proteins in flour. Data are
the effects of post-anthesis fertilizer on the
summarized from Altenbach et al. (2011).
flour proteome. Butte 86 plants were grown in
triplicate under a 24/17oC day/night temperature Change with fertilizer
regimen with or without 20-20-20 NPK fertilizer. HMW-GS Increase 33%
Total proteins from each resulting flour sample LMW-GS Decrease 15%
were analyzed in triplicate by quantitative 2-DE Alpha-gliadins Increase 31%
(Altenbach et al. 2011). Of the 373 protein spots Gamma-gliadins No change
detected in the 2-D gels, 51 showed significant Omega-gliadins Increase 144%
increases and 104 showed significant decreases. Gliadins/glutenins Increase from 1.0 to 1.3
However, when the normalized volumes of all HMW-GS/LMW-GS Increase from 0.61 to 0.95
spots identified as the same protein sequence were Chain-terminators Increase in omega-gliadins only
summed, 54 unique proteins exhibited responses
to fertilizer. The omega-gliadins showed the
largest increases in response to fertilizer,
increasing 144% overall (Table 2). All HMW-GS This study demonstrates the complex effects
also increased. In comparison, the LMW-GS and of post-anthesis fertilizer on the wheat flour
alpha-gliadins showed variable responses to proteome and the importance of considering the
fertilizer. Of the LMW-GS, three protein types nutritional status of the plant when evaluating
showed significant decreases with fertilizer. effects of temperature and drought on the flour
One beginning with the N-terminal sequence proteome.
SHIP is the most abundant protein in Butte 86
flour. Other LMW-GS that showed responses Capturing discoveries from proteomics analyses
to fertilizer had the N-terminal sequences Once proteins that change in response to growth
METSRV and METSCIP. Overall, the LMW-GS conditions are identified, transgenic approaches
decreased 15%. Of the alpha-gliadins, six showed can be used to establish links between specific
increases with fertilizer. These included alpha- proteins and flour quality. One approach is to
gliadins that contained CD epitopes as well as use RNA interference to silence the expression
those that did not. As a result, alpha-gliadins of genes encoding selected proteins in transgenic
increased 31% in the flour, but the proportions plants. The transgenic plants can then be used
of alpha-gliadins containing CD epitopes were to determine how the absence of those proteins
not altered. The gamma-gliadins showed little affects quality. This approach is most successful
change in response to fertilizer. Changes in the in a cultivar such as Butte 86 in which the
gluten protein composition resulted in increased complement of gluten protein genes is known
ratios of gliadins to glutenins and HMW-GS to and the cultivar can be genetically transformed
LMW-GS in flour from grain produced with (Altenbach and Allen 2011). The omega-gliadins
fertilizer. Of the gliadins that are likely to be were targeted first because they showed the
incorporated into the glutenin polymer (referred largest response to post-anthesis fertilizer. The
to as chain-terminators), only the omega-gliadins omega-gliadins consist of two protein types with
increased. A number of non-gluten proteins also distinct repetitive motifs. The omega5-gliadins
changed in proportion in response to fertilizer. contain FPQQQ and QQIPQQ repeats whereas
The serpins increased while the alpha-amylase/ the omega1,2-gliadins contain QQPFP repetitive
protease inhibitors decreased. Other non-gluten elements. An RNAi construct was designed in
proteins that decreased included farinin, chitinase, which a 153 bp fragment that matched all Butte
globulin-2, lipid transfer protein, thaumatin- 86 omega5-gliadin ESTs was inserted in opposite
like protein, beta-amylase, glucose and ribitol orientations on either side of an intron from a
dehydrogenase, triosephosphate isomerase, wheat starch synthase gene. This fragment was

then placed under the control of a promoter Conclusions
from the Dy10 HMW-GS gene and a terminator A detailed map of the flour proteome that links
from the Dx5 HMW-GS gene (Altenbach and each protein to a specific gene sequence makes
Allen 2011). The RNA interference construct it possible to determine the precise changes that
was introduced into immature embryos from occur in the proteome when the plant is grown
Butte 86 using biolistics and transgenic plants under different environmental conditions and to
were regenerated. Six homozygous lines were evaluate the effects of targeted gene silencing on
obtained that showed suppression of the omega5- the flour proteome. These studies are elucidating
gliadin genes. 2-DE analysis of grain proteins the roles of specific proteins in flour quality and
demonstrated that levels of omega5-gliadins in the response of the grain to the environment.
were significantly reduced in some transgenic
plants without other substantial alterations in the
proteome (Fig. 2). The quality of flour from these References
plants will be evaluated to determine the role Altenbach SB, Allen PV (2011) Transformation of the
of the omega5-gliadins in quality. Additionally, US bread wheat ‘Butte 86’ and silencing of omega-5
transgenic plants will be grown under different gliadin genes. GM Crops 2:67-74
fertilizer regimens to determine how the plant Altenbach SB, Dupont FM, Kothari K, Chan R, Johnson
responds to fertilizer in the absence of omega5- EL, Lieu D (2003) Temperature, water and fertilizer
gliadins. It is notable that the omega5-gliadins are influence the timing of key events during grain
responsible for the food allergy wheat-dependent development in a US spring wheat. J Cereal Sci 37:9-20
exercise-induced anaphylaxis. Thus, transgenic Altenbach SB, Kothari KM (2007) Omega gliadin genes
expressed in Triticum aestivum cv Butte 86: effects of
plants also contain reduced levels of a known
post-anthesis fertilizer on transcript accumulation
food allergen. during grain development. J Cereal Sci 46:169-177
A) B)
Fig. 2 2-DE analysis of grain protein from Butte 86 plants. Proteins were obtained from grain of A)
non-transgenic control plants and B) T3 transgenic plants containing an RNAi construct targeting the
omega5-gliadin genes. Omega5-gliadins are circled in Panel A.

Altenbach SB, Tanaka CK, Hurkman WJ, Whitehand LC, Cunsolo V, Muccilli V, Saletti R, Foti S (2012) Mass
Vensel WH, Dupont FM (2011) Differential effects of spectrometry in the proteome analysis of mature cereal
a post-anthesis fertilizer regimen on the wheat flour kernels. Mass Spec Reviews 31:448-465
proteome determined by quantitative 2-DE. Proteome Dupont FM, Vensel WH, Tanaka CK, Hurkman WJ,
Sci 9:46 Altenbach SB (2011) Deciphering the complexities of
Altenbach SB, Vensel WH, Dupont FM (2010a) Analysis of the wheat flour proteome using quantitative two-
expressed sequence tags from a single wheat cultivar dimensional electrophoresis, three proteases and
facilitates interpretation of tandem mass spectrometry tandem mass spectrometry. Proteome Sci 9:10
data and discrimination of gamma gliadin proteins Hurkman WJ, Tanaka CK (2004) Improved methods for
that may play different functional roles in flour. BMC separation of wheat endosperm proteins and analysis
Plant Biol 10:7 by two-dimensional gel electrophoresis. J Cereal Sci
Altenbach SB, Vensel WH, Dupont FM (2010b) Integration 40:295-299
of transcriptomic and proteomic data from a single Vensel WH, Dupont FM, Sloane S, Altenbach SB (2011)
wheat cultivar provides new tools for understanding Effect of cleavage enzyme, search algorithm and decoy
the roles of individual alpha gliadin proteins in flour database on mass spectrometric identification of wheat
quality and celiac disease. J Cereal Sci 52:143-151 gluten proteins. Phytochem 72:1154-1161

Monitoring of storage proteins: Accumulation and impact
of high temperature in wheat kernels using proteomics
A. Tasleem-Tahir1,2, E. Bancel1,2, P. Martre1,2, and G. Branlard1,2
UMR 1095 GDEC, 5 Chemin de Beaulieu, 63039 Clermont-Ferrand, France; 2Blaise Pascal University,
1INRA,
UMR1095 GDEC, F-631 77 Aubière, France
Abstract
Proteomics analysis of gluten proteins was carried out on manually dissected starchy endosperm of
three bread wheat cultivars: Récital, Arche and Tamaro. The albumins and globulins of the developing
endosperm were first extracted and 21 stages of development were analyzed for the cultivar Récital.
Proteomics analysis of storage proteins was also carried out for eight stages of development from
the late period of endosperm cell division to physiological maturity. Based on quantitative variations
in these storage proteins, principal component analysis revealed four major phases during their
accumulation. Hierarchical cluster analysis revealed 10 major profiles, providing evidence for
differences in the kinetics of protein accumulation. Identification of individual gliadins and glutenins
showed that the protein accumulation profiles differed for these two groups. The storage protein
profiles were compared to recently identified profiles of albumin-globulin proteins. A similar analysis
was performed on cultivars Arche and Tamaro (medium and very high quality, respectively) which
were subjected to two temperature regimes during the post-anthesis period: 28/15°C (high temperature
treatment) versus 23/11°C (control) day/night air temperature. Although significant differences were
observed in the gliadin:glutenin ratio between the high temperature treatment and the controls during
the grain filling period, the ratios at maturity showed no significant differences between stressed and
control for each cultivar. Interestingly, the very high quality cultivar Tamaro had far fewer fluctuations
in its gliadin:glutenin ratio than cv. Arche, indicating differences in the genetic regulation of gluten
protein synthesis among cultivars.
Introduction al. 2007). Only a few proteomics studies have

Several studies have shown that proteomics is reported the effect of heat stress on WSPs
a powerful tool to analyze the accumulation of (Majoul et al. 2003; Hurkman et al. 2009). The
proteins in developing starchy endosperm. In kinetics of the many WSPs revealed through
recent years, proteomic studies were performed 2-dimensional electrophoresis (2DEs) during
on cereal seeds (Finnie et al. 2002; Méchin et al. protein accumulation, need to be analyzed. By
2007; Tasleem-Tahir et al. 2011, 2012). Several analyzing these WSPs in the same developing
attempts have been made to characterize non- endosperm as that used for albumin and
storage proteins (namely albumins and globulins: globulin protein profiles (Tasleem-Tahir et al.
alg) in developing and mature endosperm 2012), two questions were addressed: Are the
(Wong et al. 2004; Vensel et al. 2005; Gao et al. accumulation profiles of each WSP similar or
2009). Analyses of albumins and globulins have is there any difference within and between
also focused on wheat subjected to heat stress the different gliadin classes as well as the high
during grain development (Hurkman et al. (HMW-GS) and low (LMW-GS) molecular
2009; Laino et al. 2010). Differential proteomics weight glutenin subunits when wheat is grown
of wheat storage proteins (namely gliadins under normal temperature conditions? Among
and glutenins: WSPs) of mature grain has the albumins and globulins that have already
been used to study the impact of aneuploidy been identified, which have accumulation
(Islam et al. 2002; Dumur et al. 2004) and the profiles associated with WSPs? Here, we report
effect of the 1BL/1RS translocation (Gobaa et on the use of a proteomics approach for cultivars
grown under different temperature conditions.

Materials and methods Tamaro samples. The individual storage proteins
were expressed as percent of total endosperm
Plant material proteins as well as in mg of protein per grain
The bread wheat cultivar Récital was cultivated in which was obtained by multiplying the relative
a greenhouse during the normal wheat growing percentage volume calculated from image
season at INRA, Clermont-Ferrand, France, as analysis, by the total quantity of protein per grain
previously reported (Tasleem-Tahir et al. 2012). determined using the Kjeldhal (for Arche and
Wheat grains were harvested every 50°Cdays Tamaro) or Bradford (for Récital) methods. Alg
from anthesis (0°Cd) to maturity ripeness and glutenin spots were identified using mass
(1000°Cd after anthesis) and the endosperms spectrometry, as reported in Tasleem-Tahir et al.
were manually dissected. Endosperm samples (2012). Gliadins were identified using the two-step
from 21 and 8 developmental stages were procedure (acid PAGE then 2DE) as previously
used to analyze the accumulation of alg and described in Dumur et al. (2004).
WSPs, respectively. To study the influence
of chronic moderately high temperatures on Statistics
the accumulation of WSPs, the bread wheat To compare spot volumes, differences between
cultivars Arche and Tamaro were grown at mean spot volumes were considered to be
INRA Clermont-Ferrand under semi-controlled significant when the fold change ratio was ≥ 1.8,
conditions (Triboï et al. 1996) and were subjected and the P-value and q-value (a measure of the
to two day/night thermal regimes from 4 to 5 proportion of false positives) were <0.05. Principal
days (163°Cd) after anthesis to physiological component analysis (PCA) and hierarchical
maturity (781°Cd after anthesis): 23C/11C cluster analysis (HCA) were performed based
(controls) and 28C/15C (high temperature [HT] on Pearson’s distance on spots that varied
treatment). In both treatments, grains were significantly between grain development stages.
collected at thermal time intervals from 163°Cd
(early grain filling) to 781°Cd (physiological
maturity). Results and discussion
Proteomics analysis Expression of albumins and globulins during
Albumins and globulins were extracted from grain development in cv. Récital
the developing floury endosperm using a Among the 1,780 albumin and globulin spots
low concentration salt solution as previously detected on the gels, 950 varied significantly
described (Tasleem-Tahir et al. 2012). WSPs between two or more of the 21 development
were extracted from 40 mg flour using the urea/ stages. The PCA performed on these 950 spots
thiourea/CHAPS protein extraction protocol provided clear evidence for four major phases: 1)
reported by Majoul et al. (2003). The alg and 0-93°Cd; 2) 152-192°Cd; 3) 247-764°Cd; and, 4) 804-
WSPs protein extracts of Récital were isofocussed 1006°Cd. The HCA revealed nine major protein
using the IPGPhor II apparatus (GE Healthcare, profiles throughout grain development. Among
Uppsala, Sweden) on 24 cm Immobiline dry the 950 protein spots, 487 were identified and
strips of 3-11 non-linear pH gradient. The WSPs classified according to 17 different biochemical
of Arche and Tamaro were isofocussed on 18 functions (Tasleem-Tahir et al. 2012). As expected,
cm Immobiline dry strips with a 3-11 non-linear proteins involved in cell division, transcription/
pH gradient. Equilibration of strips, SDS-PAGE translation, protein synthesis, along with
and gel staining using Coomassie blue G250, amino acid, lipid, carbohydrate, and nucleotide
were performed as previously described for alg metabolism, were highly expressed in the early
(Tasleem-Tahir et al. 2011) and WSPs (Majoul et and early to mid-stages of development. In mid-
al. 2003). Four to six 2DE replicates (2-3 biological grain development, proteins involved in folding in
× 2 technical) were performed for each protein the cytoskeleton, and storage proteins of globulin
extract at a given sampling stage. Stained gels type were highly expressed. In later stages, the
were scanned and analyzed using SameSpots most abundant proteins were those involved in
v4.1 (Nonlinear dynamics, Newcastle, UK) stress/defense, folic acid metabolism and protein
for Récital samples and Melanie-3 software turnover.
(GeneBio, Geneva, Switzerland) for Arche and

Of particular interest, the total volume of the 64 22 HMW-GS, 47 LMW-GS, 15 -gliadins,
proteins involved in redox decreased continuously 11 -gliadins and 9 -gliadins. Based on the
throughout grain development (Tasleem-Tahir quantitative variation of these 200 storage
et al. 2012). Importantly, several enzymes, proteins, principal component analysis revealed
including seven ascorbate/peroxydases and two four major phases of WSP accumulation: 1)
catalases, decreased considerably. These decreases 192-247°Cd; 2) 247-457°Cd; 3) 457-655°Cd; and
highlight the progressive oxidative stress that 4) 655-1006°Cd. When the abundance of the 200
takes place in developing endosperm. However, (104) individual spots were considered for the
some enzymes involved in redox, including three eight grain developmental stages, hierarchical
dehydroascorbate reductases, remained almost cluster analysis revealed 6 (10) major profiles.
unchanged throughout grain development and This analysis confirmed that individual gliadin
two thioredoxins increased significantly from 343 and glutenin spots had different accumulation
to 1006°Cd after anthesis (Fig. 1). These proteins profiles. Moreover, the analysis of the 10
involved in disulphide bond interchanges play a profiles revealed that each protein family may
major role in the final polymerization of gluten. have a different accumulation profile. The
22 HMW-GS spots showed three different
0.16 accumulation profiles. These profiles were each
Thioredoxins composed of spots corresponding to subunits
0.14
encoded at different HMW-GS loci: one profile
0.12 was composed of two spots identified as the
0.1 Glu-A1x2* subunit.
0.08 The second profile comprised five Glu-D1x5
0.06 spots, five Glu-B1x6 spots, five Glu-B1y8 spots
0.04 and one Glu-D1y10 spot. The third profile
was composed of four Glu-Dy10 spots. The 47
0.02
LMW-GS spots (Fig. 2A) were split into two
0 different accumulation profiles (Fig. 2B, 2C).
0
59
93
152
192
247
309
343
401
457
497
547
597
655
706
764
804
857
903
957
1006
The subunits encoded by the three LMW alleles

Fig. 1 Average normalized volume of two of Récital (Glu-A3d, Glu-B3g and Glu-D3c) were
thioredoxins expressed in the endosperm of found in these two profiles. Thirty-one spots
Récital from 0 to 1006°Cd. were grouped under the first profile (Fig. 2B)
with spots corresponding to subunits of Glu-A3,
The oxidative stress in the endoplasmic reticulum Glu-B3 and Glu-D3 loci. Surprisingly, the spot
was particularly clear in the developing characteristic of the Glu-B1g (in the D zone in
endosperm of Pin-A hard/soft near-isogenic lines 1D SDS-PAGE) was not present in this profile.
(Lesage et al. 2011). The second profile (Fig. 2C) comprised 12 spots
in the basic zone of the 2DE plus four spots
grouped with the purinin spots in the more
Expression of wheat storage proteins during
neutral zone (Fig. 2A).
grain development in cv. Récital
At 192°Cd after anthesis only a few Coomassie- Gliadins were also found in different profiles.
stained WSPs were evident. Over the 40 day The 15 - gliadin spots were grouped in
period between 192 and 1006°Cd after anthesis, only one profile. These proteins were mainly
a total of 832 spots were revealed of which only encoded by the Gli-A2 and Gli-B2 loci and
498 showed significant quantitative or qualitative shared the same accumulation profile. The 11
(presence/absence) developmental variation. -gliadin spots were grouped in two different
Some alg were also extracted during extraction profiles with 7 and 4 spots each. The -gliadins
of WSPs using urea/thiourea/CHAPS. Over 200 of Récital are mainly encoded by the Gli-
gliadin and glutenin spots were identified. Ninety A1o, Gli-B1f, Gli-D1b and Gli-B2p alleles
six of these spots were of very low abundance. (Metakovsky and Branlard 1998). Multigenic
The 104 most abundant WSP spots comprised inheritance of -gliadins showed that some

homoeologous gliadin loci had similar protein The kinetics of gliadin synthesis clearly revealed
accumulation profiles during grain development. the greater abundance of sulphur-rich proteins
The same conclusion could also be drawn from (-, - and -gliadins) compared to sulphur-poor
the 9 -gliadin spots encoded by Gli-A1o, Gli-B1f, proteins (-gliadins) from the early stages (192-
and Gli-D1b alleles and found in two profiles. 247°Cd of protein accumulation (Fig. 3A). The
greater abundance of sulphur-rich proteins was
also observed for the LMW-GS over the HMW-
GS (Fig. 3B). Variation in the gliadin:glutenin
ratio was also analyzed during endosperm
development and in relation to the amount of
total grain protein.
Some results from heat treatment

The same proteomics approach applied to Récital
was used for cv. Arche and Tamaro which were
subjected to two day/night thermal regimes:
23/11°C (controls) and 28/15°C (heat treatment
HT) from 163°Cd to 781°Cd. Heat treatment
A) throughout grain filling resulted in reduced
total amounts of gliadins and glutenins, which
7 were particularly marked for Arche. At maturity,
the total amounts of gliadins and glutenins
6.5 per grain for the HT-treated Arche were 65.7%
6 and 65.5%, respectively, of the amounts for the
control. Although gliadins and glutenins were
5.5 reduced in both cultivars, the gliadin:glutenin
5
Log normalized volume
0.7
4.5 alpha-beta
B) 0.6 gamma
4
mg protein/grain
192 247 343 457 547 655 764 1003 0.5 omega
7 0.4
6.5 0.3
6 0.2
0.1
5.5 stages of development °Cd
0
192 247 343 457 547 655 764 1006
5
1.6
4.5 1.4 HMW
C) LMW
1.2
mg protein/grain
4
192 247 343 457 547 655 764 1003 1
Thermal time in °Cdays 0.8
Fig. 2 Two-dimensional electrophoresis of proteins 0.6
extracted with urea/thiourea/CHAPS from the ma- 0.4
ture endosperm of cv. Récital (A). The two LMW- 0.2
stages of development °Cd
GS clusters resulting from HCA are highlighted 0
with 31 spots (surrounded) and 16 spots (arrows) 192 247 343 457 547 655 764 1006
(A). The five higher (lower) profiles and the average Fig. 3 Quantity of protein per grain versus thermal
profile (thick line) for each cluster are shown in (B) time after anthesis for 104 WSP spots grouped as
and (C), respectively. Vertical bar, ±1 S.D. type-, -, w-gliadins (A) and HMW-GS and
LMW-GS (B).

ratio at maturity was not affected by the HT Islam N, Woo S-H, Tsujimoto H, Kawasaki H, Hirano
treatment. These ratios, also computed for the H (2002) Proteome approaches to characterize seed
storage proteins related to ditelocentric chromosomes in
control and heat treated samples collected during
common wheat (Triticum aestivum L.). Proteomics 2:1146-
grain filling, were significantly affected by heat 1155
treatment in Arche (a medium quality cultivar), Laino P, Shelton D, Finnie C, De Leonardis AM, Mastrangelo
but did not significantly differ between the AM, Svensson B, Lafiandra D, Masci S (2010)
control and heat treatment in Tamaro (a very Comparative proteome analysis of metabolic proteins
high quality cultivar). These findings indicate from seeds of durum wheat (cv. Svevo) subjected to heat
that although chronic HT may reduce the amount stress. Proteomics 10:2359-2368
of WSPs, the gliadin:glutenin ratio is probably Lesage VS, Merlino M, Chambon C, Bouchet B, Marion D,
more genetically regulated in the high quality Branlard G (2012) Proteomes of hard and soft near-
isogenic wheat lines reveal that kernel hardness is
cultivar than in the medium quality cultivar. related to the amplification of a stress response during
Analysis of albumins and globulins could help in endosperm development. J Exp Bot 63:1001-1011
understanding this difference. Liu L, Ikeda TM, Branlard G, Peña RJ, Rogers WJ, Lerner
SE, Kolman MA, Xia X, Wang L, Ma W, Appels R,
Yoshida H, Wang A, Yan Y, He Z (2010) Comparison of
Conclusions low molecular weight glutenin subunits identified by
SDS-PAGE, 2-DE, MALDI-TOF-MS and PCR in common
Ten profiles of WSPs accumulation were wheat. BMC Plant Biol 10:124
identified, and some protein spots from the same Majoul T, Bancel E, Triboi E, Ben Hamida J, Branlard G
locus were found in different profiles. These (2003) Proteomic analysis of the effect of heat stress
profiles, which also result from the two possible on hexaploid wheat grain: Characterization of heat-
wheat storage protein trafficking pathways responsive proteins from total endosperm. Proteomics 3:
(Tosi et al. 2009), may influence the final gluten 175-183
composition. The regulatory mechanisms Méchin V, Thevenot C, Le Guilloux M, Prioul JL, Damerval
underlying well-known wheat storage protein C (2007) Developmental analysis of maize endosperm
proteome suggests a pivotal role for pyruvate
alleles, whose characterization and nomenclature orthophosphate dikinase. Plant Physiol 143:1203-1219
harmonization was undertaken (Liu et al. 2010),
Metakovsky EV, Branlard G (1998) Genetic diversity of
now need to be examined taking into account the French common wheat germplasm based on gliadin
endoplasmic reticulum oxidative stress that could alleles. Theor Appl Genet 96:209-218
influence the final size of the gluten polymers. Tasleem-Tahir A, Nadaud I, Girousse C, Martre P, Marion D,
Branlard G (2011)Proteomic analysis of peripheral layers
during wheat (Triticum aestivum L.) grain development.
References Proteomics 11:371-379
Dumur J, Jahier J, Bancel E, Lauriere M, Bernard M, Tasleem-Tahir A, Nadaud I, Chambon C, Branlard G (2012)
Branlard G (2004) Proteomic analysis of aneuploid Expression profiling of starchy endosperm metabolic
lines in the homeologous group 1 of the hexaploid proteins at 21 stages of wheat grain development. J
wheat cultivar Courtot. Proteomics 4:2685-2695 Proteome Res 11:2754-73
Finnie C, Melchior S, Roepstorff P, Svensson B (2002) Tosi P, Parker M, Gritsch CS, Carzaniga R, Martin B, Shewry
Proteome analysis of grain filling and seed maturation PR (2009) Trafficking of storage proteins in developing
in barley. Plant Physiol 129:1308-1319 grain of wheat. J Exp Bot 60:979-991
Gao LY, Wang AL, Li XH, Dong K, Wang K, Appels R, Ma Triboï E, Triboï-Blondel AM, Martignac M, Falcimagne R
WJ, Yan YM (2009) Wheat quality related differential (1996) Experimental device for studying post-anthesis
expressions of albumins and globulins revealed by canopy functioning in relation to grain quality. In:
two-dimensional difference gel electrophoresis (2-D van Ittersum MK (ed) Proc 4th European Society of
DIGE). J Proteomics 73:279-296 Agronomy Congress, Amsterdam Academic Press,
Wageningen, The Netherlands, pp68-69
Gobaa S, Bancel E, Kleijer G, Stamp P, Branlard G (2007)
Effect of the 1BL.1RS translocation on the wheat Vensel WH, Tanaka, CK, Cai N, Wong JH, Buchanan
grain proteome as revealed by proteomic analysis. BB, Hurkman WJ (2005) Developmental changes in
Proteomics 7:4349-4357 the metabolic protein profiles of wheat endosperm.
Proteomics 5:1594-1611
Hurkman WJ, Vensel WH, Tanaka CK, Whitehand L,
Altenbach SB (2009) Effect of high temperature on Wong JH, Cal N, Balmer Y, Tanaka CK, Vensel WH,
albumin and globulin accumulation in the endosperm Hurkman WJ, Buchanan BB (2004) Thioredoxin targets
proteome of the developing wheat grain. J Cereal Sci of developing wheat seeds identified by complementary
49:12-23 proteomic approaches. Phytochem 65:1629-1640

Proteomic analysis of mature kernels of Fusarium
graminearum-infected transgenic bread wheat
expressing PGIP
R. D’Ovidio1, P. Laino1,4, M. Janni1,5, E. Botticella1, M.S. Di Carli2, E. Benvenuto2, K. Lilley3, D.
Lafiandra1, and S. Masci1
1Dipartimento di Scienze e Tecnologie per l’Agricoltura, le Foreste, la Natura e l’Energia, University of Tuscia,
Viterbo, Italy; 2ENEA, Dipartimento BIOTEC, Sezione Genetica e Genomica Vegetale, C.R. Casaccia, Rome,
Italy; 3Department of Biochemistry, University of Cambridge, Cambridge, U.K.; 4Present address: Consiglio per
la Ricerca e la sperimentazione in Agricoltura-Genomics Research Center, Fiorenzuola d’Arda (PC), Italy; 5CNR
Istituto di Genetica Vegetale, Via G. Amendola 165/A, Bari, Italy
Abstract
Fusarium head blight (FHB), a worldwide wheat disease, has a detrimental effect both on productivity
and qualitative properties of wheat flour. To evaluate the impact of FHB on the wheat flour we
compared the proteomes of mature kernels obtained from Fusarium graminearum-infected spikes and
control bread wheat plants. The same analysis was also performed on mature seeds of control and
F. graminearum-infected transgenic bread wheat plants expressing a Poly-Galacturonase Inhibiting
Protein (PGIP) that possesses partial resistance to FHB. DIGE investigation of metabolic proteins and
A-PAGE and SDS-PAGE analysis of storage proteins (gliadins and glutenins) extracted from mature
seeds did not identify any component differentially accumulated between the infected plants and non-
infected controls, between infected and non-infected transgenic plants, or between infected and non-
infected transgenic and control plants. Conversely, total starch content showed a significant difference
between infected and non-infected control plants, with the infected control plants showing a reduced
total starch content. Similarly, there was a significant difference between infected transgenic and
control plants, with the transgenic plants showing higher total starch contents. These results indicate
that moderate infection of FHB does not cause significant changes in the wheat seed proteome, but
does reduce the total seed starch content. Moreover, the expression of PGIP protects the grain from
starch decrease following F. graminearum infection.
Introduction the complex wheat-Fusarium interaction, one

Wheat is one of the most important food plants in includes the disruption of the natural structural
the world and, like other crops, is affected by many barriers present in the host plant through the
plant diseases. Among them Fusarium head blight action of degradative enzymes secreted by
(FHB) occurs worldwide. Caused by fungi in the fungal pathogens during infection and spread.
genus Fusarium it is of particular importance not F. graminearum secretes a number of enzymes
only because of severe crop losses that may result, that degrade the cell wall (cell wall degrading
but also drastic deterioration of grain quality enzymes, CWDE), including polygalacturonase
that includes the presence of mycotoxins. Among (PG). The activity of PG is controlled by a specific
these, the trichothecene deoxynivalenol (DON), inhibitor, Polygalacturonase-inhibiting protein
also known as vomitoxin, is frequently found in (PGIP), localized in the cell walls of monocot and
wheat grains in association with its derivatives or dicot plants (De Lorenzo et al. 2001; D’Ovidio
similar compounds, such as nivalenol (NIV) and et al. 2004; Protsenko et al. 2008). The efficacy of
fusarenon-X (FUS) (Bai et al. 2001). PGIP in limiting host tissue colonization has been
demonstrated in several plant species, and recently
FHB is controlled by genes of small effect we proved this also in wheat, a species with a low
(quantitative trait loci (QTL)), a number of pectin content in its cell wall. Expression of a bean
which have been identified (Buerstmayr 2009). PGIP reduced the disease symptoms caused by F.
Among the various components involved in graminearum (Ferrari et al. 2012).

In the present paper, we evaluate the impacts of Results and discussion
FHB on the proteome of mature kernels of a bread Null segregant (control) and J82-23a transgenic
wheat transgenic expressing a bean PGIP and the plants showed comparable values in spikelet
corresponding null genotype. number and seed yield, measured as one-
thousand kernel weight (TKW). No significant
differences in TKW were observed between F.
Materials and methods graminearum-infected and non-infected transgenic
The transgenic line J82-23a expressing a bean plants. In contrast, significant differences in TKW
PGIP and the corresponding null segregant line were observed between infected and non-infected
(control) obtained by particle bombardment of control plants, with the infected control plants
immature embryos of the bread wheat cultivar showing a reduced seed yield. All subsequent
Bobwhite (Janni et al. 2008) were grown in single analyses were performed by comparing the
pots at 18-23°C with a 14 h photoperiod. Infection protein patterns or total starch content of whole-
experiments (five biological replicates) were grain flour of transgenic and control plants
performed using a point inoculation method with a obtained after normal growth conditions or
conidial suspension of F. graminearum as reported in following infection with F. graminearum.
Janni et al. (2008). All seeds from each experiment
Since gliadins and glutenins are the main
were pooled to obtain whole-grain flour. Gliadins
determinants of gluten quality characteristics, we
and glutenin subunits were analyzed as reported
verified whether they underwent a differential
in Scossa et al. (2008) and Masci et al. (2003),
accumulation in the grains of transgenic and
respectively. Metabolic proteins were extracted
control plants. Densitometry analysis of total
from flour as previously reported (Hurkman
(α/β, ω, γ) gliadin patterns obtained by A-PAGE
and Tanaka 2007) and analyzed by Differential
analyses did not show any significant differences
In-Gel Electrophoresis (DIGE), firstly by a three-
between the four groups of samples (Fig. 1).
dye scheme and finally with a two-dye scheme
These results are in agreement with those
(Karp and Lilley 2005). Total starch content was
reported by Dexter et al. (1996) who found no
determined using a Total Starch Amyloglucosidase/
qualitative or quantitative differences in gliadins
α-Amylase Assay Kit (Megazyme International
of healthy and FHB-damaged grains of four hard
Ireland Ltd, Bray, Ireland).
red spring wheat cultivars.
C Ci T Ti C Ci T Ti

HMW-GS
LMW-GS
α/ß
A) B)
Fig. 1 Example of endosperm protein analyses. A) A-PAGE of the gliadin extracts. B) SDS-PAGE of
glutenin subunits (HMW-GS and LMW-GS). C, control; Ci, control infected; T, transgenic; Ti, transgenic
infected. Proteins were visualized with Coomassie Blue and each sample was run in triplicate.

The accumulation of the total glutenin subunits infected control samples showing reduced total
showed a small but significant (p=0.05) decrease starch contents. Similarly, a significant difference
in the comparison between the transgenic was observed between infected transgenic and
infected and control infected samples, but control samples, with the transgenic samples
no change was observed in the HMW-GS/ showing a higher total starch content.
LMW-GS ratio. A reduction in total glutenin
subunits was also reported by previous authors,
including Dexter et al. (1996) and Boyacioglu and Conclusions
Hettiarachchy (1995) who reported a decrease in These results indicate that a moderate infection
glutenins in hard red spring wheat cultivars after of FHB does not cause significant changes in the
infection. Taken together, these results indicate wheat seed proteome, but reduces total seed starch
that F. graminearum infection or the expression of content. The lack of any significant change in grain
PGIP have no marked effects on the accumulation protein accumulation, both under normal growth
of prolamine proteins. or following F. graminearum spike infection, has
A total of about 2,000 protein spots were been demonstrated also in the transgenic plants
detected by DIGE analysis of metabolic proteins expressing a bean PGIP. Moreover, the lack of any
extracted from whole-grain flour. The Biological significant change in total starch content in this
Variation Analysis revealed 10 spots differentially latter group of plants following F. graminearum
regulated in the comparison between the control infection indicates a beneficial role of PGIP on
and control infected groups. The spots of interest grain characteristics.
were selected according to the following criteria:
t-test (p <0.01), fold change higher than 1.2, and
spot presence on at least 80% of all maps. In the References
comparison between transgenic and control, Bai G-H, Plattner R, Desjardins A, Kolb F (2001) Resistance
transgenic infected and control infected, and to Fusarium head blight and deoxynivalenol
accumulation in wheat. Plant Breeding 120:1-6
transgenic and transgenic infected groups, one
Boyacioglu D, Hettiarachchy NS (1995) Changes in some
spot, two spots and one spot, respectively, were
biochemical components of wheat grain that was
detected as differentially regulated. These spots, infected with Fusarium graminearum. J Cereal Sci 21:57-62
although very faint, were recovered from the Buerstmayr H, Ban T, Anderson JA (2009) QTL mapping
gels, digested with trypsin and submitted to MS- and marker-assisted selection for Fusarium head blight
MS analysis. However, none of them produced a resistance in wheat: a review. Plant Breeding 128:1-26
significant protein identification. D’Ovidio R, Mattei B, Roberti S, Bellincampi D (2004)
Polygalacturonases, polygalacturonase-inhibiting
The normalized volumes of spots present in at proteins and pectic oligomers in plant-pathogen
least 80% of maps were used to perform Principal interactions. Biochim Biophys Acta 1696:237-244
Component Analysis (PCA) (O’Gorman et al. De Lorenzo G, D’Ovidio R, Cervone F (2001) The role of
2007). The result did not discriminate any clusters polygaracturonase inhibiting proteins (pgips) in defense
relative to the conditions used (transgenic against pathogenic fungi. Annu Rev Phytopathol 39:313-
and control, infected and not infected), thus 335
showing that these four groups are not clearly Dexter JE, Clear RM, Preston KR (1996) Fusarium head
differentiated at the protein expression level. blight: Effect on the milling and baking of some
Canadian wheats. Cereal Chem 73:695-701
Our results are in partial agreement with Eggert
et al. (2011) who made a proteomic analysis of Eggert KA, Zörb CB, Mühling KHB, Pawelzik EA (2011)
Proteome analysis of Fusarium infection in emmer grains
Fusarium infected emmers grown in the field in (Triticum dicoccum). Plant Pathology 60:918-928
two different locations. They identified only 10
Ferrari S, Sella L, Janni M, De Lorenzo G, Favaron
polypeptides as differentially regulated, but with F, D’Ovidio R (2012) Transgenic expression of
a large variation due to different environments. polygalacturonase-inhibiting proteins in Arabidopsis
and wheat increases resistance to the flower pathogen
In contrast to the results obtained at the proteome Fusarium graminearum. Plant Biology 14 (Suppl. 1):31-38
level, total starch content of whole-grain flour
Hurkman WJ, Tanaka CK (2007) Extraction of wheat
showed a significant difference between infected endosperm proteins for proteome analysis. J
and non-infected control samples, with the Chromatogr B849:344-350

Janni M, Sella L, Favaron F, Blechl AE, De Lorenzo O’Gorman M, Beauvallet C, Lepercq P, David O, Seksik
G, D’Ovidio R (2008) The expression of a bean PH, Beaugerie L, Doré J, Martin P, Bogard P, Juste C
polygalacturonase-inhibiting protein in transgenic (2007) An investigation into Crohn’s disease using the
wheat confers increased resistance to the fungal Progenesis Same Spots analysis platform. 24th Journée
pathogen Bipolaris sorokiniana. Mol Plant-Microbe Françaises de Spectrométrie de Masse, Pau, France
Interact 21:171-177 Protsenko MA, Buza NL, Krinitsyna AA, Bulantseva EA,
Karp NA, Lilley KS (2005). Maximising sensitivity for Korableva NP (2008) Polygalacturonase inhibiting
detecting changes in protein expression: Experimental protein is a structural component of plant cell wall.
design using minimal CyDyes. Proteomics5:3105-3115 Biochemistry (Moscow) 73:1053-1062
Masci S, D’Ovidio R, Scossa F, Patacchini C, Lafiandra Scossa F, Laudencia-Chingcuanco D, Anderson OD,
D, Anderson OD, Blechl AE (2003) Production and Vensel WH, Lafiandra D, D’Ovidio R, Masci S (2008)
characterization of a transgenic bread wheat line over- Comparative proteomic and transcriptional profiling
expressing a low-molecular-weight glutenin subunit of a bread wheat cultivar and its derived transgenic
gene. Mol. Breeding 12:209-222 line over-expressing a low molecular weight glutenin
subunit gene in the endosperm. Proteomics 8:2948-2966

Genomic analysis of the expressed α/β-gliadin gene
region in hexaploid wheat
K. Kawaura1, J. Wu2, T. Matsumoto2, H. Kanamori2, S. Kagiri2, and Y. Ogihara1
1KiharaInstitute for Biological Research, Yokohama City University, 244-0813, Japan; 2National Institute of
Agrobiological Sciences, 305-8602, Japan
Abstract
BAC clones containing -gliadin genes from the loci Gli-A2, Gli-B2, and Gli-D2 were screened from
a genomic BAC library of Triticum aestivum cv. Chinese Spring. We selected five BAC clones, viz. two
clones for Gli-A2, two for Gli-B2, and one for Gli-D2, for full sequencing. Approximately 200 kb was
sequenced for each locus. In total, twelve intact-gliadin genes and four pseudogenes were found.
Annotation of these sequences revealed an abundance of retrotransposons or other transposons
accumulated around the -gliadin genes. The pattern of genome segmental duplication within each
BAC was revealed by dot-plot analysis. We calculated evolutionary time since duplication of each
set of -gliadin genes and insertion of retrotransposons based on the numbers of base substitutions
per site. Duplication of all adjacent genes within the same BAC clone took place before or after
allotetrapolyploidization, but duplication of certain genes occurred before diploid differentiation
of wheat species. Retrotransposons were also inserted before and after the segmental duplication
events. Duplication of genome segments containing -gliadin genes and retrotransposons were
brought about through unequal crossing-over or saltatory replication, but -gliadin genes were also
duplicated without recombination events. The expression of -gliadin genes annotated from the
sequence was estimated from the frequency of ESTs. Of the twelve intact -gliadin genes used in
this study, nine were expressed, although their patterns of expression were different. Since they have
similar cis-elements and promoter structures, mechanisms underlying the distinct gene expression
and possible applications are suggested.
Introduction We previously profiled the gene expression of

The major components of wheat seed storage two storage proteins in hexaploid wheat by EST
proteins are gliadins and glutenins. Wheat gluten analysis, showing that the expression patterns
is composed of a protein complex of monomeric differed among distinct -gliadin genes during
gliadins and polymeric glutenins. Gliadins the seed development stage (Kawaura et al.
are further separated into three groups, - 2005). This observation suggests that -gliadin
gliadins, g-gliadins, and w-gliadins. Glutenins genes have differentially evolved to show distinct
are generally classified as high molecular expression patterns among those genes in each
weight glutenin subunits (HMW-GS) and low homoeologous genome.
molecular weight glutenin subunits (LMW- The wheat genome is very large and characterized
GS). The -gliadin genes are located at the by repetitive sequences and transposable elements.
Gli-2 loci of the short arms of homoeologous Therefore, genome sequence information for
group 6 chromosomes, whereas other genes wheat is still limited. Wheat genome sequence
encoding seed storage proteins are located on data are powerful for isolating and characterizing
homoeologous group 1 chromosomes. All of important genes. In this study, we analyzed BAC
these seed storage proteins are encoded by clones of Triticum aestivum cv. Chinese Spring in
multigenes at each locus. The -gliadin genes order to characterize the Gli-2 loci in the hexaploid
have especially high copy numbers, estimated genome. We discuss the structures of these genes
as high as 150 in cv. Cheyenne to 60 in Chinese and how they differ expression-wise among
Spring (Okita et al. 1985). the homoeologous chromosomes as a result of
evolution.

Materials and methods duplication were calculated using a mutation rate of
A genomic BAC library of T. aestivum cv. Chinese 6.5 × 10-9 substitutions per synonymous site per year
Spring (Allouis et al. 2003) was used for screening according to San Miguel et al. (2002).
by PCR with a specific primer set of expressed
-gliadin genes (Kawaura et al. 2005). BAC
DNA was prepared by an alkaline lysis and SDS Results and discussion
method. The insert size of plasmid DNA was One or two BAC clones were screened by 11 sets
analyzed using pulsed field gel electrophoresis. of primers and eventually 15 positive BAC clones
were selected. The BAC clones were estimated to
Five BAC clones were selected for sequencing
contain one to nine -gliadin genes by Southern
determined using a shotgun approach
blot analysis. In particular, BAC clones from
(International Rice Genome Project 2005).
chromosome 6B contained high copy numbers of
Sequences were annotated by riceGAAS software
-gliadin genes. In order to analyze the genomic
(Sakata et al. 2002). Repetitive sequences were
arrangement of the -gliadin genes, five BAC
compared by BLASTN against the Triticeae
clones, viz. BAC2, BAC3, BAC5, BAC6, and BAC11,
Repeat Sequence Database. Dot plot analyses
were selected for further sequence analysis.
were performed using GENETYX software.
Phylogenetic analysis was performed using the The DNA sequences of the five BAC clones were
neighbor-joining method in MEGA4 software determined by the shotgun method. However, none
with (Tamura et al. 2007). of the five clones were connected in one contig
because of numerous repetitive sequences. The
The long terminal repeat (LTR) sequences of
self-dot plot analysis was performed using the five
retroelements were predicted by AutoPredLTR
connected BAC sequences (Fig. 1). The matrixes
and BLASTN searches against LTRdb of
of BAC3 and BAC6 showed that approximately 38
riceGAAS. The evolutionary distance between
kb and 39 kb segments, respectively, containing
two sequences was calculated by the number
an -gliadin gene, were duplicated. BAC5 from
of base substitutions per site using the Kimura
chromosome 6B harbored a complicated duplication
2-parameter method in MEGA4. The times of
structure within an approximately 107 kb sequence.
retrotransposon insertion and genome segment
BAC3 (6A): 94,535 bp BAC11 (6A): 114,651 bp BAC2 (6D): 210,855 bp
BAC5 (6B): 170, 743 bp
20 kb
BAC6 (6B): 88, 709 bp
Fig. 1 Genome structure of -gliadin

gene loci revealed by dot plot analysis.
The sites of -gliadin genes are
indicated by black arrowheads.

A. (BAC3)
Copia_Lara unknown Gypsy_Sabrina unknown CACTA Gypsy_Sabrina
Copia_Barbara unknown Gypsy_Sabrina unknown Gypsy_Sabrina CACTA

CACTA CACTA
B. (BAC11)
Mandrake
CACTA Copia_WIS
Gypsy_FaƟma
Gypsy_FaƟma Gypsy_Laura Byron
Gypsy_Sabrina CACTA
Gypsy_Sabrina Copia_TAR2CACTA CACTA Gypsy_Sabrina
C. (BAC5)
Gypsy_Derami
CACTA
CACTA Copia_unnamed
Copia_BARA1 unknown unknown Gypsy_Romani Copia_Inga Gypsy_FaƟma
unknown
D. (BAC6)
Gypsy_Sumana Gypsy_Sumana
CACTA LTR_unknown CACTA LTR_unknown CACTA Gypsy_Wilma
E. (BAC2)
Gypsy_Sabrina Gypsy_Laura Copia_WIS
CACTA CACTA Gypsy_Wilma CACTA Gypsy_Daniela Copia_WIS
Gypsy_FaƟma CACTA CACTA Copia_Valerie CACTA Gypsy_FaƟma Gypsy_Sabrina Copia_Ikeros

Copia_Angela 10kb
Fig. 2 Organization of transposons and retrotransposons. Black arrowheads indicate -gliadin genes.
Arrows above boxes indicate the directions of gene transcription. Thick arrows indicate a predicted LTR.
Upward-pointing arrows indicate the border of a major duplicated fragment.
A large number of transposable elements (TEs) Gene expression

were found in all five BAC clones (Fig. 2). Nested 0 0 0
insertions of different classes of retrotransposons A1 PA2 A3
DP D DP
occurred frequently. a/b-gliadin genes were not 67 BAC5a-gli-3 0
BAC5a-gli-1
necessarily located in gene-rich regions in the 68 BAC5a-gli-6 0 0 0
short arms of homoeologous chromosome 6, 0 0 0
and segmental duplication frequently occurred 87 BAC5a-gli-5 1 0
through unequal crossing-over or saltatory BAC5a-gli-7 4 1 0
100 BAC5a-gli-4
replication both with and independently of 5 5 0
TE insertions. 52 BAC5a-gli-8 0 0
64 BAC5a-gli-2 0 1 0
In total, 12 full-length -gliadin 64
BAC2a-gli-1 10 19 3
genes and three pseudogenes BAC3a-gli-1 4 61 8
with premature stop codons were 100 BAC3a-gli-2 13 1 1
identified from the BAC sequences. BAC6a-gli-1 0 0 0
A phylogenetic analysis of 15 100 BAC6a-gli-2 0 0 0
-gliadin genes was conducted BAC11a-gli-1 0 0 0
(Fig. 3). The expression of each 100 BAC11a-gli-2 12 0
gene was estimated by counting
the number of ESTs having 0.06 0.05 0.04 0.03 0.02 0.01 0.00
identical sequences. There
were no significant Fig. 3 Phylogenetic relationship and expression patterns of 15 -gliadin
relationships among genes inferred from their BAC sequences. DPA, days post-anthesis.

lengths in nucleotides, chromosome locations, sequences (CDS) and noncoding regions in a 1
and expression. Although expression levels were kb sequence from 200 to 1200 bp downstream
not related to the lengths of the coding sequences, of the stop codon (Table 1). We also estimated
a positive correlation was found between the times of retrotransposon insertion based
expression level and GC content. This suggests on the assumption that the two LTRs of a
that epigenetic and other complicated systems retrotransposon were identical at the time of
of gene regulation play roles in the expression of insertion. A total of 12 pairs of LTRs could be
seed storage protein genes. assigned from the BACs (Table 2). Comparing
each pair of LTRs, the insertion time was
Finally, we estimated dates of gene duplication
estimated as 9.65 million years ago (MYA) for
and retrotransposon insertion. The evolutionary
the oldest retrotransposon and 0.30 MYA for the
distances of the duplicated -gliadin gene
youngest one.
regions were determined on the basis of coding
Table 1. Estimation of evolutionary distances between duplicated genes.
Distance Million years ago

Compared gene combination Identical pairs (bp) (S.E.) (×10-2) (S.E.)
Coding region (CDS)
BAC3a-gli-1 BAC3a-gli-2 876 1.02 (0.31)
BAC11a-gli-1 BAC11a-gli-2 846 2.11 (0.50)
BAC5a-gli-1~8 (average) 871 0.60 (0.14)
BAC6a-gli-1 BAC6a-gli-2 720 1.66 (0.50)
Noncoding region (200–1200 bp downstream)
BAC3a-gli-1 BAC3a-gli-2 981 0.91 (0.29) 0.70 (0.22)
BAC11a-gli-1 BAC11a-gli-2 947 2.31 (0.52) 1.78 (0.40)
BAC5a-gli-1~8 (average) 989 1.03 (0.21) 0.79 (0.16)
BAC6a-gli-1 BAC6a-gli-2 830 2.39 (0.54) 1.84 (0.42)
Table 2. Estimation of insertion dates of retrotransposons.
5’ LTR 3’ LTR Distance Million

Locus Clone Retroelement Type (bp) (bp) (SE) (×10-2) years ago (SE)
Gli-A2 BAC3 Sabrina gypsy 1,353 718 2.56 (0.62) 1.97 (0.48)
BAC3 Sabrina gypsy 1,339 720 2.55 (0.62) 1.96 (0.46)
BAC11 Laura gypsy 1,215 1,215 2.85 (0.50) 2.20 (0.38)
BAC11 Sabrina gypsy 744 1,202 12.55 (1.37) 9.65 (1.05)
BAC11 WIS copia 1,711 1,713 4.29 (0.51) 3.30 (0.39)
BAC11 Fatima gypsy 486 486 4.04 (0.89) 3.10 (0.68)
Gli-B2 BAC5 Unknown copia 128 128 0.79 (0.72) 0.61 (0.56)
BAC5 Romani gypsy 2,840 2,850 0.39 (0.11) 0.30 (0.09)
BAC5 Inga copia 1,736 1,733 4.42 (0.55) 3.40 (0.42)
Gli-D2 BAC2 Wilma gypsy 1,522 1,520 4.28 (0.53) 3.30 (0.40)
BAC2 Valerie copia 2,177 2,178 1.11 (0.22) 0.85 (0.17)
BAC2 WIS copia 1,718 1,726 0.82 (0.20) 0.63 (0.15)
LTR = long terminal repeat

Conclusions References
Combining the estimated times of -gliadin Allouis S, Moore G, Bellec A, Sharp R, Faivre Rampant
gene duplication with those of retrotransposon P, Mortimer K, Pateyron S, Foote TN, Griffiths S,
Caboche M, Chalhoub B (2003) Construction and
insertion allows the genome dynamism of the characterization of a hexaploid wheat (Triticum
Gli-2 loci to be clarified. Duplication of genes aestivum L.) BAC library from the reference germplasm
within the same BACs took place before or after ‘Chinese Spring’. Cereal Res Comm 31:331-338
allotetraploidization, but duplication of certain International Rice Genome Project (2005). The map-based
genes occurred before diploid differentiation sequence of the rice genome. Nature 436:793-800
of wheat species. Retrotransposons were Kawaura K, Mochida K, Ogihara Y (2005) Expression
also inserted before and after the segmental profile of two storage-protein gene families in
duplication events. In this study, 15 -gliadin hexaploid wheat revealed by large-scale analysis of
expressed sequence tags. Plant Physiol 139:1870-1880
genes were identified. The overall constituents
of the cis-elements for these genes were similar. Okita T, Cheesbrough V, Reeves C (1985) Evolution and
heterogeneity of the alpha-/beta-type and gamma-type
The systems of gene regulation among these gliadin DNA sequences. J Biol Chem 260:8203-8213
multigenes should be clarified.
Sakata K, Nagamura Y, Numa H, Antonio BA, Nagasaki
H, Idonuma A, Watanabe W, Shimizu Y, Horiuchi I,
Matsumoto T, Sasaki T, Higo K (2002) RiceGAAS: an
automated annotation system and database for rice
genome sequence. Nucleic Acids Res 30:98-102
San Miguel PJ, Ramakrishana W, Bennetzen JL, Busso
CS, Dubcovsky J (2002) Transposable elements, gene
and recombination in a 215-kb contig from wheat
chromosome 5Am. Funct Integr Genomics 2:70-80
Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4:
Molecular Evolutionary Genetics Analysis (MEGA)
software version 4.0. Mol Biol Evol 24:1596-1599

Profiling gliadins from common wheat using two-
dimensional gel electrophoresis
M. Miura, K. Kawaura, M. Nakamura, H. Kouyama, and Y. Ogihara
Yokohama City University, Kihara Institute for Biological Research, Maioka-cho 641-12,
Yokohama 244-0813, Japan
Abstract
Gluten significantly contributes to the processing properties of dough. Gluten consists of gliadin
and glutenin, which are seed storage proteins in wheat. Storage proteins play an important role in
influencing the quality of flour. In addition, these proteins function as allergens. In particular, α/β-
gliadin is reported to cause celiac disease (CD). Ingestion of α/β-gliadins by people with CD induces
inflammation of the small intestine, which results in malnourishment. Genes encoding α/β-gliadins
belong to multigene families at the Gli-A2, Gli-B2, and Gli-D2 loci located on the short arms of wheat
chromosomes in homoeologous group 6. To characterize gliadins in bread wheat, we performed
profiling of gliadins from Chinese Spring using two-dimensional polyacrylamide gel electrophoresis
(2D-PAGE). We assigned certain spots to the A, B, and D genomes using the homoeologous group 6
nullisomic-tetrasomic lines and ditelosomic lines. Gliadins were extracted from mature seeds using the
sodium iodide (NaI) method. We identified 73 gliadin spots in Chinese Spring and assigned 9 spots to
the A genome, 7 to the B genome, and 16 to the D genome. Furthermore, we obtained the profiles of
gliadins from other bread wheat genotypes. Different varieties showed wide variations in the positions
of the spots, suggesting the possibility of breeding hypoallergenic bread wheat.
Introduction (Oberhuber et al. 1999). In addition, ω-gliadin

Wheat gluten consists of the seed storage causes wheat-dependent exercise-induced
proteins gliadin and glutenin. They accumulate anaphylaxis (WDEIA). In particular, ω5-gliadin
in starchy endosperm during grain filling as is reported to be a major allergen (Morita et al.
reserves to support subsequent seedling growth. 2003; Matsuo et al. 2005a; Lauriere et al. 2007;
The genes encoding these proteins are expressed Morita et al. 2007). On the other hand, glutenin
only during the stages of seed development is also associated with WDEIA (Matsuo et al.
(Kawaura et al. 2005). In addition, these proteins 2005b; Morita et al. 2007). These storage proteins
influence the processing properties of dough. consist of multigene families, and 100 (Okita et
Gliadin proteins are classified as α/β-, γ-, or al. 1985) to 150 (Anderson et al. 1997) copies of
ω-gliadin on the basis of amino acid composition the genes encoding α/β-gliadin are present in a
and molecular weight (Wieser 2007), and they single wheat genome. However, not all gliadins
determine the viscosity of the dough. Glutenins have epitopes, and although some epitopes were
are classified on the basis of molecular weight as reported as allergens, the specific gliadins that
high molecular weight glutenin subunits (HMW- have epitopes remain to be clarified. In addition
GS) and low molecular weight glutenin subunits to gliadins, some LMW-GSs have epitopes that
(LMW-GS). also cause allergy (Bouchez-Mahiout et al. 2010).
Therefore, comprehensive analyses of gliadins
Although gluten is important for processing and glutenins are required to determine their
properties, its epitopes cause allergy. α/β-gliadin allergic potentials.
causes celiac disease (CD) (Salentijn et al. 2009;
van Herpen et al. 2006). CD is an enteropathy of In this study, we aimed to assign gliadins to
gluten sensitivity (Marsh 1992) characterized by the 6A, 6B, and 6D chromosomes and to collect
malabsorption of nutrients. Histopathological profiles of gliadins from various bread wheat
examination shows damage to the small intestine strains.

Materials and methods Two-dimensional-PAGE analysis
Sample buffer (8.5 M urea, 4% CHAPS, 0.5% IPG
Plant materials
buffer [GE Healthcare, http://www.gelifesciences.
To assign gliadin spots to the A, B, and com], 25 mM DTT, 16% isopropanol, BPB) was
D genomes, we used aneuploid lines of added to the pellets, mixed for 30 min, and
Triticum aestivum cv. Chinese Spring (CS), i.e., centrifuged for 1 min at 15,000 rpm. Isoelectric
nullisomic-tetrasomic and ditelosomic lines focusing (IEF) was performed at pH 6-9. SDS-
of homoeologous group 6 chromosomes. In PAGE was performed at 200 V using 15%
addition, various strains of bread wheat were acrylamide gels for 5 h. The gels were stained
used to compare the gliadin profiles. using the dye SYPRO RUBY. Fluorescence of
the spots was detected using a Variable Image
Protein extraction Analyser Typhoon 9400 (GE Healthcare). The
Gliadins were extracted from mature seeds spots were analyzed using Image Master 2D
using the sodium iodide (NaI) method (DuPont Platinum 6.0 software (GE Healthcare).
et al. 2005; Fu and Sapirstein 1996). Mature half
seeds, excluding the embryos were crushed; each
sample consisted of 15 mg of the crushed seeds. Results and discussion
Gliadins were extracted in 300 μL of gliadin
extract buffer (0.3 M NaI and 7.5% 1-propanol). Comparative analysis of gliadins in aneuploid
After 30 min, the samples were centrifuged for lines of CS
10 min at 15,000 rpm, the supernatants were We assigned certain spots to the A, B, and
pooled; 200 μL of extract buffer was added to the D genomes using the gliadin profiles from
supenatants, and the samples were vortexed. The aneuploid lines of group 6 chromosomes. We
samples were centrifuged for 10 min at 15,000 identified 73 α/β-gliadin spots in CS (Fig. 1).
rpm and the supernatants were pooled again. We assigned 9 spots to the A genome, 7 to the
Gliadin was precipitated by adding 4 volumes of B genome, and 16 to the D genome. Although
acetone and chilling at -30°C for 4 h. The mixture 32 spots were assigned to each chromosome, 22
was centrifuged for 10 min at 15,000 rpm, and the spots were always retained irrespective of the
supernatant was discarded. The pellets were then aneuploid lines.
used for 2D-PAGE analysis.
6 pH 8.5
50 kd
37 kd
25 kd
Fig. 1 Gliadin profiles of Chinese Spring (CS) and aneuploid lines of CS obtained using 2D-PAGE.

Comparative analysis of gliadins in various Lauriere M, Pecquet C, Boulenc E, Bouchez-Mahiout I,
bread wheat varieties Snegaroff J, Choudat D, Raison-Peyron N, Vigan M,
Branlard G (2007) Genetic differences in omega-
We obtained the profiles of gliadin from 8 gliadins involved in two different immediate food
varieties of bread wheat. The gliadin profiles hypersensitivities to wheat. Allergy 62:890-896
were markedly different among genotypes. Marsh MN (1992) Gluten, major histocompatibility
Comparative analysis of gliadin profiles among complex, and the small intestine. A molecular
varieties showed that some spots were commonly and immunobiologic approach to the spectrum of
gluten sensitivity (‘celiac sprue`). Gastroenterology
present across strains, whereas others differed
102:330-354
from variety to variety.
Matsuo H, Kohno K, Morita E (2005a) Molecular
cloning, recombinant expression and IgE-binding
epitope of ω-5 gliadin, a major allergen in wheat-
Conclusions dependent exercise-induced anaphylaxis. FEBS J
The profiles we obtained suggest that these 272:4431-4438
spots may be located on other chromosomes, Matsuo H, Kohno K, Niihara H, Morita E (2005b).
Specific IgE determination to epitope Peptides of
e.g., chromosome 1. To confirm this finding, we
ω-5 gliadin and high molecular weight glutenin
are currently analyzing the 2D-PAGE profiles subunit is a useful tool for diagnosis of wheat-
by using aneuploid lines of chromosome 1. dependent exercise-induced anaphylaxis. J
We plan to perform immunoblotting and Immunology 175:8116-8122
amino acid sequence analysis of various spots. Morita E, Kunie K, Matsuo H (2007). Food-dependent
These analyses will allow us to pinpont and exercise-induced anaphylaxis. J Dermatological Sci
characterize the allergy-causing gliadins. 47:109-117
Morita E, Matsuo H, Mihara S, Morimoto K, Savage
AW, Tatham AS (2003) Fast ω-gliadin is a major
allergen in wheat-dependent exercise-induced
References anaphylaxis. J Dermatological Sci 33:99-104
Anderson OD, Litts JC, Greene FC (1997) The α-gliadin
Oberhuber G, Grandistsh G, Vogelsang H (1999)
gene family. Characterization of ten new wheat
The histopathology of celiac disease: time for a
α-gliadin genomic clones, evidence for limited
standardized report scheme for pathologists. Eur J
sequence conservation of flanking DNA, and Southern
Gastroenterol Hepatol 11:1185-1194
analysis of the gene family. Theor Appl Genet 95:50-58
Okita T, Cheegrough V, Reeves C (1985). Evolution and
Bouchez-Mahiout I, Snégaroff J, Tylichova M, Pecquet C,
heterogeneity of the alpha/beta-type and gamma-
Branlard G, Laurière M (2010) Low molecular weight
type gliadin DNA sequences. J Biol Chem 260:8203-
glutenins in wheat-dependant, exercise-induced
8213
anaphylaxis: Allergenicity and antigenic relationships
with omega 5-gliadins. Allergy Immunology 153:35-45 Salentijn EMJ, Goryunova SV, Bas N, Meer I, Broeck
H, Bastien T, Gilissen LJ, Smulders MJ (2009)
DuPont FM, Chan R, Lopez R, Vensel WH (2005) Sequential
Tetraploid and hexaploid wheat varieties reveal
extraction and quantitative recovery of gliadins,
large differences in expression of alpha-gliadins
glutenins, and other proteins from small samples of
from homoeologous Gli-2 loci. Bio Med Central
wheat flour. J Agric Food Chem 53:1575-1584
Genomics 10:48
Fu BX, Sapirstein HD (1996) Procedure for isolating
van Herpen T, Goryunova SV, van der Schoot J, Mitreva
monomeric proteins and polymeric glutenin of wheat
M, Salentijn E, Vorst O, Schenk MF, vanVeelen
flour. Cereal Chem 73:143-152
PA, Koning F, van Soest LJM, Vosman B, Bosch D,
Kawaura K, Mochida, K, Ogihara Y (2005) Expression Hamer RJ, Gilissen LJWJ, Smulders MJM (2006)
profile of two storage-protein gene families in Alpha-gliadin genes from the A, B, and D genomes
hexaploid wheat revealed by large-scale analysis of of wheat contain different sets of celiac disease
expressed sequence tags. Plant Physiol 139:1870-1880 epitopes. Bio Med Central 7:1
Wieser H (2007) Chemistry of gluten proteins. Food
Microbiology 24:115-119

Session 2: Biosynthesis, structure, and functional
analysis of storage protein
An asparagine residue affects LMW-GS maturation

processing
E. Egidi1, M. Janni1,2, F. Sestili1, A. Ceriotti3, R. D’Ovidio1, D. Lafiandra1, D.D. Kasarda4, W.H.
Vensel4, and S. Masci1
1DAFNE, Tuscia University, Viterbo, Italy; 2IGV, CNR, Bari, Italy; 3IBBA, CNR, Milan, Italy; 4USDA, ARS,
WRRC, Albany, CA, U.S.A.
Abstract
The low molecular weight glutenin subunits (LMW-GS) are represented by many heterogeneous
proteins. The most common, based on the first amino acid of the mature sequence, are known as
LMW-m and LMW-s types. These differ as a consequence of three extra amino acids at the N-terminus
of LMW-m types. The nucleotide sequences of their encoding genes are in fact nearly identical, so
that it is not easy to ascertain the exact N-terminal amino acid sequence on the basis of the nucleotide
sequence. It was hypothesized that the presence of an asparagine residue in position 23 of the
coding sequence might account for the processing of the N-terminal end of the LMW-s sequence.
The resulting cleavage by an asparagine protease would generate the observed N-terminus of the
LMW-s type. Conversely, replacement of the asparagine residue by a threonine would generate the
LMW-m type N-terminal sequence. We performed site directed mutagenesis of a LMW-s gene to
replace asparagine at position 23 with threonine and the converse to a LMW-m type gene. Next we
produced transgenic durum lines. These were transformed with the mutated versions of the LMW-m
and LMW-s genes along with the wild type counterpart of the LMW-m type gene. Western blot
analyses on 2D gels and proteomic comparisons between the transgenic and untransformed lines
enabled identification of the transgenic proteins. MS/MS analyses and Edman N-terminal amino acid
sequences confirmed that processing occurs according to our predictions.
Introduction The B group is composed primarily of typical

Glutenin subunits are divided into high (HMW- LMW-GS, whereas the majority of C and D
GS) and low (LMW-GS) molecular types. The subunits correspond to gliadin-like proteins. The
HMW-GS have been extensively studied and C subunits include mostly a- and g-gliadin types,
characterized, whereas characterization of the whereas the D group includes w-type LMW-GS.
LMW-GS lags behind due to the large number of Finally, typical LMW-GS are classified according
subunits and their heterogeneity. LMW-GS may to the N-terminal amino acid sequence of the
be classified according to their electrophoretic mature protein: serine in LMW-s, methionine
mobilities in SDS-PAGE, N-terminal amino in LMW-m and isoleucine in LMW-i types
acid sequence and primary structure. They (reviewed in D’Ovidio and Masci 2004).
are divided into “typical” types, namely those LMW-i types have a unique sequence, whereas
proteins showing a characteristic primary the LMW-s and LMW-m seem practically
structure, and “gliadin-like” types, namely those identical on the basis of the complete nucleotide
LMW-GS that are structurally similar to gliadins, sequence. The N-terminal amino acid sequences
but functionally act like glutenin subunits. The of these two proteins differ in the presence or
latter, due to the presence of odd numbers of absence of the expected first three N-terminal
cysteine residues make these gliadin-like proteins amino acids (MET-, or in some minor cases, MEN
capable of forming intermolecular disulfide (Ikeda et al. 2006)) in the mature sequence of the
bonds that would allow their incorporation LMW-s type. However, the nucleotide sequence
into gluten polymers. Based on their order of corresponding to these three amino acids is
increasing electrophoretic mobility in SDS-PAGE, present in the sequences encoding both -s and -m
LMW-GS are classified into B, C and D groups. type LMW-GS precursors.
Session 2: Biosynthesis, structure, and funcƟonal analysis of storage protein 31

According to signal sequence cleavage site Plant material and genetic transformation
prediction algorithms, cleavage by signal The durum wheat cv. Svevo was used for genetic
peptidase would generate a QMET- (or QMEN-) transformation by the biolistic procedure (Blechl
N-terminal sequence. Removal of the N-terminal and Anderson 1996) with mutated and wild-type
glutamine residue would therefore be required to LMW-GS genes. The genes were cloned into the
generate an m-type LMW-GS. On the other hand, pLRPT vector for plant expression. The three
the identification of the gene coding for a specific transgenes were put under control of the HMW-
LMW-s type (Masci et al. 1998) showed the GS Dx5 promoter. Two different tags were added
presence of an asparagine instead of a threonine at the C-terminus (His- and FLAG-tags) of each
before the start of the mature sequence (MENS- target gene. Vector pUBI::BAR was used to co-
instead of METS-). Accordingly, we postulate that transform embryos of wheat to screen and select
the presence of the asparagine in the immature transgenic plants. For identification, transformed
protein instead of the threonine residue could plants were analyzed by PCR using primers
undergo preferential processing to result in the specific for the HMW-GS Dx5 promoter. Positive
N-terminal end being of the LMW-s type. This lines were increased to the T4 generation.
would result from cleavage of the peptide MEN
by an asparaginyl endoprotease, similarly to what 2D Electrophoresis of glutenin subunits
likely occurs in w-gliadins (DuPont et al. 2004).
Glutenin subunits of durum cv. Svevo and
We thus produced transgenic durum wheat corresponding transformed lines (T4) were
lines, transformed with the mutated versions extracted according to Singh et al. (1991) and
of the LMW-m and LMW-s genes (TN for the submitted to 2D electrophoresis. Isoelectric
LMW-m gene, and NT for the LMW-s gene) and focusing (IEF) was performed using the
demonstrated this premise. IPGphorTM Isoelectric Focusing System
(Amersham Pharmacia Biotech) and carried out
on immobilized pH gradient (IPG) strips (18
Materials and methods cm, 1 mm) pH3-10 for the LMW-GS B11-33 and
pH6-11 for the LMW-GS 42K. Second dimension
Gene isolation and site directed mutagenesis separation was carried out in vertical 12%
We used a LMW-GS gene from the bread wheat polyacrylamide gels using Protean II xi Multi-cell
cultivar Yecora Rojo, isolated by using primers (Bio-Rad). Three replicas for each sample were
specific for the gene coding for the so-called 42K performed.
protein, for which correspondence between gene
Spot detection and quantification was performed
and protein showing LMW-s type N-terminal
using Progenesis SameSpots (Nonlinear
sequence was previously demonstrated (Masci
Dynamics Newcastle, UK). Replicated gels for
et al. 1998). As an LMW-m type candidate, we
each group (transformed and non-transformed)
chose the gene B11-33 (Okita et al. 1985) because
were matched to each other to identify the
it shows a cysteine at position 5 of the mature
transgenic proteins. For the same reason,
protein and no LMW-s type characterized so far at
immunoblotting was also performed on 2D
the protein level shows such a cysteine.
gels, by using Anti-FLAG polyclonal antibody
Site-directed mutagenesis of the threonine (Sigma).
residue present at position 23 of the B11-33 gene
was performed with the QuickChange II Site- Mass spectrometry analysis
Directed Mutagenesis Kit (Stratagene) to obtain an The identified transgenic proteins were
asparagine residue. The same kit was used on submitted to mass spectrometry analysis by a
the LMW-GS 42K gene to mutate the asparagine QSTAR Pulsar i quadrupole time-of-flight (TOF)
residue in threonine. The two mutated genes, mass spectrometer (Applied Biosystems/MDS
along with the wild type version of the B11-33 Sciex, Toronto, Canada) with a nano-electrospray
gene were used for plant transformation. We source and nano-flow. Coomassie-stained protein
did not perform genetic transformation with the spots were manually excised from 2-DE gels and
wild-type version of the LMW-GS 42K gene, since subjected to direct trypsin and chymotrypsin
it was already known to correspond to a LMW-s digestion (Vensel et al. 2011).
type (Masci et al. 1998).

N-terminal amino acid sequences the N-terminal peptide. Additionally, the
2D gels were also Western blotted on PVDF N-terminal sequence, METSHIP, was confirmed
membranes, stained with Coomassie R-250 (Tao by Edman degradation.
and Kasarda 1989) and spots corresponding
to the transgenic proteins submitted to Edman
degradation (DuPont et al. 2004) Conclusions
Our results clearly demonstrate that the
processing of LMW-m and LMW-s types is
Results and discussion dependent on the presence of a threonine or
Transgenic proteins corresponding to wild- asparagine residue, respectively, in position 23 of
type and mutated B11-33 were identified by the coding sequence. If an asparagine residue is
immunoblotting and proteomic analysis, whereas present, proteins are processed as LMW-s types;
the protein corresponding to the mutated conversely, when a threonine residue is present,
LMW-GS 42K was identified only by proteomic the mature proteins are LMW-m types.
comparison. Once the spots corresponding to The plant material described here will be used to
the transgenic proteins were identified, MS/ determine if the different processing of LMW-GS
MS analyses were performed on them. We first has an influence on grain quality properties.
analyzed spots corresponding to the wild-type
B11-33 to confirm that a LMW-m type was
generated. Multiple proteins with the same References
molecular weight, but slightly different pI were Blechl AE, Anderson OD (1996) Expression of a novel
generated, likely due to multiple deamidation, a high-molecular-weight glutenin subunit gene in
common artefact in 2D electrophoresis. Nineteen transgenic wheat. Nat Biotechnol 14:875-879
unique peptides, corresponding to 54% coverage, D’Ovidio R, Masci S (2004) The low-molecular-weight
including peptides corresponding to the His- and glutenin subunits of wheat gluten. J Cereal Sci
FLAG-tags, and the N-terminal peptide showing 39:321-339
the expected METSCIFGLE sequence were DuPont FM, Vensel W, Encarnacao T, Chan R, Kasarda
DD (2004) Similarities of omega gliadins from
identified. Edman degradation performed on the
Triticum urartu to those encoded on chromosome
matching protein spots, previously electroblotted, 1A of hexaploid wheat and evidence for their post-
showed the same N-terminal sequence. translational processing. Theor Appl Genet
108:1299-1308
The mutated version of the LMW-GS B11-33 also
Ikeda TM, Araki E, Fujita Y, Yano H (2006)
produced multiple spots. In this case, 22 unique
Characterization of low-molecular-weight glutenin
peptides were identified, accounting for 63% subunit genes and their protein products in common
coverage. Among the peptides identified, there wheats. Theor Appl Genet 112:327-334
was the peptide containing the two tags along Masci S, D’Ovidio R, Lafiandra D, Kasarda DD (1998)
with the N-terminal peptide SCISGLERPW, that Characterization of a low-molecular-weight glutenin
showed the expected sequence, and missing the subunit gene from bread wheat and the corresponding
first three amino acids. The N-terminus was also protein that represents a major subunit of the glutenin
polymer. Plant Physiol 118:1147-1158
confirmed by Edman N-terminal sequencing on
electroblotted protein spots. Okita TW, Cheesbrough V, Reeves CD (1985) Evolution and
heterogeneity of the alpha-/beta-type and gamma-type
Identification of the transgenic mutated LMW-GS gliadin DNA sequences. J Biol Chem 260:8203-8213
42K protein was difficult as the transgene had lost Singh NK, Shepherd KW, Cornish GB (1991) A simplified
the His- and FLAG-tags as a result of multiple SDS-PAGE procedure for separating LMW subunits of
rearrangements and deletions, during cloning. glutenin. J Cereal Sci 14:203-208
Moreover, the recipient genotype (cv. Svevo) has Tao HP, Kasarda DD (1989) Two-dimensional gel mapping
and N-terminal sequencing of LMW-glutenin
a very similar LMW-GS, with the same molecular
subunits. J Exp Bot 40:1015-1020
weight. A detailed proteomic comparison
Vensel WH, Dupont FM, Sloane S, Altenbach SB (2011)
allowed the identification of the transgenic Effect of cleavage enzyme, search algorithm and decoy
protein, by MS/MS. Twenty unique peptides were database on mass spectrometric identification of wheat
identified, covering 55% of the protein, including gluten proteins. Phytochemistry 72:154-161

Intracellular trafficking and tissue distribution of gluten
proteins: The epitope-tagging approach
P. Tosi1, J. Freeman1, C. Gritsch1, C. Sparks1, H. D. Jones1, W. Funatsuki2, K. Niwa3, E. Egidi4, A. Ceriotti5,
S. Masci4, and P. R. Shewry1
1Rothamsted Research, Harpenden, U.K; 2National Agricultural Research Center, Hokkaido Region (NARCH),
Japan, 3Tokyo AgriculturalUniversity, Tokyo, Japan; 4University of Tuscia, Viterbo, Italy; 5IBBA-CNR, Milan, Italy
Abstract
Wheat dough viscoelasticity is controlled by gluten proteins consisting of the monomeric gliadins and
polymeric glutenins, and representing the most abundant proteins in wheat grain endosperm. The
functional properties of the gluten proteins depend on their amount and composition, and also on their
assembly into polymers, which in turn depends on the specific locations of gluten proteins within the
endosperm and their segregation during deposition. It is widely accepted that two trafficking pathways
operate in wheat endosperm, leading to two mechanisms of protein body formation; one requiring
translocation via the endoplasmic reticulum (ER) and Golgi and deposition in the vacuole, and the
other, direct aggregation and deposition in the ER. It has also been suggested that individual proteins
may follow one or both routes, and that the two trafficking pathways may differ in importance during
grain development, with accumulation within the ER becoming more prominent at the later stages. Our
understanding of the dynamics of trafficking and deposition of specific gluten protein types and of the
mechanism of wheat protein body formation has been limited so far by practical difficulties in developing
monospecific antibodies for gluten proteins and by an inability to unequivocally identify the origins of
membranes in microscopy sections. We overcame both technical limitations by using epitope tagging
to specifically tag individual gluten proteins and to tag membrane proteins that can act as markers for
specific organelle compartments (ER, Golgi, vacuole) in transgenic wheat.
Introduction studies suggest that two trafficking pathways

Wheat is one of the most important food crops exist in wheat, with the proteins either being
in the world and its success is largely due to transported via the Golgi apparatus into the
the unusual functional properties of its flour, vacuole or assembling and accumulating within
which derives from the grinding of the grain the endoplasmic reticulum (ER) to form a second
starchy endosperm. In the mature wheat grain population of protein bodies. It has also been
the starchy endosperm consists of cells packed suggested that individual proteins may follow
with starch and protein, mainly gluten proteins. one or both routes and that the two trafficking
Clear quantitative and qualitative gradients pathways may differ in importance during grain
exist across the endosperm for storage proteins development, and, maybe, also between cell
(Tosi et al. 2009, 2011) and, consequently, it is layers, with accumulation within the ER being
possible to isolate high protein flour fractions more prominent in older tissues (Tosi et al. 2009).
derived from cells of the outer layers of the Our understanding of the dynamics of
starchy endosperm. These are also enriched in trafficking and deposition of specific gluten
gliadins and low protein flour fractions from protein types and of the mechanism of wheat
cells of the central part of the endosperm that are protein body formation has been limited by
enriched in the high molecular weight (HMW) practical difficulties in developing monospecific
subunits of glutenin. The functional properties antibodies for gluten proteins and by inability to
of the gluten proteins depend on their amount, unequivocally identify the origins of membranes
composition, and mechanism of deposition in sections for microscopy. This in turn has
and tissue distribution, since the combination limited our ability to relate specific molecular
of these factors determine the amount and features of gluten proteins with their mechanisms
characteristics of the gluten polymers. Several of trafficking, and our ability to discriminate

protein bodies of ER or vacuolar origin in planta. of wheat protein body formation. Preliminary
We overcame both technical limitations by results obtained by the analysis of transgenic
using epitope tagging. We produced a first set lines expressing epitope-tagged gluten protein or
of transgenic wheat lines expressing C-terminal tagged organelle markers are presented here.
epitope-tagged gluten proteins (HMW glutenin
subunit, low molecular weight (LMW) glutenin
subunit and γ-gliadin) whose location in cells Materials and methods
of the developing grain can be determined A series of 10 constructs encoding gluten
using highly specific, commercial antibodies proteins (a γ-gliadin, HMW glutenin subunit
against the tags. More recently, a second set of 1Ax1 and a B-type LMW glutenin subunit) with
transgenic bread wheat lines expressing tagged C-terminal epitope tags (either C-myc, HA or
proteins as markers for specific compartments FLAG) under control of either a LMW or a HMW
of the secretory pathway was also developed; glutenin subunit gene promoter, were prepared
these include lines expressing fluorescence- and/ as described in Shewry et al. (2007) and used to
or epitope-tagged forms of markers for the ER transform the bread wheat cv. Cadenza using
membrane, Golgi membrane and tonoplast. particle bombardment. Selected transgenic
Crossing of these two sets of transgenic lines will homozygous lines were grown in the glasshouse
provide a powerful system in which to study and developing seeds were collected and
the dynamics of trafficking and deposition of prepared for microscopy as in Tosi et al. (2009,
different gluten protein types and the mechanism 2011). Seeds were analyzed by SDS-PAGE and
Fig. 1 Constructs used for transformation of wheat for expression of fluorescence- and epitope-tagged
membrane proteins for organelle tagging. A, marker for endoplasmic reticulum (ER); B, marker for the
tonoplast (vacuole); C, marker for the Golgi.
Fig. 2 Constructs used for transformation of wheat to express epitope-tagged gluten proteins.

western blotting as described in Tosi et al. (2005). the amount of this type of gluten protein in the
Crosses between transgenic lines expressing inner cell layers of the endosperm: the pattern of
different gluten protein types and different tissue expression resembled that of endogenous
epitope tags were also performed. HMW glutenin subunits. The expression of
the transgenic subunits, in general, did not
Three further constructs encoding for
significantly change the amount of total gluten
Cherry-fused Sec61B-HA, α-mannosidase1-
protein (as measured by densitometry scanning),
GFP-HA fused, and α-TIP YFP-HA fused,
the exception being the line with silencing of
were also prepared for ER, Golgi or vacuole
endogenous HMW glutenin subunits (about
membrane tagging, respectively, and used for
15% less gluten protein). Analysis of the tissue
transformation of wheat cv Cadenza. Gene
distribution of transgenic LMW glutenin subunit
expression was driven by the endosperm-
is still ongoing.
specific PinA promoter. Screening for transgenic
expression was carried out via detection of in vivo
Trafficking and deposition of tagged gluten
fluorescence in grains at 16 days post anthesis,
proteins
using a LSM 780 confocal microscope (Zeiss).
Double immunofluorescence studies carried out
on developing grains of control and transgenic
Results and discussion lines showed segregation in deposition of
different gluten protein types within protein
Tissue distribution of tagged gluten proteins bodies of the same cell layer. Segregation was
under control of the LMW or HMW glutenin more evident in transgenic lines: tagged proteins
subunit promoters were sometimes observed to segregate into
Immunolocalization studies carried out on different populations of protein bodies. This was
developing grains from transgenic lines particularly noticeable for transgenic γ-gliadin
confirmed the presence of differential patterns of in cells two to three layers inside the aleurone.
distribution in the endosperm tissue for the main However, no clear conclusions can be drawn at
types of gluten proteins (HMW subunits being this stage on the origin (ER or vacuole) of the
enriched in the inner layer of the endosperm, protein bodies which are enriched in the different
gliadins and LMW glutenin subunits in the sub- gluten protein types and therefore on the
aleurone layer). The distribution of the tagged mechanisms responsible for their formation.
HMW glutenin subunit under the control of the
1Dx5 promoter appeared to be similar to the Tagged organelle lines
pattern of distribution of HMW glutenin subunits Confocal in vivo fluorescence studies of
in non-transgenic lines and coincided with the transgenic wheat grains from lines transformed
pattern of distribution of endogenous subunits with constructs to tag the ER, Golgi and tonoplast
in most transgenics; one exception was line 2414 confirmed the expression of the constructs and
R4P1 in which the expression of the transgenic their tagging specificity. Selected lines are now
subunit resulted in silencing of all endogenous being multiplied and screened for homozygotes
HMW glutenin subunits. Immunofluorescence which will then be crossed with each other and
labeling studies using the anti-epitope antibody with transgenic lines expressing c-myc-tagged
revealed that in this line the transgenic protein gluten proteins.
is present only in protein bodies of the central
endosperm (i.e., it is absent from the sub-
aleurone layer). In lines expressing the transgenic Conclusions
HMW glutenin subunit under control of the Immunomicroscopy studies of lines expressing
LMW glutenin promoter, transgene expression tagged gluten proteins under control of LMW
appeared highest in cells about two cell layers and HMW promoters confirm that it is possible,
from the aleurone, was absent from the cell layer using this transgenic approach, to increase the
adjacent to the aleurone, and low in the central concentrations of HMW glutenins and gliadins in
region of the endosperm. the subaleurone and the inner endosperm layer,
Expression of the transgenic γ-gliadin under the respectively. Epitope tagging of gluten proteins
control of the HMW glutenin promoter increased allowed us to follow their pattern of deposition

in protein bodies in the different cell layers of References
the endosperm, both at the light microscopy Tosi P, Masci S, Giovangrossi A, D’Ovidio R, Bekes F,
and at the TEM level. The main limitation in Larroque O, Napier J, Shewry PR (2005) Modification
these trafficking/deposition studies remains the of the low molecular weight (LMW) glutenin
difficulty in recognizing the ontogeny of the composition of transgenic durum wheat: effects on
glutenin polymer size and gluten functionality. Mol
membranes surrounding the mature protein Breeding 16:113-126
bodies. The transgenic lines expressing organelle
Tosi P, Parker M, Sanchis-Gritsch C, Carzaniga R, Martin
markers represent an important step toward B, Shewry PR (2009) Trafficking of storage proteins in
developing an in planta system to efficiently study developing grain of wheat. J Exp Bot 60:979-991
the deposition of different gluten protein types. Tosi P, Sanchis-Gritsch C, He J, Shewry PR (2011)
Distribution of gluten proteins in bread wheat
(Triticum aestivum) grain. Ann Bot 108:25-35

The trafficking of wheat prolamins in rice endosperm
M. Oszvald1, L. Tamás1, P. R. Shewry2, and P. Tosi2
1Eötvös Loránd University, Department of Plant Physiology and Molecular Plant Biology, Budapest, Hungary;
2Rothamsted Research, Harpenden, U.K.
Abstract
Transgenic rice was used to study the pattern and pathway of deposition of wheat high molecular
weight (HMW) subunit 1Dx5 within rice endosperm using specific antibodies to determine whether it
is deposited in the same or different protein bodies to rice storage proteins, and whether it is located
in the same or separate phases within them. The protein distribution and the expression pattern of
HMW subunit 1Dx5 in transgenic rice endosperm at different stages of development were also studied
using tissue printing, light and electron microscopy. The use of high molecular weight glutenin
subunits (HMW-GS) specific antibodies showed that subunit 1Dx5 was expressed mainly in the sub-
aleurone cells of the endosperm and that it was deposited in both types of protein body present in the
rice endosperm; derived from the ER and containing prolamins, and derived from the vacuole and
containing glutelins. In addition, new types of protein bodies were also formed within the endosperm
cells. Our results suggest that the HMW Dx5 protein could be trafficked by either the ER or vacuolar
pathways, possibly depending on the stage of development, and that segregation of gluten proteins
may occur both between and within protein bodies.
Introduction co-translationally inserted into the lumen. They

Wheat is the most important crop in Europe, subsequently become deposited in protein bodies,
being used for both food processing and livestock but the origin of these bodies has long been the
feed. Its success as a food crop is due in part subject of debate. Electron microscopy (EM) has
to the unique properties of the grain storage provided convincing evidence for the presence of
proteins. These form the gluten fraction which gluten proteins inside vesicles associated with the
confers the unique viscoelastic properties that Golgi apparatus, suggesting that storage proteins
allow wheat dough to be processed to form may pass through the Golgi in their transport from
bread, noodles, pasta, and many other processed the ER to the vacuoles where they form protein
foods. The well-established systems for wheat deposits. However, it is now accepted that two
grain production, storage and fractionation also routes of protein body formation (translocation
make it attractive for the production of novel via the ER and Golgi to the vacuole and direct
components, such as high value pharmaceutical accumulation in the ER) may operate in wheat
products or raw materials for industry. In order endosperm and it has been suggested that they
to exploit all of these possibilities, it is necessary may differ in their importance during grain
to understand the pathways and mechanisms development, with accumulation within the ER
which determine the synthesis, processing, becoming more prominent at later stages, and
trafficking and deposition of storage components individual proteins following one or both routes
in the developing grain (Shewry and Jones 2005; (Tosi et al. 2009; Tosi 2012).
Wiley et al. 2007). Rice differs from wheat in that two major types
A unique feature of plants is that they are able of storage proteins (prolamins and glutelins) are
to store proteins in the endoplasmic reticulum synthesized in the starchy endosperm cells and
(ER) in addition to other endomembrane deposited into distinct protein bodies (PBs) of
compartments. The accumulation of such storage different origin (Muench and Okita 1997). The
proteins also provides important sources for both prolamins (which are related to the wheat gluten
human and animal nutrition. The gluten proteins proteins) aggregate inside the ER lumen aided by
of wheat are synthesized on the rough ER and the chaperone BiP and form type-I protein bodies

whereas the glutelins (which are related to the London Resin, Berkshire, UK). For light and
11S globulin storage proteins of dicotyledonous fluorescence microscopy, approximately 1 μm
plants) are translated on separate subdomains of thick sections were cut and collected on Poly-
the ER and transported via the Golgi apparatus lysine coated multiwell slides. For electron
and vesicles into type-II protein bodies of microscopy, ultrathin sections were collected on
vacuolar origin. This clear segregation of rice nickel mesh grids and examined with a JEOL
storage proteins into two populations of protein JEM2010 transmission electron microscope.
body means that transgenic expression in rice can
Sections were pre-incubated in 3% (w/v) BSA in
be used to determine whether the wheat storage
PBS to prevent non-specific binding of antibodies,
proteins contain targeted trafficking signals that
and then incubated in primary antibodies (rabbit
are recognized by the cell secretory system.
polyclonal anti-prolamin, rabbit polyclonal anti-
We therefore developed transgenic rice lines glutelin diluted, mouse monoclonal anti-IFRN
expressing wheat high molecular weight (HMW) 1602, rabbit polyclonal anti-HMW-R2) used either
subunit 1Dx5 protein, which is present in protein individually or in combinations. After several
bodies of both vacuolar and ER origin in wheat, rinses of the sections, the binding of primary
although concentrated in the latter (Oszvald et al. antibodies were revealed by incubation with
2007), and used specific antibodies to determine secondary antibodies conjugated to a fluorescent
its distribution between and within tissues and label (Alexa 488, or/and Alexa 568) using a Zeiss
protein bodies in the developing rice grain. 780 confocal laser scanning microscope (CLSM).
A similar protocol to that described above was
used for immuno-TEM studies of ultrathin
Materials and methods sections from the same samples. Secondary
Plant materials and antibodies antibody (goat anti-rabbit conjugated to 15 nm
gold) (British BioCell International) was used
The production of transgenic rice lines expressing
at a 1:50 dilution. Gold labeled sections were
HMW subunit 1Dx5 of wheat under the control
silver-enhanced using R-Gent SE-EM Silver
of its own starchy endosperm-specific promoter
Enhancement Reagents kit from AURION
was described previously (Oszvald et al. 2007).
according to the manufacturer’s protocol.
Seeds were harvested at different developmental
stages: 7, 10, 14, 17, 20 and 25 days after flowering
(daf). Antibodies to purified rice glutelin subunits
Results and discussion
and prolamins (Krishnan et al. 1986) and were
used for the detection of the transgenic HMW Tissue printing with the anti R2-HMW antibody
glutenin subunit in rice (monoclonal IFRN 1602 showed that the HMW subunit of glutenin
(Mills at al. 2000), and polyclonal anti-R2- HMW was concentrated in the outer layers of the
(Denery-Papini et al. 1996)). endosperm. Only weak labeling was observed
in the central part of the endosperm indicating
Tissue printing a high accumulation of the HMW-GS protein
in the outer layers of this tissue. Localization
Tissue printing was carried with the anti-R2-
of the transgenic HMW-GS, both at the cellular
HMW antibody. Frozen seeds were sectioned
and subcellular levels, was carried out by
using a cryostat microtome (Leica CM 1850) and
immunofluorescence microscopy. To determine
collected on nitrocellulose coated slides (Oncyte
in which type of protein body (ER or vacuolar)
Film-slides, race Bio-Labs Inc.) according to
the transgenic wheat protein was deposited,
Wiley et al. 2007. Tissue prints were examined in
double labeling was carried out using the
a Leica MZ8 stereomicroscope.
monoclonal IFNR 1602 antibody in combination
with the polyclonal antibodies recognizing native
Preparation of specimens for microscopy
rice prolamin or glutelin proteins. In these double
Sections of developing rice grain were cut and labeling experiments the primary antibodies
fixed in 4% (v/v) paraformaldehyde and 2.5% were detected using secondary antibodies
(v/v) glutaraldehyde in phosphate buffered at conjugated to two different fluorochromes,
pH 7.2. The fixed samples were dehydrated, AlexaFluor conjugated 488 and 568. The wheat
embedded in LR White resin (medium grade; HMW-GS was often co-located with the rice

prolamin proteins in PB-I at the early stages of The presence of the recombinant wheat HMW-
development (7 daf). However, not all of the GS in the new type PB-I like structures was
protein bodies of the inner endosperm cells confirmed by labeling with the anti-R2-HMW
that contained the HMW-GS also contained rice antibody. Some of the normal PB-I containing
prolamins. Double labeling for the HMW-GS rice prolamins was also labeled by the HMW R2
and rice glutelin showed that the two proteins antibody, but differences were observed with
were largely segregated into different protein smaller bodies showing little or no labeling and
bodies. The extent of co-localization of the larger bodies having uneven patterns. A low
HMW-GS with rice prolamin also appeared to level of labeling of the PBII with the anti-R2-
increase as the grain matured. These analyses HMW antibody was also observed.
therefore demonstrated that the recombinant
Double labeling experiments were carried out
wheat HMW-GS may be co-deposited with both
using anti-R2-HMW and rice glutelin/prolamin
native rice prolamins and glutelins as well as
antibodies. Ten nm gold particles were
being present in protein bodies that apparently
conjugated to anti-rabbit secondary antibody
did not contain rice proteins.
against the anti-HMW-R2 antibody, with
The intracellular location of the wheat HMW silver enhancement to increase the size of the
subunit was studied by immunoelectron particles. The sections were then labeled using
microscopy of developing grain using the anti- antibodies specific for prolamin or glutelin
R2-HMW polyclonal antibody and a secondary with 10 nm gold particles conjugated to anti-
antibody conjugated to 10 nm gold particles. rabbit secondary antibody (Fig. 1). The PB-I and
Analyses of developing grain of transgenic rice PB-II reacted with the anti-glutelin and anti-
expressing the wheat HMW-GS showed the prolamin antibodies, respectively, in both wild
presence of PB-I and PB-II with similar sizes type and transgenic wheat (Fig. 1A, B). The
and staining properties to those in the wild type new PB-I-like structures in the transgenic rice
grain. However, a third type of protein body reacted with antibodies to the wheat HMW-GS
was also observed (Fig. 1C, D). These resemble (anti-R2-HMW) and rice prolamins (Fig. 1A,
PB-I in being spherical, but are much smaller B). The labeling also indicated that the wheat
(with 0.1-0.2 μm diameter) and lack the lamellar HMW-GS was concentrated in the peripheral
structure of typical PB-I. These new PBs also areas of the PB-I-like structures (Fig. 1D). These
appeared to be clustered and located in the ER results indicated that the wheat HMW-GS was
lumen (Fig. 1C, D). The new type of PB stained deposited in the same PB as the rice prolamins,
more densely than PB-I, but less densely than suggesting that they originated from conversion
PB-II (Fig. 1C, D). Attachment of polysomes to of PB-I into a new type of PB. Some co-labeling
the surface suggested that the new type of PB of PB-II with the wheat HMW-GS and rice
was derived from the ER in the same way as glutelin antibodies was also observed at all
PB-I (Fig. 1B). These novel PB-I-like structures stages of development, but more frequently in
were frequently observed in the endosperm the later stages (Fig. 1C, F).
cells of transgenic seeds in early and later stages
Seeds are an attractive system for the
of development.
production of recombinant proteins (Takaiwa
Such PB-I like structures were observed in the et al. 2007), with the ER, ER-derived PBs, PSVs,
endosperm cells of transgenic seeds, both in apoplast and cytosol being alternative candidate
early and later stages of development (Fig. 4, sites for the deposition and accumulation of
5) although they were more abundant in the recombinant proteins in the rice endosperm.
younger grains: the ratio of new PBs containing
Abnormal or new types of protein bodies were
the transgenic HMW glutenin subunit to
previously observed in mutant rice lines or
normal PB-I was about 1:3 in 7 and 14 daf
transgenic rice lines expressing recombinant
grains, whereas at 17 and 25 daf these new PB-I
proteins. Previous studies indicated that protein
protein bodies were quite rare as the transgenic
storage may modulate protein sorting in storage
protein was often co-localized with rice glutelin
tissues, such as endosperm, although sorting
proteins in PB-II.
signals ensure the transport of foreign proteins

A) B) C)
D) E) F)
Fig. 1 Double labeling: A-C, transgenic rice 7, 14, and 25 days after flowering (daf) labeled with anti-R2-
HMW+prolamin antibodies; D-F, transgenic rice 7, 14, and 25 daf with anti-R2-HMW+glutelin antibodies;
10 nm gold particles indicated the prolamin/glutelin proteins (large arrowheads); the larger (enhanced)
gold particles indicate the HMW Dx5 GS proteins (small arrowheads). PBS, PB-I structures.
to the expected subcellular locations in vegetable trafficking of cereal grain proteins that can be
tissues (Furukawa et al. 2007). The high rate used to improve quality for traditional end uses
of the synthesis of the storage proteins in the or to exploit cereals as systems to produce novel
endosperm tissue may have affected the site of high value products.
deposition of recombinant proteins transported
by the secretory pathway
The morphology of protein bodies was distorted
Acknowledgement
in lines accumulating the HMW-GS. These This work was supported by the Hungarian
results indicate that the transgene products can Scientific Research Fund (NKTH-OTKA, MB08A)
be transported in the most suitable form for
accumulation. Our study also suggests that the
ER of rice endosperm cells is an appropriate References
storage compartment for foreign gene products. Denery-Papini S, Popineau Y, Quillien L, Van Regenmortel
MHV (1996) Specificity of antisera raised against
synthetic peptide fragments of high Mr glutenin
subunits. J Cereal Sci 23:33-144
Conclusions Furukawa S, Itou M, Masumura T, Tanaka K, Kiyokowa
Our study provided information on differences Y, Wakai Y (2007) Ultrastructure of low-glutelin rice
in storage protein trafficking and deposition in endosperm. Plant Biotechnology 24:227-229
rice and wheat seeds, and the extent to which Krishnan HB, Franceschi VR, Okita TW (1986)
these are determined by the properties of the Immunochemical studies on the role of the Golgi
proteins. It also provided information on the complex in protein-body formation in rice seeds.
Planta 169:471-480

Mills ENC, Field JM, Kauffman JA, Tatham AS, Shewry PR, Takaiwa F, Takagi H, Hirose S, Wakasa Y (2007)
Morgan MRA (2000) Characterization of a monoclonal Endosperm tissue is a good production platform
antibody specific for HMW subunits of glutenin and for artificial recombinant proteins in transgenic rice.
its use to investigate glutenin polymers. J Agric & Food Plant Biotechnol J 5:84-92
Chem 48:611-617 Tosi P, Parker M, Carzaniga R, Martin B, Gritsch CS,
Muench DG, Okita TW (1997) The storage proteins of rice Shewry PR (2009) Trafficking of storage proteins in
and oat. In: Larkins BA, Vasil IK (eds) Cellular and developing grain of wheat. J Exp Bot 60:979-991
Molecular Biology of Plant Seed Development, Kluwer Tosi P (2012) Trafficking and deposition of prolamins in
Academic Publishers, The Netherlands, pp289-330 wheat. J Cereal Sci 56:81-90
Oszvald M, Tömösközi S, Jenes B, Békés F, Tamás L (2007) Wiley PR, Tosi P, Evrard A, Lovegrove A, Jones
Expression of the Dx5 high molecular weight glutenin HD, Shewry PR (2007) Promoter analysis and
subunit protein in transgenic rice. Cereal Res Commun immunolocalisation show that puroindoline genes
35:1543-1549 are exclusively expressed in starchy endosperm cells
Shewry PR, Jones HD (2005) Transgenic wheat: where do of wheat grain. Plant Mol Biol 64:125-136
we stand after the first 12 years? Ann Appl Biol 147:1-14

Gluten protein structures: Variation in wheat grain and for
various applications
E. Johansson1, A. H. Malik1, A. Hussain1, F. Rasheed1, W. R. Newson1, T. Plivelic2, M. Hedenqvist3,
M. Gällstedt4, and R. Kuktaite1
1Department of Agrosystems, The Swedish University of Agricultural Sciences, Box 104, SE-23053 Alnarp,
Sweden; 2MAX-lab, Lund University, SE-221 00 Lund, Sweden; 3Department of Fiber and Polymer
Technology, Royal Institute of Technology, SE-10044 Stockholm, Sweden; 4Innventia, Box 5604, SE-11486
Stockholm, Sweden
Abstract
For a number of applications, including bread-making, pasta-making, and bio-based plastics
production, gluten protein polymerization and structure are of highest importance in determining
end-use properties. Our results show that environmental factors contribute to a similar extent as the
genetic background to protein composition and polymerization behavior in mature wheat. Most
important is nitrogen availability during various phases of wheat development and the genetically
determined plant development rhythm. However, the extent of importance of nitrogen availability
and plant development rhythm is influenced by the temperature during the plant development time.
Furthermore, a genetic component is important for polymerization of proteins during mixing. In
production of bio-based plastics from wheat gluten, rearrangements of the structures and polymers of
the gluten proteins are increased even further than during bread-making. Protein polymer structure
was changed from a high content of unordered amide groups, and/or α-helix conformations, towards
a situation where β-sheet structures of both intermolecular and extended type were common. Higher
amounts of extended β-sheet structures led to outstanding characteristics of tensile strength, low
protein extraction values, very low diffusion of allergy-causing proteins from the material, and low
oxygen permeability in the material. Further, a hexagonal closed packed structure was found in this
material. Thus, understanding how proteins are aggregating and how to combine different gluten
proteins into the most suitable polymers is extremely important in both production of bread with the
best baking performance and in achieving tailored materials from gluten.
Introduction 2004; Malik 2012). The most well-known protein

Among all crops cultivated worldwide, wheat character that has been identified to determine
is the most important, covering the largest the bread-making quality properties of wheat is
cultivated area, contributing with among the the specific composition of proteins and protein
highest yields and being largely used as food and subunits in the wheat grain, i.e., the presence or
sources of proteins, calories and vitamins (FAO absence of specific proteins and protein subunits
2012; Hussain et al. 2012a). In addition to the uses such as high molecular weight glutenin subunits
for food and feed, wheat is a promising crop to 2+12 and 5+10 (e.g., Payne et al. 1987; Johansson
be used for ethanol production for fuel, and the and Svensson 1995; Johansson 1996). In recent
residual gluten from ethanol or starch production studies, gluten protein structures, protein
can be used for materials production (Gällstedt et polymerization and polymer structures have
al. 2004; Blomfeldt et al. 2011). been found of increasing importance for bread-
making properties (Gupta et al. 1993; Johansson
The proteins in the wheat grain, i.e., the gluten et al. 2001; Malik et al. 2011; Hussain et al. 2012b).
proteins, are well known to contribute to the Further, protein structures and polymerization
functional properties of wheat when wheat is were also found to be of uppermost importance
used for bread-making, materials production, for the functional properties of materials
and most likely beer production (Shewry and produced from gluten (Blomfeldt et al. 2011;
Miflin 1985; Pomeranz 1988; Gällstedt et al.

Ullsten et al. 2009, 2010; Kuktaite et al. 2011). The statistical analysis system (SAS) (SAS, 2004)
Both the wheat source (i.e., genotype) and and MS Excel were used for data analyses. The
environment (i.e., temperature and nitrogen main analyses carried out were Spearman rank
application) are known to influence the amounts correlation analysis, analysis of variance (ANOVA),
and size distributions of the polymeric and principal component analysis (PCA) and regression
monomeric proteins (Johansson et al. 2001, 2002, analysis.
2003; Malik et al. 2011).
The aim of the present paper was to evaluate how
genotype and environment interact and influence
the gluten protein composition and polymer Environmental factors contributed to a larger
structures in the wheat grain/flour, and the extent than genetic factors, measured as cultivar
impact of the obtained protein composition and determined development time (CDDT; Table 1),
polymer structures on functional properties and and for the polymerization of proteins, measured
end-use quality of wheat for various end-uses. as %UPP, at low temperature (17°C day, 14°C night
temperatures). However, at high temperatures
(24°C day, 21°C night temperatures) genetic factors
Materials and methods were more important in determining protein
polymerization compared to environmental factors
A range of spring and winter wheat cultivars
(Table 1; Malik et al. 2012). In general, temperature
grown in a number of years from 1995 to 2011
was much less important compared to nitrogen
under field conditions and/or in greenhouses
in determining protein polymerization, although
and controlled environments was used in the
the temperature was important as an interacting
investigations (Johansson et al. 2001, 2002, 2003,
factor (Table 1; Malik et al. 2012). Thus, the extent
2004, 2005, 2008; Malik et al. 2011, 2013).
of importance of other environmental and cultivar
Protein structures were evaluated primarily using factors was influenced by temperature during plant
measurements of percentage of unextractable development time.
polymeric protein in total polymeric protein
All genotypes tested for %UPP in the wholemeal
(%UPP; Gupta et al. 1993), and proteins were
flour (grain) and in the mixed dough of wholemeal
extracted by a two-step procedure and separating
flour showed significantly decreased %UPP after
the proteins by size exclusion-high performance
optimal mixing. The decrease was generally
liquid chromatography (SE-HPLC; Johansson
higher for genotypes with high %UPP in the grain
et al. 2005). Confocal microscopy (Wretfors
compared to those with low %UPP. However,
et al. 2010), transmission electron microscopy
the decrease was not consistent with the %UPP
(TEM; Blomfeldt et al. 2012), scanning electron
in the grain and a large variation was found
microscopy (SEM; Kuktaite et al. 2012) and small
among genotypes for the extent of decrease (Fig. 1;
angle X-ray spectroscopy (SAXS; Kuktaite et al.
Hussain et al. 2012b).
2011) were used for protein structure analyses.
Table 1. Percentages of explained (obtained through R-square from simple linear regression) environmental
(temperature and nitrogen) and cultivar (cultivar dependent development time) factors on percentage of
unextractable polymeric proteins into total polymeric proteins (%UPP).
Protein polymerization (%UPP)

Source Total At low temperature At high temperature
Temperature 3.76
N1 = N applied at spike formation 14.7 21.0 10.7
N2 = N applied at anthesis 10.1 15.9 5.68
Environment (N+T) 28.9 36.8 16.4
Genetic effect (CDDT1) 15.8 12.5 20.5
1CDDT, cultivar determined development time

50 gluten strength is imparted by certain subunits
High UPP and larger and more complex polymers. The
45 protein polymer size and complexity are equally
Medium UPP
40 influenced by genetic and environmental factors.
Low UPP Generally, factors affecting plant development
35
time are correlated negatively with protein
30 polymer size and complexity. Furthermore,
% UPP
25 differences in temperature resulted in various

interactions among the genotypic and
20 environmental factors investigated. During
15 dough mixing, gliadins are incorporated in the
gluten protein polymer through a number of
10
chemical reactions. Protein polymer formation
5 at dough mixing and incorporation of proteins
0 in the polymer differ in various wheat
Grain Dough genotypes are not consistent with the content
of polymers in the grain/flour. During plastics
Fig. 1 Changes in %UPP from grain to dough in production using wheat gluten, hierarchical
genotypes with high, medium and low % UPP in supramolecular structures are formed and
the mature grain. these structures seem related to functionality.
Increased functional properties were obtained
with supramolecular close-packed hexagonal
structures in the material, also incorporating the
Production of bio-based plastics from wheat gliadins in the gluten protein polymer structure.
gluten led to rearrangements of structures within Changes in the polymer structures resulted
the polymers based on gluten proteins (Gällstedt from either disulphide-sulphydryl exchange,
et al. 2004 ; Ullsten et al. 2009, 2010). A decrease isopeptide formation, or a combination thereof.
in the content of unordered amide groups, and/or Furthermore, a higher degree of β-sheets
α-helix conformations were seen with materials compared to α-helices was obtained in polymers
production, whereas increases in the content of from higher performing materials.
β-sheet structures of both intermolecular and
extended types were also seen (Ullsten et al. 2009;
Kuktaite et al. 2011, 2012). The use of NaOH as References
an additive resulted in a well ordered tetragonal Blomfeldt TOJ, Kuktaite R, Johansson E, Hedenqvist MS
structure while addition of NH4OH resulted in (2011) Mechanical properties and network structure
of wheat gluten foams. Biomacromol 12:1707-1715
a bi-dimensional hexagonal close-packed (HPC)
structure (Kuktaite et al. 2011). The increased Blomfeldt T, Kuktaite R, Plivelic TS, Rasheed F,
Johansson E, Hedenqvist MS (2012) Novel freeze-
amount of extended β-sheet structures and dried foams from glutenin- and gliadin-rich fractions.
formation of ordered structures was correlated RSC Advances 2:6617-6627
to outstanding characteristics of tensile strength, FAO (2012) FAO Statistical Yearbook 2012. FAO, Rome,
low protein extraction values, very low diffusion Italy
of allergy-causing proteins from the material, Gupta RB, Khan K, MacRitchie F (1993) Biochemical
and low oxygen permeability (Ullsten et al. 2009, basis of flour properties in bread wheats. I. Effects
2010; Kuktaite et al. 2011, 2012). of variation in the quantity and size distribution of
polymeric protein. J Cereal Sci 18:23-41
Gällstedt M, Matozzi A, Johansson E, Hedenqvist
Conclusions MS (2004) Transport and tensile properties
of compression-molded wheat gluten films.
Glutenin subunits are the proteins that are Biomacromol 5:2020-2028
mainly involved in polymer formation in wheat Hussain A, Larsson H, Kuktaite R, Johansson E (2012a)
grain/flour, and functionality is influenced by Healthy food from organic wheat: Choice of
both the specific composition of glutenin subunits genotypes for production and breeding. J Sci Food
and the polymer size and complexity. Higher Agric 92:2826-2832

Hussain A, Larsson H, Kuktaite R, Prieto-Linde ML, Kuktaite R, Plivelic TS, Türe H, Hedenqvist MS, Gällstedt
Johansson E (2012b) Towards the understanding of M, Marttila S, Johansson E (2012) Changes in the
bread-making quality in organically grown wheat: hierarchical protein polymer structure: urea and
Dough mixing behaviour, protein polymerisation and temperature effects on wheat gluten films. RSC
structural properties. J Cereal Sci 56:659-666 Advances, doi: 10.1039/C2RA21812G
Johansson E (1996) Quality evaluation of D zone omega Malik AH (2012) Governing grain protein concentration
gliadins. Plant Breeding 115:57-62 and composition in wheat and barley: use of genetic
Johansson E, Svensson G (1995) Contribution of the high- and environmental factors. Doctoral Thesis No.
molecular-weight subunit 21* to breadmaking quality 2012:55. Swedish University of Agricultural Sciences,
of Swedish wheats. Cereal Chem 72:287-290 Alnarp, Sweden.
Johansson E, Prieto-Linde ML, Jönsson J (2001) Effects of Malik AH, Prieto-Linde ML, Kuktaite R, Andersson A,
wheat cultivar and nitrogen application on storage Johansson E (2011) Individual and interactive effects
protein composition and bread-making quality. Cereal of cultivar maturation time, nitrogen regime and
Chem 78:19-25 temperature level on accumulation of wheat proteins. J
Sci Food Agric 91:2192-2200
Johansson E, Nilsson H, Mazhar H, Skerritt J, MacRitchie
F, Svensson G (2002) Seasonal effects on storage Malik AH, Kuktaite R, Johansson E (2013) Accumulation
proteins and gluten strength in four Swedish wheat of proteins in the wheat grain: The combined effect of
cultivars. J Sci Food Agric 82:1305-1311 genetic and environmental factors and their relation to
bread-making quality. J Cereal Sci (accepted).
Johansson E, Prieto-Linde ML, Svensson G, Jönsson J
(2003) Influences of cultivar, cultivation year and Payne PI, Nightingale MA, Krattiger AF, Holt LM (1987)
fertilizer rate on amount of protein groups and The relationship between HMW glutenin subunit
amount and size distribution of mono- and polymeric composition and the bread-making quality in British-
proteins. J Agric Sci Cambridge 140:275-284 grown wheat varieties. J Sci Food Agric 40:51-65
Johansson E, Prieto-Linde ML, Svensson G (2004) Pomeranz Y (1988) Composition and functionality of
Influence on nitrogen application rate and timing wheat flour components. In: Pomeranz Y (ed) Wheat:
on grain protein composition and gluten strength in chemistry and technology, vol II. AACC, St Paul, MN,
Swedish wheat cultivars. J Plant Nutr Soil Sci 167:345- pp219-370
350 Shewry PR, Miflin BJ (1985) Seed storage proteins of
Johansson E, Kuktaite R, Andersson A, Prieto-Linde economically important cereals. In: Pomeranz Y (ed)
ML (2005) Protein polymer built-up during wheat Wheat: chemistry and technology, vol II. AACC, St
development: influences of temperature and nitrogen Paul, MN, pp1-83
timing. J Sci Food Agric 85:473-479 Ullsten NH, Gällstedt M, Cho SW, Spencer G, Johansson
Johansson E, Prieto-Linde M-L, Gissén C (2008) Influences E, Hedenqvist MS (2009). Properties of extruded vital
of weather, cultivar and fertilizer rate on grain protein wheat gluten sheets with sodium hydroxide and
accumulation in field-grown wheat, and relations to salicylic acid. Biomacromol 10:479-488
grain water content and falling number. J Sci Food Ullsten NH, Gällstedt M, Spencer G, Johansson E, Marttila
Agric 88:2011-2018 S, Ignell R, Hedenqvist MS (2010) Extruded wheat
Kuktaite R, Plivelic TS, Cerenius Y, Hedenqvist MS, gluten as a barrier material. Polymer Renew Resourc
Gällstedt M, Marttila S, Ignell R, Popineau Y, 1:173-186
Tranquet O, Shewry PR, Johansson E (2011) Structure Wretfors C, Cho S-W, Kuktaite R, Hedenqvist MS, Marttila
and morphology of wheat gluten films: from S, Nimmermark S, Johansson E (2010) Effects of fiber
polymeric protein aggregates towards superstructure blending and diamines on wheat gluten materials
arrangements. Biomacromol 12:1438-1448 reinforced with hemp fiber. J Mater Sci 45:4196-4205

A model for the transcriptional regulation of low
molecular weight glutenin genes in wheat
A. Juhász1, Sz. Makai2, E. Sebestyén1, L. Tamás2 and E. Balázs1
1Applied Genomics Department, Agricultural Institute, Centre for Agricultural Research, HAS,
Martonvásár, Hungary; 2Department of Plant Physiology and Molecular Plant Biology, Eötvös Loránd
University, Budapest, Hungary
Abstract
Transcriptional regulation of LMW glutenin genes was investigated in-silico using publicly available
sequence and expression data. Genes were grouped into different low molecular weight (LMW)
glutenin types and their promoter profiles were determined using cis-acting regulatory element
databases and published results. Based on promoter sequences and expression characteristics, LMW
glutenin genes might be transcribed following two different mechanisms. Differences observed in their
expression could be related to differences in the promoter sequences. The 24 variants of 12 storage
protein-specific promoter motifs that were identified mainly belong to DOF, bZIP and Myb type
transcription factor recognition sites. The various elements belong to conserved non-coding regulatory
regions (CREs) and could act in different ways. There are regions in which regulatory elements, such
as the GCN4 motifs found in the -300 element region, could serve as key factors in tissue-specific
expression. Some other elements, like the further complete endosperm boxes in the -600 region or
the individual prolamin box variants might modulate the level of expression. Based on the relatively
small distance between different cis-acting promoter motifs, possible DOF-bZIP, DOF-Myb, bZIP-Myb
interactions can be identified.
Introduction start site in most cereal storage protein families

The storage proteins of wheat play a crucial role (Forde et al. 1985). This element, considered to
in the quality of flour-based end products. Their be primarily responsible for the endosperm-
allelic structure and expression largely determine specific expression, consists of two cis-elements,
the strength and extensibility of dough made the prolamin box (P-box) and the GCN4-motif (or
from wheat flour. N motif) (Hammond-Kosack et al. 1993). Based
on our current knowledge there are three main
Storage protein genes are specifically transcribed transcription factor families involved through
in developing wheat seeds, and their expression different levels of interactions in the endosperm-
is controlled primarily at the transcriptional level specific transactivation of the prolamin genes: the
(Sørensen et al. 1989). Synthesis and deposition bZIP, the DOF and the MYB transcription factors.
of seed storage proteins is subject to strict spatial
and temporal regulation, that occurs specifically The number and composition of cis-acting
and only in the endosperm during seed elements show some variation in different
development. Expression sensitively depends prolamin gene families. Sulphur rich prolamins
on the availability of nitrogen and sulphur (low molecular weight (LMW) glutenins,
in the grain (Duffus and Cochrane 1992) and gamma- and alpha-gliadins), as well as
shows similar profiles, indicating a remarkable sulphur-poor omega-gliadins, possess similar
conservation in the machinery responsible regulatory systems. The presence of the
for gene expression during seed maturation. bipartitate endosperm box at the -300 position
Common regulatory elements, such as cis-acting is characteristic of most of the prolamin gene
elements and trans-acting transcription factors families. Some prolamin genes contain additional
observed in monocot seed storage protein genes complete or partial endosperm boxes. The
were reviewed by Kawakatsu and Takaiwa number and position of these additional elements
(2010). The -300 element is conserved about shows some differences among the gene families.
300 nucleotides upstream of the transcriptional Studies related to the transcriptional regulation of

LMW glutenin genes have been published after #BSH, and #BSG). A blastn search with the LMW
the isolation of the LMW glutenin LMWG1D1 genes as queries against the EST sequences
gene (Colot et al. 1987). Based on similarities to was carried out and all hits were analyzed and
other cereal storage protein promoters the -300 judged individually. The blastn parameters were:
element containing both the prolamin box and -word_size: 128, -ungapped, -mismatch_penalty:
the GCN4-like motif has also been identified -30. All expression data were in a timeframe of 3
with an additional copy of each motif in close to 30 days after flowering. Expression data were
vicinity (Hammond-Kosack et al. 1993). These normalized against total EST hits found in the
two copies of endosperm boxes form a so-called single libraries and described as Transcript per
long endosperm box (LEB) (Albani et al. 1997). Million EST values (TPM).
Papers related to the promoter analysis of the
LMWD1D1 gene describe the mechanisms of Statistical analyses
transcriptional regulation of only one of the LMW glutenin sequences were characterized
LMW glutenin gene types. To better understand based on the cis-acting elements observed in
the roles of different LMW glutenin types in the the promoter regions. The presence or absence
machinery of gluten formation it is necessary to of promoter motifs was denoted with a 1 or 0
start by assessing the promoter characteristics of in a data matrix. Cluster analysis was used to
this gene family. In this study, the 5’ non-coding group promoters with similar profiles. A log-
regions of most of the LMW glutenin gene types likelihood ratio statistic (G-test) was carried out
were characterized. Expressional control of LMW to compare the abundance of transcripts with
glutenin genes were analyzed in-silico. Current the same promoter profile (Pavlidis and Noble
knowledge on cis-acting regulatory sequences 2003; Schaaf et al. 2005). Accessions possessing
and possible interacting transcription factors the same CRE patterns were clustered together.
attached to these LMW glutenin promoters were Cis-acting elements outside the CRE regions were
modeled. excluded from the analysis. Expression levels of
different CRE groups during seed development
were charted and significant differences were
Materials and methods calculated using the G tests.
Data collection and sequence characterization
Plant genomic sequence databases have been Results and discussion
screened for LMW glutenin genes from common
Twenty-four variants of 12 different SSP
wheat (Triticum aestivum L.). Only promoter
gene-specific cis-acting promoter motifs were
sequences over 100 nucleotides in length were
identified in the analyzed LMW glutenin gene
used for further analyses. Non-coding cis-
promoter dataset. Most of the cis-acting elements
acting elements were identified using published
are concentrated in the first 1,000 nucleotides
results and plant promoter databases (Higo et
upstream of the start codon. Three different
al. 1999; Lescot et al. 2002). Promoter profiles
variants of a GCN4 like motif and seven variants
were graphically represented using the Promoter
of prolamin boxes were identified in the LMW
Profiler tool developed by one of the authors.
glutenin gene promoters. Both P-box and GCN4
Accessions belonging to the same LMW glutenin
variants are present in different numbers and
gene type were aligned and conserved regulatory
positions in the different gene type promoters.
element regions (CREs) were identified manually.
Our results indicate that the duplicated
endosperm box, or LEB, motif is characteristic
Calculating gene expression based on EST
but not common in all members of the LMW
libraries
glutenin gene family (Table 1). It was present
Expression analyses of the LMW glutenin gene in gene types like IEN, MEN or MRC, whereas
types were carried out using EST sequences the remaining LMW GS gene types contain only
of developing wheat seed libraries of cultivars partial duplicates, missing either one or both of
Glenlea (NCBI EST library identifiers #A9H, the GCN4-like motif elements. However, further
#A9I, and #A9J), Chinese Spring (#10469, #19547, complete endosperm boxes were identified in
and #10479), Mercia (#FIN, #FIQ, and #FIT), and most of the gene types.
a genotype from DuPont (#BSD, #BSE, #BSF,

EB1, endosperm box at the position -300 is the m-type genes with cysteine in the fifth position,
first EB in the Long Endosperm Box; EB2, second like MRC) take a normal curve distribution in
endosperm box in the LEB; +, gene type contains expression profiles. Their transcriptional maxima
a complete LEB, -gene type does not contain a are reached at the mid-phase of grain filling.
complete LEB; no data, no sequence information Allelic effects resulted in significantly different
is available expression profiles of the same LMW GS gene
type, as in the case of MRC genes in Glenlea and
Expression levels of LMW glutenin gene types
Chinese Spring. In Glenlea, the MRC2 genes were
in the different genotype libraries varied due
strongly expressed, and the levels of expression
to the differences in their allelic compositions.
reached maxima at around 21 DPA. In cv.
Some gene types, like MSS1 or MEN were
Chinese Spring, gene type MRC1 was expressed,
underrepresented. Significant differences were
and MRC2 was present in only a negligible
identified at p = 0.05 among MRC and MSV type
amount.
genes in the Glenlea and Chinese Spring libraries.
Tendencies observed for the rest of the gene types The composition and distribution of cis-acting
indicate that LMW glutenin gene expressions elements in LMW glutenin gene promoters
follow two different profiles. A group of genes indicate that the position and order of some of
(i-type genes and most of the s-type genes) show the motifs might be more conserved compared
a continuously increasing level of transcription to other motifs. Three conserved non-coding
during development, whereas the other set (e.g., element regions were identified. The first
Table 1. Polymorphism observed in the endosperm boxes of LMW glutenin gene types.
LEB composition
LMW glutenin additional EB
type LEB -300 element (EB1) EB2
IQA - Pbox1 – GCN4 like 1 Pbox5 no data
IPP - Pbox1 – GCN4 like 1 Pbox5 Pbox3 – GCN4 like 3
IQP - Pbox1 – GCN4 like 1 Pbox5 no data
IEN + Pbox1 – GCN4 like 1 Pbox5 – GCN4 like 3 -
MEN + Pbox1 – GCN4 like 1 Pbox5 – GCN4 like 3 -
MSP1 - Pbox1 – Skn1 like Pbox 5 – GCN4 like 3 Pbox3 – GCN4 like 2-Skn1 like
MSP2 no data no data no data no data
MSP3 no data no data no data no data
MSG1 - Pbox1 – Skn1 like Pbox7 Pbox 7 – GCN4 like 3
MSG2 no data Pbox7 – Skn1 like
no data
MSH - Pbox1 – Skn1 like Pbox7
MRC1 + Pbox 1 – GCN4 like 1 Pbox 7 – GCN4 like 3 Pbox 4 – Skn1 like, Pbox3-
MRC2 + Pbox 1 – GCN4 like 1 Pbox7 – GCN4 like 3 GCN4 like 2-Skn1 like
MDS1 no data no data no data no data
MDS2 no data no data no data no data
MSC1 - Pbox1 – Skn1 like - no data
MSC2 no data no data no data no data
MSS1 no data no data no data no data
MSS2 - Pbox1 – Skn1 like Pbox5 – GCN4 like 3 no data
MSV - Pbox1 Pbox5 Pbox3 – GCN4 like 2
EB1, endosperm box at the position -300 is the first EB in the Long Endosperm Box; EB2, second endosperm box in the LEB; +, gene
type contains a complete LEB, -gene type does not contain a complete LEB; no data, no sequence information is available.

conserved regulatory region (CRE1) contains Based on these results CRE regions might have
motifs up to -120 nucleotides upstream of the different regulatory roles in transcription.
transcriptional site. This set of cis-acting elements Motif patterns in CRE1 and CRE3 resulted in
includes one or two TATA boxes, a CCAAT similar expressional profiles independent of the
motif, a MYB1AT motif, a GCAA motif2, a genotypes analyzed. Cis-acting elements in these
CAACAC motif and a P-box2 motif. The second regions are responsible for quantitative differences
group of elements contains motifs between observed among the different LMW glutenin
-300 and -420 nucleotides, and the third CRE gene types. Polymorphisms detected in the CRE2
group contains promoter motifs between -500 region did not have an impact on the observed
and -650. CRE2 included elements of the LEB differences in transcription level in the LMW
motif surrounded by several MYB core motifs genes during seed filling. This region might play
and some further elements. CRE3 included the a key role in the tissue-specific regulation and
region around position -550 containing another might have less effect on quantitative differences
complete endosperm box, some Skn-1 like motifs observed in transcript abundance.
and an SPA bZIP motif. Based on comparative
Based on the possible functions of the different cis-
analysis, accessions with the same motif pattern
acting elements and the potential TF interactions
in CRE1 followed a similar expressional pattern
the following model was proposed (Fig. 1). At
during seed development, not depending on the
the beginning of transcription, elements located
genotypes analyzed. If motifs of CRE2 region
in the first conserved non-coding region (CRE1)
were also considered, accessions with the same
might fulfill a basal promoter activity initiating
motif order did not share similar expression
transcription. At 10-13 days post-anthesis MYB
profiles in different genotypes. However, if
and DOF transcription factors might already be
cis-acting elements in CRE3 were considered
expressed and bound to their recognition sites
accessions belonging to the same CRE set showed
in LMW glutenin genes (CRE1 and CRE2). The
similar expression profiles in each genotype.
functional MYB-DOF TF complexes ensure a
Fig. 1 Model for expression regulation of LMW glutenin genes.

moderate level of transcription. This means that However, as this research is an in silico analysis,
at the earlier phases of seed filling, whereas we propose to use the results as guidelines, and
bZIP transcription factors are not yet expressed as a basis for future in vitro experiments. A more
to trigger endosperm-specific expression, DOF detailed analysis of these datasets was published
transcription factors bound to the prolamin by Juhász et al. (2011).
boxes might be responsible for trans-activation
of the storage protein promoters. This basic
level of expression might be moderated with References
the contribution of MYB transcription factors as Albani D, Hammond-Kosack MCU, Smith C, Conlan
suppressors in other vegetative tissues. These S, Colot V, et al. (1997) The wheat transcriptional
fundamental processes might be further fine- activator SPA: a seed-specific bZIP protein that
recognizes the GCN4-like motif in the bifactorial
tuned by the presence of additional elements,
endosperm box of prolamin genes. Plant Cell 9:171-184
such as CCAAT boxes. As later on, bZIP
Colot V, Robert LS, Kavanagh TA, Bevan MW, Thompson
transcription factors are bound to the GCN4- RD (1987) Localization of sequences in wheat
like motifs, Skn-1 like motifs and SPA bZIP endosperm protein genes which confer tissue-specific
recognition sites, the CRE2 region fulfills the expression in tobacco. EMBO J 6:3559-3564
tissue-specific regulator activity. This qualitative Duffus CM, Cochrane MP (1992) Grain structure and
function is accompanied by the quantitative composition. In: Shewry PR (ed) Barley: Genetics,
moderator effect of interacting bZIP-DOF and Biochemistry, Molecular Biology and Biotechnology.
bZIP-MYB TF complexes in the CRE3 region and CAB International, Wallingford, UK, pp291-317
gene expression reaches maximum level. Based Forde BG, Heyworth A, Pywell J, Kreis M (1985)
Nucleotide sequence of a B1 hordein gene and
on this model, the position of motifs located
the identification of possible upstream regulatory
in the first 1,000 nucleotides upstream of the elements in endosperm storage protein genes from
transcriptional start determines their primary barley, wheat and maize. Nucleic Acids Res 13:7327-
role as qualitative or quantitative regulators. 7339
Hammond-Kosack MCU, Holdsworth MJ, Bevan MW
(1993) In vivo footprinting of a low molecular weight
Conclusions glutenin gene (LMWG-1D1) in wheat endosperm.
EMBO J 12:545-554
The study presented here was carried out on a
Higo K, Ugawa Y, Iwamoto M, Korenaga T (1999)
statistically limited sample set, limited by the Plant cis-acting regulatory DNA elements (PLACE)
number of available promoter sequences. A total database. Nucleic Acids Res 27:297-300
of 164 accessions were used in the analyses. Some Juhász A, Makai Sz, Sebestyén E, Tamás L and E Balázs
gene types, such as MSS1, MEN or IQA, were (2011) Role of conserved non-coding regulatory
poorly represented. Using this available dataset a elements in LMW glutenin gene expression. PLoS
model for the transcriptional regulation of LMW ONE doi:10.1371/journal.pone.0029501
glutenin gene expression was developed and Kawakatsu T, Takaiwa F (2010) Cereal seed storage protein
confirmed with different statistical approaches. synthesis: fundamental processes for recombinant
protein production in cereal grains. Plant Biotech J
Our results indicate that LMW glutenin gene 8:939-953
expression follows two different patterns. One
Lescot M, Déhais P, Thijs G, Marchal K, Moreau Y, et
group of gene types, mostly belonging to m-type al. (2002) PlantCARE, a database of plant cis-acting
LMW glutenin genes follow a normal distribution regulatory elements and a portal to tools for in silico
expression pattern, starting to express at about analysis of promoter sequences. Nucleic Acids Res
the tenth day after anthesis. They reach their 30:325-327
maximum at around 21 DPA and the amount Pavlidis P, Noble WS (2003) Matrix2png: A utility for
of transcripts continuously decreases until full visualizing matrix data. Bioinformatics 19:295-296
maturity is reached. The rest of the genes, mainly Schaaf GJ, Ruijter JM, van Ruissen F, Zwijnenburg DA,
s- and i-types, show stable levels of transcription Waaijer R et al. (2005) Full transcriptome analysis of
rhabdomyosarcoma, normal, and fetal skeletal muscle:
after 21 DPA. Their promoter profiles and statistical comparison of multiple SAGE libraries.
especially the composition of their conserved FASEB J 19:404-406
regulatory element (CRE) regions also highlight Sørensen MB, Cameron-Mills V, Brandt A (1989)
some important characteristics, based on the Transcriptional and pos-transcriptional regulation of
differences observed in their promoter patterns gene expression in developing barley endosperm. Mol
which can be related to their expression profiles. & General Genet 217:195-201

Variation and classification of toxic epitopes related to
celiac disease among α-gliadin genes from four Aegilops
genomes
J. Li1,4, S.L. Wang1,4, S.S. Li1,4, P. Ge1, X.H. Li1, W.J. Ma2, F.J. Zeller3, S.L.K. Hsam3 and Y.M. Yan1
1Collegeof Life Science, Capital Normal University, Beijing 100048, China; 2Centre for Comparative Genomics,
Murdoch University and Western Australia Department of Agriculture & Food, Perth, WA 6150, Australia;
3Department of Plant Breeding, Technical University of Munich, D-85354 Freising-Weihenstephan, German
4These authors contributed equally to this work
Abstract
α-gliadins are associated with human celiac disease. A total of 23 non-interrupted full open α-gliadin
gene reading frames and 19 pseudogenes were cloned and sequenced from the C, M, N and U
genomes of 4 diploid Aegilops species. Sequence comparisons of α-gliadin genes from Aegilops and
Triticum species demonstrated extensive allelic variations in the Gli-2 loci of the 4 Aegilops genomes.
Specific structural differences included the compositions and variations of 2 polyglutamine domains
(QI and QII) and 4 T cell stimulatory toxic epitopes. The average numbers of glutamine residues in the
QI domain in the C and N genomes and the QII domain in the C, N and U genomes were much higher
than those in Triticum genomes, and the QI domain in C and N genomes. The QII domain in the C, M,
N and U genomes displayed greater length variations. Interestingly, the types and numbers of 4 T cell
stimulatory toxic epitopes in α-gliadins from the 4 Aegilops genomes were significantly less than those
in common wheat and its progenitors. Study of the relationships among the structural variations of
the two polyglutamine domains and distributions of 4 T cell stimulatory toxic epitopes resulted in the
α-gliadin genes from Aegilops and Triticum genomes being classified into three groups.
Introduction et al. 1984) are members of a multigene family

Wheat flour has the unique ability to form (Gu et al. 2004) with estimated gene copies
dough for making a wide range of food ranging from 25 to 35 to 100 or even up to 150
products. The processing quality of wheat (Harberd et al. 1985; Anderson and Greene 1997)
flour primarily depends on the structure and in individual haploid genomes. Such variations
properties of gluten. Gliadins are the major were presumably caused by duplication and
components of wheat storage proteins; they deletion of chromosome segments (D’Ovidio et
have a close relationship with wheat quality due al. 1991). However, almost 50% of the α-gliadin
to high polymorphism and heritable stability genes are inactive, or are pseudogenes, and
(Metakovsky et al. 1997). They are classified the high base substitution rate, especially CT
into α/β-, γ-, and ω-gliadins based on their substitution, contributes to the presence of stop
electrophoretic mobilities during polyacrylamide codons (Anderson and Greene 1997).
gel electrophoresis at low pH (Shewry et al. The primary structures of α-gliadin proteins
2003). Among them, α-gliadins are the most comprise a 20 amino-acid-residue short signal
abundant, accounting for 15–30% of wheat seed peptide and 5 distinct domains: a repetitive
proteins. Hence, for humans they are the most domain, a polyglutamine domain I (QI), a unique
consumed storage proteins (Gu et al. 2004; Van domain I (UI), a polyglutamine domain II (QII),
Herpen et al. 2006; Chen et al. 2008). and a unique domain II (UII) (Anderson and
The genetic loci encoding gliadins are located Greene 1997). Six conserved cysteine residues of
on the short arms of homologous group 1 (Gli-1) α-gliadins form three intramolecular disulfide
and 6 (Gli-2) chromosomes (Shewry et al. 2003; bonds (Miller and Wieser 1995). The average
Chen et al. 2008). α-gliadins encoded by the Gli-2 numbers and lengths of glutamine (Q) residues
loci on chromosomes 6A, 6B, and 6D (Anderson in two polyglutamine domains among different

α-gliadin genes present remarkable differences b). The extensive allelic variation of α-gliadin
and nearly all α-gliadin genes can be assigned to genes in Triticum and Aegilops species provide
specific genomes by using the repeat lengths of valuable resources to study the origin and
glutamine residues (Van Herpen et al. 2006). evolution of the wheat genomes, as well as a
potential gene reservoir for quality improvement.
Celiac disease (CD) is an intestinal T cell-
It is generally accepted that common wheat
mediated disease caused by an uncontrolled
was derived from hybridization between an
immune response to gluten, which contains
allotetraploid Triticum species (AABB) and the
several peptides that constitute the main toxic
diploid Aegilops tauschii (DD). Triticum urartu was
components in CD (Van de Wal et al. 1998;
shown to be the donor of the A genome, whereas
Koning 2003). The α-gliadins are considered
Aegilops speltoides probably contributed the B
to be the most important gluten component
genome (Petersen et al. 2006). Although extensive
leading to CD, which affects as many as 1 in
investigations have been performed, the origin of
300 people (Arentz-Hansen et al. 2000; Vader
the B genome remains controversial (Huang et al.
et al. 2003; Van Herpen et al. 2006; Vaccino et
2002). Although considerable work has focused on
al. 2009). HLAs are a large family and play
the origin and evolution of Triticum and Aegilops
important roles in producing the inflammatory
species, the allelic variations and molecular
response in celiac disease (CD). Among them,
characterization of toxic epitopes related to CD
only the HLA-DQ2.5 and HLA-DQ8 molecules
among the α-gliadin genes from the C, M, N and
in α-gliadins have the antigen-presenting cells
U genomes of Aegilops species are not clear. In the
to bond gluten-reactive T cells upon recognition
current study, we separated 23 α-gliadin genes
of gluten-derived peptides (Van Herpen et
from four Aegilops genomes and investigated
al. 2006; Mitea at el. 2010; Sollid et al. 2012).
the allelic variations of 4 T cell stimulatory toxic
Remarkably, the deamidated toxic epitopes are
epitopes (glia-α, glia-α2, glia-α9 and glia-α20) and
deemed to increase the bonding affinity to DQ2.5
2 polyglutamine domains (QI and QII) among the
and DQ8, when specific glutamine residues in
α-gliadin genes. Comparative and phylogenetic
epitopes sequence are converted to glutamate by
analysis among the Aegilops and Triticum genomes
the enzyme transglutaminase 2 (TG2) (Arentz-
were also conducted.
Hansen et al. 2000; Koning 2003; Camarca et
al. 2009). This means that the gluten peptides
with deamidated glutamate residues can also
affect the innate immunity system (Mitea at
Materials and methods
el. 2010; Sollid et al. 2012). It is assumed that Plant materials
a single amino acid substitution or deletion
Twelve accessions from 4 Aegilops species were
could be sufficient to diminish or abolish the T
used in this study, including Aegilops comosa
cell stimulatory capacity. But Sollid et al. (2012)
(MM) accession PI 551020, Aegilops umbellulata
list specific modifications that can still induce
(UU) PI 298906, PI 542364 and PI 573516, Aegilops
T cell stimulation. In addition, the quantity
markgrafii (CC) PI 573416, PI 551119, PI 298889 and
and distribution of toxic epitopes can be used
PI 564196), and Aegilops uniaristata (NN) PI 276996,
to assign α-gliadins to specific chromosome
PI 276996, PI 554420 and PI 554418. The accessions
loci (Vader et al. 2003; Van Herpen et al. 2006).
were kindly provided by CIMMYT, Mexico.
For example, the products of α-gliadin genes
on chromosome 6A of bread wheat almost
PCR amplification, molecular cloning and
invariably contain epitopes glia-α9 and glia-α20,
sequencing
but not intact epitopes glia-α or glia-α2. The
products of genes on chromosome 6B do not Protocols of genomic DNA extraction
contain any of the 4 T cell epitopes, whereas the from dry seeds and PCR were based on
products of genes on chromosome 6D contain all An et al. (2006). Three pairs of allele-
of the T cell epitopes in variable combinations specific PCR (AS-PCR) primers, viz. α1:
(Van Herpen et al. 2006). 5’-ATGAAGACCTTTCTCATCCTTG-3’ and
α2: 5’-TCAGTTRGTACCRAAGATG CC-3’, α3:
Aegilops, which is closely related to Triticum, 5’-GGTCAATACAAATCCATCATGAAG-3’ and
comprises 22 species and several distinct α4: 5’-TCATCGATAG TTAGTACCGAAG-3’,
genomes: C, M, N, S, T and U (Wang et al. 2011a, and α5: 5’-ATGAAGACCTTTCTCATCC-3’ and

α6: 5’-GTTAGT ACCGAAGATGCC-3’, were or CAA) was changed to a stop codon (TAG or
designed to amplify the conserved sequence TAA). The 42 separated genes were deposited in
of α-gliadin genes. The PCR products were Genzk with accession numbers from JN555615
analyzed by 1% agarose gel electrophoresis in to JN555656. To perform a comparative analysis
Tris-acetic acid-EDTA buffer. The separated of both full-ORF genes and pseudogenes, 299
fragments were purified from the gels by the α-gliadin genes from common wheat and Aegilops
Omega Gel Extraction Kit (Omega) and ligated species were selected from the GenBank database
into a pGEM-T Easy vector (Tiangen, Beijing, and Zhu et al. (2010) to pool with the 42 genes
China), then transferred into competent cells of isolated from this work (Table 1).
Escherchia coli strain DH-5α. DNA sequencing
Among the 42 genes, 3 typical genes from each of
from three clones was carried out by Sangon
the C, M, N and U genomes were selected to align
Biotech Co. Ltd., Shanghai, China.
with 6 other α-gliadin genes (Table 1) isolated
from Aegilops spp. or common wheat. According
Sequence comparison and toxic epitope
to the multiple alignments of the deduced amino
identification
acid sequences (Fig. 1), the deduced protein
Multiple alignments and analysis of complete sequences of 12 novel genes from the M, U, C,
α-gliadin nucleotide sequences as well as the and N genomes shared typical structural features
deduced amino acid sequences were carried out of previously characterized α-gliadin genes.
using BioEdit 7.0 (Ibis Therapeutics, Carlsbad, The lengths of the amino acid sequences had a
California). Identification of 4 T cell stimulatory few differences from each other mainly due to
toxic epitopes, glia-α (QGSFQPSQQ), glia-α2 variation in the glutamine number in the two
(PQPQLPYPQ), glia-α9 (PFPQPQLPY), and polyglutamine domains. The signal peptides and
glia-α20 (FRPQQPYPQ), and 2 polyglutamine repetitive domains were relatively conserved in
domains (QI and QII) among the α-gliadin genes the full sequences of the α-gliadin genes and all
was carried out according to the procedures α-gliadin subunits comprised 6 highly conserved
outlined by Cornell and Wills-Johnson (2001) cysteine residues, except for JN555635 that had
and Xie et al. (2010). Identification and analysis an arginine residue (R) instead of the last cysteine
of variations in insertions/deletions (InDels) and residue (C) in the unique II domains. Some
single nucleotide polymorphisms (SNPs) among SNPs and InDels were present in the N-terminal
the α-gliadin genes were based on the protocol repetitive domains. For instance, a deletion of a 29
described by Wang et al. (2011b). amino acid fragment occurred in JN555627, and
an extra insertion of a PFPPQ motif was present in
Phylogenetic analysis JN555632, EF561279, and EF561282.
The neighbor-joining tree of different genes or
Polyglutamine domain I (QI) and polyglutamine
genomes was analyzed using MEGA4 (Kumar et
domain II (QII) are the main characteristic
al. 2004; An et al. 2006). According to the full-
sequences of α-gliadin primary structure.
ORF of gliadin genes, the neighbor-joining tree
Variations in these two polyglutamine domains
was constructed with the evolutionary rate set to
were found between different genomes. For
6.5×10-9 (Allaby et al. 1999; Kumar et al. 2004).
example, the QI domain in many genes from the
C, M and N genomes displayed a PISQQQA/
VQQQ motif segmental duplication or triplication,
Results whereas the A, B, D, and U genomes had no such
Cloning and molecular characterization of repeats. At the same time double L residues in
α-gliadin genes from the C, M, N and U the QII domain were present in some M genome
genomes of Aegilops species genes, and some H repeats were present in the B,
D and U genomes. In addition, the great majority
A total of 42 complete α-gliadin coding sequences
of genes from the A and B genomes had a similar
were isolated from four diploid Aegilops species,
feature, including a point mutation (QA) in
of which 23 had full open reading frames (ORF)
the fourth amino acid position of the QI domain
and 19 were pseudogenes with at least one stop
and a QK mutation in the QII domain. In the
codon. Most of these pseudogenes resulted from
B genome genes, a stop codon often appeared in
C to T substitutions, i.e., a glutamine codon (CAG
front of the K substitution.

The average numbers of Q residues in the QI Identification and variation in specific toxic
and QII domains among seven genomes were epitopes related to celiac disease
calculated. As shown in Fig. 2, the average Within the genes isolated in the current study,
number across the different genomes was four previously reported (Cornell and Wills-
greater than 15 in the first repeat. Among them, Johnson 2001; Xie et al. 2010) celiac disease
the N (21.1±10.7), C (20.3±7.1), and A (20.8±3.3) (CD) toxic epitopes, viz. glia-α (QGSFQPSQQ),
genomes had more than 20 Q residues. In the glia-α2 (PQPQLPYPQ), glia-α9 (PFPQPQLPY),
second repeat four genomes had more than 15 Q and glia-α20 (FRPQQPYPQ) were positionally
residues, including U (19.9±11.2), C (19.4±16.2), conserved in most of the α-gliadin subunits as
N (19.7±9.8), and B (15.2±10.8). Two genomes, shown in Fig. 1. In particular, glia-α located at
A (9.7±4.0) and D (9.2±2.7), had less than 10 Q the unique domain II and glia-α2, glia-α9 and
residues. The N genome had the highest average glia-α20 were all present in the repetitive domain.
numbers of Q residues in both the QI and QII Toxicity epitopes of celiac disease were shown to
domains. Generally, except for the U genome, the be related to the motifs of PSQQ and QQQP in
average numbers of Q residues in QI were greater the N-terminal repetitive domains (Cornell and
than those in QII. A large variation in Q residue Wills-Johnson 2001). Sequence alignments of the
numbers was observed in the QI domain of the four CD toxic epitopes revealed several InDels
C, D and N genomes and in the QII domain of B, and SNPs among the 42 genes. For example, there
C, M, N and U genomes, indicating substantial were two genes (JN555619 and JN555620) from
length variation of the QI and QII domains the M genome with a deletion of the PYPQ motif
among different genomes. in the glia-α20 (FRPQQPYPQ) epitope while
Table 1. Summary of the 341 α-gliadin genes from different genomes of Aegilops and Triticum spp. used
in this study.
Full-ORF (GenBank Pseudogenes (GenBank

Species Genome accession number) accession number) Origin Total
Aegilops comosa M 7 (JN555615 - JN555621) 1 (JN555656) This paper 8
Aegilops umbellulata U 6 (JN555622 - JN555627) 5 (JN555643 - JN555647) This paper 11
Aegilops markgrafii C 4 (JN555628 - JN555631) 8 (JN555648 - JN555655) This paper 12
Aegilops uniaristata N 6 (JN555632 - JN555637) 5 (JN555638 - JN555642) This paper 11
Triticum monococcum Am 15 (DQ002569 - DQ002583) 39 (DQ002600 - DQ002638) Van Herpen et al. 2006 60
2 (DQ401698, DQ401700) 4 (DQ401699, DQ401701 Ma et al. 2007
- DQ401703)
Triticum uratu Au 1 (DQ401696) 1 (DQ401697) Ma et al. 2007 4
1 (M16496) 1 (M16497) Reeves and Okita.1987
Aegilops speltoides Ss 5 (DQ002584 - DQ002588) 32 (DQ002639 - DQ002670) Van Herpen et al. 2006 37
Aegilops longissima Sl 0 66 (DQ002671 - DQ002736) Van Herpen et al. 2006 66
Aegilops tauschii Dt 11 (DQ002589 - DQ002599) 62 (DQ002737 - DQ002798) Van Herpen et al. 2006 115
14 (HM188546 - HM188559) 5 (HM188560 - HM188564) Li et al. 2010
4 (GQ131527, GQ131528, 7 (GQ131529 - GQ1315331,
GQ131532, GQ131533) GQ131534 - GQ131537) Zou et al. 2009
5 (EF218656 - EF218659) 1 (EF218660) Qi et al. 2012
Triticum aestivum L. A 2 (EF561275, EF561283) 0 Xie et al. 2010 2
B 5 (EF561274, EF561279, 0 Xie et al. 2010 5
EF561282, EF561284, EF561288)
D 8 (EF561276-EF561278, EF561280, 0 Xie et al. 2010 10
EF561281, EF561285 - EF561287)
2 Zhu et al. 2010

56
Proceedings 11th InternaƟonal Gluten Workshop
Fig. 1 Multiple alignments of deduced amino acid sequences of 20 α-gliadin genes from the C, M, N and U genomes of 4 diploid Aegilops and Triticum species.
Cysteine residues are shaded in gray, deletions are indicated by dashes and the positions of four celiac disease (CD) toxic epitopes are marked by boxes. The
typical structural features, including polyglutamine domains QI and QII and 4 T cell stimulatory toxic epitopes are indicated.
another two genes (JN555632 and JN555636) from ancestral genomes were relatively more than those
the N genome had an insertion of the PFPPQ from the C, M and N genomes. Specifically, the
motif at the end of the sixth amino acid of the α-gliadin genes among the A and related genomes
glia-α9 epitope. included large numbers of glia-α9 and glia-α20, and
less glia-α. In B and D and their ancestral genomes,
Most of the full ORF genes with a gliα-α2 epitope
all 4 types of toxic epitopes can be present; the B, Sl
from the C, M and N genomes were disrupted
and Ss genomes had fewer glia-α and glia-α2.
by a mutation of the fourth amino acid (LQ)
in this epitope. In addition, the average numbers
Relationships between the four toxic epitopes and
of 4 T cell stimulatory toxic epitopes in the full
two polyglutamine domains
ORF α-gliadins genes and pseudogenes from
all studied genomes were identified (Table 2). According to the amino acid sequence variations
Apparently, the α-gliadin genes of the C, M and and conserved positions of the two polyglutamine
N genomes contained only 0-1 glia-α epitope domains and four CD toxic epitopes, the 341
whereas the genes of the U genome had no α-gliadin genes including pseudogenes (Table 1)
glia-α20. In comparison, the numbers and types could be divided into three groups. Structural
of toxic epitopes in the A, B, and D and their sketches of the groups are shown in Fig. 3.
Ninety-nine genes in group I had a QA point
mutation at the fourth amino acid residue in the
40
Q1 QI domain. In general, this group only had a
35 Q2 glia-α epitope without the other 3 toxic epitopes.
30
Number of Q-residues
Table 2. The mean number of each toxicity epitope

25
in both full-ORF α-gliadin and pseudogenes
20 among Aegilops and Triticum genomes.
15 Total
number
10 Genome of genes Gli- Gli-2 Gli-9 Gli-20
5 M 8 0.63 0 0 0
U 11 0.91 0.45 0.64 0
0
M U C N A/Am/Au B/Ss/Sl D/Dt C 12 0.33 0 0 0
N 11 0.45 0 0 0
Genome
A/Am/Au 66 0.02 0 0.85 0.88
Fig. 2 Variations in average numbers of glutamine B/Sl/Ss 108 0.18 0.03 0.48 0.50
residues in the two polyglutamine domains of D/Dt 123 0.76 0.63 0.63 0.82
α-gliadins in Aegilops and Triticum genomes.
Fig. 3 Sketches of α-gliadin genes from three groups. S, signal peptide; R, repetitive domain; QI,
polyglutamine domain I; U1, unique domain I; QII, polyglutamine domain II; U2, unique domain II; *,
stop codon; , position of point mutation in the two polyglutamine domains; dashed box, segment might
occur as an insertion or a deletion.

The absence of glia-α9 and glia-α2 epitopes was Discussion
due to a fragment containing P/LFPQPQPFP/
LP/SQLPYPQ or PFPP/SQLPYPQ/* instead of Molecular characterization and phylogenetic
glia-α9 (PFPQPQLPY) and glia-α2 (PQPQLPYPQ) evolution of α-gliadin genes among 4 Aegilops
sequences. The second amino acid residue R genomes
(arginine) of the glia-α20 epitope was changed With extensive allelic variation, the wheat
to a S (serine) or P (proline) residue, resulting gliadin genes are closely related to bread-making
in the absence of glia-α20. Group II included quality attributes, mainly imparting extensibility
125 genes, among which the typical feature was to dough (Metakovsky et al. 1997). A single
that a Q was changed to a K at the fourth or exception was the decrease in GMP volume in
fifth amino acid residue in the QII domain. The Chinese Spring deletion line 6DS-2 that resulted
glia-α9 and glia-α20 epitopes were present in in decreased elasticity rather than decreased
this group whereas both the glia-α2 and glia-α dough strength (Van den Broeck et al. 2009).
epitopes were absent because the eighth amino The number of α-gliadin genes was estimated to
acid S (serine) in glia-α2 and the fifth amino range from 25 to 150 copies in different wheat
acid Q (glutamine) in glia-α were changed to P genotypes. However, nearly half of the α-gliadin
(proline) and R (arginine), respectively. Group genes from the A, B and D genomes were
III had 117 genes, including most genes from the pseudogenes, mainly resulting from changes of C
D genome, but no A or K point mutation at the to T (Anderson and Greene 1997). In the present
position corresponding to those in the genes in study, we isolated 42 α-gliadins genes from the
groups I and II. In this group, almost all genes C, M, N and U genomes in 4 diploid Aegilops
generally had 3–4 epitopes, but some genes from species, of which 19 were pseudogenes with
the M genome, such as JN555619 and JN555620, similar frequency and generation mechanism
contained only the glia-α epitope. Some genes to those in hexaploid wheats. This suggests that
from the D genome had two glia-α2 epitopes there exists a high frequency of pseudogenes
caused by an insertion of a PQLPYPQ motif after in both Triticum and related Aegilops species.
the glia-α2 epitope sequence. In the evolutionary process, a high-frequency
appearance of pseudogenes may be necessary to
Phylogenetic relationships within the α-gliadin restrain the gene redundancy by gene silencing
gene family from the Aegilops and Triticum or other variations (Galili 1989). However, a
genomes recent study revealed that some prolamin gene
To further investigate evolutionary relationships copies with premature stop codons are still
among the α-gliadin genes from various genomes, transcribed; this appeared to be controlled at the
31 complete coding sequences were used to transcriptional and/or post-transcriptional levels
construct a neighbor-joining tree, including 23 (Xu and Messing 2009).
new genes isolated in the current study, 6 genes The current study demonstrated that α-gliadins
(Table 1) previously cloned from common wheat genes from the C, M, N and U genomes in
and several Aegilops species, one γ-gliadin gene diploid species Ae. comosa, Ae. uniaristata,
(HQ327440) and one ω-gliadin gene (AY667097). Ae. markgrafii and Ae. umbellulata had some
As shown in Fig. 4, except for two separate genome-specific structural features, particularly
one-gene branches, one being γ-gliadin and in the compositions and variations in two
the other being ω-gliadin, the phylogenetic tree polyglutamine domains and four toxic epitopes
was clustered into two main subgroups. Within related to CD. The two polyglutamine domains
the first group, genes from the C, M, N and QI and QII encoded by microsatellite-like
U genomes were clustered together and were sequences are typical regions in α-gliadins, and
slightly more distant from those from A, B and D their numbers of amino acid residues range from
genomes of common wheat. This suggests that the 5 to 107 (Anderson and Greene 1997). A high
α-gliadin sequences obtained from these genomes proportion of glutamine residues in the QI and
are more similar to each other than to those from QII domains might be helpful for improving
the A, B and D genomes. In the other subgroup, visco-elasticity properties of wheat dough
M and U genomes displayed close phylogenetic through inter-molecular interactions of large
relationships to the D genome. numbers of glutamine side chains that act as
hydrogen bond donors and acceptors (Masci et

al. 2000). As shown in Fig. 2, the average numbers closely related and could have a common
in the QI domains of the C, N, and A genomes, ancestor, or that one was derived from the other
and the QII domains in C, N and U are much (Tanaka 1985). Previous study also suggested that
higher than those in other genomes. Concurrently the D genome of Ae. tauschii was closely related
there are highly significant length variations to the M and N genomes (Wang et al. 1997).
of the QI domain in the C and N genomes and From analyses of Glu-3 loci the C and U genomes
the QII domain in the C, M, N and U genomes, were recently found to be closely related to the
suggesting that the four Aegilops genomes A genome whereas the M and N genomes had
generally possessed more glutamine residues and closer relationships with the D genome (Wang
higher length variations in the two polyglutamine et al. 2011b). It was also reported that the C and
domains. Thus it may be worthwhile to attempt U genomes were closely related to each other
to use new gene sources from Aegilops genomes and both the C and N genomes had a close
for wheat quality improvement. relationship with the A genome (Vanichanon et
al. 2003). By study of the Gli-2 loci our results
The C, M, N and U genomes constitute the
showed that the four diploid Aegilops genomes
basic genomes in Aegilops species, from which
were closely related (Fig. 4). In particular, several
polyploid Aegilops species were derived by
genes of the M and U genomes were more similar
hybridization (Dvořák et al. 2006). Previous
to genes in the D genomes than to those in the A
research revealed that the diploid species Ae.
and B genomes.
comosa (MM) and Ae. uniaristata (NN) were
98 JN555628 Aegilops markgrafii (C)

67
JN55630 Aegilops markgrafii (C)
JN555631 Aegilops markgrafii (C)
57 JN55629 Aegilops markgrafii (C)
JN555635 Aegilops uniaristata (N)
100
95 49 JN555633 Aegilops uniaristata (N)
77 JN55634 Aegilops uniaristata (N)
99 JN55618 Aegilops comosa (M)
JN555621 Aegilops comosa (M)
95
JN55616 Aegilops comosa (M)
99
100 JN55636 Aegilops uniaristata (N)
JN555624 Aegilops unbellulata (U)
100 99 DQ401696 Triticum urartu (Au)
99 EF561283 Triticum aestivum (A)
46 EF561282 Triticum aestivun (B)
EF561279 Triticum aestivum (B)
96 67 JN555620 Aegilops comosa (M)
EF561276 Triticum aestivum (D)
99 HM188558 Aegilops tauschii (D)
90 JN555627 Aegilops umbellulata (U)
JN555623 Aegilops umbellulata (U)
100
JN555622 Aegilops umbellulata (U)
68
HQ327440 g-Gliadin gene
AY667097 w-Gliadin gene
30,000,000 20,000,000 10,000,000 0
0.20 0.15 0.10 0.05 0.00
Fig. 4 Neighbor-joining tree of 23 cloned α-gliadin genes and 8 other genes from common wheat and
several Aegilops species. The suffixes of GenBank accession numbers indicate different types of genes.

Toxic epitope variation and its application for References
wheat quality improvement Allaby RG, Banerjee M, Brown TA (1999) Evolution of the
Toxic epitopes present in the α-gliadin genes high molecular weight glutenin loci of the A, B, D, and
G genomes of wheat. Genome 42:296-307
appeared to be genetically diverse (Vader et al.
An X, Zhang Q, Yan Y, Li Q, Zhang Y, Wang A, Pei Y,
2003; Van Herpen et al. 2006). Four antigenic
Tian J, Wang H, Hsam SLK, Zeller FJ (2006) Cloning
peptides derived from α-gliadins are known and molecular characterization of three novel LMW-i
to be involved in CD, namely the HLA-DQ2.5 glutenin subunit genes from cultivated einkorn
epitopes Gli-α2, Gli-α9 and Gli-α20, and the (Triticum monococcum L.). Theor Appl Genet 113:383-395
HLADQ8 T-cell epitope Gli-α (Mitea at el. 2010; Anderson OD, Greene FC (1997) The α-gliadin gene
Sollid et al. 2012). In this study, for the first time, family. II. DNA and protein sequence variation,
we identified the type and quantity of the four subfamily structure, and origins of pseudogenes.
Theor Appl Genet 95:59-65
toxic epitopes in the C, M, N and U genomes
(Fig. 1 and Table 2). Considerable variation in the Anderson OD, Litts JC, Gautier MF, Greene FC (1984)
Nucleic acid sequence and chromosome assignment
composition of the four CD-related toxic epitopes of a wheat storage protein gene. Nucleic Acids Res
was found among different Aegilops and Triticum 12:8129-8144
species. Interestingly, the types and numbers Arentz-Hansen EH, McAdam SN, Molberg O, Kristiansen
of toxic epitopes among α-gliadins from the C, C, Sollid LM (2000) Production of a panel of
M, N and U genomes are less than those from recombinant gliadins for the characterisation of T cell
the A, B, D and their progenitor genomes. As reactivity in coeliac disease. Gastroenterology 46:46-51
shown in Table 2, C, M and N genomes have no Camarca A, Anderson RP, Mamone G, Fierro O, Facchiano
Gli-α2, Gli-α9 or Gli-α20, while Gli-α20 is absent A, Costantini S, Zanzi D, Sidney J, Auricchio S,
Sette A, Troncone R, Gianfrani C (2009) Intestinal
in the U genome. This suggests that α-gliadin
T cell responses to gluten peptides are largely
subunits in Aegilops have no, or a low, possibility heterogeneous: implications for a peptide-based
of stimulating CD4 T cells, possibly providing a therapy in celiac disease. J Immunol 182:4158–4166
valuable resource for decreasing and preventing Chen F, Xu C, Chen M, Wang Y, Xia G (2008) A new
various CDs. alpha-gliadin gene family for wheat breeding: somatic
introgression line II-12 derived from Triticum aestivum
Important relationships between the 4 T and Agropyron elongatum. Mol Breeding 22:675-685
cell stimulatory toxic epitopes and two
Cornell HJ, Wills-Johnson G (2001). Structure-activity
polyglutamine domain variations were identified relationships in coeliac-toxic gliadin peptides. Amino
in the current study where we were able to Acid 21:243-253
classify α-gliadins into three groups. Group D’Ovidio R, Lafiandra D, Tanzarella OA, Anderson
I had only one toxic epitope whereas groups OD, Greene FC (1991) Molecular characterization
II and III generally contained 2 and 3–4 toxic of bread wheat mutants lacking the entire cluster of
epitopes, respectively. Furthermore, most of chromosome 6A controlled gliadin components. J
Cereal Sci 14:125-129
the α-gliadins from the C, M and N genomes
Dvořák J, Akhunov ED, Akhunov AR, Deal KR, Luo MC
were located in group I, indicating fewer toxic
(2006) Molecular characterization of a diagnostic DNA
epitopes are present in these genomes. These marker for domesticated tetraploid wheat provides
results could inevitably lead to a lowering of the evidence for gene flow from wild tetraploid wheat to
toxicity level of wheat flour in wheat breeding hexaploid wheat. Mol Biol Evol 23:1386-1396
program. Specific SNPs and InDels identified Galili G (1989) Heterologous expression of a wheat high
in the 4 T cell stimulatory toxic epitopes and molecular weight glutenin gene in Escherichia coli. Proc
the two polyglutamine domains could be used Natl Acad Sci USA 86:7756-7760
as effective molecular markers for screening Gu YQ, Crossman C, Kong XY, Luo MC, You FM,
α-gliadin genes with no toxic epitopes during the Coleman-Derr D, Dubcovsky J, Anderson OD (2004)
Genomic organization of the complex alpha-gliadin
transfer of valuable genes from Aegilops species to
gene loci in wheat. Theor Appl Genet 109:648-657
bread wheat.
Harberd NP, Bartels D, Thompson RD (1985) Analysis
of the gliadin multigene loci in bread wheat using
nullisomic-tetrasomic lines. Mol Genet Genomics
198:234-242

Huang S, Sirikhachornkit A, Su X, Faris J, Gill B, Vaccino P, Becker H-A, Brandolini A, Salamini F, Kilian
Haselkorn R, Gornicki P (2002) Genes encoding plastid BA (2009) Catalogue of Triticum monococcum genes
acetyl-CoA carboxylase and 3-phosphoglycerate encoding toxic and immunogenic peptides for celiac
kinase of the Triticum–Aegilops complex and the disease patients. Mol Genet Genomics 281:289-300
evolutionary history of polyploid wheat. Proc Natl Vader W, Stepniak D, Kooy Y, Mearin L, Thompson A,
Acad Sci USA 99:8133-8138 van Rood JJ, Spaenij L, Koning F (2003) The HLA-DQ2
Koning F (2003) The molecular basis of celiac disease. J gene dose effect in celiac disease is directly related to
Mol Recogni 16:333-336 the magnitude and breadth of gluten-specific T cell
Kumar S, Tamura K, Nei M (2004) MEGA3: integrated responses. Proc Natl Acad Sci USA 100:12390-12395
software for molecular evolutionary genetics analysis Van de Wal Y, Kooy Y, Van Veelen P, Pena S, Mearin
and sequence alignment. Brief Bioinform 5:150-163 L, Molberg O, Lundin L, Mutis T, Benckhuijsen W,
Li G, Zhang T, Wei P, Jia J, Yang Z (2010) Sequence Drijfhout JW, Koning F (1998) Small intestinal T
analysis of α-gliadin genes from Aegilops tauschii cells of celiac disease patients recognize a natural
native to China. Asian J Agric Sci 2:128-135 pepsin fragment of gliadin. Proc Natl Acad Sci USA
95:10050-10054
Ma ZC, Wei YM, Yan ZH, Zheng YL (2007)
Characterization of α-gliadin genes from diploid Van den Broeck H, Van Herpen T, Schuit C, Salentijn E,
wheats and the comparative analysis with those from Dekking L, Bosch D, Hamer R, Smulders M, Gilissen L,
polyploid wheats. Genetika 43:1534-1541 Van der Meer I (2009) Removing celiac disease-related
gluten proteins from bread wheat while retaining
Masci S, D’Ovidio R, Lafiandra D, Kasarda DD (2000)
technological properties: a study with Chinese Spring
A 1B coded low-molecular-weight glutenin subunit
deletion lines. BMC Plant Biol 9:41
associated with quality in durum wheats show strong
similarity to subunits present in some bread wheat Van Herpen T, Goryunova SV, Van der Schoot J, Mitreva
cultivars. Theor Appl Genet 100:396-400 M, Salentijn E, Vorst O, Schenk MF, Van Veelen
PA, Koning F, Van Soest LJM, Vosman B, Bosch D,
Metakovsky EV, Annicchiarico P, Boggini G, Pogna NE
Hamer RJ, Gilissen L, Smulders MJM (2006) Alpha-
(1997) Relationship between gliadin alleles and dough
gliadin genes from the A, B, and D genomes of wheat
strength in Italian bread wheat cultivars. J Cereal Sci
contain different sets of celiac disease epitopes. BMC
25:229-236
Genomics 7:1-13
Miller S, Wieser H (1995) The location of disulphide bonds
Vanichanon A, Blake NK, Sherman JD, Talbert LE (2003)
in a-type gliadins. J Cereal Sci 22:2127
Multiple origins of allopolyploid Aegilops triuncialis.
Mitea C, Salentijn E, Van Veelen P, Goryunova S, Van Theor Appl Genet 106:804-810
der Meer I, Van den Brock H, Mujico J, Monserrat V,
Wang G, Miyashita NT, Tsunewaki K (1997) Plasmon
Gilissen L, Drijfhout, Dekking L, Koning F, Smulders
analyses of Triticum (wheat) and Aegilops: PCR-single-
M (2010) A universal approach to eliminate antigenic
strand conformational polymorphism (PCR-SSCP)
properties of alpha-gliadin peptides in celiac disease.
analyses of organellar DNAs. Proc Natl Acad Sci USA
PLoS ONE, 5(12): e15637
94:14570-14577
Petersen G, Seberg O, Yde M, Berthelsen K (2006).
Wang K, Gao L, Wang S, Zhang Y, Li X, Zhang M, Xie
Phylogenetic relationships of Triticum and Aegilops
Z, Yan Y, Belgard M, Ma W (2011a) Phylogenetic
and evidence for the origin of the A, B, and D genomes
relationship of a new class of LMW-GS genes in the
of common wheat (Triticum aestivum). Mol Phylogenet
M genome of Aegilops comosa. Theor Appl Genet
Evol 39:70-82
122:1411-1425
Qi PF, Wei YM, Chen GY, Jiang QT, Liu YX, Li W (2012)
Wang S, Li X, Wang K, Wang X, Li S, Zhang Y, Guo G,
Development of chromosome 6D-specific markers for
Friedrich JZ, Hsam SLK, Yan Y (2011b) Phylogenetic
α-gliadin genes and their use in assessing dynamic
analysis of C, M, N, and U genomes and their
changes at the Gli-2 loci. Mol Breeding 29:199-208
relationships with Triticum and other related genomes
Reeves CD, Okita TW (1987) Analyses of α/β-type gliadin as revealed by LMW-GS genes at Glu-3 loci. Genome
genes from diploid and hexaploid wheats. Gene 54:273-284
52:257-266
Xie Z, Wang C, Wang K, Wang S, Li X, Zhang Z, Ma W,
Shewry P, Halford N, Lafiandra D (2003) Genetics of Yan Y (2010) Molecular characterization of the celiac
wheat gluten proteins. Adv Genet 49:111-184 disease epitope domains in α-gliadin genes in Aegilops
Sollid LM, Qiao S, Anderson R, Gianfrani C, Koning F tauschii and hexaploid wheats (Triticum aestivum L.).
(2012) Nomenclature and listing of celiac disease Theor Appl Genet 121:1239-1251
relevant gluten T-cell epitopes restricted by HLA-DQ Xu J, Messing J (2009) Amplification of prolamin storage
molecules. Immunogenetics, (open access) protein genes in different subfamilies of the Poaceae.
Tanaka M (1985) The relationships of the M and Mu Theor Appl Genet 119:1397-1412
genomes of Aegilops. Wheat Inf Serv 60:39 Zhu X, Li X, Li Y, Yan Y (2010) Cloning and localization
of α-gliadin genes from common wheat and Aegilops
tauschii and their molecular evolution. Acta Agron Sin
36:580-589

The interactive effects of wheat storage proteins in a
gluten free model system
M. Oszvald1, G. Balazs2, S. Tömösközi2, F. Békés3, and L. Tamás2
1Deptartment of Plant Physiology and Molecular Plant Biology, Eötvös University, Hungary; 2Dept of Applied
Biotech and Food Science, Budapest University of Technology and Economics, Hungary; 3FBFD Pty. Ltd.,
Beecroft, NSW 2119, Australia
Abstract
The aim of this work was to compare the effects of incorporated wheat storage proteins on the
functional properties of rice and wheat flour. The advantage of rice as a base flour compared to
wheat is that it does not contain any wheat flour components, and therefore has no interactive effect
between wheat glutenin proteins. The incorporation of individual high molecular weight (HMW)
glutenin subunit proteins (Bx6, Bx7, and By8) in different ratios had significant positive effects on the
mixing requirements of both rice and wheat doughs. Reconstitution experiments using two x+y type
high molecular weight glutenin subunits (HMW-GS) pairs together with a bacterially expressed low
molecular weight glutenin subunit (LMW-GS) was also carried out in this study. The largest effects of
polymer formation and mixing properties of rice flour dough were observed when Bx and By subunits
were used in 1:1 ratios and when LMW-GS were in 1:3 ratios. However, with the same ratios when
using wheat flour (instead of rice flour) as the base flour, these synergistic effects were not observed.
Introduction protein of interest provided information about

A unique property of wheat flour is its ability the role of that protein. However, the principal
to form dough when water is added and mixed. limitation of incorporation studies is that
The mixing requirement of wheat flour and the the results of such experiments are largely
rheological properties of the resulting dough dependent on the allelic composition of the base
largely depend on the composition of gluten flour. The resulting changes in the rheological
proteins. It is generally accepted that both properties of the wheat dough are partly
qualitative and quantitative aspects of protein dependent on the direct contribution of the
composition are relevant in determining wheat incorporated protein, but also on its interactions
dough properties. The large variation of storage with other prolamin type proteins present in
protein composition of wheat is caused by the base flour.
significant polymorphisms of the 12 prolamin Rice flour does not contain wheat prolamin
coding loci. Alleles at the Glu-1 and Glu-3 loci type proteins, and it therefore provides a
in the A, B, and D genomes define the presence new approach to investigate the functional
of high molecular weight (HMW) and low properties of wheat proteins. By incorporating
molecular weight (LMW) glutenin subunits (GS), the proteins into rice dough, we can investigate
whereas the Gli-1 and Gli-2 loci code for the the direct effect of the protein on functional
gliadin proteins (Shewry et al. 2009). properties of the rice dough in the absence of
By using the direct measurement approach, interactions with other prolamin type proteins.
significant information about the effects of The aim of this work was to use rice and
individual GS proteins on dough properties was wheat flours in incorporation experiments
obtained by a specially designed reconstitution to determine and compare the effects of
study: alteration of the rheological properties purified wheat GS proteins individually or in
of a so-called “base-flour” dough after combination on the functional properties of
supplementation or incorporation of a glutenin reconstructed rice and wheat doughs.

Materials and methods the extractable protein fractions of the dough
Flours produced from rice (Orysa sativa L.) cv. according to Oszvald et al. (2007).
Illadong with 7% protein content and bread
Statistical analysis
wheat (Triticum aestivum L.) cv. Hombar, a
medium quality cultivar with 12.1% protein, All measurements were carried out in triplicate.
were used as base flours in this study. Analysis of variance (ANOVA) was then carried
out on mean values. The Statistica 10.0 program
Wheat proteins (StatSoft, Inc. 2006, USA) was used for statistical
Samples of wheat near isogenic lines Galahad 6, evaluations.
Galahad 7 and Galahad 8 were kindly provided
by P.I Payne, PBI Cambridge Ltd., UK. This
set of wheat lines contained null alleles for Results and discussion
the Glu-1 genes in the A and D genomes and For functional studies, individually purified Bx6
expressed only x type 6 or 7, or y type 8, HMW- or Bx7 together with By8 HMW-GS proteins were
GS coded by the B-genome (Payne and Seekings incorporated into 4.0 g flour in different ratios,
1996). LMW-GS protein was expressed in E. coli representing 10% of the total protein content
according to Ciaffi et al. (1999). of the rice and wheat flours. Incorporation of
HMW-GS at any x:y ratio increased the amounts
Isolation of HMW and LMW glutenin subunits of larger polymers in both doughs resulting in
Individual HMW-GS polypeptides were isolated increased mixing requirements and more stable
from Galahad flours as previously published doughs. To better compare the data obtained in
by Oszvald et al. (2009). LMWG-1D1 subunit the different types of experiments, the measured
protein was expressed, extracted and purified as values were expressed as a percentage of those
described by Ciaffi et al. (1999). from the control experimental sample. The
responses of the different parameters to changes
Dough mixing in the x:y ratio are shown in Fig. 1. The dough
development time to reach the maximum
Micro-scale mixing experiments were carried
consistency of reconstructed rice dough increased
out using rice and wheat base flours on a
the most (by 32%) when the Bx7 to By8 ratio was
prototype micro z-arm mixer (METEFEM Ltd,
1:1 (Fig. 1B). The incorporation of x+y HMW-GS
Hungary). Individual glutenin proteins and
also had a positive effect on tolerance to over-
their mixtures were incorporated into the base
mixing as indicated by the significantly lower BD
flours, systematically altering the composition
values of the rice base flour. The largest effect on
of the supplemented proteins (Békés et al. 1994).
this parameter was observed when using a 1:1
The x:y ratios of the HMW-GSs were altered
ratio of Bx7 and By8 subunits for incorporation
using either Bx6 or Bx7 subunits combined with
into rice flour (Fig. 1C). The stability of the
the By8 subunit. The HMW-GS to LMW-GS
reconstructed rice dough increased with
ratio was also varied using the particular x:y
increasing x:y ratios, when 7+8 HMW subunits
ratio where the highest mixing requirement
were incorporated, compared to the control
was measured in the previous experiments. In
dough (Fig. 1D). The incorporation of 6+8
each of these experiments, the supplemented
HMW subunit proteins showed similar effects
protein represented 10% of the original protein
on stability to the 7+8 protein mixture (data not
content of the base flours. The following mixing
shown).
parameters were determined: maximum
resistance (PR), dough development time (DDT), SE-HPLC analyses were carried out to study
breakdown in resistance (BD), and stability the effects of the incorporated glutenin subunits
(ST). Each mixing experiment was carried out in on the distribution of polymeric protein in
triplicate. rice dough. Compared to the control dough,
the UPP% increased with incorporation of the
SE-HPLC analysis mixed HMW-GS. The amount of unextractable
Size exclusion liquid chromatography was polymeric protein fractions showed the highest
carried out to determine the protein size values when the incorporated HMW-GS ratio
distribution in both the unextractable and was 1:1 (Fig. 1A). Similar effects (improvements

to the functional properties of the dough), but the LMWG-1D1 glutenin subunit expressed in
with remarkable differences, were detected on bacterial systems to alter the HMW-GS/LMW-
the mixing parameters of wheat dough compared GS ratio. The x:y ratio was chosen to be 1:1,
to rice dough after incorporation of Bx and By because it was found to have the maximum
subunit proteins. The DDT, ST and UPP% values effect on the rice dough properties. Results of
in the wheat flour experiments monotonically using the Bx7+By8 pairs mixed with LMW-GS
increased when increasing the x:y ratio and each are shown in Fig. 1B.
parameter in the rice flour experiments had
The dough development time of reconstructed
a maximum value at the 1:1 x:y ratio. The BD
rice dough increased most when the HMW:
figures, however, showed an inverse effect, i.e.,
LMW-1D1 ratio was 1:3. By incorporating only
a local minimum for rice flours and an almost
the LMW-1D1 subunit, the DDT value was
monotonic decrease for the wheat flours (Fig. 1A).
drastically reduced (Fig. 2B). With both HMW-
GS pairs the LMW-GS had a positive effect
Altering the HMW-GS:LMW-GS ratio in the
on tolerance to over-mixing indicated by the
reconstituted dough
decreasing BD values. The minimum value of
Incorporation experiments were carried out with the curve occurred when the HMW:LMW ratio
native HMW-GS purified from wheat flour and was 1:3.
180 160
A) B)
150
160
140
UPP [%]
UPP [%]
140
130
120 Wheat Wheat

Rice 120
Rice
100 110
0 25 50 75 100 % LMW GS 0 25 50 75 100 % LMW GS
100 75 50 25 0 % Bx7+By8 100 75 50 25 0 % Bx7+By8
100 180
C) D)
160
80
UPP [%]
BD [%]
Wheat 140
Rice
60
120 Rice
Wheat
40 100
0 25 50 75 100 % LMW GS 0 25 50 75 100 % LMW GS
100 75 50 25 0 % Bx7+By8 100 75 50 25 0 % Bx7+By8
Figure 1

160 180
A) B)
150
160 Wheat
Wheat
140
DDT [%]
UPP [%]
140
130 Rice
120
120
Rice
110 100
0 25 50 75 100 % Bx7 0 25 50 75 100 % Bx7
100 75 50 25 0 % By8 100 75 50 25 0 % By8
100 160
C) D)
90
150
80
BD [%]
ST [%]
140
70 Rice
Rice 130
60
Wheat
120 Wheat
50
0 25 50 75 100 % Bx7 0 25 50 75 100 % Bx7
100 75 50 25 0 % By8 100 75 50 25 0 % By8
Figure 2
Fig. 1 and 2. Effects of individually incorporated HMW-GS subunits Bx7 and By8 (Fig. 1), and individual
Bx7 and By8 subunits plus LMW glutenin subunits (Fig. 2) on mixing properties and the polymer size
distribution in the protein matrix in rice ( ) and wheat doughs ( ). Bx7 and By8 were isolated native
HMW-GS; the LMW-GS was bacterially expressed and purified LMWG-1D1 protein. For each figure: A,
un-extractable polymeric protein (UPP); B, dough development time (DDT); C, breakdown in resistance
(BD); D, stability (ST). 100% value represents the parameters regarding the control dough.
With both HMW-GS pairs the LMW glutenin increasing the proportion of the LMW subunit
subunit had a positive effect on tolerance to in the wheat base flour resulted in a slight
over-mixing as indicated by decreasing BD monotonic decrease in DDT and an increase
values. The minimum value of the curve was in BD values (Fig. 2B, C). The incorporated
reached when the HMW:LMW ratio was 1:3. The LMW-1D1 subunit did not significantly modify
stability of the reconstituted rice dough revealed the strength of the wheat dough according to
similar effects from the LMW subunit as the DDT the stability (ST) data (Fig. 2D). The amount of
mixing parameter. The peak value was reached polymeric proteins in the reconstructed wheat
at the 1 HMW:3 LMW ratio. Incorporation of dough did not change significantly during the
the LMW subunit alone drastically reduced the increase in the LMW:HMW subunit ratio, but
strength of the dough in both cases (Fig. 2B). was monotonic (Fig. 2A). It should be noted
The amount of polymeric proteins in rice dough that the bacterially expressed and purified
was also increased after incorporation of the LMW-1D1 subunit itself had positive effects on
LMW-1D1 glutenin subunit. The UPP% was at the mixing parameter and UPP% of the wheat
its maximum value at a 1:3 ratio of HMW:LMW- dough compared to the control.
1D1 proteins in the mixture (Fig. 2A). Conversely,

Rice flour, due to the absence of wheat type dough than HMW-GS when applied in the same
storage proteins, provides a suitable background amounts and at the same polymer/monomer
to observe physical/functional variations between protein ratio. A possible explanation is that the
wheat and rice flour. Because of these different magnitude of the effect is greater on polymers
features of the rice flour, a rice-based model with longer ‘backbones’ (caused by greater
system was recently developed in our laboratory numbers of HMW-GS) than longer branches
to study the functional properties of wheat (caused by greater numbers of LMW-GSs).
storage proteins.
The largest effects on polymer formation and
mixing properties of reconstructed rice flour
Conclusions
dough were observed when Bx and By HMW-GS Variation of wheat and rice based flours provides
were used in a 1:1 ratio. While in agreement with a way to study the effects of a particular protein
previous studies, the synergistic effect of these and to determine the individual and interactive
two incorporated protein types was not observed effects on two related but different aspects;
when wheat flour was used as base flour. In namely, polymer formation and contribution
the case of rice flour, the increased UPP% and to the visco-elastic properties of the dough.
mixing requirements clearly indicated the This new information provides a deeper
individual effects of the incorporated protein on understanding of the structural/functional
the polymeric structure. Using wheat flour, the relationships of wheat glutenin proteins and can
effects of interactions of the incorporated proteins be applied to improve wheat quality.
affected the internal prolamin interactions in the
base flour. Thus, the synergistic effects of x and
y HMW-GS can be monitored only in the case of Acknowledgement
very strong subunits of the Glu-D1 locus while This work was supported by the Hungarian
using subunits coded by the Glu-B1 locus. The Scientific Research Fund (PD 101330).
interactive effects are hidden by the complex
effects observed in in vitro studies when using
wheat flour as the base flour (Uthayakumaran References
et al. 2000). By replacing wheat with rice flour, Békés F, Gras PW, Gupta RB (1994) Mixing properties
the synergistic effects can be clearly monitored as measure of reversible reduction and oxidation of
during functional studies because of the lack doughs. Cereal Chem71:44-50
of other wheat prolamin proteins. In light of Ciaffi M, Lee YK, Tamas L, Gupta R, Skerritt J, Appels R
the different results provided by using rice and (1999) The low-molecular-weight glutenin subunit
proteins of primitive wheats. III. The genes from
wheat flour doughs to observe the synergistic D-genome species. Theor Appl Genet 98:35-148
effects of co-incorporated x and y type HMW
Oszvald M, Tomoskozi S, Laszlo TL, Bekes F (2009) Effects
subunits, apparently conflicting results of of wheat storage proteins on the functional properties
using D or B genome-derived HMW-GS can be of rice dough. J Agric Food Chem 57:10442-10449
explained. Oszvald M, Tomoskozi S, Larroque O, Keresztenyi E,
Tamas L, Bekes F (2007) Characterization of rice
The real effects of linear LMW glutenin polymers
storage proteins by SE-HPLC and micro z-arm mixer. J
can be measured after their incorporation into Cereal Sci 48:68-76
rice flour due to the absence of HMW subunits. Payne PI, Seekings JA (1996) Characterisation of
By incorporating both HMW- and LMW- Galahad-6, Galahad-7 and Galahad-8, isogenic lines,
GS proteins, these subunits are able to build that contain only the one HMW glutenin subunit. In:
a polymer resembling the native form. The Wrigley CW (ed) RACI, North Melbourne, Victoria,
maximum value in mixing parameters of rice Australia, pp14-17
dough (by the incorporation of x and y type Shewry PR, Ovidio R, Lafiandra D, Jenkins JA, Mills
HMW-GSs at a 1:1 ratio and LMW- and HMW- ENC, Békés F (2009) Wheat grain proteins. In:Khan
K, Shewry PR (eds) Wheat chemistry and technology.
GS at a 3:1 ratio) indicate that either ‘head-to- AACC Press, St Paul, MN, USA pp 223-298
tail’ or ‘backbone’ polymerization of glutenin
Uthayakumaran S, Stoddard FL, Gras PW, Bekes F
subunits is possible in rice flour. LMW-GS had (2000) Effects of incorporated glutenins on functional
a smaller effect on functional properties of rice properties of wheat dough. Cereal Chem 77:737-743

Genetic variation of dough properties independent of high
and low molecular weight glutenin subunit genes in wheat
H. Ohta1, T. M. Ikeda2, A. Torada1, M. Hayashi1, I. Tsutsui1, Y. Tanaka1, and T. Kita1
1HOKUREN Federation of Agricultural Cooperatives, Hokkaido, 069-1316, Japan; 2NARO, Western Region
Agricultural Research Center, Fukuyama, 721-8514, Japan
Abstract
Allelic differences in glutenin subunit genes influence bread-making quality; therefore it is important
to characterize the glutenin subunit alleles using DNA and protein markers, and to understand the
relationships between different alleles and dough properties for developing new varieties. In this
study, the glutenin alleles in Japanese cv. Haruyokoi with good bread-making quality and advanced
lines were determined. One of the advanced lines, HW5, proved to have the same alleles as Haruyokoi.
Although the protein content of HW5 was similar to Haruyokoi, the dough properties were clearly
different. Based on SDS-sedimentation volume and SE-HPLC analysis, the quantity of glutenin
polymers in HW5 was less than in Haruyokoi. 2D-PAGE was carried out focusing on gliadin proteins,
and at least three spots were specifically detected only in the flour protein of HW5. In addition, the
cell walls of HW5 seemed to be thick when observed by electron microscopy. This indicated that the
arabinoxylan content of HW5 might be more abundant. In this study, varietal differences controlling
dough properties and glutenin polymer quantity were independent of the glutenin genes. This
suggests a need to pay attention to genetic factors additional to glutenin alleles for selecting lines with
high bread-making quality in the early generation stages in breeding programs.
Introduction However, in the process of breeding, one

The unique functional properties of wheat dough advanced line showed different dough properties
are due to gluten, which contains glutenins and from Haruyokoi although both possessed the
gliadins. Glutenins are mainly composed of same glutenin alleles. In this study we tried to
two types of subunits, high and low molecular determine the additional factors influencing
weight glutenin subunits (HMW-GS and LMW- dough properties.
GS, respectively). Many studies have shown
that particular protein subunits are associated
with bread-making quality (Branlard and Materials and methods
Dardevet 1985; Payne et al. 1987). For example, Japanese cultivar Haruyokoi, with good bread-
many cultivars with HMW-GS 5+10 have strong making quality, and advanced line HW5 were
dough properties contributing to good bread- used in this study. Grains were milled using
making quality. In addition, LMW-GS are also a Buhler test mill and the flour was prepared
involved in dough properties and bread-making at a 60% extraction rate. Flour protein content
quality (Branlard et al. 2001; Gupta et al. 1994; was measured using a KJELTEC AUTO 1305
Maruyama-Funatsuki et al. 2004). HMW-GS are instrument.
encoded by the Glu-A1, Glu-B1, and Glu-D1 loci,
DNA markers were utilized for identification of
and LMW-GS are encoded by Glu-A3, Glu-B3,
Glu-D1, Glu-A3 and Glu-B3 alleles using primers
and Glu-D3. The allelic composition of glutenin
developed by Suzuki et al. (unpublished data)
subunit genes influences bread-making quality.
and Zhang et al. (2004). Total DNA extraction
The Japanese cultivar Haruyokoi is a leading and PCR were carried out according to Torada
variety with good bread-making quality, and et al. (2008). SDS-PAGE and 2D-PAGE were
has been acceptable to end users. HMW-GS and performed according to the method of Ikeda et al.
LMW-GS alleles are the targets for selecting lines (2005). The gliadin protein fraction extracted with
with high quality equivalent to Haruyokoi at the 50% (v/v) 1-propanol was used for 2D-PAGE.
early generation stages in our breeding program. IEF was carried out for 18kWh using Immobiline

Dry-Strip pH6-11 (Amersham Bioscience, USA). Glu-D1d allele. Contrary to expectations, flour
Gluten index (GI) was determined using a characteristics and dough properties of the two
Glotomatic System 2202 according to the AACC lines were clearly different (Table 2). The flour
38-12 procedure. Mixograph parameters were protein contents of the two lines were similar.
determined using 10g mixograph following On the other hand, the gluten index of HW5
the AACC 54-40A procedure. Farinogragh and was lower, and the mixograph peak time (PT),
extensogragh tests were performed according to development time (DT), and stability (stab)
AACC 54-21 and 54-10, respectively. The SDS- in the farinogragh were shorter than those of
sedimentation volume analysis was performed Haruyokoi. Furthermore, the values for resistance
according to the prolonged micro-SDS volume (R) and the ratio of R and extensibility (E) (R/E)
method using a 0.7g flour sample (Takata et al. in extensograph were clearly lower than those of
1999). Size-exclusion high-performance liquid Haruyokoi. These data indicated that the dough
chromatography (SE-HPLC) analysis was of HW5 was not as elastic as that of Haruyokoi.
carried out according to the method of Singh
To analyze factors controlling differences in
et al. (1990) as modified by Batey et al. (1991).
dough properties between the two lines, we
Scanning electron microscope (SEM) analysis was
focused on the glutenin quantity, glutenin/
performed according to the procedure of Araki
gliadin ratio and gliadin profiles. The SDS-
et al. (2009). Grains were snapped in the middle
sedimentation volume and proportion of
using a straight razor blade, and transverse
unextractable polymeric protein relative to
sections were examined using an S-3400N
the soluble monomeric protein (UPP/SMP)
instrument (Hitachi High-Technology, Tokyo,
determined by SE-HPLC were lower than
Japan). The SEM analysis was conducted without
those of Haruyokoi. These results suggested
a coating at an accelerating voltage of 8.00 kV
that the quantity of glutenin polymers of HW5
under low vacuum conditions of 70 Pa at -25˚C.
was less than Haruyokoi and the proportion
of glutenin to gliadin of HW5 was lower than
that for Haruyokoi. According to Zhang et al.
Results and discussion (2007) the ratio of gliadin to glutenin determines
SDS-PAGE and PCR were performed to detect dough properties and bread-making quality. The
alleles at the HMW-GS and LMW-GS loci, variation in dough properties between HW5 and
respectively. HW5 proved to have the same Haruyokoi might therefore be due to the quantity
alleles as Haruyokoi (Table 1); both lines had of glutenin polymers and/or the proportion of
the HMW-GS 5+10 subunit encoded by the glutenin to gliadin.
Table 1. High and low molecular weight glutenin alleles determine by DNA marker and SDS-PAGE.
Line Glu-A1 Glu-B1 Glu-D1 Glu-A3 Glu-B3 Glu-D3

Haruyokoi b c d c h a
HW5 b c d c d a
Table 2. Flour characteristics and dough properties of Haruyokoi and HW5.

Flour protein Farinograph Extensograph SDS SE-HPLC
Gluten Index PT in sedimentation analysis
Mixograph DT stab R E volume
Line (%) (%) (min) (min) (min) (BU) (min) R/E (ml) UPP/SMP1
Haruyokoi 13.3 97.8 4.25 19.6 28.1 871 199 4.37 12.7 0.43
HW5 13.1 75.8 3.15 7.7 14.4 556 205 2.72 11.2 0.37
1 UPP/SMP indicates the proportion of unextractable protein to soluble monomeric protein

The quantitative variation in gliadin fractions Another factor that may influence dough
and allelic differences of gliadin genes are so far properties independently of glutenin genes is
unknown in our breeding materials. As one of the polysaccharides which are considered to inhibit
approaches for gliadin protein characterization, glutenin polymer formation. Comparison of
the proteins extracted using 50% (v/v) 1-propanol the seed microstructures of the two lines by
were subjected to 2D-PAGE. Comparingthe low-vacuum scanning electron microscopy
2D-PAGE profiles between the two lines, at without pre-treatment (N-SEM) indicated that
least three spots were detected only in the flour the endosperm cell walls of HW5 were slightly
protein of HW5 (Fig.1). These spots seemed to be thicker than those of Haruyokoi (Fig.2). This
α/β-gliadins. Different gliadin proteins appeared indicates that the arabinoxylan content of HW5
to be expressed in the flour protein of HW5, might be more abundant than in Haruyokoi, and
although it is unclear if these spots contributed as a result, the dough properties of HW5 might
to the variation in glutenin/gliadin ratio, gliadin not be as elastic due to inhibition of glutenin
quantity and dough properties. polymer formation by the increased amount of
arabinoxylan in the flour.
pH10 IEF pH6 pH10 IEF pH6
1) 2)
SDS-PAGE
Fig. 1 2D-PAGE of gliadin fraction: (1) Haruyokoi, (2) HW5. Arrows indicate spots detected only in HW5.
1) 2)
Fig. 2 SEM images of transverse sections of wheat seeds: (1) Haruyokoi, (2) HW5. Arrows indicate
endosperm cell wall.

Conclusions Gupta RB, Paul JB, Cornish GA, Bekes F, Rathjen AJ (1994)
Allelic variation at glutenin subunit and gliadin loci,
We found an example where varietal differences Glu-1, Glu-3 and Gli-1, of common wheat. I. Its additive
in dough properties and glutenin polymer and interaction effects on dough properties. J Cereal Sci
quantity were not related to the glutenin alleles 19:9-17
in the expected manner. This study suggests that Ikeda TM, Ohnishi N, Nagamine T, Oda S, Hisatomi T, Yano
we need to give attention not only to glutenin H (2005) Identification of new puroindoline genotypes
alleles when selecting in early generations using and their protein products among wheat cultivars. J
Cereal Sci 41:1-6
SDS-PAGE and DNA markers, but also to other
genetic factors. The extra gliadin proteins and Maruyama-Funatsuki W, Takata K, Nishio Z, Tabiki T,
Yahata E, Kato A, Saito K, Funatsuki H, Saruyama H,
difference in cell wall thickness may be some of Yamauchi H (2004) Identification of low-molecular
the genetic factors affecting the dough properties weight glutenin subunits of wheat associated with
and polymer contents, and may need to be bread-making quality. Plant Breeding 123:355-360
considered when selecting lines with high bread- Payne PI, Nightingale MA, Krattiger AF, Holt LM (1987)
making quality. The relationship between HMW glutenin subunit
composition and the bread-making quality of British-
grown wheat varieties. J Sci Food Agric 40:51-65
References Singh NK, Donovan R, MacRitchie F (1990) Use of
sonication and size-exclusion high-performance liquid
chromatography in the study of wheat flour proteins.
AACC International (2000) Approved Methods of AACC II. Relative quantity of glutenin as a measure of bread-
International, 10th ed, American Association of Cereal making quality. Cereal Chem 67:161-170
Chemists, St. Paul, MN Takata K, Yamauchi H, Iriki N, Kuwabara T (1999)
Araki E, Ikeda TM, Ashida K, Takata K, Yanaka M, Iida S Prediction of bread-making quality by prolonged
(2009) Effects of rice flour properties on specific loaf swelling SDS-sedimentation test. Breeding Sci 49:221-223
volume of one-loaf bread made from rice flour with Torada A, Koike M, Ikeguchi S, Tsutsui I (2008) Mapping
wheat vital gluten. Food Sci Technol Res 15:39-448 of a major locus controlling seed dormancy using
Batey IL, Gupta RB, MacRitchie F (1991) Use of size- backcrossed progenies in wheat (Triticum aestivum L.).
exclusion high-performance liquid chromatography Genome 51:426-432
in the study of wheat flour proteins: an improved Zhang PP, He ZH, Chen DS, Zhang Y, Larroque OR, Xia XC,
chromatographic procedure. Creal Chem 68:207-209 (2007) Contribution of common wheat protein fractions
Branlard G, Dardevet M (1985) Diversity of grain proteins to dough properties and quality of northern-style
and bread wheat quality correlation between high- Chinese steamed bread. J Cereal Sci 46:1-10
molecular-weight subunits of glutenin and flour Zhang W, Gianibelli MC, Rampling LR, Gale KR (2004)
quality characteristics. J Cereal Sci 3:345-354 Characterisation and marker development for low
Branlard G, Dardevet M, Saccomano R, Lagoutte F, molecular weight glutenin genes from Glu-A3 alleles of
Gourdon J (2001) Genetic diversity of wheat storage bread wheat (Triticum aestivum. L.). Theor Appl Genet
proteins and bread wheat quality. Euphytica 199:59-67 108:1409-1419

What causes extreme gluten quality variation in winter
wheat in Norway?
S. Koga1, U. Böcker2, A. K. Uhlen1,2, E. F. Mosleth2, B. Hoel3, and A. Moldestad2
1Departmentof Plant and Environmental Science, Norwegian University of Life Sciences, P.O. Box 5003,
NO-1432 Ås, Norway; 2Nofima AS, P.O. Box 210, NO-1431 Ås, Norway; 3Bioforsk Øst Apelsvoll, Rute 509,
NO-2849 Kapp, Norway
Abstract
Large variations in Rmax between seasons, varieties and locations were observed. Seasons with lower
temperature and higher precipitation during grain development seemed to be associated with lower
gluten resistance. Extremely low Rmax values in 2011 were associated with protein degradation after 4
h of incubation indicating enzyme activity
Introduction Extensibility Rig (Kieffer et al. 1998). Resistance to

Analyses of Norwegian wheat varieties from stretching (Rmax) was used as the main parameter
different fields in south-eastern Norway during for gluten quality.
2005-2011 showed large variations in gluten
viscoelastic properties. Norwegian wheat SDS-PAGE
producers faced additional challenges during the Total gluten proteins were extracted with 50%
2011 season, with frequent, heavy rainfall during 1-propanol and 4.5% DTT from flours prior to
the maturation and harvesting period which led albumin and globulin extraction with 62.5 mM
to complete losses in some fields. Some winter Tris-HCl, 2% SDS and 1.5% DTT buffer (pH 6.8).
wheat samples from several fields were harvested Gluten protein extracts were freeze dried and
before the extreme rainfall events and therefore pellets were re-dissolved in albumin/globulin
achieved acceptable falling number (>200). extracts. These mixtures were then incubated for 4
However, they showed remarkable quality and 16 h. For electrophoresis, NuPAGE®NOVEX
deficiencies regarding viscoelastic properties of 4-12% Bis-tris Mini Gels (Invitrogen AS) were used
gluten and therefore also posed a challenge to the and the gels were stained by Coomassie Brilliant
milling industry. Blue-R 250.

Field experiments with commercial winter wheat Years 2005 and 2006 were “good weather” years
varieties were carried out at several locations in with higher average temperatures and less
south-eastern Norway from 2005 to 2011. Field precipitation throughout the grain development
trials were complete block design with two period (Table 1). Large variations in average Rmax
replications. Sprout damage was determined by values were observed between seasons. The 2005
the falling number method. Samples with falling and 2006 seasons showed high Rmax, whereas 2011
numbers lower than 200 were eliminated from showed low Rmax. Similar seasonal variations were
further study, leaving analyses of 3-9 fields per also observed in spring wheat (Moldestad et al.
season. 2011). High Rmax for all four winter wheat varieties
grown in 2005 and 2006 indicated strong gluten.
Kieffer dough and gluten extensibility Rig
Winter wheat varieties grown in the 2011 season
Gluten dough was prepared in a Glutomatic showed extremely low Rmax, except for Olivin,
2100 (Perten AB, Huddinge, SE). Viscoelastic which had relatively normal Rmax (Fig. 1). Alpha-
properties of gluten dough were then determined amylase activities in that year were low compared
by using the SMS/Kieffer Dough and Gluten to 2005 and 2006 (Fig. 1).

1.0 327 1.0
A) B)
0.9 0.9
0.8 290 0.8
348 300
0.6 0.6
0.5 0.5
Rmax (N)
321
0.4 0.4
0.3 0.3
0.2 0.2 398
242 296
0.1 257
0.1
0.0 0.0
r
n
ifik
rke
e
ifik
vis
ivi
e
øln
an
ivi
gn
Bjø
øln
Ol
Ell
gn
Fin
Ol
Mj
Ma
Mj
Ma
Fig. 1 Mean Rmax of varieties from different fields in season 2005 and 2006 (A) and- 2011 (B).
Numbers in the figure show average falling.
Finans Olivin Cultivars with the highest (Olivin) and one of

the lowest (Finans) Rmax values in 2011 were
4h 16h C 4h 16h C
selected to further investigate gluten protein
properties (Fig. 2). The gluten proteins of
the variety with extremely low Rmax seemed
highly degraded after 4hrs incubation at
30˚C. Degradation was more severe in Finans
compared to Olivin. This degradation of proteins
was not observed when the extracts were heated
at 60˚C for 15 min prior to incubation (control).
This indicates that enzymes may be the cause of
degradation and weakening of gluten proteins.
More research is required to explore these
phenomena, which are of high importance to
ensure good quality wheat for industry.
References
Kieffer R, Wieser H, Henderson MH, Graveland A (1998)
Correlations of the breadmaking performance of
Fig. 2 2SDS-PAGE of Olivin and Finans wheat flour with rheological measurements on a
from one of the fields in 2011 after 4 and 16 h micro-scale. J Cereal Sci 27:53-60
incubation (C: control). Moldestad A, Færgestad EM, Hoel B, Skjelvåg AO, Uhlen
AK (2011) Effect of temperature variation during
grain filling on wheat gluten resistance. J Cereal Sci
53:347-354

Changes in protein components and gluten content in
noodle processing and cooking
R. Liu, Y.M. Wei, and B. Zhang
Institute of Agro-Food Processing Science and Technology, Chinese Academy of Agriculture Sciences/Key
Laboratory of Agro-Products Processing, Ministry of Agriculture, Beijing 100193, China
Abstract
To learn about the changes in protein components during noodle processing and cooking, three kinds
of wheat cultivars with large differences in quality were selected. Protein content, protein components
content, glutenin macro-polymer (GMP) content, gluten content and gluten index of the flour, dough
crumbs, dough sheets, dried noodles and cooked noodles were investigated. Protein content was
unchanged from wheat flour to cooked noodles. Changes in protein components differed with wheat
cultivars. After dough mixing, the content of gliadin and GMP significantly decreased. During the process
from dough crumbs to dried noodles, glutenin content, GMP content, content of wet and dry gluten
underwent no significant changes. After cooking, the contents of albumin+globulin and gliadin were
significantly decreased, while the glutenins content and GMP increased substantially, and the gluten
structure disappeared. In conclusion, protein content underwent no significant change during noodle
processing and cooking, but gluten protein components changed significantly after mixing and cooking.
Introduction investigated, and the effect of changes of protein

Content and quality of wheat protein have a characteristics on white Chinese noodle quality
large influence on white Chinese noodle quality. has not been reported.
Protein content has negative effects on color The objective of this research was to investigate
and appearance of noodles, but contributes changes in protein components and gluten
significantly and positively to cooked noodle characteristics in noodle processing and cooking
texture properties, especially in improving for indicating the effects of protein characteristics
hardness and elasticity (Morris et al. 2000; Liu on noodle sensory quality and the noodle-
et al. 2003; Park et al. 2003; Park and Baik 2004; making process.
Wang et al. 2004; Liu et al. 2011). Hardness,
chewiness, cohesiveness and tensile strength
of cooked noodles tend to become high when Materials and methods
glutenin content increases (Wang and Kovaks
2002; Hu et al. 2007; Zhang et al. 2011). To Wheat and flour
a certain extent, increased protein quality Samples of three leading Chinese wheat cultivars,
contributes positively to noodle sensory scores including one hard winter wheat (Zhengmai 366),
and cooking quality. Excessive protein quality one medium hard winter wheat (Xiaoyan 22),
(wet gluten content >35%, sedimentation value and one medium hard spring wheat (Yongliang
>60 mL, stability time >16 min) decreases the 4) were used in this study. Grain samples of
appearance quality of noodles (Crosbie et al. the three cultivars were obtained from Henan,
1999; Liu et al. 2003; Park et al. 2003). In general, Shanxi and Inner Mongolia, respectively. A 50 kg
the protein quality criteria of good noodle- sample of each wheat cultivar was cleaned and
making quality wheat are as follows: flour tempered overnight. Hard and medium wheats
protein content 12-14%, wet gluten content were tempered to ≈16.0 and 15.0% moisture
28-34%, sedimentation value 40-45 ml, stability content, respectively. All samples were milled
time 5-15 min (Liu et al. 2011). Changes in protein with a Brabender Quadrumat Senior laboratory
components and gluten characteristics during mill (Duisburg, Germany) according to Approved
noodle processing and cooking have not been Method 26-50 (AACC 2000).

Noodle preparation dryer (CHRIST ALPHA 1-2LDPLUS, Osterode,
Flour (200 g, 14% moisture basis) was mixed Germany). Then the dried samples were
with water (65 mL) and salt (1%) in a laboratory powdered on a Retsch mill (Ultra Centrifugal
mixer (Dongfangfude JHMZ-200, Beijing, China) ZM200, Germany) using a 0.5 mm sieve.
for 4 min. Salt was dissolved in distilled water
and added to flour before mixing. Dough was Analytical methods
passed through the rolls of a noodle machine Moisture was determined by Approved Method
(Dongfangfude JMTD-168/140, Beijing, China) at 44-15A (AACC 2000). Protein content (N×5.7) was
a 3 mm gap. The noodle dough sheet was folded estimated using the Kjeldahl method according
and put through the sheeting roll two more to Approved Method 46-12 (AACC 1999). Protein
times. The dough sheet was rested in a PE plastic was fractionated according to the procedure of
bag for 30 min at room temperature, and then Weegels et al. (1996) with some modifications.
sheeted four times at progressively reduced roll Wet gluten and the gluten index were determined
gap settings of 2.4, 1.8, 1.4 and 1 mm. Finally, the using the Glutomatic system according to ICC
dough sheet was cut into 2 mm wide noodles. standard 155 (ICC 1994).
Raw noodles (20 g) were cooked in 660 mL of GMP content was determined according to
boiling distilled water. The optimum cooking the procedure of Don et al. (2003) with some
time of noodles was from the time the noodles modifications. Flour was suspended in 1.5%
were placed in boiling water until the white core SDS (0.05 g flour in 1 ml) and centrifuged at
disappeared when cooked noodle strands were 15,500 g for 15 min at 25°C in a SIGMA3-30k
squeezed between transparent plates. ultracentrifuge (Germany). The supernatant (SUP)
was decanted and the gel-layer collected as GMP.
Samples preparation
Protein content, protein components content, Statistical analysis
glutenin macro-polymer (GMP) content, wet The data obtained in this study were expressed as
gluten and gluten index of the flour, dough means of at least three replicated determinations.
crumbs after mixing, sheets before cutting, dried One-way analysis of variance was used to
noodles and cooked noodles were investigated analyze the data and significant differences
(Fig. 1). among means were compared by the LSD test
The samples were put into PE plastic bags and using SPSS ver. 16.0 for Windows. Differences
pre-frozen for 12 h until frozen. The drying was were considered significant at P < 0.05 and 0.01.
carried out in a laboratory-scale vacuum freeze
Milling Mixing Sheeting Cutting and drying Boiling
Wheat Flour Dough crumbs Sheets Dried noodles Cooked noodles
Vacuum freeze-drying
Kjeldahl method Continuous extraction SDS digestion Glutomatic system
Protein content Protein composition GMP content Gluten content and gluten index
Changes in protein components and gluten

content in noodle processing and cooking
Fig. 1 Methods of sampling and sample analysis in the noodle-making process.

Results and discussion (2003) suggested that mechanical force is the
The results showed that protein content had predominant cause of GMP breakdown during
no significant change in noodle processing and mixing.
cooking (Fig. 2). After cooking, protein content During the process from dough crumbs to dried
increased slightly for Yongliang 4 and Xiaoyan 22, noodles, glutenin content and GMP content
because small amounts of starch dissolved into underwent no significant changes, but gliadin
soup, leading to increases in the relative contents content increased after sheeting and resting.
of protein. After cooking, the content of salt-soluble protein
After dough mixing, the contents of gliadin and (albumin+globulin) and gliadin decreased
GMP decreased significantly (Fig. 4 and 6). The significantly, while the contents of glutenin and
change of GMP quantity was consistent with the GMP increased substantially (Fig. 3, 4, 5, 6). This
reports of Don et al. (2003) proposed that the is related to protein denaturation during noodle
changes in GMP quantity were due to chemical cooking. Cooking changed the solubility of
breakdown of glutenin structure. Don et al. protein in different solvents.
16 6
14 5
Crude protein content (%)
12
Salt-soluble protein (%)
4
10
8 3
6
2
Zhengmai 366 Zhengmai 366
4
Yongliang 4 1 Yongliang 4
2 Xiaoyan 22 Xiaoyan 22
0 0
1 2 3 4 5 1 2 3 4 5
Sample Sample
Fig. 2 Changes in protein content in noodle Fig. 3 Change in salt-soluble protein content in
processing and cooking; Sample 1, flour; Sample noodle making.
2, dough crumbs after mixing; Sample 3, sheets
before cutting; Sample 4, dried noodles; Sample 5,
cooked noodles.
Table 1. Analysis of protein properties in three wheat cultivars.
Zhengmai 366 Yongliang 4 Xiaoyan 22

Protein component
Albumin+globulin (%) 4.88 4.20 5.02
Glutenin (%) 4.51 3.87 4.15
Gliadin (%) 2.56 2.54 1.94
GMP (%) 4.48 3.35 2.84
Total protein (%)
Gluten properties
Wet gluten content (%) 13.41 27.3 29.9
Dry gluten content (%) 11.7 9.2 10.3
Gluten index 93 97 79

4.0 Changes in protein components were different for
each wheat cultivar. For Zhengmai 366 with high
3.5
protein content and strong gluten, the salt-soluble
3.0 protein content was significantly decreased after
Gliadin content (%)
2.5 mixing and drying, and increased after sheeting

and resting. For Yongliang 4 and Xiaoyan 22 there
2.0 were no significant changes during dried noodle
1.5 making (Fig. 3). Changes in protein components in
noodle making are related to wheat flour quality
1.0 Zhengmai 366 and type, especially the protein characteristics of
Yongliang 4 the wheat flour.
0.5
Xiaoyan 22
0.0 The wet gluten contents were slightly decreased
1 2 3 4 5 after dough mixing (Fig. 7) due to damage to the
Sample gluten network of dough crumbs after freeze-
drying and crushing. During the process from flour
Fig. 4 Changes in gliadin content in noodle making.
to dried noodles, gluten index decreased gradually
for Yongliang 4 and Xiaoyan 22, whereas there was
no significant change for Zhengmai 366 (Fig. 8).
12
After cooking, the gluten structure disappeared.
10 Zhengmai 366
Yongliang 4 35
Glutenin content (%)
8 Xiaoyan 22
30
Wet gluten content (%)
6
25
4 20
2 15
10 Zhengmai 366
0
1 2 3 4 5 Yongliang 4
5 Xiaoyan 22
Sample
Fig. 5 Changes in glutenins content in noodle making. 0
1 2 3 4 5
Sample
14 Fig. 7 Changes in wet gluten content in noodle making.
12 Zhengmai 366 120

Yongliang 4
10
Xiaoyan 22 100
GMP content (%)
8
80
Glutenin index
6
60
4
2 40
Zhengmai 366
0 20 Yongliang 4
1 2 3 4 5 Xiaoyan 22
Sample 0
Fig. 6 Changes in GMP content in noodle making. 1 2 3 4 5
Sample
Fig. 8 Changes in gluten index in noodle making.

Conclusions ICC (1994) Standard 155. Determination of wet gluten
quantity and quality (Gluten Index ac. to Perten) of
Protein content underwent no significant changes whole wheat meal and wheat flour (Triticum aestivum).
during noodle processing and cooking. Changes The Association of Cereal Chemistry, Vienna, Austria
in protein composition differed with wheat Liu JJ, He ZH, Zhao ZD, Peña RJ, Rajaram S (2003) Wheat
cultivars that have distinct protein characteristics. quality traits and quality parameters of cooked dry
The contents of gliadin and GMP were white Chinese Noodles. Euphytica 131:147-154
significantly decreased after dough mixing. The Liu R, Wei YM, Zhang B (2011) Review on relationship
contents of albumin+globulin and gliadin were between wheat protein and noodle quality (in
significantly decreased after cooking, whereas Chinese). J Triticeae Crops 31:1183-1187
the contents of glutenin and GMP increased Morris CF, Jeffers HC, Engle DA (2000) Effect of
processing, formulae and measurement variables
substantially and gluten structure disappeared.
on alkaline noodle color — Toward an optimized
laboratory system. Cereal Chem 77:77-85
Park CS, Baik BK (2004) Cooking time of white salted
References noodles and its relationship with protein and amylose
AACC (1999) Approved Methods of the AACC, 10th Ed. contents of wheat. Cereal Chem 81:165-171
Methods 46-12. The Association of Cereal chemists, Park CS, Hong BH, Baik BK (2003) Protein quality of
St. Paul, MN wheat desirable for making fresh white salted noodles
AACC (2000) Approved Methods of the AACC, 10th Ed. and its influences on processing and texture of
Methods 26-50, 44-15A. The Association of Cereal noodles. Cereal Chem 80:297-303
Chemists, St. Paul, MN Wang C Kovacs MIP (2002) Swelling index of glutenin Test
Crosbie GB, Ross AS, Moro T, Chiu PC (1999) Starch and II. Application in prediction of dough properties and
protein quality requirements of Japanese alkaline end-use quality. Cereal Chem 79:190-196
noodles (Ramen). Cereal Chem 76:328-334 Wang C, Kovacs MIP, Fowler DB, Holley R (2004) Effects
Don C, Lichtendonk WJ, Plijter JJ, Hamer RJ (2003) of protein content and composition on white noodle
Understanding the link between GMP and dough: making quality: Color. Cereal Chem 81:777-784
from glutenin particles in flour towards developed Zhang SB, Lu QY, Yang HS, Meng DD (2011) Effects of
dough. J Cereal Sci 38:157-165 protein content, glutenin-to-gliadin ratio, amylose
Hu XZ, Wei YM, Wang C, Kovacs MIP (2007) Quantitative content, and starch damage on textural properties of
assessment of protein fractions of Chinese wheat Chinese fresh white noodles. Cereal Chem 88:296-301
flours and their contribution to white salted noodle Weegels PL, Pijpekamp AM, Graveland A, Hamer
quality. Food Res Int 40:1-6 RJ, Schofield JD (1996) Depolymerisation and
repolymerisation of wheat glutenin during dough
processing. I. Relationships between glutenin
macropolymer content and quality parameters. J
Cereal Sci 23:l03-l14

In silico analysis of transcriptional regulation of HMW
glutenin genes in wheat
S.Z. Makai1,2, G.Y. He2, L. Tamás1, and A. Juhász3
1Department of Plant Physiology and Molecular Plant Biology, Eötvös Loránd University, Budapest, Hungary;
2China-UK HUST-Rres Genetic Engineering and Genomics Joint Laboratory, The Genetic Engineering
International Cooperation Base of Chinese Ministry of Science and Technology, Chinese National Center of Plant
Gene Research (Wuhan) HUST Part, College of Life Science and Technology, Huazhong University of Science
& Technology (HUST), Wuhan, China; 3Applied Genomics Department, Agricultural Institute, Centre for
Agricultural Research, HAS, Martonvásár, Hungary
Abstract
The storage proteins of wheat have a crucial role in the development of end-product quality; their
allelic composition and expression determine the strength and extensibility of dough made from
wheat flour. Prolamins consisting of monomeric gliadins and polymeric high molecular weight
(HMW-GS) and low molecular weight glutenin subunits (LMW-GS), stabilized through inter- and
intra molecular disulfide bonds, form a complex protein matrix, the gluten network. The composition
and size of this network and its physico-chemical characteristics depend on the expression profile of
the different components. An in silico analysis was carried out to identify various motif compositions
in the promoter regions of HMW genes showing significant differences in gene expression. Our
study was limited to HMW sequences and expressed sequence tags (EST) data available in public
databases. Results show that no single motif is responsible for high or low expression but rather it is
the composition of different motifs.
Introduction (EST)-based expression analysis together or

The synthesis and deposition of seed storage paired with promoter sequence profiling was
proteins of wheat occur specifically in the used to gain a deeper insight into the (differing)
endosperm and only at a specific time period expression rates of different types of HMW-
during seed development. In addition, they GS genes encoded at different loci. The HMW
are influenced by the availability of nitrogen glutenin subunit gene family is encoded by
and sulphur in the grain (Duffus and Cochrane pairs of paralogous x- and y-type genes in
1992). Their expression shows similar profiles, complex loci denoted as Glu-1, on the long arms
indicating a remarkable conservation in the of homoeologous chromosomes 1A, 1B and 1D.
machinery responsible for scheduling storage The x-type encodes a higher molecular weight
protein gene expression. Common regulatory subunit protein than the y-type. The order and
cis-acting elements and trans-acting transcription orientation of the genes are strictly conserved.
factors observed in monocot seed storage protein The different loci show significant differences
genes have already been described (Fauteux and in expression level and the y-type present at the
Strömvik 2009; Kawakatsu and Takaiwa 2010). Glu-A1 locus is always silent in hexaploid wheat.
Comparison of promoter sequences revealed
the identity of several cis-elements and common
trans-activating transcription factors involved Materials and methods
in the transcription/expression of wheat storage HMW-GS sequences were collected from the
protein genes. In silico analysis carried out NCBI nucleotide database. The genes involved
on LMW glutenin subunits earlier revealed/ in this promoter profiling were only those where
identified statistical correlations between the at least 490 nucleotide long promoter sequences
promoter profiles of different types of LMW- were available. Promoter motifs were collected
GS genes and their transcription rates (Juhász from published results and from an in-house
et al. 2011). A similar, expressed sequence tags motif collection. Profiling was performed by

software developed in our laboratory. Alignment parameters, optimized by an earlier study
of promoter sequences and the construction of on the expression of LMW-GSs where more
the phylogenetic tree was inferred by MEGA data were available (Juhász et al. 2011). The
software (Tamura et al. 2011). Due to high BLAST results were processed by a judgment
similarities in HMW-GS sequences, calculation of algorithm where all hits were either qualified
expression levels based on EST may be difficult. or disqualified. Because of the high sequence
We therefore developed an electronic hybridization similarities of the query sequences, there were
algorithm specifically for this project/task. many ESTs that showed up in more than one
Publicly available EST data from developing query result. Evidently one EST can represent
wheat seed libraries of cv. Glenlea (NCBI EST only one transcribed allele in the library;
library identifiers #A9H, #A9I and #A9J), or therefore all hits had to be individually judged.
Chinese Spring (#19555, #10469, #19547, #10479) For example if an EST showed up with multiple
were used for the analyses. A BLAST search queries (genes) it was only accounted as a
was performed against EST libraries with each sequence where it had the highest hit identity
HMW-GS gene as a query, applying strict BLAST and longest alignment length.
gb/AJ577815/Glu-1Dx2 0
gb/DQ208971/Glu-1Dx2
gb/AJ577815/Glu-1Dx2
gb/DQ537337/Glu-1Dx
gb/M2220/Glu-1Ax2 Group 3
gb/DQ537335/Glu-1Ax
gb/DQ533690/Glu-1A-2
gb/X12929/Glu-1Dy10?
gb/DQ537337/Glu-1Dy
gb/X03042/Glu-A1-2
gb/AJ566661/Glu-A1-2
gb/AJ566673/Glu-A1-2 Group 1
gb/DQ537336/Glu-1By
gb/EU137874/Glu-1By15
gb/X61026/Glu-1By9
gb/AY486486S1/Glu-1Dy12
gb/AJ567961
gb/X13927/Glu-B1-1b
gb/AJ567950/Glu-B1-1
gb/DQ537336/Glu-1Bx
gb/EU157184/Glu-1Bx7 0
gb/DQ119142/Glu-1Bx7 Group 2
gb/AJ567936/Glu-B1-1
gb/AY367771/Glu-1Bx14
gb/AY553933/Glu-1Bx23
gb/EU157184/Glu-1Bx7
gb/FJ561336/y-type
gb/X12928/Glu-1Dx5?
Fig. 1 Phylogenetic tree of promoter profiles and related gene expressions. The tree shows the result
of multiple alignment of promoter sequences of high molecular weight glutenin subunits (HMW-GS)
that were longer than 490 bp. Their accession numbers and genome locations are shown. Promoter
sequences of HMW glutenins are clustered into groups 1, 2 and 3.

Results and discussion and polymorphisms are different. G2 have a
The alignment of promoter sequences longer than well conserved P-box (TGCAAAC) at -312.
490 nucleotides was used to create a phylogenetic This feature is not fully conserved in the
tree. The topology of the tree provided a other 2 groups. G3 has a variant of the P-box
grouping as shown in Fig. 1. Group 1 has only (TGCAAAAG) at around -417, while G1 has a
y-type HMW-GSs localized on all three genomes; mixture of both locations and polymorphisms.
Group 2 consists of x-type subunit genes from the The most characteristic difference in the motif
B genome, and group 3 includes x-type subunit compositions of the groups is the Cereal-box
genes from the A and D genomes. (C-box). G2 has a well conserved double C-box
(5’- GACATGGTTAGAAGTTTTGAGTG-3’)
The promoter compositions of the three groups at -447 and -501. The “A” genes of G3
show characteristic differences (data not shown have a single C-box with three SNPs: 5’-
due to color restriction). For easier reading, GACATGCTTAGAAGCATTGAGTG-3’. An
promoter compositions of group 1, 2 and 3 expression analysis was carried out for HMW-
will be referred as G1, G2 and G3, respectively. GS sequences belonging to the allele set of both
The 3’-end of the promoter sequences show cultivars. In both cultivars the highest numbers
similarities up to the TATA-box (at -93 nt). of ESTs belong to Bx genes, followed by the D
Almost all sequences belonging to G3 have genome genes (Tables 1 and 2). Ax genes are silent
an AACA/TA (AACAAA) motif upstream in Chinese Spring, but expressed in Glenlea. The
from the TATA-box at -142 nt. At -247, G3, known over-expressing Bx7 (EU157184) gene has
G2 and most G1sequences have an HMW an identical promoter to the normal expressing
enhancer box, with three single nucleotide one (DQ537336), indicating that the difference in
polymorphisms (SNPs). G2 has a sequence of expression level must be mainly due to the gene
5’-GTTTTGCAAGCACCAATTGCTCCTTACTTA duplication (Ragupathy et al. 2008). The different
TCCAGCT-3’ at -245 and most G1 have a variant motif composition of the promoter regions may
of it: 5’-GTTTTGCAAAGCTCCAATTGCTCCTTG be the cause of the different expression levels
CTTATCCAGCT-3’ at -246 (polymorphisms of the genes. SNPs in HMW enhancers may be
highlighted). The D genome genes of G3 contain responsible for a less effective binding capacity,
also the enhancer of G1 but genes located on therefore slowing the transcription rate of the
chromosome 1A have a third variant of the motif: genes in group 3 (Fig. 1). The complete lack of a
5’-GTTTTACAAAGCTCCAATTGCTCCTTGCTT Cereal-box in all sequences of group 2 might also
ATCCAGCT-3’ at -247, also with 3 SNPs but be related to the lower expression levels in y-type
located differently. All three compositions HMW glutenin genes.
have prolamin boxes (P-box) but their locations
Table 1. Numbers of expressed sequence tag (EST) hits of selected sequences in Chinese Spring EST
libraries. 5-30 stand for 5, 10, 20, and 30 DPAs, respectively. The highest expressing allele is Bx7 as
expected. Ax2 has zero expression that is in harmony with earlier findings.
Accession Allele 5 DPA 10 DPA 20 DPA 30 DPA Cultivar

gb|DQ537336 Glu-1Bx7 0 46 11 15 Renan
gi|262205142|tpg|BK006459 Glu-1Dy12 0 34 16 3 Chinese Spring
gi|X03041 Glu-1Dx2 0 20 15 2 Chinese Spring
gb|DQ537336 Glu-1By8 0 10 6 6 Renan
gb|AY486484 Glu-1Dy12 0 8 7 2 Dongnong 7742
gb|X03042 Glu-1Ay pseudo 0 1 0 0 Cheyenne
gb|AJ577815 Glu-1Dx2 0 0 0 0 Unknown

Table 2. Numbers of expressed sequence tag (EST) hits of selected sequences in Glenlea EST libraries.
The highest expressed genes are x-type from genome B, followed by both x and y-types from chromosome
1D. In Glenlea, Ax is expressing. Glenlea 5-25 denotes the libraries at 5, 15, 25 DPAs.
Accession Allele 5 DPA 15 DPA 25 DPA Cultivar

gb|DQ119142 Glu-1Bx7 9 76 17 Glenlea
gb|EU157184 Glu-1Bx7 9 76 17 Glenlea
gb|X12929 Glu-1Dy10 10 45 4 Cheyenne
gb|DQ537337 Glu-1Dx5 6 43 2 Renan
gb|X12928 Glu-1Dx5 3 42 9 Cheyenne
gb|DQ537337 Glu-1Dy10 3 42 9 Renan
gb|M22208 Glu-1Ax2* 6 22 7 Chinese Spring
gb|DQ537335 Glu-1Ax2* 4 20 1 Renan
gb|DQ537336 Glu-1By8 0 9 5 Renan
gb|FJ561336 Glu-1Dy10 1 7 0 Soissons
gb|X03042 Glu-1Ay 0 2 0 Cheyenne
This study was based on publicly available Duffus CM, Cochrane MP (1992) Grain structure and
data and due to the small number of HMW composition. In: Shewry PR (ed) Barley: Genetics,
Biochemistry, molecularbiology and biotechnology.
promoter sequences, our data may not satisfy CAB International, Wallingford, UK, p291-317
the stringency of a robust statistical analysis but
Fauteux F, Strömvik MV (2009) Seed storage protein
we believe it provides insights to the biology gene promoters contain conserved DNA motifs in
behind HMW-GSs expression profiles. Both the Brassicaceae, Fabaceae and Poaceae. BMC plant
promoter profiles and expression data show biology 9:126. http://www.pubmedcentral.nih.gov/
some variability related to alleles. The highest articlerender.fcgi?artid=2770497&tool=pmcentrez&ren
expressed Glu-1B genes have a distinct branch dertype=abstract
of their own on the phylogenetic tree. Alleles at Juhász A, Makai S, Sebestyén E, Tamas L, Balazs E (2011)
Role of conserved non-coding regulatory elements
Glu-1A and Glu-1D loci have different promoter
in LMW-glutenin gene expression. PLoS ONE 6(12):
motif compositions depending on their HMW- e29501. doi:10.1371/journal.pone.0029501
GS type (x or y). Nevertheless due to the fact Kawakatsu T, Takaiwa F (2010) Cereal seed storage
that only 6 HMW-GS genes are present in each protein synthesis: fundamental processes for
cultivar, EST-based expression analysis cannot recombinant protein production in cereal grains. Plant
be directly linked to different promoter profiles. Biotech J 2008:1-15. http://www.ncbi.nlm.nih.gov/
pubmed/20731787 (September 22, 2010)
Ragupathy R, Naeem HA, Reimer E, Lukow OM,
Sapirstein HD, Cloutier S (2008) Evolutionary origin
of the segmental duplication encompassing the
wheat GLU-B1 locus encoding the overexpressed Bx7
(Bx7OE) high molecular weight glutenin subunit.
Tamura K, Peterson D, Peterson N, Stecher G, Nei M,
Kumar S (2011) MEGA5: Molecular evolutionary
genetics analysis using maximum likelihood,
evolutionary distance, and maximum. parsimony
methods. Molecular Biol & Evol 28:2731-2739

Session 3: Improvement of end use qualities by genetic
and alternative approaches
Transferring research findings into industry application

C.F. Morris
USDA-ARS Western Wheat Quality Lab, USA
One objective of wheat quality research is to solve problems for the milling, baking and food
processing industries, and ultimately to benefit consumers. A second objective is to provide these
same industries and consumers new and innovative products and ingredients. In this regard, research
information flows in both directions – up and down the wheat research-processor-consumer chain.
This presentation will focus on two such examples. The first deals with refining an end-use quality
model for soft wheat, the second involves the development of a ‘new’ type of wheat – soft kernel
durum. In the US and elsewhere many ‘soft’ wheat products include low moisture cookies and
crackers. These products may be prepared from ‘lean’ formulas with little more than wheat flour,
water, salt and leavening, to ‘rich’ formulas with high ratios of sugar and sometimes fat. For most
cookies, gluten development is undesirable as it imparts a tough or hard (‘flinty’) product texture.
However, in many instances, the low moisture content of cookie dough and the high formula sugar
content prevent gluten development. Conversely, crackers, most notably ‘saltines’, typically require
some level of gluten development and are prepared from a fermented dough which is sheeted,
laminated and docked (Morris and Rose 1996). These doughs may contain a considerable amount of
hard wheat flour to attain the necessary level of gluten strength. A key feature of cookies and crackers
is their very low moisture content and relatively long shelf life (stability). Consequently, one aspect of
baking is to drive out dough water down to less than about 4% moisture. This requirement places great
importance on dough water relations, especially damaged starch and arabinoxylans. Damaged starch
results mostly from the milling process and can be controlled genetically through the deployment
of the ‘soft’ Puroindoline genes (Pina-D1a/Pinb-D1a). Among soft wheat varieties and germplasm, the
‘carbonate solvent retention capacity test’ (carbonate SRC) is effective in estimating this parameter.
Arabinoxylans (AX), often sub-divided into total and water-extractable fractions, appear to be more
complexly inherited, with an approximately twofold range among Pacific Northwest soft (Finnie et
al. 2006) and hard (Li et al. 2009) wheat varieties. Direct measurement of AX may be accomplished via
phloroglucinol colorimetric (Kiszonas et al. 2012) or gas chromatography-flame ionization detection
(Courtin et al. 2000). Similar to before, the sucrose SRC test is targeted at estimating the effect of AX to
overall water absorption of flours. The last point to be made here is that there is a marked premium
associated with the hard ‘bread’ wheat classes of wheat in the US and elsewhere. For example, local
wheat prices in eastern Washington at the time of this writing were $248/mt soft white vs. $320/
mt for ‘Dark Northern Spring’ (14% protein). Obviously, bakers are willing (forced?) to pay for
gluten strength, and differences in ‘strength’ are implicit features of the US Classes. This generalized
landscape is driving a trend to increase gluten strength in US soft wheat. In the western US, club
wheat is bred to have very weak gluten and thus this sub-class provides an excellent option for
“blending down” strength in soft white. The second example involves the development of soft kernel
durum wheat. We successfully transferred through ph1b-mediated recombination the Hardness locus,
including Pina-D1a/Pinb-D1a, into Langdon durum, and subsequently crossed the soft kernel trait
into the Italian cultivar Svevo (Morris et al. 2011). Why did we do it? 1) Because we could, and 2) to
examine the effect that kernel texture has on durum milling, pasta and bread processing. To date, we
have found no detrimental effects of soft kernel, only neutral-to-positive influences on milling, baking
and pasta processing.
Session 3: Improvement of end use qualiƟes by geneƟc and alternaƟve approaches 83

References Kiszonas, A.M., Courtin, C.M., Morris, C.F. (2012) A
critical assessment of the quantification of wheat grain
Courtin, C.M., Van den Broeck, H., Delcour, J.A. (2000)
arabinoxylans using a phloroglucinol colorimetric
Determination of reducing end sugar residues in oligo-
assay. Cereal Chemistry 89:143-150
and polysaccharides by gas-liquid chromatography.
Journal of Chromatography 866:97-104 Li, S., Morris, C.F., Bettge, A.D. (2009) Genotype and
environment variation for arabinoxylans in hard winter
Finnie, S.M., Bettge, A.D., Morris, C.F. (2006) Influence
and spring wheats of the U.S. Pacific Northwest. Cereal
of cultivar and environment on water-soluble and
Chemistry 86:88-95
water-insoluble arabinoxylans in soft wheat. Cereal
Chemistry 83:617-623 Morris, C.F., Rose, S.P. (1996) Chapter 1. Wheat (In) Cereal
Grain Quality. R.J. Henry and P.S. Kettlewell (eds.),
Chapman Hall, London, pp. 3-54

Wheat quality improvement in China: Progress and prospects
Z.H. He1,2, X.C. Xia1, Y. Zhang1, Y. Zhang1, and X. M. Chen1
1Institute
of Crop Science/National Wheat Improvement Center, Chinese Academy of Agricultural Sciences
(CAAS), 12 Zhongguancun South Street, Beijing 100081, China; 2CIMMYT China Office, C/O CAAS, 12
Zhongguancun South Street, Beijing 100081, China
Abstract
Genetic improvement of wheat quality in China was initiated in the 1980s and significant progress
was made in three aspects during the last ten years: (1) Standardized laboratory testing and evaluation
methods were established for traditional Chinese noodles, and three key traits responsible for noodle
quality were identified. (2) A comparative genomics approach was employed to understand the allelic
variations at loci for low molecular weight glutenin subunits (LMW-GS) and for traits associated with
color for various products; 33 functional markers were developed and validated. Conventional quality
testing and molecular markers were used to characterize major Chinese wheat varieties. (3) Improved
varieties with excellent pan bread-making quality and noodle quality were released and sown on large
areas. Development of new varieties with high yield potential, excellent quality, and broad adaptation
remains a challenge, but it is expected that more markers and new technologies will be used in
breeding programs.
Introduction Significant progress has been achieved in the

China is the largest wheat producer and last ten years. This paper reviews the progress
consumer in the world, with a planting area of achieved on quality improvement, with emphasis
24 million ha and production of more than 110 on standardization of testing methodology
million tonnes. In general, Chinese wheat has for traditional products and development and
acceptable protein content, but has weak gluten validation of functional markers.
quality, and thus is not suitable for mechanized
Noodle quality
food production. Therefore, improvement of
gluten quality is a primary objective for breeding Laboratory noodle preparation and quality
programs targeting quality. Noodles and steamed evaluation was reviewed by He et al. (2010), and
bread make up around 80% of Chinese wheat a brief summary is presented here. To establish
consumption. Bright white color is preferred for a standardized noodle preparation formula,
traditional Chinese products, thus low yellow the effects of flour extraction rate (50%, 60%,
pigment and low polyphenol oxidase (PPO) 70%), added water (33%, 35%, 37%) including
activity are desirable traits. the moisture available in the flour, and salt
concentration (0, 1%, 2%) on color and texture of
To meet the demand for high quality wheat
raw white noodles were investigated using flour
in China, an integrated CAAS-CIMMYT joint
samples from five leading Chinese wheat varieties.
wheat quality improvement program was
The recommended composition for laboratory
established at the National Wheat Improvement
preparation of raw Chinese noodle quality is 60%
Center of the Chinese Academy of Agricultural
flour extraction, 35% water addition, and 1% salt
Sciences (CAAS) in 1997, and integration of
concentration (Ye et al. 2009).
conventional quality testing and breeding,
and molecular technologies was employed A standardized laboratory noodle preparation
to understand and improve Chinese wheat protocol was established (Zhang et al. 2005).
quality, with emphasis on traditional products Desirable attributes of cooked Chinese white
such as noodles. Close collaborations were also noodles include white and bright color, smooth
established with the USDA-ARS Wheat Quality appearance, a medium level of firmness, good
Laboratory at Pullman, Washington, U.S.A. viscoelasticity, a feeling of smoothness in the
and Murdoch University in Western Australia. mouth, and a pleasant taste. Although the official

method for the sensory evaluation of dry white suitable for making Chinese noodles. Major
Chinese noodles (SB/T10137-93, Chinese Ministry traits associated with DWCN quality were
of Commerce, 1993) was released in 1993, it identified, i.e., gluten strength and extensibility,
needed improvement. We therefore adopted starch viscosity, and flour color associated traits,
the evaluation system used in Japan and other although grain hardness and protein content
Asian countries for white-salted noodles, with also have significant effects on noodle quality.
major modifications in the score assigned to each The association between SDS sedimentation
noodle trait (Zhang et al. 2005). In this method, value, farinograph stability, and extensograph
evaluations of elasticity and stickiness were maximum resistance, extension area, and DWCN
combined into viscoelasticity. The weight given score fitted a quadratic regression model. Starch
to each noodle trait was modified according to peak viscosity contributed positively to DWCN
differences in consumer preferences for noodle quality; flour ash content and PPO had moderate
attributes in China vs. Japan. Color was assigned negative effects on noodle color, and protein
a higher weighting (15) than in the previous content and grain hardness were negatively
Chinese method (10), but was lower than in the associated with noodle color, appearance, and
Japanese method (20). The score assigned to smoothness. To summarize, SDS sedimentation
appearance (10) was also lower than that in the value or mixing time from mixograph, RVA peak
Japanese method (15) because this parameter viscosity or flour swelling volume, PPO activity,
is less important for evaluating noodle quality and yellow pigment content, can be used to
in China. The score given to viscoelasticity (30) screen for DWCN quality in the early generations
was lower than the combined value (25+25) of a wheat breeding program (He et al. 2010).
assigned to elasticity and stickiness in the
previous Chinese method. The score given to Molecular marker development and validation
smoothness (15) was higher than in the previous As already stated, improving gluten quality is
method (5) because Chinese consumers believe the primary objective for quality improvement
the sensory mouth feel, including smoothness in China. Composition of high molecular weight
and viscoelasticity, is essential for evaluating raw glutenin subunits (HMW-GS) and low molecular
Chinese noodle quality. weight glutenin subunits (LMW-GS) has an
To improve consistency among panel members, important role in determining gluten quality.
a new scoring method was developed (Zhang HMW-GS has been successfully used in breeding
et al. 2005). In this method, each attribute was programs since they can be easily separated both
classified into seven classes, i.e., excellent, by SDS-PAGE and molecular markers. However,
very good, good, fair, poor, very poor, and selection for LMW-GS in breeding is very limited,
unacceptable, and a score was assigned to each largely due to the non-availability of a simple
class based on comparison with a reference method for separating them. Genomics and
sample at each panel session. The well-known proteomics approaches were employed to develop
commercial Xuehua Flour, with 5% sweet potato new methods for rapid identification of LMW-GS.
starch added to it, showed a relatively good A LMW-GS gene molecular marker system was
and stable noodle quality and each attribute developed and all LMW-GS genes in the Chinese
had a good score. This flour blend was used as wheat core collection were determined by the
a reference sample in our evaluation. Panelists Chinese Academy of Science. This provided a
compared six parameters (i.e., color, appearance, new insight into understanding the gene structure
firmness, smoothness, viscoelasticity, and taste- although it is not economic to use the system in
flavor) and assigned a score to each. breeding programs (Zhang et al. 2011b, c). A set of
standardized varieties for classification of LMW-
A large number of varieties from different parts GS was recommended based on the comparison of
of China, Mexico, and Australia were used to SDS-PAGE, functional makers, MALI-TOF MS and
establish the association between flour traits 2-D methods, through international collaboration
and noodle quality performance. In general, among six institutes from five countries and seeds
flour from medium-hard wheat genotypes with of these genotypes are available upon request (Liu
low ash content, high flour whiteness, medium et al. 2010). Functional markers (gene-specific or
protein content, medium to strong gluten diagnostic) for LMW-GS at Glu-A3 and Glu-B3 loci
type, and good starch viscosity is considered were developed and are widely used (Table 1),

making it possible to select for particular LMW-GS with the insertion have a significantly high level
in wheat breeding programs (Wang et al. 2009b, of yellow pigment than those with Psy-A1a. The
2010). Markers were not developed for separating PCR profiles of YP 7A are associated with yellow
LMW-GS at the Glu-D3 locus as these have little pigment phenotypes.
importance for breeding. MALDI-TOF MS was
At present, 55 functional markers are available for
used for rapid identification of HMW-GS and some
quality improvement (Liu et al. 2012). Thirty three
LMW-GS by workers at Capital Normal University
functional markers at 10 loci were developed and
(An et al. 2006; Liu et al. 2009, 2010), but the high
validated by the CAAS-CIMMYT wheat program
cost of the equipment limits its application in
for manipulation of noodle color and gluten
many breeding programs. Much more work is still
quality in breeding programs (Table 1). In addition
needed to identify additional LMW-GS in current
to these markers, significant progress was also
germplasm.
made to understand the distribution of various
Improving flour color and products is another alleles at the Pina/Pinb locus in Chinese wheats
important objective for noodle and steamed (Chen et al. 2006). These markers (listed in Table 1)
bread quality. A comparative genomics approach together with markers for HMW-GS were used to
was used to clone genes for color-associated characterize more than 300 varieties from China as
traits, to identify allelic variation, and then to well as varieties from various other countries (Liu
develop and validate functional markers derived et al. 2012). The use of these functional markers
from polymorphic sites within genes directly will greatly improve breeding efficiency. For
associated with phenotypic variation. Full-length example, three advanced lines including CA996
genomic DNA sequences were characterized for and CA998 with improved quality and yield
genes associated with noodle color, including potential, developed from our molecular marker
three phytoene synthase genes (Psy-A1, Psy-B1, program, are expected to be released within the
and Psy-D1) (He et al. 2008, 2009; Wang et al. next 1-2 years.
2009a), one gene for zeta-carotene desaturase
(ZDS) controlling yellow pigment content (Zhang Development of quality varieties
et al. 2011a; Dong et al. 2012), three polyphenol A regional quality classification based on climatic
oxidase genes (Ppo-A1, Ppo-B1, and Ppo-D1) (temperature/rainfall), soil type, farming system,
(Sun et al. 2005; He et al. 2007) responsible for use of fertilizers, and quality data collected
color degradation during storage, and one gene in China for the last 15 years was released in
for lipoxynase (LOX) activity (Geng et al. 2012) 2002 (He et al. 2002). In general, three regions
associated with color and processing quality. are recognized; winter and facultative wheats,
The associations of these genes with phenotypic including Zones I and II, which focus on hard
variation were clarified. For example, we isolated white and medium-hard types targeting bread,
and sequenced the Psy 1 gene on chromosome noodle, and steamed bread qualities; autumn-
7A to develop a functional marker to assist sown spring wheat in Zones III, IV, and V, with
breeding of yellow pigment in wheat (He et al. a focus on soft red wheat, although medium-
2008). Sequence analysis of Psy-A1 in wheat hard red types for steamed bread and noodles
revealed a similar intron-exon structure to Psy1 are also recommended. Sprouting tolerance is
homologs in maize, rice, and tomato. Comparing needed in this region because it is a high rainfall
the sequences at alleles Psy-A1a and Psy-A1b, environment. Spring-sown spring wheat in Zones
there was a difference of an insertion of 37 bp VI, VII, and VIII focus on hard and medium-hard
in the second intron of Psy-A1b and differences red types, targeting bread, steamed bread, and
in a number of SNPs. As the particular SNP noodles. Sprouting tolerance is also needed to
mutations did not affect the protein sequence, it is ensure processing quality.
likely that the 37 bp insertion disrupts the intron
structure resulting in an alternative splicing of the Breeding for noodle quality in China was
mRNA and a truncated Psy-A1b protein. A set of reviewed by He et al. (2010). Breeding efforts in
polymorphic primers were designed around the quality improvement started in the late 1980s.
37 bp insertion and a marker named YP 7A was The objectives of wheat quality improvement
developed. The marker was tested on Chinese programs are to combine high yield potential
wheat cultivars and results confirmed that lines and excellent processing quality. Improvement

in gluten strength was the primary objective for varieties with improved quality. In addition
all products including pan bread, noodles, and to the final products testing, gluten strength
steamed bread although color is also important parameters such as HMW-GS composition,
for noodles and steamed bread. Two approaches absence of the 1B/1R translocation, SDS or Zeleny
were employed to improve quality. Firstly, a sedimentation values, farinograph stability, and
significant effort was put into screening current starch parameters, such as flour swelling volume
varieties and advanced lines to identify quality and rapid viscoanalyzer (RVA) peak viscosity,
varieties for production. Secondly, Chinese biochemical and molecular markers for the Wx-
varieties with outstanding pan bread and noodle 7A null allele, and flour color parameters, such
quality, or introductions from USA, Canada, as PPO activity or yellow pigment, are employed
Australia, and CIMMYT, were crossed with to select for desirable noodle quality at various
high-yielding Chinese wheats to develop new
Table 1. Functional markers developed and validated by the CAAS-CIMMYT wheat program.
Trait Locus Marker Allele Chromosome Reference

Polyphenol oxidase activity Ppo-A1 PPO18 Ppo-A1a/Ppo-A1b 2AL Sun et al. 2005
PPO33 Ppo-A1a/Ppo-A1b He et al. 2007
Ppo-D1 PPO16 Ppo-D1a 2DL He et al. 2007
PPO29 Ppo-D1b
Lipoxygenase activity TaLox-B1 LOX16 TaLox-B1a 4BS Geng et al. 2012
LOX18 TaLox-B1b
Yellow pigment content Psy-A1 YP7A PsyA1a/PsyA1b 7AL He et al. 2008
YP7A-2 PsyA1a/PsyA1b/PsyA1c He et al. 2009
Psy-B1 YP7B-1 Psy-B1a/Psy-B1b 7BL He et al. 2009
YP7B-2 Psy-B1c
YP7B-3 Psy-B1d
YP7B-4 Psy-B1e
Psy-D1 YP7D-1 Psy-D1a/Psy-D1g 7DL Wang et al. 2009a
YP7D-2 Psy1-D1a/Psy1-D1g
TaZds-A1 YP2A-1 TaZds-A1a/TaZds-A1b 2AL Dong et al. 2012
Low molecular weight TaZds-D1 YP2D-1 TaZds-D1a/TaZds-D1b 2DL Zhang et al. 2011a
glutenin subunits (LMW-GS) Glu-A3 gluA3a Glu-A3a 1AS Wang et al. 2010
gluA3b Glu-A3b
gluA3d Glu-A3d
gluA3e Glu-A3e
gluA3f Glu-A3f
gluA3g Glu-A3g
gluA3ac Glu-A3a/Glu-A3
Glu-B3 gluB3a Glu-B3a 1BS Wang et al. 2009b
gluB3b Glu-B3b
gluB3c Glu-B3c
gluB3d Glu-B3d
gluB3e Glu-B3e
gluB3g Glu-B3g
gluB3h Glu-B3h
gluB3i Glu-B3i
gluB3bef Glu-B3b/Glu-B3e/Glu-B3f
gluB3fg Glu-B3f/Glu-B3g

stages (He et al. 2010). Outstanding quality Genomic technologies have provided excellent
varieties for noodle and/or pan bread used for tools to understand the molecular basis of
large-scale production after 2000 are listed in various quality traits, and completion of the
Table 2. wheat genomic sequence will speed up gene
cloning and molecular marker development.
Jinnan 17 and Jinmai 20, representative varieties
Integration of conventional breeding and
with excellent pan bread quality and high
molecular markers will greatly promote quality
yield potential, used to be leading varieties
improvement. Development of new varieties
in Shandong province. Yumai 34, conferring
combining high yield potential, excellent quality,
excellent pan bread and noodle qualities, was
and broad adaptation is still challenging. It is
a leading variety in Henan province for about
expected that increasing numbers of markers
10 years. However, the area of quality varieties
and new technologies will be applied in breeding
has been largely reduced over the last few years
programs. Meanwhile, more work is needed to
since lower quality wheats attracted relatively
understand other traditional Chinese products.
high prices. Therefore, quality breeding is not
improvement of quality alone; high quality must
be combined with yield potential and broad
adaptation before farmers will grow them.
References
An X, Zhang Q, Yan Y, Li Q, Zhang Y, Wang A, Pei Y,
Tian Y, Wang H, Hsam SLK, Zeller FJ (2006) Cloning
and molecular characterization of three novel LMW-I
Conclusion and future perspectives glutenin subunit genes from cultivated einkorn
Significant progress has been achieved in (Triticum monococcum L.). Theor Appl Genet 113:383-395
standardizing a methodology for noodle Chen F, He ZH, Xia XC, Xia LQ, Zhang XY, Lillemo
quality evaluation, for development and M, Morris C (2006) Molecular and biochemical
characterization of puroindoline a and b alleles in
validation of functional markers for LMW-GS Chinese wheat landraces and historical cultivars.
and traits associated with product color, and in Theor Appl Genet 112:400-409
development, promotion and adoption of quality Chinese Ministry of Commerce (1993) Noodle flour. SB/
varieties. T10137-93
Table 2. Outstanding quality varieties grown on large areas after 2000.
Variety Cross Quality type Year of release

Zhongyou 9507 8131-1 selection Pan bread 2000
Jing 9428 Jing 411/German Introduction Noodle 2000
Zhongmai 175 Jing 411/BPM 27 Noodle 2007
Gaocheng 8901-11 77546-2/Linzhangmai Pan bread 1998
Jishi 02-1 9411/9430 Pan bread 2004
Jinan 17 Linfen 5064/Lumai 13 Pan bread 1999
Jinmai 19 Linfen 5064/Lumai 13 Noodle 2001
Jinmai 20 Lumai 14/Shandong 84187 Pan bread/noodle 2003
Yumai 34 775-1/Yumai 2 Pan bread/noodle 1994
Yumai 47 Yumai 2/Baiquan 3199 Noodle/pan bread 1997
Yumai 49 Wen 2540 selection Noodle 1998
Zhengmai 366 Yumai 47/PH82-2 Pan bread/noodle 2005

Dong CH, Ma ZY, Xia XC, Zhang LP, He ZH (2012) Allelic Liu YN, He ZH, Appels R, Xia XC (2012) Functional
variation at the TaZds-A1 locus on wheat chromosome markers in wheat: current status and future prospects.
2A and development of a functional marker in Theor Appl Genet 125:1-10
common wheat. J Integrative Agriculture 11:1067-1074 Sun DJ, He ZH, Xia XC, Zhang LP, Morris CF, Appels
Geng HW, He ZH, Zhang LP, Qu YY, Xia XC (2012) R, Ma WJ, Wang H (2005) A novel STS marker for
Development of functional markers for TaLox-B1 on polyphenol oxidase activity in bread wheat. Mol
chromosome 4BS in common wheat. Crop Sci 52:568-576 Breeding 16:209-218
He ZH, Lin ZJ, Wang LJ, Xiao ZM, Wan FS, Zhuang QS Wang JW, He XY, He ZH, Wang H, Xia XC (2009a)
(2002) Classification on Chinese wheat regions based Cloning and phylogenetic analysis of phytoene
on quality performance. J Chinese Agric Sci 35:359-364 synthase I (Psy 1) genes in common wheat and related
(in Chinese) species. Hereditas 146:208-256
He XY, He ZH, Zhang LP, Sun DJ, Morris CF, Fuerst EP, Wang LH, Zhao XL, He ZH, Ma W, Appels R, Peña RJ, Xia
Xia XC (2007) Allelic variation of polyphenol oxidase XC (2009b) Characterization of low-molecular-weight
(PPO) genes located on chromosomes 2A and 2D and glutenin subunit Glu-B3 genes and development of
development of functional markers for the PPO genes STS markers in common wheat (Triticum aestivum L.).
in common wheat. Theor Appl Genet 115:47-58 Theor Appl Genet 118:525-539
He XY, Zhang YL, He ZH, Wu YP, Xiao YG, Ma CX, Xia Wang LH, Li GY, Peña RJ, Xia XC, He ZH (2010)
XC (2008) Characterization of phytoene synthase 1 Development of STS markers and establishment of
gene (Psy1) located on common wheat chromosome multiplex PCR for Glu-A3 alleles in common wheat
7A and development of a functional marker. Theor (Triticum aestivum L.). J Cereal Sci 51:305-312
Appl Genet 116:213-221 Ye YL, Zhang Y, Yan J, Zhang Y, He ZH, Huang S, Quail
He XY, He ZH, Ma W, Appels R, Xia XC (2009) Allelic KJ (2009) Effects of flour extraction rate, added water
variants of phytoene synthase 1 (Psy1) genes and salt on color and texture of Chinese white noodles.
in Chinese and CIMMYT wheat cultivars and Cereal Chem 86:477-485
development of functional markers for flour colour. Zhang Y, Nagamine T, He ZH, Ge XX, Yoshida H, Pena
Mol Breeding 23:553-563 RJ (2005) Variation in quality traits in common wheat
He ZH, Xia XC, Zhang Y (2010) Breeding noodle wheat as related to Chinese fresh white noodle quality.
in China. In: Hou G (ed) Asian Noodles: Science, Euphytica 141:113-120
Technology, and Processing. John Wiley & Sons, Zhang CY, Dong CH, He XY, Zhang LP, Xia XC, He ZH
Hoboken, New Jersey, USA, pp1-24 (2011a) Allelic variation at the TaZds-D1 locus on
Liu L, Wang Al, Appels R, Xia XC, He ZH, Bekes F, Yan wheat chromosome 2DL and their association with
YM, Ma WJ (2009) A MALDI-TOF based analysis of yellow pigment content. Crop Sci 51:1580-1590
high molecular weight glutenin subunits for wheat Zhang XF, Liu DC, Jiang W, Guo XL, Yang WL, Sun JZ,
breeding. J Cereal Sci 50:295-301 Ling HQ, Zhang AM (2011b) PCR-based isolation and
Liu L, Ikeda TM, Branlard G, Peña R, Rogers WJ, Lerner identification of full-length low-molecular-weight
SE, Kolman MA, Xia XC, Wang LH, Ma W, Appels glutenin subunit genes in bread wheat (Triticum
R, Yoshida H, Wang AL, Yan YM, He ZH (2010) aestivum L.). Theor Appl Genet 123:1293-1305
Comparison of low molecular weight glutenin Zhang XF, Liu DC, Yang WL, Liu KF, Sun JZ, Guo XL,
subunits identified by SDS-PAGE, 2-DE, MALDI-TOF- Li YW, Wang DW, Ling HQ, Zhang AM (2011c)
MS and PCR in common wheat. BMC Plant Biol 10:124 Development of a new marker system for identifying
the complex members of the low-molecular-weight
glutenin subunit gene family in bread wheat (Triticum
aestivum L.). Theor Appl Genet 122:1503-151

Transformation of common wheat (Triticum aestivum L.)
with an Avenin-like b gene improves flour mixing
properties
G.Y. He, Y.S. Wang, F.Y. Ma, M. Li, Y. Li, T.T. Li, W. Liu, K.X. Li, and G.X. Yang
The Genetic Engineering International Cooperation Base of Chinese Ministry of Science and Technology, Chinese
National Center of Plant Gene Research (Wuhan) HUST Part, The Key Laboratory of Molecular Biophysics
of Chinese Ministry of Education, College of Life Science and Technology, Huazhong University of Science &
Technology (HUST), Wuhan 430074, China
Abstract
Avenin-like b proteins, a small family of wheat storage proteins, each containing 18 or 19 cysteine
residues, could contribute to the functional properties of wheat flour via inter-chain disulphide bonds.
To clarify the effect of the Avenin-like b proteins on functional properties of wheat flour, functional
and biochemical properties of wheat flour were analyzed using transgenic wheat lines overexpressing
Avenin-like b proteins by Mixograph and size exclusion-high performance liquid chromatography (SE-
HPLC) analyses. Mixograph analysis showed that the overexpression of the Avenin-like b proteins in
two transgenic lines led to improved flour mixing properties. SE-HPLC analysis of the gluten proteins
in wheat flour demonstrated that improved flour properties of transgenic lines were due to increased
proportions of large polymeric proteins compared with the ratios of %F1, %F1 /%F2, (%F3 + %F4)/%F1
and %UPP. These results demonstrated that Avenin-like b proteins could contribute to the functional
properties of wheat flour and provided a basis to confirm the hypothesis that Avenin-like b proteins
could be integrated into polymeric subunits by inter-chain disulphide bonds. This work helps to clarify
the influence and mechanism of Avenin-like b proteins on functional properties of wheat flour.
Introduction and do not correspond to known protein

The viscoelastic properties of wheat dough are sequences. When Avenin-like b proteins were
primarily determined by the glutenin fraction of specifically overexpressed in wheat grain, it was
the gluten proteins (Shewry 2009). However, the still unclear if they could improve the functional
existence of a further class of storage proteins, properties of wheat flour.
named Avenin-like proteins by Kan et al. (2006), In this study, an Avenin-like b gene was
could also contribute to the functional properties introduced into wheat (Triticum aestivum L.
of wheat flour. The distinguishing feature of cv Zhengmai 9023) by particle bombardment.
Avenin-like proteins is that they contain more Mixograph and SE-HPLC analyses were
cysteine residues with the potential to form performed to characterize the biochemical and
intra-chain and inter-chain disulphide bonds functional properties of wheat flours of two
(Kan et al. 2006; Chen et al. 2008). The molecular transgenic wheat lines.
mass of Avenin-like a proteins is about 18 kDa;
each Avenin-like a protein contains 14 cysteine
residues (Kan et al. 2006). Because of the high Materials and methods
homology in sequence similar to a previously
reported “low-molecular weight gliadin” Plant materials
monomer, it may be assumed that cysteine Common wheat (T. aestivum L.) Zhengmai 9023
residues of Avenin-like a proteins mediate seven is a commercial wheat cultivar of the lower
intra-chain disulfide bonds (Clarke et al. 2003). central Yangzi River region. It contains four high
In contrast, Avenin-like b proteins contain 19 molecular weight glutenin subunits (HMW-GS),
(typ-b1 and -b2) or 18 (typ-b3) cysteine residues, viz. Bx7, By8, 1Dx2 and Dy12.

Vector construction Mixograph analysis
Plasmid pLRPT was used for the construction of Dough mixing properties were determined with
the transgenic expression cassette. Primers were a 10 g Mixograph (National Manufacturing Co.,
designed to amplify the avenin-like b gene from Lincoln, NE) based on AACC method 54-40A.
genomic DNA of Zhengmai 9023. The amplified Mixing was carried out in triplicate.
products were ligated to cohesive termini of
linearized pLRPT resulting in insertion between SE-HPLC
the endosperm-specific 1Dx5 promoter and the The soluble proteins and insoluble proteins from
CaMV35S terminator. The recombinant vector flour for SE-HPLC analysis were extracted using
named pLRPT-avel was co-bombarded with the the method described by Tosi et al. (2005). Three
plasmid pAHC25 containing the β-glucuronidase replicate separations were performed on each
(uidA) gene and the selectable marker gene bar flour sample. The proportion of total polymeric
that confers tolerance to phosphinothricin (PPT) proteins (%UPP) was determined according to
under the control of the maize ubiquitin promoter. Gupta et al. (1993).
Genetic transformation and plant regeneration

The transformation procedure was performed Results and discussion
following the method of Sparks and Jones (2004).
Immature scutella of cv. Zhengmai 9023 were Production of transgenic wheat plants
used as targets for transformation by particle The Avenin-like b gene sequence in this research
bombardment with the plasmids pLRPT-avel and was 855 bp long and encoded a protein with
pAHC25. Plants were regenerated and selected in 284 amino acid residues containing 18 cysteine
the presence of PPT (3 mg/l). residues, and grouped into Avenin-like b (typ-b3)
protein. Positive transgenic T0 plants were
PCR and Southern blotting confirmed by PCR to amplify the uidA gene and
As the uidA gene and CaMV35S terminator CaMV35S terminator sequence (Fig. 1A, B). A
were the only sequences in the pAHC25 and total of 30 positive transgenic T0 plants were
pLRPT-avel vectors, respectively, without any obtained. Southern blotting analysis showed
similarity to the genomic DNA of common that the selected T0 transgenic wheat lines
wheat, PCR amplifications for the uidA gene and contained relatively simple insertion sites (Fig.
the CaMV35S terminator were used to confirm 1C), in having a single band on the blot, with
the transgenic plants. Southern blotting analysis exception of lane 4 which had no hybridization
was performed to determine the transgene band. The banding patterns in lanes 5 and 6
copy numbers in T0 plants. The membrane was were very similar. The banding patterns in lanes
hybridized with a DIG-labeled 417 bp probe 3 and 7, however, were different, confirming
generated by PCR using primers designed to the that the plants were derived from independent
CaMV35S terminator sequence. transformation events and could be therefore
considered as independent lines.
SDS-PAGE and western blotting analysis After analysis for the presence and expression
Total proteins of wheat seeds were extracted of transgenes by PCR, SDS-PAGE and western
and separated by SDS-PAGE. Western blotting blotting over four generations (T0-T3), two lines
was performed using a rabbit anti-avenin-like (M3 and M6) considered homozygous for the
b protein polyclonal antibody raised from the foreign genes were used for further analysis.
recombinant Avenin-like b proteins expressed The contents of Avenin-like b proteins in the two
in E. coli. Housekeeping protein GAPDH transgenic lines were increased compared to non-
antibody was used to normalize equal amounts transformed Zhengmai 9023 (Fig. 2A, B). Western
of proteins across samples and to calculate the blotting showed that the relative contents of
relative content. The relative contents of Avenin- Avenin-like b proteins in the transgenic (M3
like b proteins in the transgenic wheat plants, and M6) lines were increased 1.7- and 1.9-fold,
compared to the non-transformed line, were respectively, compared to the non-transformed
measured by western blotting analysis. line, as calculated by densitometry (Fig. 2C).

M 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7
1,200 bp
A
M 1 2 3 4 5 6 7 8 9 10 11 12
400 bp
B) C)
Fig. 1 PCR (A and B) and Southern blotting analysis (C) of the transgenic plants. A & B, PCR
amplification of the uidA gene (A) and CaMV35S terminator sequence (B). Lane M, DNA Marker III (A)
or Marker I (B); lane 1, plasmid pAHC25 (A) or pLRPT-avel (B) for positive control; lane 2, water for
negative control; lane 3, DNA of Zhengmai 9023 for negative control; lanes 4-12, DNA of regenerated
plants. The samples in lanes 4-12 of Fig. 1B correspond to the samples in lanes 4-12 in Fig. 1A. (C)
Southern blotting analysis of the transgenic plants. Lane 1, positive control of pLRPT-avel digested with
BamHI; lane 2, genomic DNA of Zhengmai 9023 digested with BamHI and HindIII; lanes 3-7, genomic
DNA of transgenic plants digested with BamHI and HindIII. The PCR results for lanes 3-7 in C are
shown in 1B lanes 5, 6, 7, 9 and 12, respectively.
A) kDa M 1 2 3
Fig. 2 SDS-PAGE (A) and Western blotting analyses (B)
95
of gluten proteins extracted from flours of transgenic and
72 non-transformed plants. (A) Lane M, Protein marker; lane
1, Zhengmai 9023; lane 2, line M3; lane 3, line M6. The
arrow indicates the position of the transgenic Avenin-like
55 b proteins. (B) Lane 1, Zhengmai 9023, lane 2: line M3;
lane 3, line M6. Housekeeping protein GAPDH was used
43 as control to normalize for equal amounts of protein and
to calculate the relative content. (C) Relative amounts of
the Avenin-like b proteins in the transgenic plants were
densitometrically quantified with respect to the non-
34 transformed plants.
26
1 2 3
B)
Aveni
GAPI
C) 4
Relative Expression
(Arbitrary units)
3
2
1
0
1 2 3

Mixing properties analysis
A) Zhengmai 9023
A number of parameters of the Mixograph 90
curve can be measured, including mixing time
(MT), peak resistance (PR) (both positively 70
Resistance (AU)
related to strength), resistance to breakdown
(RBD) (positively related to stability), maximum 50
bandwidth during the mixing (MBW), the
bandwidth of midline after mixing time (MRW) 30
and bandwidth at peak resistance (BWPR) (all
positively related to resistance to extension) (Tosi 10
et al. 2005). As shown in Table 1, the PR, MRW 0
and BWPR of transgenic wheat lines M3 and M6 0 1 2 3 4 5 6 7 8 9 10
Time (min)
were increased relative to the non-transformed B) M3
line, suggesting that extensibility of the 90
transgenic wheat dough was improved. Based on
the RBD (Table 1) the stability of the transgenic 70
wheat dough was improved. The transgenic Resistance (AU) 50
lines showed a significant increase in PR, which
together with decreases in RBD indicated greater 30
dough strength and stability. In addition, slight
positive changes were observed in MT. 10
0
0 1 2 3 4 5 6 7 8 9 10
SE-HPLC Time (min)
SE-HPLC analysis can be used to fractionate C) M6
90
gluten proteins based on molecular mass without
reduction of the inter-chain disulphide bonds
70
Resistance (AU)
that stabilizes the glutenin polymers. Based on

the SE-HPLC analysis (Table 2), overexpression 50
of Avenin-like b proteins in the transgenic lines
had significant effects on the proportion of the 30
polymeric gluten protein fractions. It is known
that large polymeric proteins are associated with 10
the degree of cross-linking in the seed storage 0
protein. The results of SE-HPLC analysis also 0 1 2 3 4 5 6 7 8 9 10
Time (min)
indicated that the Avenin-like b proteins affected
the degree of cross-linking. Based on these Fig. 3 Mixograph curves of the dough of transgenic
results, increased Avenin-like b protein contents wheat lines M3 and M6 and non-transformed
had significant effects on the proportion of Zhengmai 9023.
polymeric proteins.
Table 1. The 10-g Mixograph parameters for flours of transgenic lines M3 and M6 and non-transformed
Zhengmai 9023.
Bandwidth of Maximum
Mixing Peak Resistance to Bandwidth at midline after bandwidth
time resistance breakdown peak resistance mixing time during mixing
Flour (min) (AU) (%) (AU) (AU) (AU)
M3 3.46±0.04 45.67±0.78b 14.44±0.67b 26.44±0.65b 28.98±0.66b 31.73±1.26b
M6 3.56±0.04 46.16±0.67b 13.16±0.44b 24.92±0.48b 25.77±2.27b 35.91±3.5b
Zhengmai 9023 3.42±0.09 40.28±0.14a 16.42±0.76a 17.2±0.43a 18.62±1.81a 21.17±0.21a
LSD0.05 NS 2.07 1.98 1.82 5.96 7.44

Table 2. Molecular size distribution of gluten proteins in flours of transgenic and non-transgenic wheat
lines determined by SE-HPLC.
Flour sample %F1 %F1/%F2 (%F3+%F4)/%F1 %UPP

M3 46.31±1.63b 2.38±0.11b 0.74±0.04b 43.53±1.32b
M6 47.29±0.63b 2.6±0.15b 0.73±0.03b 44.1±1.54b
Zhengmai 9023 36.64±0.3a 2.11 ±0.17a 1.19±0.03a 35.57±0.41a
LSD0.05 2.17 0.29 0.14 2.2
We clearly demonstrated that lines testing PCR- Chen P, Wang CD, Li KX, Chang JL, Wang YS, Yang GX,
positive for the CaMV35S terminator were also Shewry PR, He GY (2008) Cloning, expression and
characterization of novel avenin-like genes in wheat
positive for transgene expression, unequivocally and related species. J Cereal Sci 48:734-740
determined by using anti-Avenin-like b proteins
Clarke B, Phongkham T, Gianibelli M, Beasley H, Bekes F
polyclonal antibody. (2003) The characterisation and mapping of a family of
After analysis for the presence and expression LMW-gliadin genes: Effects on dough properties and
bread volume. Theor Appl Genet 106:629-635
of the transgene by PCR, Southern blotting,
Gupta RB, Khan K, MacRitchie F (1993) Biochemical
SDS-PAGE and western blotting in four
basis of flour properties in bread wheats. I. Effects
generations (T0-T3), two transgenic wheat lines of variation in the quantity and size distribution of
overexpressing the Avenin-like b gene were polymeric protein. J Cereal Sci 18:23-41
studied. Mixograph analysis showed that the Kan YC, Wan YF, Beaudoin F, Leader DJ, Edwards K,
increased content of Avenin-like b protein Poole R, Wang DW, Mitchell RAC, Shewry PR (2006).
in transgenic lines improved flour mixing Transcriptome analysis reveals differentially expressed
properties. Analysis of gluten proteins extracted storage protein transcripts in seeds of Aegilops and
wheat. J Cereal Sci 44:75-85
from wheat flour demonstrated that this was
associated with an increased proportion of large Shewry PR (2009) Wheat. J Exp Bot 6:1537-1553
and medium-sized glutenin polymers, with the Sparks CA, Jones HD (2004) Transformation of wheat
by biolistics. In: Curtis IS (ed) Transgenic crops of
Avenin-like b proteins overexpressed in the
the world-essential protocols. Kluwer Academic
transgenic wheat plants. Publishers, Dordrecht. pp. 19-34.
Tosi P, Masci S, Giovangrossi A, D’Ovidio R, Bekes F,
Larroque O, Napier J, Shewry PR (2005) Modification
of the low molecular weight (LMW) glutenin
composition of transgenic durum wheat: effects on
glutenin polymer size and gluten functionality. Mol
Breed 16:113-126

Update on low-molecular-weight glutenin subunit
identification
T.M. Ikeda1, W.J. Rogers2, G. Branlard3, R.J. Peña4, S.E. Lerner5, A. Arrigoni5, W.J. Ma6, R. Appels6, O.
Lukow7, W. Hurkman8, M. Appelbee9, M. Sissons10, J.M. Carrillo11, and Z.H. He12
1NARO, WARC, Fukuyama, Japan; 2CIISAS, CICPBA-BIOLAB AZUL, Facultad de Agronomía, Azul,
UNCPBA, Argentina; CONICET INBA - INBIOTEC; 3INRA UMR1095 GDEC -UBP, Clermont-Ferrand,
France; 4CIMMYT, El Batan, Mexico; 5CRESCAA, Facultad de Agronomía, Azul, UNCPBA, Argentina;
6Western Australia Department of Agriculture and Food, State Agriculture Biotechnology Center, Murdoch
University, Murdoch 6150, Western Australia; 7Agriculture and Agri-Food Canada, Cereal Research Center,
Winnipeg, Canada; 8USDA Agricultural Research Service, Western Regional Research Center, Albany, USA;
9South Australian Research and Development Institute, Adelaide, and LongReach Plant Breeders, Lonsdale
5160, South Australia; 10Tamworth Agricultural Institute, Calala, NSW 2340, Australia; 11Unidad de Genética,
ETSIA, Madrid, Spain; 12Institute of Crop Science, National Wheat Improvement Center/The National Key
Facility for Crop Genetic Resources and Genetic Improvement, Chinese Academy of Agricultural Sciences,
Beijing, and CIMMYT China Office, Beijing, China
Abstract
Allelic variation for the low-molecular-weight glutenin subunits (LMW-GS) is a major determinant
of differences in dough viscoelastic properties observed between cultivars of both bread wheat and
durum. Technical difficulties in allelic identification due to the complexity of the protein profile
produced by each cultivar and the use of different nomenclature systems in different laboratories
has historically interfered with information exchange between research groups, a situation
exacerbated by the vast number of possible profiles found in different cultivars due to the multi-
allelic nature of the principal loci encoding LMW-GSs (Glu-A3, Glu-B3 and Glu-D3). These various
difficulties prompted research workers at CAAS, CIMMYT, INRA, NARO and UNCPBA to form an
international collaborative group aimed at unifying criteria across laboratories and comparing four
different methods of allelic identification (SDS-PAGE, 2-DE, MALDI-TOF-MS and PCR). The current
contribution summarizes progress to date made by this group in studies on bread wheat, seeks to
address remaining challenges and places the findings in the context of the wheat gene catalog. We
also propose the formation of a wider international group aimed at facilitating the resolution of the
remaining problems in the field.
Introduction observed in protein electrophoresis in which

Allelic identification of the low molecular different alleles frequently produce bands of
weight glutenin subunits (LMW-GS), major very similar mobility. Added to these technical
determinants of genetic differences in quality in difficulties is the fact that different nomenclature
wheat, has historically been highly problematic, systems have been used in different laboratories
far more so than that of the other major group and that the Glu-3 loci are multi-allelic, the latter
of glutenins influencing quality, the high of which means there is an enormous number
molecular weight glutenin subunits (HMW-GS). of protein profiles observed in germplasm
Although both groups are encoded by series collections. Indeed, these proteins were used to
of complex homoeologous loci (Glu-1 in the assist discrimination between cultivars in such
case of the HMW-GS and principally Glu-3 in collections; for example, Lerner et al. (2009)
the case of the LMW-GS), the protein profile provisionally found 93 allelic combinations in a
associated with each cultivar is considerably collection of 119 Argentinean cultivars for these
more complex for the LMW-GS than for the and other protein groups, a figure that, due to
HMW-GS: multiple overlapping bands are the above technical difficulties, may well be an
underestimation.

According to work published by our group (Liu et SDS-PAGE (one-dimensional sodium dodecyl
al. 2010), the following ambiguities have resulted sulphate - polyacrylamide gel electrophoresis)
from the different classification systems hitherto was performed in all five laboratories, with
available: 1) at the Glu-A3 locus, both Glu-A3a differences across them in three methodological
and Glu-A3c were reported for the same cultivar aspects: (i) the concentrations of separation gel
and Glu-A3a, Glu-A3b, Glu-A3c and Glu-A3d were were 14.0% concentration (T) with 1.3% cross
reported to be identical to Glu-A3e; 2) at the Glu-B3 linker (C), 15.0% T with 1.3% C, 12.5% T with
locus, results differed for Glu-B3b and Glu-B3g, 0.97% C, 15.0% T with 1.4% C, and 13.5% T with
and for Glu-B3f and Glu-B3g in the same cultivars; 0.8% C in the laboratories of CAAS, CIMMYT,
and 3) at the Glu-D3 locus, there was uncertainty INRA, NARO and UNCPBA, respectively;
for Glu-D3a and Glu-D3c, and for Glu-D3a and (ii) the pH for separation gel was pH 8.8 in all
Glu-D3b, in some cultivars. As a consequence of laboratories except CIMMYT with pH 8.5; and
these problems, reports of correlations between (iii) running gel currents were 16, 12.5, 30, 30 and
certain allelic forms of LMW-GS and quality 40 mA in CAAS, CIMMYT, INRA, NARO and
parameters in common wheat have often been UNCPBA, respectively. LMW-GS compositions
contradictory (see Results and discussion), which were identified according to Singh et al. (1991)
points towards the need for a simple and uniform and Jackson et al. (1996) and the gliadins were
classification system and for a set of standard used as indicators of LMW-GS based on the
cultivars for each LMW-GS allele. These various linkage between LMW-GS and gliadin because
difficulties prompted research workers at CAAS, the gliadin composition can be screened more
CIMMYT, INRA, NARO and UNCPBA to form readily than specific LMW-GS in some cases.
an international collaborative group aimed at The nomenclature system of LMW-GS followed
unifying criteria across laboratories and defining Gupta and Shepherd (1990), Jackson et al. (1996),
such a standard cultivar set, and the current Branlard et al. (2003), Ikeda et al. (2008), Appelbee
contribution summarizes progress made by et al. (2009) and the catalog of gene symbols for
this group to date, seeks to address remaining wheat (McIntosh et al. 2011).
challenges and places the findings in the context
2-DE (two-dimensional gel electrophoresis by
of the wheat gene catalog (McIntosh et al. 2011).
(IEF) isoelectric focusing x SDS-PAGE) was
only performed at CAAS and NARO, with
the protocols and differences described in Liu
Materials and methods et al. (2010). At least three gel images of each
Our studies to date have been based upon a sample were taken and compared. The LMW-
worldwide collection of bread wheat cultivars GS compositions were identified by distinctive
widely used in glutenin subunit composition spot patterns according to Ikeda et al. (2006). The
studies and relationships with processing quality, nomenclature system of the LMW-GS was the
which were analyzed in different laboratories for same as the above SDS-PAGE separation. In some
LMW-GS allelic composition with each laboratory cases the 2-DE was run without glutenin protein
applying its own particular range of techniques. alkylation.
The collection consisted of 103 cultivars from 12
countries (21 from China, 19 from Argentina, 15 MALDI-TOF-MS (matrix assisted laser desorption/
from Australia, 14 from France, 10 from Japan, ionization time-of-flight mass spectrometry) was
8 from Mexico, 7 from Canada, 3 from the USA, performed at the State Agriculture Biotechnology
2 from Italy, 2 from the Netherlands, 1 from Center, Murdoch University, Australia, using the
Finland, and 1 from Germany). protocol described by Liu et al. (2010); sample
preparation was carried out according to the
Protein extraction was similar in all five dried droplet method, using sinapinic acid as
laboratories and involved sequential extraction matrix and analyses performed on a Voyager
from wholemeal flour based upon, for gliadins, DEPRO TOF mass spectrometer (Applied
propanol-1-ol, and, for glutenins, a propanol-1-ol, Biosystems, Foster City, CA, USA) equipped with
Tris-HCl solution containing 1% w/v dithiothreitol a 337 nm nitrogen laser and delayed extraction.
(DTT) and subsequently 4-vinylpyridine. Details Identification of LMW-GS alleles was established
are found in Liu et al. (2010). using a set of 19 near-isogenic lines of cultivar
Aroona (Wang and Appels, unpublished data).

PCR (polymerase chain reaction) was performed band was usually lighter and thinner; (c) allele
only at CAAS; genomic DNA was extracted Glu-B3f could not be readily distinguished from
from seeds using a modified CTAB procedure. Glu-B3g, although a gliadin band that differed
Details of allele-specific markers for the between the two could be used to discriminate
discrimination of Glu-A3 and Glu-B3 alleles and them; and (d) alleles classified as Glu-B3b,
PCR conditions were reported previously by Glu-B3g and Glu-B3i were often identified as
Wang et al. (2009, 2010). Glu-B3ab, Glu-B3ac and Glu-B3ad by 2-DE,
Bands of Glu-D3 can be faintly stained and not
always easily distinguished, although technical
Results and discussion improvements have often allowed discrimination
Results obtained from the different laboratories of, for example, Glu-D3a, Glu-D3b and Glu-D3d.
were compared in order to understand the As previously mentioned, ambiguities such as
basis of existing differences in interpretation. these imply that recourse to 2-DE, MALDI-TOF-
Minor differences in the conditions used during MS or PCR is often required.
SDS-PAGE resulted in separations that were
sufficiently different to result in contrasting 2-DE
conclusions regarding the allelic composition of
At Glu-A3, alleles Glu-A3d and Glu-A3g,
the cultivars; hence the other three procedures
which could only be reliably distinguished
(2-DE, MALDI-TOF-MS and PCR) were applied
by using linked gliadins in one dimensional
in order to refine the analyses. For example,
electrophoresis (see above), could be
in the collection of cultivars analyzed, all
distinguished from each other by 2-DE.
alleles observed at the Glu-A3 locus could be
distinguished by 2-DE or PCR, whereas there At Glu-B3, alleles Glu-B3a and Glu-B3b, which
were specific cases where particular alleles could were difficult to distinguish in SDS-PAGE, could
not be distinguished by 1-DE or MALDI-TOF- be distinguished in 2-DE. New alleles Glu-B3ab,
MS. At the Glu-B3 locus, some alleles were clearly Glu-B3ac and Glu-B3ad, which have an additional
identified by all four methods, whereas only spot, were identified by Ikeda et al. (2009).
2-DE was effective for others. At the Glu-D3 locus, At Glu-D3, alleles Glu-D3c and Glu-D3l could be
allelic identification was at times problematic for distinguished.
1-DE, 2-DE and PCR; one of the marker bands
detected by SDS-PAGE was not a LMW-GS, but Nonetheless, there were Glu-3 alleles that were
a gliadin that contaminated the glutenin fraction; not readily distinguished by this method.
some of the alleles at this locus were detected Some of them were gliadin alleles, which were
only by SNPs in PCR; and it was found that contaminated in glutenin fractions.
MALDI-TOF-MS has considerable potential for
allelic identification at this locus. MALDI-TOF-MS
The following specific results were obtained in At all three Glu-3 loci, cases were identified of
the set of cultivars studied: alleles difficult to distinguish in SDS-PAGE, but
were readily distinguished by MALDI-TOF-MS;
SDS-PAGE for example, at Glu-A3, alleles Glu-A3e and Glu-
A3f; at Glu-B3, alleles Glu-B3f and Glu-B3g; and at
At Glu-A3, (a) alleles Glu-A3a, Glu-A3b, Glu-A3c Glu-D3, alleles Glu-D3b and Glu-D3c.
and Glu-A3f could be readily distinguished, but
it was difficult to distinguish Glu-A3e (null) and PCR
Glu-A3f since both tended to be identified as null;
and (b) alleles Glu-A3d and Glu-A3g could only PCR made valuable contributions to allelic
be distinguished by the gliadin encoded by Gli- identification. For example, in this collection of
A1o linked to Glu-A3d. cultivars, all alleles thought to be present at the
Glu-A3 locus could be distinguished (Glu-A3a
At Glu-B3, (a) alleles Glu-B3d, Glu-B3h and Glu- to Glu-A3g); and of the Glu-B3 alleles, only three
B3i each carried slow bands that were not always (Glu-B3ab, Glu-B3ac and Glu-B3ad) do not yet
easy to distinguish; (b) allele Glu-B3b almost have allele specific primers.
coincided with Glu-B3a, although the Glu-B3b

Liu et al. (2010) provide a table summarizing the best combinations for Glu-3 were bbb, bbc and
the allelic variants that can be distinguished by cbc. Branlard et al. (2001) also compared allelic
the four different methods. Furthermore, they effects on quality parameters, finding that, for
give the following standard set of 30 cultivars dough strength, the rankings were as follows: at
that include all allelic variants identified in this Glu-A3: a=d=f≥e, at Glu-B3: b’≥d=c=c’=b=g>i>f≥j and
study: at locus Glu-A3, allele a: Neixiang 188, at Glu-D3: a≥b=d=c. For extensibility at Glu-A3:
Chinese Spring; b: Gabo, Pavon 76; c: Pitic, Seri d=a=f≥e, at Glu-B3: i≥b’≥c=c’=g>b=f=d>j, and, at
82; d: Nidera Baguette 10, Cappelle-Desprez; Glu-D3, no significant differences were found (see
e: Amadina, Marquis; f: Kitanokaori, Renan; the wheat gene catalog for an explanation of the
g: Bluesky, Glenlea; at locus Glu-B3, allele a: allelic terminology used in this study). Luo et al.
Chinese Spring; b: Renan, Gabo; c: Insignia, (2001) found that, in New Zealand cultivars: (i)
Halberd; d: Pepital, Ernest; f: Fengmai 27; g: the Glu-A3 alleles ranked: d>c=e, coinciding with
Splendor, Cappelle-Desprez; h: Aca 303, Pavon Gupta et al. (1990a) for Rmax; (ii) the Glu-B3 alleles
76; i: Norin 61; j: Grebe, Seri 82; ab: Nanbu- ranked: b>g, which coincides both with Gupta et
komugi; ac: Thesee, Aca 801; ad: Heilo, Opata al. (1990a) and Cornish et al. (1993); and (iii) the
85; at locus Glu-D3, allele a: Chinese Spring, Glu-D3 alleles ranked: b>a.
Neixiang 188; b: Gabo, Avocet; c: Insignia,
Thus it can be seen from these studies that not all
Cappelle-Desprez; m: Darius; l: Amadina, Heilo;
the published allelic rankings are consistent, hence
n: Fengmai 27. Of these, four cultivars (Chinese
the need for the aforementioned tools and further
Spring, Opata 85, Seri 82 and Pavon 76) are
work on this general research area.
recommended as a core set to be included in
each SDS-PAGE gel when identifying LMW-
GS alleles. The 30 cultivars have been placed in Conclusions
the germplasm banks of CIMMYT, Mexico, and In general, it was shown that the four methods
INRA Clermont Ferrand, France, with the aim of LWM-GS identification can be regarded as
of making the set publicly available. Seeds of the complementary and, together, provide powerful
cultivars are available by request. tools for allelic identification. As an aid to allelic
Use of these tools should help to resolve the identification across laboratories, a standard set
contradictions between different published of cultivars was defined to represent all allelic
reports of correlations between certain allelic variants in the collection analyzed, and the results
forms of LMW-GS and quality parameters in of this study have been coordinated with those
bread wheat, referred to in the Introduction. presented in the wheat gene catalog (McIntosh
Such contradictions can be discerned from the et al. 2011) in order to ensure consistency in
following summary of some of these reports. the availability of information to the wheat
community.
In Australian cultivars (Gupta and Shepherd
1988; Gupta et al. 1989, 1990a,b, 1991, 1994; Although there are currently over 90 described
Gupta and MacRitchie 1991; Metakovsky et al. alleles at the three Glu-1 loci, only a handful have
1990), for Rmax (maximum dough resistance), been assessed for quality. We believe that only
the Glu-A3 alleles ranked b>d>e>c, the Glu-B3 a true collaboration between many groups will
alleles ranked i>b=a>e=f=g=h>c and the Glu-D3 provide sufficient resources to allow all or at least
alleles ranked: e>b>a>c>d. The allele b of both the majority of allelic variants to be evaluated, and
Glu-A3 and Glu-D3 seemed to be associated hence we should also like to propose the formation
with more extensible genotypes. Cornish et al. of a wider International Gluten Research Group,
(1993) found that the Glu-3 allelic pattern bbb (at as a part of the G20 Wheat Initiative (previously
Glu-A3, Glu-B3 and Glu-D3, respectively) gave the International Research Initiative for Wheat
the best extensibility, especially when combined Improvement). The group would be aimed at
with the Glu-1 pattern bba (at Glu-A1, Glu-B1 and achieving the following:
Glu-D1, respectively). Glu-3 bbc also had excellent • Sharing materials and methods, and studying
extensibility. They also concluded that Glu-A3e together to improve wheat quality.
was detrimental to extensibility by virtue of being
null and that Glu-B3 c, d and g had medium to • Unifying the nomenclature of gluten protein
weak dough properties. They suggested that alleles of bread and durum wheat (Glu-1, Glu-3,
Gli-1 and Gli-2, etc.).

• Unifying the standard methods to analyze • Studying the usefulness of particular gluten
gluten proteins and genes (SDS-PAGE, 2-DE, compositions for various food products by
PCR and MS). evaluating landraces and modern cultivars
for quality under various environmental
• Identifying standard lines and wheat cultivars
conditions considering the unpaired influence
from worldwide origins to form reference sets
of drought and heat stress associated with
as genetic stocks representing diverse gluten
climate change on wheat grain yield and
protein alleles.
quality attributes.
• Storing the reference sets in national and
A flow diagram summarizing the project is
international seed banks to maintain and
facilitate the availability of the existing range provided as Fig. 1. We believe the project is
of biodiversity in wheat (near-isogenic lines, important since, with increasing globalization,
landraces and modern cultivars). increasing urban populations, and the demand
for more healthy and convenient wheat-based
• Exchanging and producing materials having foods, there is a need for the development of
a unique gluten protein composition (near- wheat cultivars with more specific grain quality
isogenic lines and mapping populations). attributes.
Gluten study
Importance Increasing needs for the development of wheat cultivars with more specific grain
quality attributes.
Gluten proteins plays a major role in determining the end use of wheat under
different environmental conditions.
The degree of complexity of gluten proteins has led to the misclassification of
several alleles.
Problems Identification methods of gluten alleles are different among laboratories.
Stand cultivars to identify the gluten alleles are not shared.
We use different wheat growing conditions to evaluate the effects of particular
alleles on end-use properties.
It is difficult to evaluate Glu-3 allelic effects without considering tightly linked Gli-1
alleles, which seem to contribute dough extensibility.
Further international collaboration for gluten research

Proposal of International Gluten Research Group as a part of IRIWI (International
Research initiative for Wheat Improvement).
Activities Exchanging and sharing valuable wheat cultivars.
Unifying the nomenclature of gluten protein alleles and the standard methods to
examine gluten proteins.
Studying the usefulness of particular gluten compositions for quality under

various environmental conditions.
Goal
Development of wheat cultivars with specific grain quality attributes under
different environmental conditions, including heat and drought stresses.
Fig. 1 Flow diagram of the project proposed by the International Gluten Research Group.

References Ikeda TM, Araki E, Fujita Y, Yano H (2006)
Characterization of low-molecular weight glutenin
Appelbee MJ, Mekuria GT, Nagasandra V, Bonneau JP,
subunit genes and their protein products in common
Eagles HA, Eastwood RF, Mather DE (2009) Novel
wheats. Theor Appl Genet 112:327-334
allelic variants encoded at the Glu-D3 locus in bread
wheat. J Cereal Sci; 49: 254-261 Ikeda TM, Branlard G, Peña RJ, Takata K, Liu L, He
Z, Lerner SE, Kolman MA, Yoshida H, Rogers WJ
Branlard G, Dardevet M, Saccomano R, Lagoutte F,
(2008) International collaboration for unifying Glu-3
Gourdon J (2001) Genetic diversity of wheat storage
nomenclature system in common wheats. In: Appels
proteins and bread wheat quality. Euphytica
R, Eastwood R, Lagudah E et al. (eds) Proc 11th Int
2001;119:59-67
Wheat Genet Symp, Sydney University Press, Sydney,
Branlard G, Dardeve, M, Amiour N, Igrejas G (2003) NSW, Australia, pp130-132
Allelic diversity of HMW and LMW glutenin subunits
Ikeda TM, Branlard G, Peña RJ, Takata K, Liu L, He
and omega-gliadins in French bread wheat (Triticum
Z, Lerner SE, Kolman MA, Rogers WJ (2009)
aestivum L.). Genet Resour Crop Evol 50:669-679
Characterization of New Glu-3 alleles in bread wheat.
Cornish GB, Burridge PM, Palmer GA, Wrigley CW In: Gluten Protein 2009, Proceedings of the 10th
(1993) Mapping the origins of some HMW and LMW International Gluten Workshop, Clermont-Ferrand,
glutenin subunit alleles in Australian germplasm. France, pp180-184
In: Proc 42nd Austr Cereal Chemistry Conf RACI,
Jackson EA, Morel MH, Sontag-Strohm T, Branlard
Parkville, Vic, Australia, pp166-170
G, Metakovsky EV, Redaelli R. (1996) Proposal for
Gupta RB, Bekes F, Wrigley CW (1990a) Predicting combining the classification systems of alleles of Gli-1
values of LMW glutenin alleles for dough quality of and Glu-3 loci in bread wheat (Triticum aestivum L.). J
bread wheat. In: Bushuk W, Tkachuk R (Eds.), Gluten Genetics and Breeding 50:321-336
Proteins. American Association of Cereal Chemists, St.
Lerner SE, Kolman MA; Rogers WJ (2009) Quality and
Paul, Minnesota, USA, pp. 615–621
endosperm storage protein variation in Argentinean
Gupta RB, Bekes F, Wrigley CW (1991) Prediction of grown bread wheat. I. Allelic diversity and
physical dough properties from glutenin subunit discrimination between cultivars. J. Cereal Sci 49:337-345
composition in bread wheats: correlation studies.
Liu L, Ikeda TM, Branlard G, Peña RJ, Rogers WJ, Lerner
Cereal Chem 68:328-333
SE, Kolman, MA, Xia X, Wang L, Ma W, Appels R,
Gupta RB, Bekes F, Wrigley CW, Moss HJ (1990b) Yoshida H, Yan Y, He Z (2010) Comparison of low
Prediction of wheat dough quality in breeding molecular weight glutenin subunits identified by SDS-
on the basis of LMW and HMW glutenin subunit PAGE, 2-DE, MALDI-TOF-MS and PCR in common
composition. In: Proceedings of the Sixth Australian wheat. BMC Plant Biol 10:124 (24 pages)
Wheat Breeders Society Assembly, Tamworth,
Luo C, Griffin WB, Branlard G, McNeil DL (2001)
Australia, pp. 217–225
Comparison of low- and high molecular-weight
Gupta RB, MacRitchie F (1991) A rapid one-step one- glutenin allele effects on flour quality. Theor Appl
dimensional SDS-PAGE procedure for analysis of Genet 102:1088-1098
subunit composition of glutenin wheat. J Cereal Sci 14,
McIntosh RA, Dubcovsky J, Rogers WJ, Morris CF, Appels
105–109
R, Xia XC (2011) Ann Wheat Newsl 57:303-321
Gupta RB, Paul JG, Cornish GB, Palmer CA, Bekes F,
Metakovsky EV, Wrigley CW, Bekes F, Gupta R (1990)
Rathien AJ (1994) Allelic variation at glutenin subunit
Gluten polypeptides as useful genetic markers of
and gliadin loci, Glu-1, Glu-3 and Gli-1, of common
dough quality in Australian wheats. Austr J Agric Res
wheats. 1. Its additive and interaction effects on dough
41:289-306
properties. J Cereal Sci 19:9-17
Singh NK, Shepherd KW, Cornish GB (1991) A simplified
Gupta RB, Shepherd KW (1990) Two-step one-dimensional
SDS-PAGE procedure for separating LMW subunits of
SDS-PAGE analysis of LMW subunits of glutenin.
glutenin. J Cereal Sci 14:203-208
I. Variation and genetic control of the subunits in
hexaploid wheats. Theor App. Genet 80:65-74 Wang LH, Li GY, Peña RJ, Xia XC, He ZH (2010)
Development of STS markers and establishment of
Gupta RB, Singh NK, Shepherd KW (1989) The cumulative
multiplex PCR for Glu-A3 alleles in common wheat
effect of allelic variation in LMW and HMW glutenin
(Triticum aestivum L.). J Cereal Sci 51:305-312
subunits on dough properties in the progeny of two
bread wheats. Theor Appl Genet 77:57-64 Wang LH, Zhao XL, He ZH, Ma W, Appels R, Peña RJ, Xia
XC (2009) Characterization of low-molecular-weight
Gupta RB, Shepherd KW (1988) Low-molecular-weight
glutenin subunit Glu-B3 genes and development of
glutenin subunits in wheat: their variation, inheritance
STS markers in common wheat (Triticum aestivum L.).
and association with bread-making quality. In: Miller
TE, Koebner RMD (eds) Proc 7th Int Wheat Genet
Symp, IPSR, Cambridge, UK, pp943-949
Gliadin composition of 137 doubled haploid wheat lines
derived from a cross between Kariega and Avocet
M.T. Labuschagne and A. Van Biljon
Department of Plant Sciences, University of the Free State, P.O. Box 339, Bloemfontein 9300, South Africa
Abstract
Gliadins are seed storage proteins coded by a series of genetic loci on several chromosome arms. The
frequency of alleles encoding the gliadins often differs among countries, with cultivars from the same
country often having similar patterns. In this study a doubled haploid (DH) population of 137 lines
was developed from a cross between Kariega (a hard red South African bread wheat with excellent
quality) and Avocet S (a standard white Australian spring wheat), using the wheat-maize technique.
Reverse phase high performance liquid chromatography (RP-HPLC) was used to determine the gliadin
composition of the parents and the population. Protein extraction was done and absorbance units
under the different peaks were calculated. There were 12 distinct gliadin peaks in Avocet and 14 in
Kariega. Three peaks absent in Avocet were present in Kariega, and one was absent in Kariega but
present in Avocet. The gliadin peaks of the DH population were not intermediate between those of the
parents, indicating some interaction between genes coding for these loci. Missing peaks recorded in
either parent were expressed in the DH population in a much higher percentage than expected in two
cases and much lower than expected in two cases. Highly significant positive or negative correlations
between pairs of gliadins indicated that they were sometimes consistently expressed together, or that
the presence of one peak led to the absence of another.
Introduction Materials and methods

Wheat gliadins are a highly polymorphic group of A DH population of 137 lines was developed
seed storage proteins. They are the most abundant from an F1 of Kariega (a hard, red, South
storage protein in wheat seed (about 40% of the total African bread wheat with excellent quality)
weight of protein in flour). They are controlled by and Avocet S (a standard, white, Australian,
Gli loci on the short arms of chromosomes 1 and spring wheat), using the wheat-maize technique
6, called the Gli-1 and Gli-2 loci on the A, B, and D (Laurie and Bennett 1988) modified by Pienaar
genomes. Gliadins γ and ω are coded by the Gli-A1, et al. (1997). RP-HPLC was used to determine
Gli-B1, Gli-D1 loci on the short arm of chromosomes the gliadin composition of the parents and the
1A, 1B and 1D. The α and β gliadins are coded by DH population. Protein extraction was done
group 6 chromosomes, by the Gli-A2, Gli-B2, Gli-D2 according to Wieser et al. (1998) using 70%
loci (Payne 1987; Caballero et al. 2004). Most wheat ethanol. A Shimadzu HPLC system was used
cultivars show similar overall patterns of gliadin with a Phenomenex Jupiter Su C18 column (300
subunits with characteristic regions for each type. Å, 250 x 4.60 mm). Injection volume was 50 μl.
The coincidence of two unrelated individuals Quantification was achieved by using a detection
having the same Gli alleles is unlikely due to the wavelength of 210 nm. Absorbance units under
high level of polymorphism (Metakovsky et al. the different peaks were calculated according to
2006). Therefore gliadins are also very useful for Wieser et al. (1998). The expression of peaks in
cultivar identification. The ω gliadins are low in the parents and the DH progeny was determined,
cysteine content compared to the other polypeptides and correlations between peaks were calculated.
which are cysteine rich (Tatham and Shewry 1995).
Gliadins exist as monomers which can be extracted
with 70% ethanol. Several gliadins are often Results and discussion
inherited as a unit or “block”. The aim of the study Gliadin ω peak 22.3 was absent in Avocet, but
was to study gliadins in a DH population of a cross was present in 50 of 137 DHs. The ω gliadin peak
between Kariega and Avocet. 23 was absent in Avocet but was present in 127

of 137 DHs (Table 1). The α gliadin peak 25.1 was of one led to the suppression of the other. Peaks
absent in Avocet, but was present in only 41 of 137 20.8 and 21.4; and 27.5 and 28.6 were also highly
DHs, α gliadin peak 28.6 was absent in Kariega, negatively correlated indicating that the presence
but was present in 118 of 137 DHs, α gliadin peak of one usually suppressed expression of the other.
26 was much higher in Avocet (13.12) than in The gliadin peaks of the DH population were
Kariega (8.67) but the DH average was 13.16 which not intermediate between those of the parents,
was similar to the Avocet parent. Gliadin α peak indicating that gliadins as single peaks were not
26.5 was higher for the DH average (11.29) than additively inherited and that there was interaction
either parent Avocet (10.35) and Kariega (7.89), α between genes coding for these loci. The DH
gliadin peak 27.5 was lower for the DH average population expressed the peak missing in either
(6.69) than for either Avocet (8.00) or Kariega parent to a much higher extent than expected
(7.49). The DH averages for the other peaks were in two cases and much lower than expected in
close to the averages of the two parents. two cases. One can speculate on the reason for
this. One of the disadvantages of gliadin analysis
Gliadin peaks ω 21.4 and 22.3 were highly
is that it covers only 6 of the 21 chromosomes
significantly correlated (r = 0.87) indicating that
(Metakovsky and Graybosch 2006), which may
they usually occurred together (Table 2). Gliadin
also be a restrictive factor in this case where a
peaks α 25.1 and 26 were highly significantly
cross is made and DH progeny are studied. Yet it
negatively correlated (r = -0.89), indicating that
still has advantages over DNA based techniques
the expression of one led to the absence of the
where the amount of polymorphism may not be
other. The same was found for ω gliadin peaks
sufficient and where costs are relatively high.
20.8 and 21.4 (r = -0.69) and α gliadin peaks 27.5
Intra cultivar polymorphism may also complicate
and 28.6 (r = -0.69). There were two additional
the use of DNA markers (Metakovsky and
positive correlations and one negative correlation
Graybosch 2006).
which were also highly significant, but of lower
magnitude, between gliadin peak pairs.
Table 2. Highly significant (P≤0.0001) correlations
between gliadin peaks. Gliadin type is in
Conclusions parentheses.
In terms of the gliadins, three peaks absent in
Avocet were present in Kariega, and one was Peak 1 Peak 2 Correlation
absent in Kariega but was present in Avocet. Peak P20.8 (ω) P21.4 (ω) -0.693
22.3 absent in Avocet was present in 36% of the DH P20.8 (ω) P22.3 (ω) -0.432
population. Peak 23 absent in Avocet was present P21.4 (ω) P22.3 (ω) 0.872
in 93% of the DH population. Peak 25.1 absent in P22.3 (ω) P24 (α) 0.376
Avocet was present in 30% of DH population. Peak P22.3 (ω) P31 (α) -0.376
28.6 absent in Kariega was present in 86% of the P23 (α) P36 (γ) 0.346
DH population. Peaks 21.4 and 22.3 almost always P25.1 (α) P26 (α) -0.886
occurred together. Peaks 25.1 and 26 were highly P27.5 (α) P28.6 (α) -0.685
negatively correlated indicating that the presence
Table 1. Gliadin peak area (%) at specific elution times (min) of parents and the doubled haploid (DH)
population.
Name P20.8 P21.4 P22.3 P23 P24 P25.1 P26 P26.5 P27 P27.5 P28.6 P31 P34.5 P36
Avocet 2.91 6.13 0.00 0.00 2.4 0.00 13.12 10.35 4.25 8.00 2.66 18.47 6.89 1.27
Kariega 3.99 3.63 1.94 1.3 2.3 2.65 8.67 7.89 4.02 7.49 0.00 18.49 5.03 1.33
DH 3.44 3.86 1.28 1.77 2.36 0.900 13.16 11.29 5.13 6.69 2.59 16.57 6.77 1.74
Type ω ω Ω α α α α α α α α γ γ γ
References Payne PI (1987) Genetics of wheat storage proteins and
the effect of allelic variation on bread-making quality.
Caballero L, Martín LM, Alvarez JB (2004) Variaton and
Annu Rev Plant Physiol 38:141-153
genetic diversity for gliadins in Spanish spelt wheat
accessions. Genet Res Crop Evol 51:679-686 Pienaar R de V, Horn M, Lesch AJG (1997) A reliable
protocol for doubled haploid accelerated wheat
Laurie DA, Bennett MD (1988) The production of haploid
breeding. Wheat Inf Serv 85:9-51
wheat plants from wheat x maize crosses. Theor Appl
Genet 76:393-397 Tatham AS, Shewry PR (1995) The S-poor prolamins of
wheat, barley and rye. J Cer Sci 22:1-6
Metakovsky EV, Branlard GP, Graybosch RA (2006)
Gliadins of common wheat: polymorphism and Wieser H, Antes S, Seilmeier W (1998) Quantitative
genetics. In: Wrigley C, Békés F, Bushuk W (eds) determination of gluten protein types in wheat
Gliadin and glutenin, the unique balance of wheat flour by reversed-phase high-performance liquid
quality. AACC, St Paul, MN, pp35-84 chromatography. Cereal Chem 75:644-650
Metakovsky EV, Graybosch RA (2006) Gliadin alleles in
wheat: identification and applications. In: Wrigley C,
Békés F, Bushuk W (eds) Gliadin and glutenin, the
unique balance of wheat quality. AACC, St Paul, MN
pp85-114

Comparison of glutenin subunit composition among
Australian and North American wheat classes
T.M. Ikeda and K. Takata
NARO, Western Region Agricultural Research Center, Fukuyama, Japan
Abstract
Glutenin subunit composition is one of the most important determinants of wheat end-use properties.
We compared Australian and North American wheat classes [Australian Standard White (ASW),
Australian Prime Hard (APH), No.1 Canada Western (1CW), US Dark Northern Spring (DNS), US
Hard Red Winter (HRW) and US Western White (WW)] exported to Japan in 2009. APH, DNS, HRW,
and 1CW are hard wheat classes mainly used for bread and Chinese alkaline noodles. ASW is a
mixture of hard and soft wheat used for Japanese white salted noodles. WW is a soft wheat class used
for cakes and cookies. We analyzed 60 individual seeds of each genotype by protein (SDS-PAGE and
2DE) and DNA (PCR) analyses. For the Glu-1 loci, most seeds comprising DNS, HRW and 1CW had
Glu-D1d, which increases gluten strength, whereas about 20% of ASW, PH and WW had this allele. On
the other hand, most WW seeds had a null allele (Glu-A1c) or lacked one of the two Glu-B1 subunits
(Glu-B1a or Glu-B1bl) that decrease gluten strength. For Glu-3 loci, most seeds of ICW had Glu-A3e or
Glu-A3f, which decrease gluten strength, but only half of DNS had this allele. Most 1CW, DNS, PH and
WW seeds had Glu-B3h, but most HRW seeds had Glu-B3b or Glu-B3g, which increase gluten strength.
In the combination of Glu-D1d and Glu-B3b or Glu-B3g, dough becomes very strong (Tabiki et al. 2006;
Ito et al. 2011). HRW should have genetic potential for very strong dough properties. We also found
a new Glu-A3 allele in HRW and WW, and a new Glu-B3 allele in WW (Ikeda et al. 2008; Ikeda et al
2009; Liu et al. 2010). We propose a matrix based on glutenin subunit alleles and their effects on gluten
strength in Fig. 1. Each class was plotted at different positions based on glutenin allelic compositions.
The protein amount and glutenin composition of these classes should correspond to their end-product
uses. This matrix should also be useful for selecting breeding lines for particular uses.
Strong
Glu-B1al Strong Extra strong
Glu-D1d DNS
1CW
HRW
Glu-1
ASW
(HMW-GS)
APH
Glu-A1a/b
Glu-D1a Medium
Glu-A1c Club WW
Glu-B1a
Weak
Weak Strong
Glu-A3e Glu-A3e/f Glu-A3c/d
Glu-B3j Glu-B3d/h/i Glu-B3b/g
GLU-3 (LMW-GS)
Fig. 1 A matrix indicating relationships between glutenin alleles and gluten strength. The seven wheat
classes are plotted based on glutenin allelic compositions. HMW-GS = high molecular weight glutenin
subunits; LMW-GS = low molecular weight glutenin subunits; ASW = Australian Standard White;
APH = Australian Prime Hard; 1CW = No.1 Canada Western; DNS = US Dark Northern Spring; HRW = US
Hard Red Winter; WW = US Western White.
References Ito M, Fushie S, Maruyama-Funatsuki W, Ikeda TM,
Nishio Z, Nagasawa K, Tabiki T, Yamauchi H (2011)
Effect of allelic variation in three glutenin loci on
Z, Lerner SE, Kolman MA, Yoshida H, Rogers WJ
dough properties and bread-making qualities of
(2008) International collaboration for unifying Glu-3
winter wheat. Breeding Sci 61:281-287
nomenclature system in common wheats. In: Appels R,
Eastwood R, Lagudah E et al. (eds) Proc 11th Int Wheat Liu L, Ikeda TM, Branlard G, Pena RJ, Rogers WJ, Lerner
Genet Symp, Sydney University Press, Sydney, NSW, SE, Kolman MA, Xia X, Wang L, Ma W, Appels R,
Australia, pp130-132 Yoshida H, Wang A, Yan Y, He Z (2010) Comparison
of low molecular weight glutenin subunits identified
by SDS-PAGE, 2-DE, MALDI-TOF-MS and PCR in
Z, Lerner SE, Kolman MA, Rogers WJ (2009)
common wheat. BMC Plant Biol 10:124
Characterization of new Glu-3 alleles in bread wheat.
In: Proc 10th Gluten Protein 2009, INRA, Clermont- Tabiki T, Ikeguchi S, Ikeda TM (2006) Effects of high-
Ferrand France, pp180-184 molecular-weight and low-molecular-weight glutenin
subunit alleles on common wheat flour quality.
Breeding Sci 56:131-136

Understanding the effects of HMW-GS and LMW-GS
on processing quality using near-isogenic lines and
development of functional markers for LMW-GS in wheat
X.C. Xia1, Z.H. He1,2, H. Jin1, L.H. Wang1, Y. Zhang1, and G.Y. Li3
1Institute
of Crop Science/National Wheat Improvement Center, Chinese Academy of Agricultural Sciences
(CAAS), 12 Zhongguancun South Street, Beijing 100081, China; 2CIMMYT China Office, C/O CAAS,
12 Zhongguancun South Street, Beijing 100081, China; 3Crop Research Institute, Shandong Academy of
Agricultural Sciences, Jinan 250100, Shandong, China
Abstract
HMW-GS and LMW-GS play important roles in wheat processing quality. In the present study, the
relationships between glutenin subunits and processing qualities of pan bread, Chinese steamed bread
and fresh white Chinese noodles were investigated in variety Aroona and its near-isogenic lines (NILs)
with different high molecular weight glutenin subunits (HMW-GS) and low molecular weight glutenin
subunits (LMW-GS). For dough strength-related traits such as farinograph stability and energy,
7+9, 17+18, 5+10, Glu-A3b, Glu-A3d, Glu-A3f, Glu-B3b, Glu-B3g were correlated with superior dough
properties. The NIL with the Glu-A1 null subunit conferred a lower pan bread total score (PBTS)
than the other two NILs with alleles at this locus. For dough extensibility, a significant difference was
found among Glu-A3 NILs; Glu-A3e was the most inferior allele, whereas the other five Glu-A3 alleles
had similar effects on Rmax, ST, %UPP and extensibility. Glu-B3g and Glu-B3b were associated with
higher quality parameters than the other Glu-B3 alleles, and the Glu-B3c NIL performed the worst in
PBTS, Rmax, ST, SV, %UPP and some other parameters. Five Glu-D3 alleles produced similar quality
parameters except for farinograph water absorption. For Chinese steamed bread quality, Glu-B3a
and Glu-B3b showed significantly better skin color. Glu-B3a exhibited a higher total score than other
allelic variants. For fresh white Chinese noodles, Glu-B3a and Glu-B3d showed significantly better
smoothness than the other lines, and Glu-A3b, Glu-B3g, Glu-B3h exhibited better whiteness than the
others. Functional markers for LMW-GS were developed and validated on the Aroona NILs and 141
CIMMYT wheat varieties and advanced lines with different Glu-A3 and Glu-B3 alleles. These results
provide very useful information for wheat quality improvement in breeding programs.
Introduction the composition of HMW-GS and LMW-GS and

Wheat glutenin is composed of high molecular Chinese steamed bread and fresh white Chinese
weight (HMW-GS) and low molecular weight noodle qualities.
glutenin subunits (LMW-GS) (Gianibelli et al. Traditionally, SDS-PAGE (Jackson et al. 1996)
2001), encoded by loci on the long and short arms is used to determine allelic compositions of
of group 1 chromosomes, respectively (Payne et LMW-GS in wheat. Nevertheless, difficulties in
al. 1987; Singh and Shepherd 1988; Shewry et al. resolving the multigene families and overlapping
1992; D’Ovidio and Masci 2004). The effects of fractions of LMW-GS hinder their routine use,
HMW-GS and LMW-GS on dough properties and particularly for testing large populations in the
end-use quality have been extensively studied. early generations of wheat breeding programs.
However, correlations between HMW-GS and Therefore, it is important to develop functional
LMW-GS and wheat quality reported by different markers to identify different LMW-GS genes
researchers are often contradictory (Mao et al. (Andersen and Lübberstedt 2003).
1995; Payne et al. 1987; Liu et al. 2005), largely
The objectives of this study were to (1) determine
due to different materials with different genetic
the effects of HMW-GS and LMW-GS on the
backgrounds. In addition, little information is
processing qualities of Chinese steamed bread
available regarding the relationship between
and fresh white Chinese noodles using variety replications during the 2010 cropping season.
Aroona and its derived near-isogenic lines, and Twenty varieties from our laboratory collection
(2) clone LMW-GS genes at the Glu-A3 and Glu-B3 and 141 wheat varieties and advanced lines from
loci from common wheat, and develop allele- the International Maize and Wheat Improvement
specific STS markers for different Glu-A3 and Center (CIMMYT) with different Glu-A3 and
Glu-B3 alleles. This will benefit the improvement Glu-B3 protein alleles detected by SDS-PAGE
of wheat quality in breeding programs. were employed to validate the allele-specific
markers.
Materials and methods Characterization of grain and flour properties

Plant materials Grain protein contents were measured using a
near infrared transmittance (NIT) analyzer Foss-
The 25 near-isogenic spring wheat lines differing
Tecator 1241 (Foss, Höganas, Sweden). Grain
at the Glu-A1, Glu-B1, Glu-D1, Glu-A3, Glu-B3,
hardness and moisture were determined using
and Glu-D3 loci used in this study (Table 1) were
single kernel characterization (SKCS 4100, Perten,
kindly provided by Dr. Marie Appelbee and
Sweden). The samples were tempered overnight
Prof. Ken Shepherd at the SARDI Grain Quality
to moisture contents of 14.5%, 15.5% and 16.5%
Research Laboratory, Adelaide, South Australia.
for soft, medium and hard wheat, respectively.
They were planted at the Xinjiang Academy of
All samples were milled on a Buhler MLU
Agri-Recalamation Sciences in Shihezi and at
202 laboratory mill (Buhler Bros, Ltd, Uzwil,
Xinjiang Academy of Agricultural Sciences in
Switzerland) based on AACC method 26-21A
Urumqi, in randomized complete blocks with two
Table 1. Allelic constitutions of 25 near-isogenic lines with different high molecular weight glutenin
subunits (HMW-GS) and low molecular weight glutenin subunits (LMW-GS).
Accession
number Line Glu-A1 Glu-B1 Glu-D1 Glu-A3 Glu-B3 Glu-D3 Donor parent
1 Aroona A c a c b c
2 Aril2-4 (2*) b c a c b c Sonalika
3 Aril3-2 (null) c c a c b c Chinese Spring
4 Aril5-2 (7*) a a a c b c Orca
5 Aril7-1 (7*+8) a b a c b c India 115
6 Aril9-3 (6+8*) a d a c b c Rendezvous
7 Aril10-1 (17+18) a i a c b c Gabo
8 Aril12-3 (3+12) a c b c b c Rendezvous
9 Aril13-3 (5+10) a c d c b c Halberd
10 Aril14-3 (2.2+12) a c f c b c Norin 61
11 Aril16-1 (A3b) a c a b b c Gabo
12 Aril18-5 (A3d) a c a d b c Orca
13 Aril19-2 (A3e) a c a e b c Lerma Rojo
14 Aril20-1 (A3f) a c a f b c Bungulla
15 Aril21-2 (B3a) a c a c a c Chinese Spring
16 Aril23-4 (B3c) a c a c c c Halberd
17 Aril24-3 (B3d) a c a c d c Orca
18 Aril26-1 (B3f) a c a c f c Gawain
19 Aril27-6 (B3g) a c a c g c Millewa
20 Aril28-4 (B3h) a c a c h c Sonalika
21 Aril29-4 (B3i) a c a c i c Jufy 1
22 Aril30-1 (D3a) a c a c b a Chinese Spring
23 Aril36-2 (D3b) a c a c b b Bungulla
24 Aril33-1(D3d) a c a c b d Jufy 1
25 Aril35-1 (D3f) a c a c b f India 115

to give flours with straight-run extraction rates. specific primers were 94°C for 5 min followed by
Farinogragh and extensograph parameters were 38 cycles of 94°C for 45 s, 56-61°C for 45s, 72°C
obtained according to AACC Approved Methods for 90 s, and a final extension at 72°C for 8 min.
54-21 and 54-10, respectively.
Allele-specific PCR marker design and
Preparation of Chinese steamed bread and fresh validation
white Chinese noodles Allele-specific PCR markers were designed based
Chinese steamed bread (CSB) was prepared on the allelic variants of Glu-A3 and Glu-B3
and evaluated according to Chen et al. (2007) following the method of Zhang et al. (2003).
with minor modifications. The optimum water These markers were firstly validated with the
addition was modified to 80% of farinograph Aroona NILs and Cheyenne, and then with
water absorption for flour samples. Flour (200 161 wheat varieties and advanced lines from
g) mixed with yeast and water was slurried in a CIMMYT, Australia and France with different
National mixer for 1.5-2.0 min. The dough was Glu-A3 and Glu-B3 protein mobility alleles
divided into two parts with the same weight previously identified in SDS-PAGE by other
and sheeted by passing ten times through a pair workers.
of rollers set with a gap of 9/32 inch (National
Mfg. Co., Lincoln, NE). After each pass, the Statistical analysis
dough was folded end-to-end and re-sheeted in Data analysis was performed by SAS software
a uni-directional manner. The dough was gently 9.0 (SAS Institute, Cary, NC) using analysis of
shaped by hand to form a rounded dough piece variance (GLM) and Duncan’s shortest significant
and then placed in an Extensograph rounder ranges (DSSR).
and rounded 20 times. The rounded dough
pieces were proofed for 20 min in a National
fermentation cabinet (35°C, 85% RH) and Results and discussion
steamed for 25 min in a steamer containing cool
water. The CSB score includes specific volume Dough properties
(weighting, 20), skin color (10), smoothness (10), The Glu-A3 locus showed a significant effect
shape (10), structure (15), and stress relaxation on extensograph extensibility, accounting for
(35). Stress relaxation (SR) was measured by a 47% of the phenotypic variance. The values
Texture Analyzer TA-XT2i (Stable Micro Systems, for individual alleles ranked as Glu-A3c>Glu-
Ltd., Surrey, England). A3d>Glu-A3f>Glu-A3b>Glu-A3e. Glu-B1, Glu-D1,
Fresh white Chinese noodles (FWCN) were Glu-A3 and Glu-B3 had significant effects on
prepared and evaluated according to the farinograph stability time, accounting for 9, 44, 11
method described by Zhang et al. (2005). FWCN and 16% of the phenotypic variance, respectively.
score included color score (weighting 15), Subunit alleles 17+18, 5+10, Glu-A3b, Glu-A3d,
appearance (10), firmness (20), viscoelasticity Glu-A3f, Glu-B3b, Glu-B3g showed significantly
(30), smoothness (15) and taste flavor (10). The higher farinograph stability times than other
experiment was performed at room temperature alleles. This is generally consistent with previous
(20-25oC) and 50-60% RH. For sensory evaluation studies indicating that the contributions of
of noodle quality, commercial flour (Hetao Glu-D1 and Glu-B3 to mixing tolerance were the
Xuehua product) was used as a control. largest and 5+10, Glu-A3d had the largest positive
effects on dough properties compared with
DNA extraction and PCR amplification other alleles (Liu et al. 2005). The contribution of
Genomic DNA was extracted from seedlings or Glu-A1 to dough properties was not significant
seeds using a modified CTAB procedure (Gale and no significant differences were observed
et al. 2001). PCR was performed using TakaRa regarding the effects of different allelic variants
Taq DNA polymerase (1.0 unit) in 20 μl reaction on dough properties. However, previous studies
volumes containing approximately 50 ng of found that subunits 1 and 2* had significantly
genomic DNA, 1×PCR buffer (1.5 mM MgCl2), better effects on quality parameters than the null
100 μM of each of dNTPs and 10 pmoles of each (Luo et al. 2001; He et al. 2005; Liu et al. 2005).
PCR primer. PCR cycling conditions for gene- Zhang et al. (2009) found that epistatic effects
were important among loci. In this study, the
lines with 1, 2* and N plus the inferior subunit Fresh white Chinese noodle quality
2+12 showed no significant differences in dough Glu-B1 had a significant effect on viscoelasticity,
properties due to the epistatic effect of 2+12. taste flavor and total score, accounting for 35, 37,
The contribution of Glu-D3 to dough properties and 19% of the phenotypic variances, respectively.
was not significant and there were no significant Glu-A3 made a high contribution to color and
differences between the effects of individual firmness, accounting for 13 and 27% of the
Glu-D3 alleles on dough properties, consistent phenotypic variances, Glu-B3 had a higher effect
with Liu et al. (2009). on smoothness than other loci, accounting for 26%
of phenotypic variances, respectively, and Glu-D3
Chinese steamed bread showed a higher effect on appearance, accounting
Glu-D1 had a significant effect on skin color of for 26% of the phenotypic variance. At the Glu-A1
Chinese steamed bread, accounting for 24% of the and Glu-D3 loci, no significant differences were
phenotypic variance. Glu-A3 showed a significant observed in fresh white Chinese noodle quality.
effect on stress relaxation of Chinese steamed Noodle sheet b* values differed among lines with
bread, accounting for 25% of the phenotypic different Glu-B1, Glu-D1, and Glu-A3 alleles. At
variance. Glu-B3 made the highest contributions Glu-B1 subunits 7*+8 and 17+18 were associated
to shape, specific volume and steamed bread total with superior whiteness. At Glu-D1, 5+10 conferred
score. Glu-D3 made the highest contribution to significantly better whiteness than other subunits.
smoothness and structure. We concluded that the Significant differences occurred at the Glu-B3
contribution of LMW-GS exceeded HMW-GS in locus for smoothness and a* and b* values of
determination of Chinese steamed bread quality. noodle sheets; Glu-B3a and Glu-B3d conferred
At the Glu-A1 locus, a significant difference in significantly better smoothness than others. Glu-
bun shape was observed for different lines. The B3f was correlated with an inferior noodle sheet a*
genotype with subunit 1 showed significantly value. Glu-B3g and Glu-B3h had significantly higher
better shape. At the Glu-B1 locus, a significant whiteness. This was in general agreement with
difference in stress relaxation was observed; previous studies in which no significant difference
7*+8 and 6+8* were associated with better stress in fresh white Chinese noodle quality was observed
relaxation than other subunits. At the Glu-D1 for HMW-GS at Glu-1, and Glu-A3d and Glu-B3d
locus, there was a significant difference in skin were slightly better than other alleles (He et al. 2005).
color and structure; lines with 5+10 showed
Gene-specific PCR markers for Glu-A3
significantly better skin color and structure
protein alleles
than others. At the Glu-A3 locus, a significant
difference was observed in stress relaxation for To distinguish genotypes containing Glu-A3
different lines; Glu-A3e had the highest stress alleles a, b, c, d, e, f and g, seven STS markers were
relaxation. At the Glu-B3 locus, a significant developed based on the SNPs detected among
difference was observed in skin color, stress them; the primers are listed in Table 2. In the six
relaxation and steamed bread total score with NILs and Glenlea, marker gluA3a amplified a 529 bp
Glu-B3a making a positive contribution to skin specific PCR product in Aroona-A3a with the Glu-
color, stress relaxation and steamed bread A3a protein allele. Marker gluA3b designed for the
total score, whereas Glu-B3d and Glu-B3g were Glu-A3b allele in Aroona-A3b generated an 894 bp
associated with inferior steamed bread total score. band. Marker gluA3d identified Glu-A3d in Aroona-
There were no significant differences at Glu-D3 A3d, producing a 967 bp PCR product. For Glu-A3e
associated with Chinese steamed bread quality. in Aroona-A3e, marker gluA3e generated a 158 bp
band. To discriminate Glu-A3f from other alleles,
The DSSR multiple comparison of the effect of marker gluA3f was developed, generating a 552 bp
genotypes for Chinese steamed bread indicated band in Aroona-A3f. In Glenlea with the Glu-A3g
that 1/7+9/5+10/Glu-A3c/Glu-B3b/Glu-D3c had allele, a 1,345 bp PCR product was amplified with
better skin color than other genotypes. Subunits marker gluA3g. Marker gluA3ac was developed to
and allele combination 1/7+9/2+12/Glu-A3e/Glu- amplify both Glu-A3a and Glu-A3c in both Aroona-
B3b/Glu-D3c had the highest stress relaxation A3a and Aroona because it was difficult to design
and steamed bread total score. Stress relaxation a specific primer set for Glu-A3c; this marker in
determined the steamed bread total score. combination with the marker gluA3a can be used to
identify Glu-A3c.

Gene-specific PCR markers for Glu-B3 Validation of Glu-A3 and Glu-B3
protein alleles allele-specific markers
As for Glu-A3, specific markers were developed Thirteen NILs, Glenlea and Cheyenne used in
to distinguish Glu-B3 alleles (Table 3). Again, it cloning the Glu-A3 and Glu-B3 genes, and an
was necessary to combine two or three different additional 161 wheat varieties and advanced
markers to identify some alleles. lines were used to validate the Glu-A3 and Glu-B3
Table 2. Allele-specific PCR markers for the discrimination of Glu-A3 alleles defined by protein mobility
in common wheat.
Marker Primer Target
name set Sequence (5’3’)a Glu-A3 allele Fragment size
gluA3a LA1F AAACAGAATTATTAAAGCCGG a 529
SA1R GGTTGTTGTTGTTGCAGCA
gluA3b LA3F TTCAGATGCAGCCAAACAA b 894
SA2R GCTGTGCTTGGATGATACTCTA
gluA3ac LA1F AAACAGAATTATTAAAGCCGG acb 573
SA3R GTGGCTGTTGTGAAAACGA
gluA3d LA3F TTCAGATGCAGCCAAACAA d 967
SA4R TGGGGTTGGGAGACACATA
gluA3e LA1F AAACAGAATTATTAAAGCCGG e 158
SA5R GGCACAGACGAGGAAGGTT
gluA3f LA1F AAACAGAATTATTAAAGCCGG f 552
SA6R GCTGCTGCTGCTGTGTAAA
gluA3g LA1F AAACAGAATTATTAAAGCCGG g 1,345
SA7R AAACAACGGTGATCCAACTAA
a Mismatched nucleotide is underlined; b Specific for both Glu-A3a and c.
Table 3. Allele-specific PCR markers for the discrimination of Glu-B3 alleles defined by protein
mobility in common wheat.
Marker Primer Target
name set Sequence (5’3’)a Glu-A3 allele Fragment size
gluB3a SB1F CACAAGCATCAAAACCAAGA a 1,095
SB1R TGGCACACTAGTGGTGGTC
gluB3b SB2F ATCAGGTGTAAAAGTGATAG b 1,570
SB2R TGCTACATCGACATATCCA
gluB3c SB3F CAAATGTTGCAGCAGAGA c 472
SB3R CATATCCATCGACTAAACAAA
gluB3d SB4F CACCATGAAGACCTTCCTCA d 662
SB4R GTTGTTGCAGTAGAACTGGA
gluB3e SB5F GACCTTCCTCATCTTCGCA e 669
SB5R GCAAGACTTTGTGGCATT
gluB3fg SB6F TATAGCTAGTGCAACCTACCAT fgb 812
SB6R CAACTACTCTGCCACAACG
gluB3g SB7F CCAAGAAATACTAGTTAACACTAGTC g 853
SB7R GTTGGGGTTGGGAAACA
gluB3h SB8F CCACCACAACAAACATTAA h 1,022
SB8R GTGGTGGTTCTATACAACGA
gluB3i SB9F As SB6F i 621
SB9R TGGTTGTTGCGGTATAATTT
gluB3bef SB10F GCATCAACAACAAATAGTACTAGAA befc 750
SB10R GGCGGGTCACACATGACA
aMismatched nucleotides are underlined; bSpecific for both Glu-B3f and g; cSpecific for Glu-B3b, e and f.
allele-specific markers (Tables 2 and 3). With four Acknowledgements
exceptions, the results were consistent with those This study was supported by the National
detected by SDS-PAGE Basic Research Program (2009CB118300), the
National Science Foundation of China (30830072),
international collaboration project from the
Conclusions
Ministry of Agriculture (2011-G3), and the China
Glu-B1, Glu-D1, Glu-A3, and Glu-B3 had large Agriculture Research System (CARS-3-1-3).
effects on dough strength, such as energy,
extensograph maximum resistance, farinograph
development time and farinograph stability time. References
HMW-GS subunits 7+9, 17+18, 5+10, and alleles Andersen JR, Lübberstedt T (2003) Functional markers in
A3b, A3d, A3f, B3b and B3g showed significantly plants. Trends in Plant Sci 8:554-560
superior effects on dough strength compared to Chen F, He ZH, Chen DS, Zhang CL, Zhang Y, Xia XC
other genotypes. (2007) Influence of puroindoline alleles on milling
performance and qualities of Chinese noodles,
For Chinese steamed bread, Glu-D1 and Glu-B3 steamed bread and pan bread in spring wheats. J
had significant effects on skin color of Chinese Cereal Sci 45:59-66
steamed bread, with 5+10, Glu-B3a and Glu-B3b D’Ovidio R, Masci S (2004) The low-molecular-weight
showing the highest values. Allelic variants at glutenin subunits of wheat gluten. J Cereal Sci
Glu-A1 showed a difference in shape and subunit 39:321-339
1 conferred the highest value. Allelic variants Gale KR, Ma W, Zhang W, Rampling L, Hill AS, Appels
at Glu-D1 showed differences in structure R, Morris P, Morrel M (2001) Simple high-throughput
and subunit 3+12 was associated with inferior DNA markers for genotyping in wheat. In: Eastwood
R, Hollamby G, Rathjen T, Gororo N (eds) Proc 10th
structure. Allelic variants at Glu-B1, Glu-A3
Australian Wheat Breeding Assembly, Wheat Breeding
and Glu-B3 loci showed differences in stress Society of Australia, Mildura, Vic, pp26-31
relaxation, with 7*+8, 6+8*, Glu-A3e and Glu-B3a Gianibelli MC, Larroque OR, MacRitchie F, Wrigley
having the highest values. Allelic variants at CW (2001) Biochemical, genetic, and molecular
Glu-B3 showed differences in steam bread total characterization of wheat glutenin and its component
score and Glu-B3a had a higher score than others. subunits. Cereal Chem 78:635-646
He ZH, Liu L, Xia XC, Liu JJ, Peña RJ (2005) Composition
Allelic variants at Glu-B3 had different effects of HMW and LMW glutenin subunits and their effects
on smoothness of fresh white Chinese noodles, on dough properties, pan bread, and noodle quality of
with Glu-B3a and Glu-B3d producing higher Chinese bread wheats. Cereal Chem 82:345-350
smoothness than other alleles. Allelic variants Jackson EA, Morrel MH, Sontag-Strohm T, Branlard
at Glu-B1, Glu-D1, Glu-A3 and Glu-B3 conferred G, Metakovsky EV, Radaelli R (1996) Proposal for
different noodle sheet b* values, and 7*+8, 17+18, combining the classification systems of alleles of Gli-1
and Glu-3 loci in bread wheat (Triticum aestivum L.). J
5+10, Glu-A3b, Glu-B3g, Glu-B3h showed superior
Genet & Breed 50:321-336
whiteness relative to others.
Liu L, He ZH, Ma WJ, Liu JJ, Xia XC, Peña RJ (2009) Allelic
Seven and 10 allele-specific markers developed variation at the Glu-D3 locus in Chinese bread wheat
for Glu-A3 and Glu-B3 loci, respectively, were and effects on dough properties, pan bread and noodle
qualities. Cereal Res Commun 37:57-64
validated on 13 NILs, Glenlea and Cheyenne,
and 161 wheat varieties and advanced lines. The Liu L, He ZH, Yan J, Zhang Y, Xia XC, Peña RJ (2005)
Allelic variation at the Glu-1 and Glu-3 loci, presence
results were in general accordance with those of the 1B·1R translocation, and their effects on
detected by SDS-PAGE. mixographic properties in Chinese bread wheats.
Euphytica 142:197-204

Luo C, Griffin WB, Branlard G, McNeil DL (2001) Singh NK, Shepherd KW (1988) Linkage mapping of genes
Comparison of low- and high-molecular-weight wheat controlling endosperm storage proteins in wheat. 1.
glutenin alleles effects on flour quality. Theor Appl Genes on the short arms of group 1 chromosomes.
Genet 102:1088-1098 Theor Appl Genet 75:628-641
Mao P, Li ZZ, Lu SY (1995) The effect of high molecular Zhang W, Gianibelli MC, Ma W, Rampling L, Gale KR
weight glutenin subunits on bread quality. Acta Agric (2003) Identification of SNPs and development of AS-
Boreali-Sin 10 (Suppl):55-59 PCR markers for γ-gliadin alleles in Triticum aestivum.
Payne PI, Nightingale MA, Kattiger AF (1987) The Theor Appl Genet 107:130-138
relationship between HWM glutenin subunit Zhang Y, Nagamine T, He ZH, Ge XX, Yoshida H, Peña
composition and the bread-making quality of British RJ (2005) Variation in quality traits in common wheat
grown wheat varieties. J Sci Food & Agric 40:51:65 as related to Chinese fresh white noodle quality.
Shewry PR, Halford NG, Tatham AS (1992) High Euphytica 141:113-120
molecular weight subunits of wheat glutenin. J Cereal Zhang Y, Tang JW, Yan J, Zhang YL, Zhang Y, Xia XC, He
Sci 15:105-120 ZH (2009) The gluten protein and interactions between
components determine mixograph properties in an F6
recombinant inbred line population in bread wheat. J
Estimating dough properties and end-product quality
from flour composition
F. Békés1, W. Ma2, and S. Tömösközi3
1FBFD PTY LTD,Beecroft, NSW, Australia;2State Agricultural Biotechnology Center, Murdoch University,
Perth, WA, Australia; 3Budapest University of Technology and Economics, Department of Applied Biotechnology
and Food Science, Budapest, Hungary
Abstract
Most of our knowledge about the ‘genetics of quality’ derives from two different experimental
approaches: i) direct measurements of quality traits on samples with systematically altered chemical
compositions; and, ii) relating quality and chemical composition/genetics of large sample populations
using statistical methods. An overview on the recent achievements of these two approaches will
be given illustrating the advantages/limitations of small- and micro-scale methodology and in vitro
reconstitution/incorporation methods in basic research and in different applications as well as
introducing three prediction procedures for dough properties (Protein Scoring System; PSS), loaf
volume (PQI) and water absorption. The PSS model relates the individual and interactive contributions
of high molecular weight (HMW) and low molecular weight (LMW) glutenin alleles to specific dough
parameters, predicting the genetic potential of dough strength and extensional properties of dough
of wheat flour with 12% protein content and with the ratios of glutenin to gliadin and HMW to LMW
glutenin subunits (GS) of 1.0 and 0.2, respectively. The most advanced version of the model is capable
of considering the effects of expression levels of different storage protein genes, allowing prediction of
actual dough parameters. The input of this model is the allelic composition and quantitative protein
composition (including UPP%), while the output provides a good estimate of actual dough strength
and extensibility of a given sample (r2 >0.85 and r2 >0.75, respectively). Further applications of the
models are also useful in blending formulations and in combining data to predict the technology
specific bread-making potential (PQI) of samples using a nonlinear model.
Introduction Results and discussion

One of the most important tasks of cereal science
Direct in vitro methodology
is to relate end-product quality to genes involved
in determining certain attributes of quality. It The concept of in vitro direct methodologies
requires a good understanding of the complex is based on the reconstitution technique of
nature of quality that can lead to the proper separating flour components and putting them
measurements of these attributes. Most of our back without changing the functionality of the
knowledge about the ‘genetics of quality’ derives isolated components. This allows replacement
from two different experimental approaches: i) of components with those isolated from other
direct measurements of quality traits on samples samples, and/or to modify the relative amounts
with systematically altered chemical composition; in reconstituted samples such that chemical
and, ii) relating quality and chemical compositions can be altered both qualitatively
composition/genetics of large sample populations and quantitatively (Gras and Békés 1996).
using statistical methods. An overview of the The capabilities of in vitro methodologies have
recent achievements of these two approaches is been multiplied by the development of the so-
provided to illustrate the advantages/limitations called incorporation procedure which allows
of small- and micro-scale methodology and in building the supplemented glutenin subunits
vitro reconstitution/incorporation methods in (GS) into the polymeric glutenin structure
basic research and in different applications as through a partial reduction/re-oxidation
well as introducing three prediction procedures process (Békés et al. 1994a). Two basic in vitro
for dough properties (Protein Scoring System; methodologies have been developed and
PSS), loaf volume (Protein Quality Index; PQI) successfully applied in recent years: (i) the ‘base
and water absorption.

flour method’ where the chemical composition i=1
of a base flour is altered systematically by Q = αi * (qH)i ..................................................................... (1)
supplemented (simply added or incorporated) 13
compounds (Békés et al. 1994b), or (ii) the
method where a ‘model dough’ is formed by Where: Q is the Payne score estimating the dough
putting isolated flour components (starch, soluble strength of the sample; and αi is the contribution of
proteins, lipids, gliadins and in vitro polymerized individual HMW-GS alleles. The value of (qH)i is
glutenin subunits) together (Beasley et al. 2001). either one or zero depending on whether the actual
Reconstitution studies with potential to give allele is present or not. Since the significant success
direct evidence about individual proteins are of using the Payne score in breeding programs by
hindered, however, by the difficulty of isolating applying HMW-GS composition as protein markers
single polypeptides. In an alternative approach, for dough properties, there have been several
genes for the desired wheat proteins have been attempts to also involve the LMW-GS alleles in
inserted into foreign genomes such as Escherichia similar mathematical formulas (Gupta et al. 1991;
coli, yeast or insect cell lines. This technique allows Eagles et al. 2002; Cornish et al. 2006). The PSS of
the production of single polypeptides in large Békés et al. (2006) applying sophisticated statistical
quantities and makes the protein purification/ approaches of multiple regression is capable of
isolation easier. A further advantage of applying describing the individual effects of HMW-GS and
heterologous expression is the ability to alter the LMW-GS, as well as the pair-wise interactions
chemical composition of gene products by genetic among the alleles, thus providing an excellent
engineering. The role of structural features in prediction tool for dough strength and extensibility
determining certain functional parameters such with r2 values of 0.873 and 0.559, respectively,
as the number and location of cysteine residues between measured and predicted values:
in polypeptides analogous to glutenin subunits
(ANG proteins) was successfully investigated i=1 j=1 i=1 j=1
using this approach (Tamas et al. 2002).
 
Q = αi *(qH)i + αi * (qL)j +  i,j *(qH)j *(qL)j 
The in vitro nature of these approaches provides 17 16 17 16
a much wider opportunity to alter the chemical
................................................................................................................... (2)
composition of wheat flour and to monitor the
consequences in relation to functional properties.
On the other hand, the use of these techniques where αi and βi describe the individual and
– again because of the in vitro nature of the interactive contributions of the alleles to Q quality
experiments – requires very careful consideration parameters (dough strength or extensibility). The
in relation to extrapolating the results to in vivo terms (qH)i and (qL)i indicate the presence of a
situations. certain HMW or LMW glutenin allele. Only 3 (qH)
i and 3 (qL)i values are equal to 1 in an ordinary
The most reliable evidence for applicability
wheat sample, representing the gene products of
of reconstitution and addition/incorporation
Glu-1 and Glu-3 at the A, B and D loci.
type experiments to evaluate the functionality
of polypeptides in wheat doughs is the full The αi and βi,j parameters determined with the
agreement in functional properties between PSS model on large sample populations (n>5000)
the results for transgenic wheat samples with confirmed previous well-established observations
introduced Glu1-Ax1 and Glu-Dx5 and those of about the role of HMW-GS alleles and extend work
using in vitro methods (Barro et al. 1997). on LMW-GS alleles (Békés et al. 2006). HMW-GS
alleles in general make a larger contribution to
Indirect correlative methods to predict dough strength than LMW-GS alleles, whereas
dough properties for extensibility, LMW-GS alleles make a slightly
The most frequently used tool relating protein larger contribution than HMW-GS. Comparing the
composition to quality is the Payne score (Payne effects of different alleles from the same loci, the
et al. 1987) providing a single number to estimate differences among alleles are significantly larger for
dough strength from the HMW-GS glutenin allelic dough strength than for extensibility.
composition.
The application of the PSS model on different and partly on the protein’s interaction with other
sample populations importantly point to the prolamin type proteins present in the base flour.
impact of allelic interactions as major variables A novel approach, introducing incorporation into
determining dough properties. The realization of rice flour dough can now clearly monitor only
the large influences of allele–allele interactions the direct effects of individual polypeptides on
on wheat flour doughs may alter our way of mixing properties while the same experiments
utilizing our knowledge of relating genetic/ using wheat flour doughs result in the direct plus
chemical information to quality attributes in interactive effects (Oszvald et al. 2011).
wheat breeding, the grain industry and in
Rice flour contains no wheat prolamin type
basic research. In breeding, the real value of a
proteins and therefore, in principle, it can provide
particular allele has to be investigated in several
a new approach to investigate the functional
backgrounds before its interaction-potential can
properties of wheat proteins. By incorporating
be utilized. Thus, different allelic combinations,
these proteins into rice flour dough, the direct
rather than individual glutenin alleles, should
effects of the proteins on the functional properties
be targeted in breeding situations to develop
of rice dough can be investigated in the absence
new lines with particular quality attributes, but
of interactions with other prolamin type proteins.
especially to improve extensibility.
Previous studies describe a reduction/oxidation
The interactive effects of the alleles present in procedure, optimized for rice flour, which is
commercial wheat flour blends are responsible suitable for incorporating purified wheat protein
for the well-known industrial problem where fractions into rice dough and monitoring their
dough properties, such as dough strength effects on the mixing properties of rice dough.
and extensibility, are not simply additive It was confirmed that dough with reasonable
characteristics, but usually show non-linear strength and stability can be made from rice
relationships with the blend formulation (Békés flour supplemented by wheat gluten (Oszvald
et al. 1998). In applying equation (2) to flour et al. 2008, 2009). Initial results of incorporating
blends; if the αi and βi,j parameters are known, different wheat storage proteins in different
the quality attributes of blends can be estimated, proportions into both rice and wheat flour
providing an efficient way of using samples in doughs demonstrate the potential of using rice
stock (Haraszi et al. 2004), even to develop non- flour as a model system to characterize functional
linear optimization model methods for blend properties of wheat storage proteins (Oszvald et
formulation (Békés 2011). al. 2011, 2012).
Further genetic and biochemical investigations
Estimating end product quality
are needed in the future to satisfactorily
explain the molecular aspects of allele-allele Since the pioneering work of Finney and
interactions. This requirement underlines the Barmore (1948), various researchers have
potential for more intensive application of attempted to predict bread quality by combining
in vitro methodologies in the initial stages of measurements made from grain, flour, or dough
pre-breeding and germplasm development. and combining them into prediction models
Using the bulk isolated product of the gene of (reviewed by Békés 2012b). Among these
interest (produced either from the flour of the predictive models the most successful tool for
source of the allele or bacterial expression) in estimating loaf volume seems to be PQI, a new
‘incorporation’ or ‘model dough’ experiments, generation predictive model (Békés et al. 2006).
the interaction-potential of a certain gluten Previous models did not deal with the fact that
protein (glutenin allele) can be monitored in a most of the relationships between grain or flour
reasonably cheap and rapid set of experiments. parameters and loaf volume are not linear and
there is an optimum level of energy required to
However, the principal limitation of these produce the largest volume loaf. Under- or over-
incorporation studies is that the results of such mixing will produce loaves of inferior volume and
experiments are largely dependent on the allelic quality. For any set of technological parameters
composition of the base flour (Békés 2012a, b). and ingredients, there is an optimal dough
The resulting changes in rheological properties strength and extensibility. The PQI model applies
of the wheat dough are partly dependent on the the Morup–Olesen transformation (Morup and
direct contribution of the incorporated protein

Olesen 1976) (originally developed for describing Conclusions
the effects of individual essential amino acid Mathematical models for the estimation of
levels in food or feedstuffs on biological value) dough properties and end-product quality
on certain dough parameters prior to multiple can successfully be adopted to utilize genetic/
regression. Using only such parameters, an r2 biochemical and functional data from high
value greater than 0.85 can be achieved with a low throughput methodologies to intensify quality
standard error of prediction on loaves produced oriented selection in breeding programs and
using commercial bread-making formulations. blending formulations in the grain industry.
The model can be applied by using data on
dough property parameter predictions based on
spectrophotometric techniques, making the end- References
product quality estimation incomparably cheaper Barro F, Barcelo P, Rooke L, Tatham AS, Békés F, Shewry
than carrying out baking tests. Such models have PR, Lazzeri P (1997) Improvement of the processing
potential for the baking industry to evaluate properties of wheat by transformation with HMW
source materials and/or be part of the quality subunits of glutenin. Nature Bio/Technology,
control in the bakery. 15:1295-1299
Beasley HL, Blanchard CL, Békés F (2001) Preparative
Beyond gluten proteins – estimating water method for in vitro production of functional polymers
from the glutenin subunits of heat. Cereal Chem
absorption
78:464-470
Together with the new level of knowledge on Békés F (2011) Studying the protein-protein interactions
traditionally considered quality parameters, and functional properties of the wheat storage proteins
such as dough strength and extensibility, several in a gluten free model system. Agro Food Ind High-
new aspects of ‘quality’ become more important Tech 22:1-5
and require deeper understanding in relation Békés F (2012a) New aspects in quality related wheat
to the genetics and to the composition of wheat research: I. Challenges and achievements (Review).
Cereal Res Comm 40:159-184
flour. Traits, directly related to the economics of
Békés F (2012b) New aspects in quality related wheat
production such as milling yield, the amounts of
research: II: New methodologies for better quality
water and energy required to make dough with wheat (Review). Cereal Res Comm DOI: 10.1556/
optimal consistency become significantly more CRC.2012.0003
important than earlier. These alterations in the Békés F, Gras PW, Gupta RB (1994a) Mixing properties
definition of good quality require a more complex as a measure of reversible reduction/ oxidation of
procedure of quality evaluation of wheat, among doughs. Cereal Chem 71:44-50
which the traditional gluten-protein based Békés F, Gras PW, Gupta RB, Hickman DR, Tatham AS
characterization covering the monitoring of the (1994b) Effects of 1Bx20 HMW glutenin on mixing
chemical composition of soluble components will properties. J. Cereal Sci 19:3-7
be supplemented by both non-gluten proteins Békés F, Kemény S, Morel M (2006) An integrated
approach to predicting end-product quality of wheat.
and non-starchy polysaccharides, as well as the
Eur J Agron 25:155-162
study of starch-related characteristics. Based
Békés F, Larrocque O, Hart P, O’Riordan B, Miskelly D,
on both in vivo and in vitro studies it seems that
Baczynski M, Wrigley CW (1998) Non-linear behavior
polymeric glutenin and monomeric gliadin of grain and flour blends made from dissimilar
proteins make almost identical contributions to components. In: Tarr AW, Ross AS, Wrigley CW (eds)
water absorption. Hence, this very important Proc. 48th RACI Conf. RACI, Melbourne, Australia,
quality attribute does not depend on glutenin pp123-127
or gliadin alleles, or on the glutenin to gliadin Békés F, Rakszegi M, Balázs G, Kovács A, Ma W, Láng
ratio. By applying analytical data related to the L, Tömösközi S, Bedő Z (2012) The relationships
of breadmaking properties of wheat with the
non-gluten components of the flour, a reasonably
composition of prolamins and soluble proteins and
good prediction model can be developed by non-starchy carbohydrates. In: Bedő Z, Láng L (eds)
using the soluble protein (albumin plus globulin) Proc. 11th Eucarpia General Congress, Budapest,
and total AX content of the flour, together with MTAKI, Martonvásár, Hungary, p171.
starch damage (r2>0.75) (Békés et al. 2012)
Cornish GB, Békés F, Eagles HA, Payne PI (2006) Oszvald M, Tömösközi S, Tamás L, Keresztényi E, Békés
Prediction of dough properties for bread wheats. In: F (2008) Characterization of rice proteins by size
Wrigley CW, Békés F, Bushuk W (eds) Gliadin and exclusion high performance liquid chromatography. J
glutenin. The unique balance of wheat quality. AACCI Cereal Sci 48:68-76
Press, St Paul, MN, pp243-280 Oszvald M, Tömösközi S, Békés F, Tamás L (2009) Effects
Eagles HA, Hollamby GJ, Gororo NN, Eastwood RF (2002) of wheat storage proteins on the functional properties.
Estimation and utilisation of glutenin gene effects from J Agric Food Chem 57:10442-10449
the analysis of unbalanced data from wheat breeding Oszvald M, Balázs G, Tömösközi S, Békés F, Tamás L
programs. Aust J Agric Res 53:1047-1057 (2011) Comparative study of the effect of incorporated
Finney KF, Barmore MA (1948). Loaf volume and protein individual wheat storage proteins on mixing
content of hard winter and spring wheats. Cereal properties of rice and wheat dough. J Agric Food
Chem 25:291-312 Chem 59:9664-9672
Gras PW, Békés F (1996) Small-scale testing: The Oszvald M, Balázs G, Tömösközi S, Békés F, Tamás L
development of instrumentation and application as a (2012) Studying the protein-protein interactions and
research tool. In: Wrigley CW (ed), Proc. 6th Int. Gluten functional properties of wheat prolamins in gluten
Workshop. RACI, Melbourne, Australia, pp506-510 free model system. In: He ZH (ed) Proc. 11th Int Gluten
Gupta RB, Békés F, Wrigley CW (1991) Prediction of Workshop, Beijing, CIMMYT
physical dough properties from glutenin subunit Payne PI, Nightingale MA, Krattiger AF, Holt LM (1987)
composition in bread wheats: correlation studies. The relationship between HMW glutenin subunit
Cereal Chem 68:328-333 composition and the bread-making quality of British-
Haraszi R, Békés F, Ruggiero K, Gale KR, Anderssen RS grown wheat varieties. J Sci Food Agric 40:51-65
(2004). Analysis of wheat grain blends. In: Black CK, Tamás L, Solomon RG, Békés F, Gras PW, Morell MK,
Panozzo JF, Rebetzke GJ (eds) Proc. 54th Australian Cer Appels R (2002) Modification of chain termination
Chem Conf and 11th Wheat Breeders Assembly, RACI, and chain extension properties by altering the density
Melbourne, Australia, pp362-365 and distribution of cysteine residues in a model seed
Morup IK, Olesen ES (1976) New method for prediction of storage protein: Effects on dough quality. J Cereal Sci
protein value from essential amino acid patterns. Nutr 36:313-325
Rep Int 13:355-365

Reliability of gluten-related small-scale tests to estimate
dough viscoelasticity and bread loaf volume
R.J. Peña, N. Hernandez-Espinosa, G. Posadas-Romano, and C. Guzman
CIMMYT, Km 45 Carr. Mexico-Veracruz, El Batan, Texcoco, Mexico, CP 56130, Mexico
Abstract
Dough visco-elasticity is the main factor defining the bread-making quality of wheat. In breeding to
improve bread-making quality it is necessary to apply selection pressure as early as possible (at least
at the early-advanced stages). The Mixograph and the Alveograph are commonly used methods to
determine dough mixing properties (development time, DDT and work input, %T), dough visco-
elasticity (deformation energy or dough strength; ALVW, and tenacity/extensibility; ALVPL ratio).
Bread loaf volume is the ultimate parameter indicating bread-making quality. However, all these
methods are time consuming and slow, with little application in early stages of selection for quality
traits. The present study evaluated three small-scale tests: sodium dodecyl sulfate-sedimentation
(SDS-S), lactic acid retention capacity (LARC), and swelling index of glutenin (SIG), for their value in
estimating dough visco-elastic properties (DDT; %T; ALVW; ALVPL) and bread loaf volume (LV). A
large population (242) of advanced wheat lines was used. SIG and LARC were strongly interrelated (r
= 0.92) and both showed larger correlation coefficients than SDS-S with dough strength parameters and
bread LV, although SIG showed a larger correlation coefficient than LARC to screen for dough strength
(r: 0.77 vs. 0.83, respectively) and bread LV (r: 0.54 vs. 0.62, respectively). Testing for LARC is simpler
and faster than for SIG. Therefore, the use of either LARC or SIG to screen for dough strength and bread
LV depends on the strength of the selection pressure applied to early advanced wheat germplasm.
Introduction There are rapid, small-scale methods requiring

In breeding for bread-making quality it is small quantities of flour that are or could be
necessary to screen experimental lines with used to roughly estimate dough visco-elasticity
respect to several grain quality attributes, and bread-making quality. These methods are
especially for dough visco-elastic properties, mainly focused on the quantity and quality of
which are the main factors defining processing- the insoluble glutenin fraction of the gluten
and end-product quality. Among the several protein, the most important compositional factors
instruments used to determine dough visc- determining differences in bread-making quality
oelastic properties, the Mixograph (National among cultivars. Glutenin has the ability to swell
Mfg. Co., USA) determining dough mixing in various non-reducing solvents such as diluted
properties, and the Chopin Alveograph (Tripette weak acids and sodium dodecyl sulfate (SDS)
& Renaud, France) determining dough strength solutions. The swelling volume of flour or flour-
and extensibility, have demonstrated their high derived suspensions is directly related to the
value in breeding programs to identify advanced quantity and quality of the glutenin protein in the
lines possessing desirable dough strength, dough flour (Weegels et al. 1996). Among these small-
extensibility, and bread-making quality, as results scale methods, the SDS-sedimentation (SDS-S)
are confirmed when the lines are evaluated with test is based on the flour being suspended in
actual bread-making tests. However, Mixograph, a SDS-lactic acid solution. The SDS-S volume
Alveograph, and the bread-making tests take shows a good correlation with loaf volume and
time and require flour in amounts of 35, 60, and other rheological parameters (Peña et al. 1990).
100 g, respectively; amounts that are often not Another small-scale test that has shown a strong
available for segregating or early advanced lines relationship with glutenin content and with
of a wheat breeding program. Therefore, these bread-making quality-related parameters is the
tests may not be useful in the early stages of the swelling index of glutenin (SIG), in which the gel
breeding process. protein formed mainly by the swollen insoluble
glutenins in an isopropanol-lactic acid solution Lactic acid retention capacity (LARC)
is quantified (Wang and Kovacs 2002). The lactic LARC was determined in flour samples using
acid retention capacity (LARC) test (AACC (2000) a scaled-down version of the standard LARC
Method 56-11, 2000) is another small-scale method method of the AACC Method (56-11). In the
that has gained some recognition as a valuable miniaturized method, 300 mg of flour was placed
predictor of dough visco-elastic properties (Bettge into a 2.0 ml previously weighed Eppendorf
et al. 2002; Duyvejonck et al. 2011), and can be tube; followed by the addition of 1.5 ml of 5%
considered an additional tool for predicting lactic acid solution and vortex-mixed for 10-
dough rheological properties and bread-making 20 sec to fully suspend the flour. The tubes
quality in breeding programs. were immediately put in a Thermomixer with
The aim of this study was to determine and mixing action at 1400 rpm for 5 min (25°C);
compare the reliability of three different gluten- and centrifuged at 4000 x g for 2 min. After
quality related small-scale tests (SDS-S, LARC, centrifugation the supernatant was decanted
and SIG) as tools to predict dough rheological and the tube was left to drain on tissue paper
parameters and bread loaf volume. for 10 min. Finally, the tube was weighed, and
the LARC was calculated following the formula:
[(Tube and gel weight – empty tube weight)/
Materials and methods Flour weight)] (86/100 flour moisture) -1 * 100.
Plant material and grain characteristics Swelling index of glutenin (SIG)

A large set (242 genotypes) of bread wheat This test, based on the method described by
(Triticum aestivum) advanced lines belonging to Wang and Kovacs (2002), was scaled down as
various elite yield trials of the International Maize follows. Forty mg of flour was placed into a
and Wheat Improvement Center (CIMMYT) previously weighed 2.0 ml Eppendorf tube;
wheat breeding program were grown under followed by the addition of 0.8 ml of distilled
diverse climatic and management conditions (but water and vortex-mixed for 5 sec to fully hydrate
with no limitation in N fertilizer) at Cd. Obregón, the flour. The tubes were immediately put in a
Sonora, Mexico, during the 2010-11 crop cycle. The Thermomixer with mixing action at 1400 rpm
samples were selected from large populations and for 10 min at 25°C; 0.4 ml of isopropanol-lactic
different nurseries with the purpose of including acid (lactic acid 4.5% v/v, isopropanol 20% v/v)
genotypes representing a wide range of variability solution was added. The tubes were mixed and
with respect to grain hardness (soft to hard), placed again in the Thermomixer (1400 rpm)
dough visco-elasticity (strength and extensibility), for 10 min, and then centrifuged at 100 x g for 5
and bread-making quality. min. After centrifugation the supernatant was
Flour samples were obtained with a Brabender decanted and the tube was drained on tissue
Quadrumat Sr. mill. Grain hardness, moisture, paper. Finally, the tube was weighed, and the SIG
and grain and flour protein were determined was calculated following the formula: [(Tube and
by near-infrared spectroscopy (NIRS) (Foss gel weight – empty tube weight)/Flour weight)]
NIRSystems INFRATEC 1255, Sweden) calibrated (86/100 flour moisture).
based on AACC (2000) methods for particle size
Dough rheological parameters and bread-
index (Method 55-30); moisture (Method 44-15A);
making quality
and protein (Method 46-11A). Lower hardness
index (%) values correspond to harder cultivars. Mixograph (National MFG. Co.) dough mixing
Grain protein and flour protein were adjusted to properties were determined according to AACC
12.5% and 14% moisture bases, respectively. Method 54-40A with 35 g of flour, and Chopin
Alveograph dough parameters were determined
SDS-sedimentation (SDS-S) using AACC Method 2000, but was adjusted
This test was carried out following the for dough water absorption of 52-53% when the
protocol of Peña et al. (1990), using a 1 g flour grain hardness values indicated medium-hard to
sample suspended in 25 ml of solution. SDS- hard grain. Mixograph dough development time
sedimentation volume achieved with this test (DDT) and %Torque (%T), Alveograph strength
ranges from 8 to 23 ml. (ALVW) and tenacity-extensibility ratio (ALVP/L)

were recorded. Bread-making was performed correlation coefficients were obtained between
using 100 g flour samples according to AACC protein content and ALVW with all the small-
Method 10-09 and bread loaf volume (LV) was scale tests (Fig. 2). LARC and SIG also showed
measured by rapeseed displacement. a highly significant correlation with the dough
extensibility parameter, ALVPL, although the
correlations were not high enough (0.27 and
Results and discussion 0.24) to consider LARC and SIG as good ALVPL
The wheat population included samples predictors. This indicates that dough extensibility
showing a wide range of values for the grain is influenced by gluten proteins, and also by
characteristics examined. Mean grain hardness other grain components contributing to dough
was 46.1%, with a range of 30–62%. Mean flour viscosity. Until now, there has been no small-
protein was 11.5%, with a range of 8.8–17.0%. scale test or indirect parameter that can reliably
The small-scale tests, SDS-S, LARC, and SIG, also predict dough extensibility.
showed a wide range of values (9.5 to 23 ml, 72.2 In choosing the use of a rapid-small-scale
to 187.5%, and 3.28 and 6.7 for swelling index, method to screen for quality attributes in a wheat
respectively). The population studied included breeding program, it is important to consider the
advanced lines varying from weak to very number of samples that can be screened per unit
strong gluten as well as limited to large dough time (hours or days). In regard to the small-scale
extensibility, resulting in very low to high bread methods compared in this study, the SIG method
loaf volumes (Table 1). Thus, the population is more laborious and takes more time than the
examined covered a very wide range of quality
attributes.
The correlations among the three small-scale 1
methods were determined to examine how
closely these three tests were interrelated. Both,
0.8
LARC and SIG showed similar highly significant
correlations with SDS-S (Fig. 1), explained by
the strong relationship (r = 0.92) between LARC 0.6
and SIG, and indicating that both methods are
influenced by similar gluten protein fractions, 0.4
and are therefore expected to show similar value
in predicting gluten-related quality traits. 0.2
Correlation analysis was performed to compare
the predictive values of SDS-S, LARC, and 0
SIG. All three small-scale methods showed SD-S vs. LARC SDS-S vs. SIG LARC vs. SIG
highly significant (p>0.01) relationships with
flour protein, dough rheological parameters Fig. 1 Pearson’s correlation coefficients. SDS-S vs.
and bread loaf volume (Fig. 2). SIG showed LARC, SDS-S vs. SIG and LARC vs. SIG were
larger correlation coefficients than LARC highly significant (p >0.01). SDS-S = sodium dodecyl
and SDS-S with flour protein, all the gluten sulfate-sedimentation; LARC = lactic acid retention
strength parameters, and bread LV. The largest capacity; SIG = swelling index of glutenin.
Table 1. Mean and range of values for dough rheological parameters and bread loaf volumes of diverse
advanced bread wheat lines (n = 242).
Mixograph dough Mixograph Alveograph Alveograph Bread

development % torque strength Tenacity-extensibility loaf volume
Value time (min) (W x 10-4 J) ratio (ml)
Highest 8.0 284.3 790 2.2 975
Lowest 1.1 27.9 42 0.3 575
Mean 3.5 127.4 343 0.9 808
0.90 Conclusions
0.80 SDS-S The small-scale tests, SDS-sedimentation, lactic
LARC
0.70 SIG acid retention capacity and swelling index of
glutenin, showed different prediction values for
0.60 bread-making quality parameters. SIG appeared
0.50 to be the best predictor (followed by LARC and
then by SDS-S), particularly of dough strength
0.40
and bread loaf volume. LARC is a simpler and
0.30 faster small-scale test, and has only slightly lower
0.20 prediction value than SIG in estimating dough
strength and bread LV. The choice of either test
0.10 over SDS-S depends on the selection pressure
0.00 being applied on early advanced lines.
FLRPRO MIXTIM %TQ ALVW ALVPL LOFVOL
Fig. 2 Pearson correlation coefficients between small-
scale tests (SDS-S, LARC and SIG) and flour protein References
(FLRPRO), mixograph dough development time
(MIXTIM) and %Torque (%TQ), alveograph strength AACC (2000) Approved methods of the American
(ALVW) and tenacity/extensibility ratio (ALVPL), and Association of Cereal Chemists, (10th ed). The
bread loaf volume (LV). SDS-S = sodium dodecyl American Association of Cereal Chemists, St Paul, MN
sulfate-sedimentation; LARC = lactic acid retention Bettge AD, Morris CF, Demacon VL, Kidwell KK (2002)
capacity; SIG = swelling index of glutenin. Adaptation of AACC Method 56-11, Solvent Retention
Capacity, for use as an early generation selection tool
for cultivar development. Cereal Chem 79:670-674
Duyvejonck AE, Lagrain B, Pareyt B, Courtin CM,
Delcour JA (2011) Relative contribution of wheat flour
LARC and the SDS-S tests. Between LARC and constituents to Solvent Retention Capacity profiles of
SIG, LARC is faster, because it is easier to weigh European wheats. J Cereal Sci 53:312-318
300 mg than 40 mg, and because the mixing and Peña RJ, Amaya A, Rajaram S, Mujeeb-Kazi A (1990)
centrifugation times are smaller and the steps are Variation in quality characteristics associated with
some spring 1B/1R translocation wheats. J Cereal Sci
fewer. Sample weighing (1 g of flour) is much
12:105-112
faster in SDS-S. Considering testing time and
Weegels PL, Hamer RJ, Schofield JD (1996) Functional
simplicity, LARC seems to be a better screening properties of wheat glutenin. J Cereal Sci 23:1-18
method than SIG; however, if the prediction
Wang C, Kovacs MIP (2002) Swelling Index of Glutenin
value is critical, then SIG should be preferred Test. I. Method and comparison with Sedimentation,
over LARC and SDS-S to predict dough strength Gel-Protein, and Insoluble Glutenin Tests. Cereal
and bread-making quality. Chem 79:183-189

New possibilities in micro-scale wheat quality
characterization: Micro-gluten determination and starch
isolation
S. Tömösközi1, S.Z. Szendi1, A. Bagdi1, A. Harasztos1, G. Balázs1, D. Diepeveen2, R. Appels3, and F. Békés4
1Budapest University of Technology and Economics (BUTE), Department of Applied Biotechnology and Food
Science, Budapest, Hungary; 2Department of Agriculture and Food, South Perth, Western Australia; 3Center for
Comparative Genomics, Murdoch University, Perth, Western Australia; 4FBFD Pty Ltd, Beecroft, NSW 2119,
Australia
Abstract
When sample size is limited, small-scale dough testing methods are needed. Since the development
of the 2 g Mixograph, the first commercially available small-scale dough testing instrument, followed
by the appearance of micro-extension instruments and micro-dough inflation procedures, intense
research activities have been carried out to evaluate different micro-scale systems as well as studying
the effects of reduced sample quantities on the measured parameters. During the last decade, a family
of micro-scale instruments have been developed through long-term cooperation between Hungarian
and Australian partners; resulting in prototypes of a micro-mill and a micro-sieve, a Farinograph/
Valorigraph-type micro-Z-arm mixer, and a combined macro- and micro-sedimentation tester
(Sedicom®) followed by commercialization of these instruments. Recently, a new type of gluten and
starch washer (GluStar®, BME-Labintern, Hungary) was constructed. The novelty of this system is its
capability of determing wet gluten content and isolating eluted starch and soluble components of flour
from one cycle. Both macro- and micro-modes of application are possible, with sample requirements
of 10 g or 4 g of flour, respectively. The modularity of the measuring system ensures the separate
utilization of the basic modules (gluten washer, and carbohydrate collector). The gluten washer alone
is applicable for measuring wet gluten content. The method was validated, and results were found to
be well correlated with generally used standard methods. The equipment has been used in various
research studies, such as screening large sample populations of Hungarian and Australian wheat
lines, with satisfactory reproducibility in wet gluten content and providing starch samples for further
chemical and functional tests. Some examples for the application of the new equipment are given,
including an investigation of the impact of the AM/AP ratio and the arabinoxylan (AX) content on
rheological properties of samples for producing blends of conventional bread wheat flours with high-
amylose and waxy wheat flours.
Introduction parameters. In the last ten years, a family of

In the early stages of breeding programs, only a micro-scale instruments has been developed in
few seeds are available for the determination of cooperation between our Department and other
chemical, physical and rheological properties. Hungarian firms and Australian partners, namely
Since the development of 2 g Mixograph, the first the Wheat Quality CRC, Newport Scientific Ltd,
commercially available small-scale dough testing Metefém Ltd and LAB-INTERN Ltd. (Tömösközi
instrument (Rath et al. 1990; Gras and Békés et al. 2000, 2009) In this work, a micro-mill,
1996), followed by the appearance of the micro- micro-sieve, farinograph/valorigraph-type micro-
extension instruments (Békés et al 2001; Kieffer et Z-arm mixer, sedimentation tester (SediCom
al. 1998) and micro-dough inflation procedures System®), and recently, a combined gluten
(Dobraszczyk 1997), intense research activities washer (GluStar System®) were developed. Most
were carried out in evaluating different micro- of the small-scale methods were carried out
scale instruments as well as studying the effects following the standard procedures. The only
of reduced sample quantities on the measured exception is development of the GluStar system,
a simple but new measuring procedure. In this For determination of wet-gluten content and
case, a combined instrument was developed in starch (carbohydrate) isolation, the new gluten
which gluten washing and starch (unsoluble washer system (GluStar® System) was used (Fig. 1).
carbohydrate) isolation can run separately, or Standard macro- and micro-scale procedures were
in parallel on macro- and micro-scales. This developed for measuring 10 and 4 g flour samples,
flexible instrument and associated methods allow respectively. The gluten washing procedures
both routine analytical and research activities. were validated against the ICC-Standard Method
A study of the roles of starch and non-starchy No. 106/2. AM/AP ratio was measured by the
carbohydrates in wheat dough formation, and the iodine binding method (Hoover and Ratnayake
rheological behavior of the dough, was one of the 2001) and AX content was determined by
focus areas in the present research. gas chromatography (Gebruesrs et al. 2009).
Viscosimetric characterization of wheat flours and
isolated carbohydrates were performed with a
Materials and methods Rapid Visco Analyser (RVA, Perten Instruments.
Various Hungarian and Australian wheat CITY). Standard statistical evaluation was
samples were used in the experiments. A total performed on the basis of the results of parallel
of 740 flour samples from lines in a MAGIC measurements using the Statistica Ver 6 package.
population (4-way cross) established in CSIRO
Plant Industry (Cavanagh et al. 2008) were
analyzed using the GluStar equipment. For the Results and discussion
study of amylose/amylopectin (AM/AP) and The novelty of the newly developed analytical
arabinoxylan (AX) effects, 26 old Hungarian equipment is its capability of simultaneously
landraces and varieties recently bred at the determining wet gluten content and separating
Agricultural Institute, Center for Agricultural the starch from wheat flour samples of 10 g or 4 g.
Research, Hungarian Academy of Sciences, Both the macro- and micro–methods take the same
Martonvásár, were investigated. time. The modularity of the measuring system
ensures separate utilization of the basic modules.
Fig. 1 GluStar®, the new member of small-scale instruments for determining gluten contents and
separating wheat carbohydrates.

The gluten washer is applicable for measuring viscous properties of wheat flours and from
wet gluten content in one step. The GluStar carbohydrates eluted from the same flours was
macro- and micro-methods were validated; carried out. As expected, the RVA profiles were
the results are well correlated (r = 0.993 and 0. different in each case: the breakdown after the
988, respectively) with generally used standard peak viscosity and the setback at the final stage
methods (Fig. 2). being lower in the case of the isolated starches.
Differences among the RVA viscosity profiles
As part of the complex genetic, proteomic and
of starchy phases isolated from different wheat
functional evaluation of the MAGIC population
varieties showed more variation than in cases of
(Cavanagh et al. 2008) wet gluten content of
wheat flours (Fig. 3).
samples were determined, with samples of cv.
Carnamah used as controls at every tenth run.
Wet gluten content of the 74 Carnamah controls
ranged from 1.295 to 1.405 g/4.5 g flour and 4,500
hence the reproducibility of the measurement Isolates
4,000
was good. The variation in wet gluten content Flours
assayed by GluStar for the set of samples from 3,500
the structured MAGIC population ranged from 3,000
0.95 g to 2.8 g wet gluten/4.5 g flour and within
Viscosity (cP)
2,500
this analysis, the reproducibility was consistent
with that defined by the Carnamah control 2,000
samples. A particularly useful aspect of the small- 1,500
scale analyses, in addition to the % wet gluten
1,000
measurement, was the fast qualitative assessment
of the dough formed by the flour samples. 500
The GluStar filter module connected to a vacuum 0

0 100 200 300 400 500 600 700 800
system (Fig. 1) is suitable for separating wheat time (s)
starch and other carbohydrates from the effluent
of the gluten washing process. As one of the Fig. 3 Application of GluStar® system in research:
first examples for the scientific application of Comparison of viscosity (RVA) profiles in flours and
the GluStar system, an investigation of the separated (isolates) starch.
A) Standard methods vs Macro GluStar System B) Standard methods vs Micro GluStar System
Correlation: r= ,99298 Correlation: r= ,98815
42 42
40 40
38 38
Micro GluStar System
Micro GluStar System
36 36
34 34
32 32
30 30
28 28
26 26
24 24
22 22
24 26 28 30 32 34 36 38 40 42 44 24 26 28 30 32 34 36 38 40 42 44
Glutomatic 2,200 0,95 Conf. Int. Martonvás ás 0,95 Conf. Int.
Fig. 2 Method validation: Comparison of wet-gluten content measured with the macro- and micro-
GluStar® methods and the ICC-Standard No. 106/2 procedure.
The composition of the eluted starchy materials the viscous behaviour of these starchy isolates
was also investigated. One of the interesting was determined not only by the AM/AP rates and
observations was that the water unextractable sizes of molecules, but also by the NSPs.
arabinoxilane (WUAX) content of starchy
Simple correlation studies showed different
isolates is higher than, or almost equal to,
relationships between viscosity parameters
those for wheat flours (Fig. 4). Naturally, the
determined by the RVA method and the
water extractable arabinoxilanes (WEAX) in
composition of starchy isolates. For example,
washed out starches are much lower than in
it is clear in Table 1 that the breakdown after
wheat flours. These results indicate that the
peak maximum and the time required to reach
non-starchy carbohydrates (NSPs) are (partly)
maximum viscosity is significantly correlated
washed out from the gluten during gluten
with amylose content whereas setback apparently
washing and remain in the effluent. Therefore,
depends more on WEAX.
WUAX Flour Isolate WEAX
TF-RETSAG TF-RETSAG
TF-GYULAVARI TF-GYULAVARI
MV-MAZURKA MV-MAZURKA
MV-MARSALL MV-MARSALL
FLEISCHMANN FLEISCHMANN
-481 -481
BANKUTI-1205 BANKUTI-1205
BANKUTI-1201 BANKUTI-1201
MV-KOLOMPOS MV-KOLOMPOS
0 0,5 1 1,5 0,0 0,2 0,4 0,6 0,8 1,0

g/100g(dm) g/100g(dm)
Fig. 4 Application of the GluStar® system in research: Water extractable (WEAX) and unextractable (WUAX)
arabinoxilane content in wheat flours from various cultivars and carbohydrates separated with GluStar®.
Table 1. Application of GluStar® system in research: Correlation between carbohydrate constitution and
Rapid Visco Analyser (RVA) parameters of separated carbohydrates.
Person’s correlation Water A/X ratio

coefficients. Marked correlations extracteable A/X ratio Total AX in Total Amylose
are significant at p <0,05 AX (DM%) in WEAX (dm%) AX (dm%)
Peak viscosity (cP) 0,25 -0.57 -0,19 -0,1 -0,42
Trough (cP) 0,2 -0,55 -0,22 -0,13 -0,33
Break down (cP) 0,47 -0,46 0,07 0,13 -0,87
Final viscosity (cP) 0,28 -0,61 -0,15 -0,06 -0,38
Setback (cP) 0,61 -0,76 0,2 -0,26 -0,53
Peak time (min) 0,29 -0,14 0,06 0,07 -0,79
Pating temperature (ºC) 0,05 0,39 0,15 0,07 0,09
AX = arabinoxylan; WEAX = water extractable arabinoxilanes.

Conclusions Gebruesrs K, Courtin CM, Delcour JA (2009) Quantification
of arabinoxylans and their degree of branching using
The results from preliminary studies reported gas chromatography. Healthgrain methods, analysis
here show the potential of newly developed of bioactive components in small grain cereals. AACC
equipment expected to become a routine tool in International, Laboratory of Food Chemistry and
breeding and in fundamental research. Biochemistry and Leuven Food Science and Nutrition
Research Centre (LFoRCe), Katholike Universiteit
Leuven, Kasteelpark Arenberg 20, Boksz 2463, 3001
Leuven, Belgium, pp177-189
Acknowledgements Gras PW, Békés F (1996) Small-scale testing: The
This research work was supported by the development of instrumentation and application as a
project “Development of Breeding, Agricultural research tool. In: Wrigley CW (ed) Proc. 6th Int. Gluten
Production and Food Industrial Processing Workshop. RACI, Melbourne, Australia, pp506-510
System of Pannon Wheat Varieties (TECH_09_ Hoover R, Ratnayake WS (2001) Determination of total
A3-2009-0221)”. This work is also connected amylose content of starch. Current Protocols in Food
Analytical Chemistry 1-5. doi:10.1002/0471142913.
to the scientific program of the basic research fae0203s00
project supported by the National Research
Kieffer R, Wieser H, Henderson MH, Graveland A (1998).
Foundation (OTKA-80334). Correlations of the breadmaking performance of wheat
flour with rheological measurements on a micro-scale. J
Cereal Sci 27:53-60
References Rath, CR, Gras PW, Wrigley CW, Walker CE (1990)
Békés F, Gras PW, Anderssen RS, Appels R (2001) Quality Evaluation of dough properties from two grams of flour
traits of wheat determined by small-scale dough using the mixograph principle. Cereal Foods World
testing methods. Austr J Agric Res 52:1325-1338 35:572-574
Cavanagh C, Morell M, Mackay I, Powell W (2008) From Tömösközi S, Varga J, Gras PW, Rath C, Salgó A, Nánási
mutations to MAGIC: resources for gene discovery, J, Fodor D, Békés F (2000) Scale down possibilities in
validation and delivery in crop plants. Cur Opin Plant development of dough testing methods. In: Shewry PS,
Biol 11:215-221 Tatham AS (eds), Wheat gluten. The Royal Society of
Chemistry, Cambridge, UK, pp 321-326. ISBN 0-85404-
Dobraszczyk BJ (1997) Development of a new dough
865-0l
inflation system to evaluate doughs. Cereal Foods
World 42:516-519 Tömösközi S, Nádosi M, Balázs G, Gergely SZ, Cavanagh
A, Morgounov A, Salgó A, Békés F (2009) Revival of
sedimentation value - method development, quality
prediction and molecular background. In: Gluten
Proteins 2009. Proc 10th Int Gluten Workshop, pp104-108
Characteristics and evaluation parameters associated with
cooking quality of Chinese fresh noodles
Y. Zhang1, J. Yan2, Y.G. Xiao1, D.S. Wang1, and Z.H. He1,3
1Instituteof Crop Sciences/National Wheat Improvement Center, Chinese Academy of Agricultural Sciences,
Beijing 100081, China; 2Cotton Research Institute, Chinese Academy of Agricultural Sciences, Anyang 455000,
China; 3CIMMYT China Office, Beijing 100081, China
Abstract
Noodle cooking quality is an important aspect in assessing processing quality of Chinese fresh noodles.
Forty-six Chinese wheat cultivars and advanced lines from the Northern Plain and the Yellow and
Huai River Valleys Winter Wheat Regions were used to determine the relationship between wheat
quality characters and evaluation parameters of Chinese fresh noodle cooking quality, including total
organic matter (TOM), cooking losses, water absorption, and noodle stickiness. Large variations were
observed in milling quality, dough rheology characteristics, starch properties, and noodle cooking
quality parameters, including TOM value, cooking losses, and noodle stickiness. Extensogram energy
and maximum resistance contributed negatively to TOM value, with correlation coefficients of -0.66
(P <0.01) and -0.56 (P <0.01), respectively. Correlation coefficients between farinogram stability,
extensogram energy and maximum resistance and cooked noodle weights with optimal cooking (6 min)
and overcooking (10 min) ranged from -0.55 to -0.63 (P <0.01). Farinogram mixing tolerance index was
significantly and positively correlated with the weights of noodles cooked for 6 min and 10 min, with
correlation coefficients of 0.67 (P <0.01) and 0.69 (P <0.01), respectively. Starch pasting temperature was
significantly and positively correlated with noodle weights when cooked for 10 min (r = 0.60, P <0.01).
This indicated that increased flour protein content and dough gluten strength contributed positively to
noodle cooking quality. Flour protein properties were the major factor in determining noodle cooking
quality. Noodle cooking quality was only slightly affected by starch pasting parameters. TOM value
was significantly and positively correlated with noodle weights after cooking for 6 min and 10 min,
with correlation coefficients of 0.66 (P <0.01) and 0.69 (P <0.01), respectively. The correlation coefficient
between noodle weights after cooking for 6 min and 10 min was 0.86 (P <0.01). It is recommended that
the weight of noodles cooked for 10 min should be an important parameter for evaluation of noodle
cooking quality. Noodle weight (from 10 g fresh noodles) after 10 min of cooking should be no more
than 21.0 g for good noodle cooking quality Chinese wheat samples.
Introduction quality parameters and molecular markers for

Wheat flour noodles, the most popular traditional evaluating noodle quality in Chinese wheat
Chinese food, are also widely consumed in Japan, breeding programs (Wei et al. 1998; Li et al. 2001;
Korea and other Southeast Asian countries. Liu et al. 2002; Lei et al. 2004; He et al. 2007;
Noodle consumption in European and American Zhang et al. 2007). A few Chinese wheat cultivars
countries is rapidly increasing. The various with better noodle quality, such as Jimai 19,
noodles consumed in China include dry, instant, Jimai 20, and Yumai 34, were released recently.
frozen and other types, but Chinese fresh noodles However, many consumers and industrialized
are the most popular. processors have recognized that Chinese wheat
cultivars generally have unsatisfactory noodle
In recent years, many studies have been made on cooking quality. The undesirable cooking
methods for laboratory processing and sensory characteristics are expressed in terms of low
evaluation of Chinese white noodles and the noodle elasticity, excessive stickiness, and more
relationship between flour characteristics and viscous cooking water after slight overcooking.
noodle quality in Chinese wheat cultivars. A Noodles cooked in hot soup are preferred in
number of researchers reported on important China. Lower resistance to softening in hot soup

is also an important factor affecting noodle are still unconfirmed, and a clear major parameter
quality in Chinese wheat cultivars. Hence, for evaluating noodle cooking quality in wheat
understanding noodle cooking quality will breeding programs is not available. Therefore,
help wheat breeding programs to respond to further research is needed to better confirm
consumer and market demands. relationships among parameters for evaluating
noodle cooking quality, and to identify a simple
Quality parameters for evaluating noodle cooking
rapid estimator of noodle cooking quality for use
quality include total organic matter (TOM),
in wheat breeding programs. Currently, there
loss of solids in cooking, water absorption,
are differences in measuring procedures, and
and cooked noodle stickiness and firmness
TOM value, cooking loss of solids and water
(Dexter et al. 1985; D’Egidio et al. 1990). TOM
absorption are included among those that are
determines the surface material released from
used. All are time-consuming. For these reasons,
cooked noodles after exhaustive rinsing. Cooking
46 leading Chinese wheat cultivars and advanced
loss of solids tests the amount of residue in the
lines from major wheat growing areas were used
cooking water. Low TOM value and cooking loss
to investigate relationships with the objective
are indicators of high noodle cooking quality.
of identifying a major parameter for evaluating
The water absorption test is based on cooked
noodle cooking quality in wheat breeding
noodle weight. Low cooked noodle weight
programs.
indicates low noodle water absorption and high
noodle cooking quality. Early studies on pasta
cooking quality found that protein content
and gluten strength were significantly and
Materials and methods
negatively correlated with TOM value, cooking Wheat and flour
loss, and stickiness of pasta (D’Egidio et al. 1990;
Forty six leading Chinese winter wheat cultivars
Malcolmson et al. 1993; Del Nobile et al. 2005).
and advanced lines from the North China Plain
Pasta water absorption decreased as amylase
Winter Wheat Region and Yellow and Huai
content increased (Soh et al. 2006). Some studies
Valley’ Facultative Wheat Region were used in
on Chinese white noodle cooking quality found
this study (Table 1). All grain samples were from
that TOM value can be a useful parameter for
the 2009-2010 cropping season and were free of
assessing noodle cooking quality. Protein content
sprouting damage. These materials represent most
is the major factor affecting TOM value (Zhang et
of the current elite and widely grown varieties.
al. 1998; Zhao et al. 1998). High grain hardness,
Each grain sample was cleaned and tempered
flour yield, wet gluten content, and Farinograph
overnight. Hard, medium and soft wheats were
water absorption contribute negatively to noodle
tempered to around 16.5, 15.5, and 14.5% moisture
water absorption, and flour quality properties
content, respectively. All samples were milled on a
have no association with noodle cooking losses
Buhler MLU 202 laboratory mill (Buhler Bros, Ltd,
(Zhang et al. 2000). However, Zhang et al. (2008)
Uzwil, Switzerland) according to AACC method
reported that flour sedimentation value, dough
26-21A to give flours at 60% extraction rates.
development time, and stability have negative
associations with cooking loss. Cooking loss
Analytical methods
is also slightly affected by total starch content.
Gluten strength is the major factor affecting Grain hardness and moisture were determined
noodle cooking quality. Other researchers using the Single Kernel Characterization System
found that damaged starch and amylose and (SKCS 4100, Perten Instruments, Sweden).
amylopectin contents have significant effects on Flour protein content (14% MB) was recorded
cooking losses (Yin et al. 2005; Wang et al. 2010). with a NIR Analyzer Foss-Tecator 1241 (Foss,
Högänas, Sweden). Ash content, farinogragh,
All of the above studies have two primary and extensograph parameters were obtained
problems: (a) low numbers of samples producing according to AACC Approved Methods 08-01,
limited results regarding the relationship 54-21 and 54-10, respectively. Starch peak viscosity
between flour quality characteristics and noodle and breakdown were obtained using a Rapid
cooking quality; and (b) associations among ViscoAnalyzer (RVA-3D super type, Newport
parameters for evaluating noodle cooking quality Scientific) as described by Batey et al. (1997).
Noodle preparation Parameters of noodle cooking quality
Noodle preparation and quality evaluation TOM value: For each test, 100 g of noodles (20
were performed according to Ye et al. (2010). cm length) were cooked in 1,000 mL of distilled
Flour (200 g, 14% MB) was mixed with enough water previously heated to 100°C for 6 min. The
water to achieve 35% water absorption and 1% TOM value that could be rinsed from the drained
salt using a mixer (Kenwood, Havant, UK). Salt cooked noodles was determined as described by
was dissolved in distilled water and added to Zhang et al. (1998).
the flour during mixing. For sensory evaluation
Cooking losses: The cooking loss test was a
of noodle quality, six samples were tested at
slight modification of that described by Zhang
each panel session including a well-known
et al. (2008). Loss of solids during cooking was
Chinese commercial flour (Xuehuafen) as the
determined by collecting the cooking water
control, which showed a relatively consistent
following drainage of the noodles for the TOM
noodle quality score in sensory tests, and at
value test; 5 ml of cooking water was dried in an
least 5 panelists compared the samples with the
oven at 105°C. The dried residue was weighed
control for six weighted parameters (viz. color
and results were expressed as a proportion of the
15, appearance 10, firmness 20, viscoelasticity
uncooked noodles.
30, smoothness 15, and taste-flavor including
taste and mouth feeling 10) from which a total Water absorption test: The water absorption test
score for each sample was derived. The scores of was a slight modification of that described by
the six parameters for the control were always Matsuo et al. (1986). Seven separate 10 g samples
10.5, 7, 14, 21, 10.5 and 7, respectively. All work of noodles were cooked in 2,000 ml of boiling
was performed under uniform temperature and distilled water for 4, 5, 6, 7, 8, 9 and 10 min,
humidity conditions of the noodle laboratory, respectively. Samples were immersed in stainless
that is, 22±2°C; and 50-60% RH. strainers. At the end of cooking, samples were
Table 1. Name and origin of wheat cultivars used in this study.
Code Cultivar Origin Code Cultivar Origin

1 Beijing 0045 Beijing 24 Shannong 055843 Shandong
2 Zhongmai 175 Beijing 25 Yannong 19 Shandong
3 Zhongmai 415 Beijing 26 Yannong 23 Shandong
4 Zhongmai 548 Beijing 27 Linmai 4 Shandong
5 Zhongyou 206 Beijing 28 Zimai 12 Shandong
6 Zhongyou 335 Beijing 29 Weimai 8 Shandong
7 Zhongyou 629 Beijing 30 Wennong 14 Shandong
8 Lunxuan 987 Beijing 31 Lingxing 66 Shandong
9 Shi 05-7338 Hebei 32 Zhongmai 155 Shandong
10 Jishi 02-1 Hebei 33 Zhongmai 349 Henan
11 Han 6172 Hebei 34 Zhongmai 895 Henan
12 Shun 1718 Shanxi 35 08CA101 Henan
13 Taishan 21 Shandong 36 10CA006 Henan
14 Taishan 23 Shandong 37 10CA11 Henan
15 Taishan 24 Shandong 38 Zhengmai 366 Henan
16 Taishan 223 Shandong 39 Zheng 9023 Henan
17 Tainong 2413 Shandong 40 Taikong 6 Henan
18 Jinan 17 Shandong 41 Aikang 58 Henan
19 Jimai19 Shandong 42 Zhoumai 18 Henan
20 Jimai 20 Shandong 43 Zhoumai 23 Henan
21 Jimai 22 Shandong 44 Xinmai 18 Henan
22 Jining 16 Shandong 45 Xinong 979 Shaanxi
23 Lumai 23 Shandong 46 Xinong 9871 Shaanxi

removed from the strainers, rapidly blotted dry, quality of tested samples (Table 2). The lowest
and accurately weighed. Two determinations ranges occurred in grain protein content
were made for each sample. Cooked noodle and farinograph water absorption, and their
water absorption was calculated from the cooked coefficients of variance (CV) were less than 10%.
noodle weight. However, development time, stability, mixing
tolerance index, energy, and maximum resistance
Stickiness: For each test, 100 g of noodles (20
exhibited the highest range, and their CVs varied
cm length) were cooked in 1,000 ml of boiling
from 60.7 to 85.8%. Higher ranges occurred for
distilled water for 6 min. The cooked noodles
grain hardness and starch pasting parameters
were cooled in the rinse water. Then five strands
such as trough, breakdown, and final viscosity;
per sample were measured for stickiness using a
CVs were more than 20%. This confirmed the
TA-XT2i texture analyzer (Stable Micro System,
need to increase quality screening in Chinese
Surrey, UK) by TPA procedures.
wheat breeding programs.
Statistical analysis
There were also variations in noodle processing
Means, coefficients of variance (CV), and simple quality of tested samples (Table 3). The larger
correlations of cooking quality factors were ranges were shown in noodle sensory color and
calculated on the basis of samples. Multiple noodle cooking quality parameters such as TOM
LSD comparison was made to examine cooked value, cooking loss, and stickiness; and their CVs
noodle weight differences among samples. were more than 15%. Less variation occurred for
All computations were completed using SAS noodle sensory evaluation parameters including
8e software (Statistical Analysis System, SAS appearance, firmness, viscoelasticity, smoothness,
Institute, Cary, NC). flavor and total score, and water absorption
(cooked weights at 6 min and 10 min); their
CVs were less than 10%. Generally, variations
Results and discussion in noodle processing qualities of tested samples
were less than in milling quality, protein and
Overall grain and flour quality performance starch properties.
There were wide variations in milling quality,
dough rheology properties, and starch
Table 2. Mean, range, and coefficients of variance (CV) for quality parameters of tested samples.
Type Parameter Mean Range CV (%)

Milling quality Grain hardness 54.3 9.4–76.0 33.9
Grain protein content (14% MB, %) 13.1 11.4–15.4 7.4
Flour ash (%) 0.46 0.38–0.60 11.5
Dough rheology Water absorption (%) 64.5 53.9–72.5 6.4
Development time (min) 4.4 1.3–19.0 70.8
Stability (min) 6.6 1.1–27.0 85.8
Mixing tolerance index (BU) 63.9 4.0–195.0 66.1
Energy (cm2) 57.3 2.7–158.2 60.7
Extensibility (mm) 165.3 64.1–234.0 16.3
Maximum resistance (BU) 254.3 23.9–722.0 64.9
Starch pasting properties Peak viscosity (RVU) 166.9 82.3–230.1 17.9
Trough (RVU) 127.5 35.3–189.2 23.2
Breakdown (RVU) 39.5 20.8–61.8 22.0
Final viscosity (RVU) 194.7 74.3–270.8 20.9
Pasting temperature (°C) 73.3 66.8–89.8 10.9
Noodle water absorption Zhoumai 23 was apparent with 10 min cooking;
their values were 22.2 and 20.2 g. This suggests
Increasing cooking times at 4, 5, 6, 7, 8, 9, and 10
that noodle water absorption and cooking quality
min led to increases in cooked noodle weights of
for Taikong 6 was similar to Zhoumai 23 at the
tested samples (data not shown). Noodle water
optimal cooking time, but noodle cooking quality
absorption increased, but noodle cooking quality
for Zhoumai 23 was clearly better than for Taikong
decreased, as cooking time increased. Cooked
6 when overcooked. Cooked noodle weights for
noodle weights of four samples at different
Xinong 979 were lowest with values of 16.2, 17.7,
cooking times are listed in Table 4. There were
and 19.8 g at 4 min, optimal, and overcooking
larger differences in noodle water absorption of
times, respectively, indicating the best noodle
different samples at 4 min cooking time, optimal
cooking quality. These results showed that noodle
cooking time (6 min), and with overcooking (10
cooking qualities for Xinong 979 and Zhoumai 23
min). Cooked noodle weight for Linmai 4 was
were better than those for Taikong 6 and Linmai 4.
highest with the values of 18.2, 19.9, and 23.5 g at
4, 6, and 10 min cooking, respectively. The cooked The 46 tested varietal samples were classified
noodle weight for Linmai 4 was significantly into three groups based on cooked noodle weight
higher than those for the other three genotypes, at 10 min cooking time. Nineteen varieties,
indicating lower noodle cooking quality. On the including Xinong 979, Zhoumai 23, Shi 05-7338,
other hand, there were no significant differences Zhongyou 335, Zhongmai 349, Wennong 14, Jishi
between Taikong 6 and Zhoumai 23 with respect 02-1, Zhengmai 9023, Shun 1718, Zhongyou 206,
to cooked noodle weights at 4 and 6 min of 10CA11, Xinmai 18, Xinong 9871, Beijing 0045,
cooking, the values being 17.1 g and 17.0 g versus 10CA006, Zhongmai 895, Shannong 055843,
18.6 g, and 18.5 g, respectively. A significant Zhongyou 629, and Jimai 20, conferred good
difference at P = 0.05 between Taikong 6 and noodle cooking quality with cooked noodle
weights of less than 21.0 g. Among them, only
Table 3. Means, ranges, and coefficients of variance (CV) for processing parameters of noodles.
Trait Parameter Mean Range CV (%)

Cooking quality TOM value 1.00 0.09–2.12 54.2
Cooking loss (%) 5.58 3.73–7.65 16.3
Noodle stickiness -1.96 -4.30 to -1.05 35.4
Cooked weight at 6 min (g) 18.5 17.2–20.6 4.2
Cooked weight at 10 min (g) 21.3 19.8–23.5 4.0
Sensory evaluation Color (15 score) 9.2 6.0–11.8 15.2
Appearance (10 score) 7.1 5.5–8.3 7.7
Firmness (20 score) 13.2 11.5–16.0 9.8
Viscoelasticity (30 score) 19.4 15.8–22.9 8.2
Smoothness (15 score) 10.1 7.8–12.3 9.5
Flavor (10 score) 7.0 5.7–7.9 8.2
Total score (100 score) 66.0 54.4–75.5 7.5
Table 4. Cooked noodle weights (g) of varietal samples after different cooking times.
Variety Noodle cooking time1

4 min 5 min 6 min 7 min 8 min 9 min 10 min
Linmai 4 18.2 a 18.8 a 19.9 a 20.7 a 21.4 a 22.1 a 23.5 a
Taikong 6 17.1 b 17.6 c 18.6 b 19.8 b 20.2 b 21.0 b 22.2 b
Zhoumai 23 17.0 b 17.9 b 18.5 b 18.9 c 19.5 c 20.1 c 20.2 c
Xinong 979 16.2 c 16.7 d 17.7 c 18.2 d 18.5 d 19.2 d 19.8 d
1Values in columns followed by different letters are significantly different (P <0.05) according to t-tests

Wennong 14 had weak gluten strength, Zhongmai content and gluten extensibility have effects
349, Beijing 0045, and Zhongmai 895 exhibited on both cooking loss and noodle stickiness.
medium gluten strength, and the other samples The procedure for determining TOM value
conferred strong gluten strength. Seventeen involves cooking and determining the amount
varieties including, eight weak gluten, six of material rinsed off the noodle surface,
medium gluten, and three strong gluten types drying the rinse water, and dissolving the
conferred fair noodle cooking quality with cooked amount of residue, followed by titration. The
noodle weights of 21.0 to 22.0 g. Ten varieties main shortcoming of the TOM test is that it is
including seven with weak gluten, two medium time-consuming. The method for determining
gluten, and one strong gluten were identified as cooking loss is to determine the residue in the
having poor noodle cooking quality with cooked cooking water by freeze-drying or drying at
noodle weights of more than 22.0 g. The average 130°C. The method is time-consuming and in
cooked noodle weight for the good, fair, and poor some instances gives results that are difficult
noodle cooking quality groups were 17.9, 18.5, to reproduce. Noodle stickiness is generally
and 19.5 g for 6 min and 20.6, 21.4, and 22.6 g for measured using the TA-XT2i texture analyzer, a
10 min cooking times, respectively. Therefore, very expensive instrument that is not affordable
increasing flour gluten strength appears to for some wheat breeding units. However, the
increase noodle cooking quality. procedure for determining cooked noodle
weight is only to weigh the cooked noodles after
Noodle cooking quality parameters: TOM
blotting dry. The procedure is quick and simple.
values were significantly and positively
Assessing noodle cooking quality by cooked
associated with cooked noodle weight at 6 min
noodle weight at 10 min of cooking time could
(r = 0.66, P <0.01) and 10 min (r = 0.69, P <0.01)
be more relevant than 6 min.
cooking time (Fig. 1A). The correlation coefficient
between cooked noodle weights at 6 min and Therefore, it is recommended that cooked
10 min was 0.86 (P <0.01, Fig. 1B). However, noodle weight at 10 min could be an appropriate
TOM values and cooked noodle weights had parameter for evaluation of noodle cooking
no significant relationship with cooking loss quality. The cooked noodle weight at 10 min
and noodle stickiness. The main reason for this (from 10 g fresh noodles) should be no more
may be that TOM value and cooked noodle than 21.0 g for good noodle cooking quality
weight are largely influenced by gluten strength Chinese wheat samples.
and starch pasting properties, whereas protein
6 min cooking time 10 min cooking time

24 24
A) B)
Cooked noodle weight for 10 min (g)
r = 0.69 P < 0.01

23 r = 0.86, P < 0.01
Cooked noodle weight (g)
22
22
20
21
18
r = 0.66 P < 0.01 20
16 19
0.0 0.5 1.0 1.5 2.0 2.5 17 18 19 20 21
TOM value (%) cooked noodle weight for 6 min (g)
Fig. 1 Relationship among evaluation parameters of noodle cooking quality. (A) Relationship between
total organic matter (TOM) value and cooked noodle weight; (B) Relationship of noodle weights between
optimum cooking time and overcooking.
Relationship between wheat quality trait and 2.5
A)
noodle cooking quality
Relationships between TOM value and 2.0 r = 0.66 P < 0.01
parameters farinograph, extensograph, and starch
TOM value
pasting properties are shown in Table 5 and 1.5
Fig. 2. The correlation coefficient between TOM
value and energy and maximum resistance were 1.0
-0.66 (P <0.01) and -0.56 (P <0.01), respectively,
indicating that increasing gluten strength leads 0.5
to decreased TOM. Protein content and dough
extensibility were negatively associated with 0.0
cooking loss, with correlation coefficients of -0.51 0 50 100 150
(P <0.01) and -0.51 (P <0.01), respectively. There Energy (cm2)
were significant relationships between grain 24.5
hardness, gluten strength, and starch pasting B)
Noodle weight (10 min, g)

parameters and cooked noodle weight. The 23.5 r = 0.69 P < 0.01
correlation between grain hardness and cooked
noodle weight at 6 min cooking time was -0.47 (P
22.5
<0.01). Correlation coefficients between stability,
energy, and maximum resistance and cooked
21.5
noodle weight at 6 min and 10 min were from
-0.55 to -0.63 (P <0.01). Mixing tolerance index
20.5
was positively associated with cooked noodle
weight at 6 min (0.67, P <0.01) and 10 min (0.69,
P <0.01), respectively. This shows that increased 19.5
0 50 100 150 200
grain hardness and gluten strength should
Mixing tolerance index (BU)
result in decreased noodle water absorption and
increased cooking quality. These results are in Fig. 2 Relationships between gluten strength
agreement with previous studies (D’Egidio et and noodle cooking quality parameters. (A)
al. 1990; Malcolmson et al. 1993; Zhang et al. Relationship between energy and total organic
1998, 2000, 2008; Zhao et al. 1998; Del Nobile et matter (TOM) value of cooked noodles; (B)
al. 2005). Correlation coefficients between starch Relationship between mixing tolerance index and
pasting trough and final viscosity and cooked overcooked noodle weight.
Table 5. Correlation coefficients between wheat quality traits and noodle cooking quality parameters.
Total organic Cooking Cooked noodle weight

Trait matter value loss At 6 min At 10 min Noodle stickiness
Grain hardness ns ns -0.37* -0.47** ns

Protein content ns -0.51** ns ns -0.41**
Development time -0.33* ns -0.34* -0.36* ns
Stability -0.42** ns -0.59** -0.58** ns
MTI 0.49** ns 0.67** 0.69** ns
Energy -0.66** ns -0.55** -0.61** ns
Extensibility -0.31* -0.51** ns ns ns
Max. resistance -0.56** ns -0.60** -0.63** ns
Peak viscosity -0.44** ns -0.36* ns ns
Trough viscosity -0.36* ns -0.50** -0.30* ns
Final viscosity -0.43** ns -0.52** -0.45** ns
Pasting temperature ns ns 0.48** 0.60** ns
*, ** Significant at P = 0.05 and P = 0.01, respectively; MTI, Mixing tolerance index of Farinograph; ns, not significant

noodle weight at 6 min were -0.50 (P <0.01) and evaluation of noodle cooking quality. Cooked
-0.52 (P <0.01), respectively. Pasting temperature noodle weight at 10 min (from 10 g fresh noodles)
was significantly associated with cooked noodle should be no more than 21.0 g for good noodle
weight at 10 min (r = 0.60, P <0.01), indicating that cooking quality in Chinese wheat samples.
higher starch pasting viscosity and lower pasting
temperature may contribute to lower noodle
water absorption. However, starch pasting Acknowledgments
parameters were not significantly correlated with The work was supported by National Natural
cooking loss, a result not consistent with previous Science Foundation of China (31171547) and
reports (Yin et al. 2005; Wang et al. 2010; Huang an international collaboration project on wheat
et al. 2012). Yin et al (2005) and Wang et al. (2010) improvement from the Ministry of Agriculture of
reported that one sample of commercial flour China (2011-G3).
was used to determine the relationship between
starch properties and noodle cooking quality,
and it was confirmed that damaged starch, total References
starch content, and amylose and amylopectin Batey IL, Curtin BM, Moore SA (1997) Optimization of
significantly affected noodle water absorption Rapid Visco Analyser test conditions for predicting
and cooking loss. Huang et al. (2012) found that Asian noodle quality. Cereal Chem 74:497-501
addition of dry gluten significantly improved the D’Egidio MG, Mariani BM, Nardi S, Novaro P, Cubadda R
resistance of noodles to softening after soaking, (1990) Chemical and technological variables and their
relationships: a predictive equation for pasta cooking
and the addition of 6% potato starch resulted
quality. Cereal Chem 67:275-281
in a significant improvement in firmness of
Del Nobile MA, Baiano A, Conte A, Mocci G (2005)
cooked noodles made from one flour after 20 Influence of protein content on spaghetti cooking
min soaking, but not for noodles made from a quality. J Cereal Sci 41:347-356
different flour. The reason for these differences Dexter JE, Matsuo RR, MacGregor AW (1985) Relationship
may be due to sampling differences between this of instrumental assessment of spaghetti cooking
study and previous reports, and the cooking time quality to the type and the amount of material rinsed
difference between 6 min optimal cooking time from cooked spaghetti. J Cereal Sci 3:39-53
for this research and the 20 min soaking time He ZH, Xia XC, Chen XM, Zhang Y, Zhang Y, Wang
used by Huang et al. (2012). The above results DS, Xia LQ, Zhuang QS (2007) Wheat quality
improvement: history, progress, and prospects. Sci
suggest that increased flour protein content,
Agric Sin 40 (Suppl1):91-98 (in Chinese with English
dough gluten strength and extensibility may abstract)
improve noodle cooking quality. Flour protein Huang SD, Liu JJ, Quail K (2012) Effect of starch quality
properties were the major factor in determining and gluten and starch fortification on noodles’
noodle cooking quality, and noodle cooking resistance to softening. In: Proc 61st Austr Cereal
quality was also slightly affected by starch Chemistry Conf, RACI, Melbourne, Australia. (in
pasting parameters. press)
Lei J, Zhang Y, Wang DS, Yan J, He ZH (2004) Methods
for evaluation of quality characteristics of dry white
Conclusions Chinese noodles. Sci Agric Sin 37: 2000-2005 (in
Chinese with English abstract)
Increased flour protein content and dough gluten Li SB, Shan MZ, Wang Y, Li BY, Zhang SG (2001)
strength contributed positively to noodle cooking Evaluation of wheat quality for wet noodle making.
quality. Flour protein properties were the major Acta Agron Sin 279: 334-339 (in Chinese with English
factors in determining noodle cooking quality, abstract)
and noodle cooking quality was also slightly Liu JJ, He ZH, Zhao ZD, Liu AF, Song, JM, Peña RJ (2002)
affected by starch pasting parameters. The Investigation on relationship between wheat quality
procedure used for determining cooked noodle traits and quality parameters of dry white Chinese
noodles. Acta Agron Sin 28:738-742 (in Chinese with
weight is quick and simple. Assessing noodle English abstract)
cooking quality by cooked noodle weight at 10
Malcolmson LJ, Matsuo RR, Balshaw R (1993) Textural
min seems to be superior to 6 min. It is therefore optimization of spaghetti using response surface
recommended that cooked noodle weight at methodology: effects of drying temperature and
10 min could be an important parameter for durum protein level. Cereal Chem 70:417-423
Matsuo RR, Dexter JE, Boudreau A, Daun JK (1986) The Zhang L Wang XZ, Yue YS (1998) TOM being a new
role of lipids in determining spaghetti cooking quality. assessment method for Chinese noodle cooking
Cereal Chem 63:484-489 quality and effects of wheat quality characteristics on
Soh HN, Sissons MJ, Turner MA (2006) Effect of starch size it. J Chin Cereals Oils Assoc 13(1):49-53 (in Chinese
distribution and elevated amylose content on durum with English abstract)
dough rheology and spaghetti cooking quality. Cereal Zhang GQ, Wei YM, Ou Yang SH, Xi ML, Hu XZ, Sietz W
Chem 83:513-519 (2000) Noodle quality characters of wheat varieties.
Wang XX, Lei H, Qu Y, Liu X, Shi JF (2010) Effect of starch J Chin Cereals Oils Assoc 15(3):5-8 (in Chinese with
composition in flour on noodle cooking quality. J English abstract)
Henan Univ Technol (Nat Sci Edn) 31(2):24-27 (in Zhang Y, Yan J, Yoshida H, Wang DS, Chen DS, Nagamine
Chinese with English abstract) T, Liu JJ, He ZH (2007) Standardization of laboratory
Wei YM, Zhang GQ, Ou Yang SH, Xi ML, Hu XZ, Sietz W processing of Chinese white salted noodle and
(1998) Influence of processing parameters on quality its sensory evaluation system. J Triticeae Crops
properties of spaghetti. J Chin Cereals Oils Assoc 27(10):158-165 (in Chinese with English abstract)
13(5):42-45 (in Chinese with English abstract) Zhang J, Li MQ, Gong XZ, Jiang LY (2008) Relationship
Ye YL, He ZH, Zhang Y (2010) Effect of different water between wheat flour traits and quality indexes of
addition levels on Chinese white noodle quality. Sci fresh-wet noodles. J Chin Cereals Oils Assoc 23(2):20-
Agric Sin 43:795-804 (in Chinese with English abstract) 24 (in Chinese with English abstract)
Yin SW, Lu QY, Yang XG (2005) Study on effect of Zhao ZD, Liu JJ, Dong JY, Zhang L, Li Q (1998) The
damaged starch on cooking quality of noodles. relation between TOM and cooking quality of Chinese
Food Sci Technol (10):68-70 (in Chinese with English noodle. Acta Agron Sin 24:738-741 (in Chinese with
abstract) English abstract)

Optimization and application of capillary electrophoresis
for rapid separation and characterization of water-soluble
proteins in wheat grains
C. Han1, Z. Yu1, S. Feng1, D. Lv1, X. Yan1, G. Chen1, X. Li1, W. Ma2, and Y. Yan1
1College of Life Science, Capital Normal University, Beijing 100048, China; 2Western Australian Department
of Agriculture and Food and State Agriculture Biotechnology Center, Murdoch University, Murdoch, WA
6150, Australia
Abstract
Separation and analysis of water-soluble proteins (WSP) are important in understanding the
fundamentals of the wheat grain proteome. However, due to their high degrees of heterogeneity and
complexity in composition, separating WSP is generally difficult and relevant methodologies are not
yet fully developed. Capillary electrophoresis (CE) is one of the analytical methods currently used for
protein separation and characterization. In the present study, a CE method was optimized for rapid
separation and characterization of water-soluble proteins (WSP) in wheat grains. The established
method was tested in various applications, including wheat varieties and germplasm identification,
as well as protein synthesis and accumulation during different grain development stages and
under different environmental conditions. The characteristic CE patterns of a range of bread wheat
cultivars and related species were readily identified. The synthesis and accumulation patterns of
wheat WSP during grain development, as well as their stabilities in different environments, were also
investigated. The technical advances presented in this paper appear to be useful for wheat cultivar
and germplasm identification, as well as for genetic and breeding research.
Introduction Compared with the gluten proteins, albumins

Wheat is one of the most important grain crops and globulins contain higher levels of essential
in the world. It is an important protein source amino acids such as lysine and methionine
for human consumption and is also used (Singh and Skerritt 2001) that are important for
as raw material for beer, fodder, medicine, balanced diets. Their main roles are to regulate
spices and other products (Shewry et al. 2002). various metabolic processes during wheat growth
Traditionally, wheat grain proteins are classified and development. WSP also have important
into two major groups; the main storage functions and influences on starch and storage
proteins of the endosperm called prolamins protein synthesis as well as their modification,
including gliadins and glutenins, and the transportation, and accumulation (Singh and
nonprolamins consisting of water-soluble (WS) Skerritt 2001). Zawistowdka et al. (1986) found
albumins and salt-soluble globulins (Osborne that the amount of highly polymerized WSP
1907). Albumins and globulins are mainly significantly reduced the bread-making quality,
deposited in the embryo and aleurone layer as probably due to its high affinity to polar lipid
well as partly in the endosperm, accounting materials, thus affecting the loaf volume. To a
for about 9% and 15%, respectively, of the seed certain extent, WSP appear to be associated with
proteins (Singh and Skerritt 2001). In recent the hardness of wheat, and its content in durum is
years, water-soluble proteins (WSP) have higher than in hexaploid bread wheat (Wang et al.
attracted considerable attention and significant 2006). Some WSP are food allergens that can cause
advances have been made in developing anaphylactic reactions, such as baker’s asthma,
methodologies for separating and analyzing urticaria and vomiting, especially in young
them by using a range of instruments (Shomer et children (Weiss et al. 1997). It is generally accepted
al. 1995; Weiss et al. 1997). that wheat albumins and globulins are present in
monomeric form. However, the specific molecular
structural characteristics are still not clear (Wang 8, Jing 411, Gaocheng 8901, Jimai 20, Sunstate,
et al. 2006). During kernel development, albumins Zhongyou 9507 and CB037, which were obtained
and globulins are synthesized earlier than other from the Chinese Academy of Agricultural
proteins, rapidly accumulating in the early filling Sciences (CAAS), Beijing. In addition, four
stages and gradually declining thereafter, but durum cultivars (T. durum, AABB, 2n=4x=28),
slightly rebounding at the mature stage (Zhao and viz. Durum-1, Bonadur, Grandwr and Ofanto,
Yu 2005). and accessions of seven wheat-related species,
viz. tetraploids T. dicoccum (AABB, 2n=4x=28)
Traditional separation methods of wheat grain
TD-05, T. dicoccoides (AABB, 2n=4x=28) TD-91,
proteins include sodium dodecyl sulfate-
and Aegilops peregrina (UUSS, 2n=4x=28) AP-1,
polyacrylamide gel electrophoresis (SDS-PAGE),
and diploids T. urartu (AuAu, 2n=2x=14) TU-7
acidic-polyacrylamide gel electrophoresis
and T. monococcum (AmAm, 2n=2x=14) TM-2,
(A-PAGE) and reversed-phase high-performance
TM-6 and Einkorn-1 (AA, 2n=2x=14), which were
liquid chromatography (RP-HPLC). These
collected from different countries and stored
methods, however, have considerable limitations,
in the Laboratory of Molecular Genetics and
and cannot meet the needs of in-depth dissection
Proteomics, Capital Normal University (CNU),
of the complexity required by modern scientific
Beijing, were used. Three CS alien chromosome
research. In recent years, capillary electrophoresis
substitution lines CS-1C/1A, CS-1Sl/1B and
(CE), two-dimensional electrophoresis (2-DE)
CS-1U/1B, developed at the Plant Breeding
and particularly a variety of biological mass
Department, Technical University of Munich,
spectrometry techniques have been developed in
were kindly provided by Dr. S. Hsam.
the field of structural and functional proteomics
(Gao et al. 2009). CE has also been used to study
WSP extraction and CE
WSP in wheat, allowing a better understanding
of their functions. Grain samples were ground into powder and 200
mg of flour from each were oscillated for about
CE has many advantages compared to traditional 1 h with 0.2 ml ethanol at room temperature
separation methods, including higher resolution, based on a method outlined by Wang et al.
lower reagent consumption and simpler sample (2008). After centrifugation for 5 min at 13,000 g,
requirements, and a high degree of automation. the supernatant was collected, and centrifuged
It provides a powerful tool for studying the again at 13,000 g for 15 min. Finally, 40 μl
genetic and biochemical properties of WSP as supernatant was transferred to a 1 ml centrifuge
well as their contributions to processing quality. tube and used for CE analysis. A Beckman
Nevertheless, after extensive application in PA800 with uncoated fused-silica capillaries
various separation tasks, it is often criticized for and the accompanying 32 Karat operating
its poor reproducibility. In recent years, with software was used for WSP separations. The
the availability of MALDI-TOF technology, CE buffer was 50 mM sodium phosphate, pH 2.5
seems to have lost attention and has become less with 20% (v/v) acetonitrile (ACN), 0.05% (w/v)
used in separation of wheat grain proteins. In the hydroxypropylmethylcellulose (HPMC) and 50
current study, attempts were made to increase mM hexane sulfonic acid (HSA) based on Bean
CE reproducibility in analyzing wheat grain and Tilley (2003). Different separation conditions,
WSP by using a phosphoric acid-glycine buffer including performance voltages, temperature,
system. A fast and efficient separation method inner and outside diameter, and injection time,
was established and its potential application was were optimized. All samples were detected by
investigated. UV absorbance at 200 nm by the traditional way
of using NaOH-HCl as a balanced system for
regulation of pH. However, sodium and chloride
Materials and methods ions may easily interfere with the experimental
Plant materials results at 200 nm of UV absorption wavelength.
The system has high conductivity and therefore
Wheat samples used in this study included produces high currents and joule heating that
ten bread wheat cultivars (Triticum aestivum, causes baseline problems. We therefore used
AABBDD, 2n=6x=42), viz. Chinese Spring phosphoric acid to adjust pH and buffer to
(CS), Zhengmai 9023, Neixiang 188, Jingdong minimize such interference.

To obtain a high reproducibility, new capillaries in unstable peak shapes and repeatability. At
were washed for 5 min with Milli-Q water, 55°C or above, visible dust started to accumulate
20 min with 1 M NaOH, 5 min with Milli-Q inside the instrument, affecting current stability
water, 20 min with 0.1 M NaOH, 5 min with and resulting in failure of electrophoresis.
Milli-Q water, 20 min with 20% H3PO4, 5 min Based on these results, an optimal separation
with Milli-Q water and 20 min with buffer. A temperature for the WSP was set at 35°C.
voltage of 11 kV was then applied for 20 min
Capillary internal diameter generally has two
to the capillary filled with CE buffer to activate
sizes: 20 μm and 50 μm. The column diameter
the capillary. Between each sample, capillaries
directly affects separation efficiency, retention
were rinsed with 0.1 M NaOH (5 min), Milli-Q
characteristics and sample capacity. Small-bore
water (5 min), and buffer (5 min) to ensure high
capillaries have higher column efficiency than
repeatability. After all samples were analyzed,
wide-bore capillaries. However, the capacity
the capillary was flushed with Milli-Q water, 0.1
of a small-bore capillary is too low for efficient
M NaOH and Milli-Q water for 5 min, and then
separation of proteins. Resolution is proportional
air-dried for 3 min.
to the square root of the column length. Longer
columns mean higher resolution and longer
migration times. Usually, capillary columns of
Results and discussion 30 cm (detection length 20 cm) will give optimal
Optimization and reproducibility of CE for resolution and are sufficient for separating
separation of grain WSP water-soluble proteins. In summary, the optimal
capillary column for the separation of WSP is
A high voltage means short component
50 μm i.d. and 30 cm. In theory, the injection
migration times and increased joule heating, but
time only affects the sample capacity; our results
increased column efficiency. However, when the
showed that injection time had no significant
voltage reaches a certain level, excessive heat
effects on the migration time and repeatability.
will be produced and consequently reduce the
Seven seconds or more resulted in an unstable
column efficiency. In addition, high medium
current or short-circuit; demonstrating that the
temperatures can lead to protein denaturation.
most suitable injection time was 6 seconds. In
In our study, operating voltages were optimized
addition, considering the presence of proteases
to 7-15 kV. Considering the instrument lifespan,
in the WSP extracts and possible degradation of
separation degree and migration time, 11 kV
proteins, WSP from samples stored at 4oC for
appeared to be optimal for water-soluble protein
1-2 days were tested, and the CE patterns were
separation.
identical with those from non-stored samples,
Temperature is one of the important parameters suggesting that the protein samples were
that affect electrophoretic patterns. As the stable. The reproducibility of WSP separation
temperature increases, the migration time by optimized CE conditions was tested. Ten
shortens. In general, an increase of 1°C results consecutive runs of WSP from wheat cultivar
in a 2% increase in protein mobility. But the Sunstate displayed high reproducibility.
performance temperature can affect the pH value, Particularly, RSD% of peak migration times, peak
viscosity and other electrophoretic factors, and height and peak area of protein peaks which
further change the electro-osmotic flow, solute were less than 1%, 2% and 4%, respectively,
charge distribution (including the structure of consistently better than previous reports (Wang
the protein) and ionic strength of the solution et al. 2008).
(Francois et al. 1996), leading to low repeatability
(Strege and Lagu 1993), low column efficiency Application of the optimized CE method
and inadequate separation (Guttman 1996). Thus, Cultivar and germplasm identification: As shown
it is critical to select an appropriate temperature. in Fig. 1, the WSP from 5 diploid (Fig. 1a),
A temperature optimization experiment was and 7 tetraploid (Fig. 1b) bread wheat-related
performed. Starting from 15°C, the protein species and 7 hexaploid bread wheat cultivars
separation efficiency gradually increased along (Fig. 1c), as well as 3 CS alien chromosome
with increasing temperature. At 40°C or above, substitution lines (Fig. 1d) were separated and
there were dramatic current fluctuations resulting characterized by the optimized CE method.
Apparent differences in WSP compositions were could be readily differentiated based on WSP peak
present in different species and genotypes. In migration, height and area. Although hexaploid
general, the polymorphisms of WSP in diploid bread wheats showed similar CE patterns of
and tetraploid species were higher than those WSP composition, different cultivars were
in hexaploid bread wheat. The main WSP distinguishable from their peak characteristics and
components were present in 5-6 min in diploid numbers (Fig. 1c). Analysis of CS substitution lines
species, 7-9 min in tetraploid species and 6-8 showed that some WSP peaks could be assigned
min in bread wheat cultivars. Our results to chromosomes 1A, 1B, 1S1, 1C and 1U (Fig. 1d).
indicated that different species and cultivars This shows that CE can be used as a tool to study
the genetic control of WSP.
A) Einkorn-1 B) Durum-1
0.14 0.18
Bonadur
TU-7 0.16
0.12
0.14 Grandwr
0.10
TM-6 0.12
0.08 Ofanto
0.10
0.06 0.08 TD-91

TM-2
0.06
0.04 TD-05
0.04
0.02 Einkorn-2 0.02 AP-1
0.00 0.00
Absorbance at 200 nm
2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10
C) Zhengmai 9023 D)
0.250 0.16
0.225
Jingdong 8 0.14 1B
0.200 1B CS
Neixiang 188 0.12 1B 1B
0.175
0.150 0.10 1S1 1S1 1S1 CS-1S1/1B
Jing 411
0.125 0.08
0.100 Gaocheng 8901 1C
0.06 1C 1C
CS-1C/1A
0.075
Jimai 20 0.04
0.050 1U 1U
1U 1U
0.02 CS-1U/1B
0.025 Sunstate
0.000 0.00
2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10
Tim, min
Fig. 1 Separation and identification of grain WSP from diploid (a), tetraploid (b), hexaploid
bread wheat (c), and CS alien substitution lines (d).

Dynamic accumulation patterns of WSP during stages such as the peaks marked by a. Pattern
grain development revealed by CE: Wheat grain b displayed a stable increase over the first four
protein content and composition are influenced developmental stages and reached the maximum
by both genetic and environmental conditions. level at the fourth stage, but was down-regulated
Temperature is one of the most important to absence at stage 5. In pattern c, WSP had no
environmental factors that can shorten growing apparent expression in the early development
time and influence protein synthesis. In order stages 1 and 2, but was up-regulated from
to minimize experimental error associated stage 3 to mature grain. Pattern d had the
with sampling strategy, we used the method highest expression level at the first period, then
of Majoul et al. (2003) in sampling, for which gradually decreased in the later stages and
samples are classified according to cumulative finally disappeared in the last stage. In addition,
temperatures during growth instead of the WSP expression patterns during grain
number of days after flowering. This sampling development in the two cultivars appeared to be
strategy can avoid protein variations due to similar in the early stages. However, clear WSP
environmental conditions and various genotypic differences occurred from stage 3 to 5.
and physiological traits.
Variation in WSP detected by CE in grain produced at
Through CE analysis of WSP expression different locations: Studies have shown that grain
differences during five grain developmental quality properties are influenced by genotype
stages of two Chinese bread wheat cultivars (G), environment (E) and G x E interactions
Zhongyou 9507 and Jingdong 8, four expression (Basset et al. 1989; Rharrabti et al. 2001). To
patterns (Fig. 2A, a~d) corresponding to different further understand whether CE can be used
WSP components were found. In pattern a, WSP as a tool for investigating variation in WSP
gradually increased over the five development in grain produced in different environments,
stages and reached the highest expression level spring wheat cultivar CB037, grown in Beijing,
at stage 5, indicating that these proteins were Yinchuan, Zhangjiakou and Yuanmou was
mainly synthesized during later embryogenesis analyzed by the optimized CE method (Fig. 2B).
Zhongyou 9507 Jingdong 8 Beijing

0.30 0.225
0.10
0.200
0.25
0.175 0.08
Yinchuan
0.20 0.150
0.06
0.125
0.15 Zhangjiakou
0.100
0.04
0.10 0.075
0.050 0.02 Yuanmou

0.05
0.025
0.00
0.00 0.00
2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10 2 3 4 5 6 7 8 9 10
Tim, min Tim, min
Fig. 2 Applications of capillary electrophoresis (CE). A, Dynamic accumulation analysis of water-soluble

proteins (WSP) in Zhongyou 9507 and Jingdong 8; B, Effects of growing environment on grain WSP.
Grain produced at the four locations generally Gao L, Wang A, Li X, Dong K, Wang K, Appels R, Ma
displayed similar WSP patterns, but clear W, Yan Y (2009) Wheat quality related differential
expressions of albumins and globulins revealed by
quantitative and qualitative differences were
two-dimensional difference gel electrophoresis (2-D
also present. For example, most of the main DIGE). J Proteomics 73:279-296
WSP components, such as the protein peak at Guo G, Lv D, Yan X, Subburaj S, Ge P, Li X, Hu Y, Yan Y
about 7 min, showed greater variations in peak (2012) Proteome characterization of developing grains
height and area, suggesting that the amounts in bread wheat cultivars (Triticum aestivum L.). BMC
of grain WSP were affected by environmental Plant Biol 12:147
conditions. Moreover, the WSP peak at about 7 Guttman A (1996) Effect of the temperature on the peak
min from Yinchuan was divided into major and efficiency in capillary gel electrophoresis. Trends in
minor components and two minor peaks at about Anal Chem 15:194-198
5 min and 7.6 min from Beijing were absent as Majoul T, Bancel E, Triboi E, Hamida JB, Branlard G
(2003) Proteomic analysis of the effect of heat stress
indicated in Fig. 2b, probably resulting from
on hexaploid wheat grain: characterization of heat-
genetic variations induced by environmental responsive from non-prolamins fraction. Proteomics
factors. Additionally, the growing environment 3:175-183
may cause post-translational modifications of Osborne TB (1907) The proteins of the wheat kernel.
protein (Guo et al. 2012), and consequently lead Carnegie Inst Washington Pub 84, 119pp
to qualitative differences in WSP compositions. Rharrabti Y, Villegas D, Garcia D, Moral LF, Aparicio N,
Elhani S, Royo C (2001) Environmental and genetic
determination of protein content and grain yield in
Conclusions durum wheat under Mediterranean conditions. Plant
Breed 120:381-388
We optimized experimental parameters such Shewry PR, Halford NG, Belton PS, Tathan AS (2002) The
as temperature, voltage, buffer, injection time, structure and properties of gluten: an elastic protein
and capillary column type that influence CE from wheat grain. Phil Trans Royal Soc London Ser B -
resolution. An efficient CE procedure was Biol Sci 357:133-142
established for analyzing wheat grain WSP, Shomer I, Lookhart G, Salomon R, Vasiliver R, Bean S
including capillary length; 30.2 cm (detection (1995) Heat coagulation of wheat flour albumins
length 20 cm), inner diameter; 50 μm, operating and globulins, their structure and temperature
fractionation. J Cereal Sci 22:237-249
voltage; 11 kV, capillary temperature; 35°C,
Singh JS, Skerritt J (2001) Chromosomal control of
injection time; 6 s, and detection wavelength;
albumins and globulins in wheat grain assessed using
200 nm. The CE procedure was then applied different fractionation procedures. J Cereal Sci 33:163-
to differentiate various wheat cultivars and 181
genotypes. WSP accumulation patterns across Strege MA, Lagu AL (1993) Studies of migration time
5 grain development stages, and various reproducibility of capillary electrophoretic protein
environmental conditions, were also successfully separations. J Liquid Chromatogr 16:51-68
studied by our optimal CE procedure, indicating Wang A, Pei Y, Zhang Q, Yan Y (2006) Advances on
its potential as a powerful alternative tool for wheat non-prolamins proteins. J Capital Normal Univ
studies on genetic, physiological, biochemical (Natural Science Edition) 27:5-8
and proteomic characterization of grain WSP. Wang A, Gao L, Pei Y, Zhang Y, He Z, Xia X, Appels
R, Ma W, Huang X, Yan Y (2008) Rapid separation
and characterization of grain water-soluble proteins
in bread wheat cultivars (Triticum aestivum L.) by
References
capillary electrophoresis. Can. J Plant Sci 88:843-848
Basset LM, Allan RE, Rubenthaler GL (1989) Genotype ×
Weiss W, Huber G, Engel KH, Pethran A, Dunn MJ,
environment interaction on soft white winter wheat
Gooley AA, Gorg A (1997) Identification and
quality. Agron J 81:955-960
characterization of wheat grain albumin/globulin
Bean SR, Tilley M (2003) Separation of water-soluble allergens. Electrophoresis 18:826-833
proteins from cereals by high-performance capillary
Zawistowska U, Bietz JA, Bushuk W (1986)
electrophoresis (HPCE). Cereal Chem 80:505-510
Characterization of low molecular weight protein with
Francois C, Morile PH, Dreux M (1996) Adjusting the high affinity for flour lipid from two wheat classes.
selectivity of inorganic anion separation by capillary Cereal Chem 63:414-419
electrophoresis. J High Resol Chromatogr 19:5-16
Zhao J, Yu Z (2005) Review on relationship between
metabolism of protein and starch in grains of wheat
and formation of quality. J Triticeae Crops 25:106-111

The effects of bug (Eurygaster spp. and Aelia spp.)
damaged flours on quality characteristics of cakes, cookies
and breads
S. Özçandır, and D. Sivri Özay
Hacettepe University, Department of Food Engineering, Beytepe, Ankara, Turkey
Abstract
Certain heteropterous insects (Eurygaster spp., Aelia spp., and Nysius huttoni) cause pre-harvest damage
to wheat in Europe, the Middle East, North Africa and New Zealand. Wheat bugs attack developing
wheat kernels by piercing them with their mouthparts and injecting their salivary secretions containing
proteases. Dough prepared from bug-damaged flour is very sticky and weak and produces loaves of
poor volume and unsatisfactory texture due to gluten degradation. Loaves baked from bug-damaged
wheat have low volume and coarse texture. The presence of 3-5% damaged kernels seriously affects
rheological and baking quality. Grain with higher levels of damage (>10%) must be rejected from milling
grades of bread wheat, generally being down-graded to feed grade. In this study, the effects of high
protease activity flour (HPAF) due to Eurygaster spp. and/or Aelia spp. damage to wheat on rheological
properties and quality characteristics of bakery products such as bread, cookies and cakes were
determined. High protease activity flour (HPAF) due to Eurygaster spp. and/or Aelia spp. damage was
added at different levels in cookies (0%, 20%, 40%, 60%, 80% and 100%), cakes (0%, 25%, 50%, 75% and
100%) and bread (25%) produced by AACC and modified AACC (2.5%, 5%, 10% and %15) formulations.
High level addition of HPAF reduced the quality of bread (volume and texture) and cookies (spread ratio
and texture), but there were no effects on cake quality (volume, texture and symmetry).
Introduction gluten contents were determined according to

Certain Pentatomid insects known as “wheat AACC Methods 08-01 and 38-11, respectively
bug” (Eurygaster spp., Aelia spp. and Nysius (AACC 1990). Falling number was determined
huttoni) cause pre-harvest damage to wheat in according to ICC Standard No: 107/1 (ICC 2012)
Europe, the Middle East, North Africa and New by using a Falling Number 1900 instrument
Zealand. Wheat bugs attack wheat by piercing (Perten Instruments, Sweden), and Zeleny
the developing grains with their mouthparts sedimentation and modified sedimentation tests
and injecting their salivary secretions containing were performed according to ICC Standard No:
proteases (Paulian and Popov 1980). Dough 116/1 (Köksel et al. 2000). High protease activity
prepared from this flour is very sticky and flour (HPAF1) was added at 2.5%, 5%, 10%, 15%
weak and produces loaves of poor volume and (w/w) levels in bread-making formulations.
unsatisfactory texture due to gluten degradation Cookies and cakes were produced with flours
(Sivri et al. 1998; Sivri and Köksel 2002; Koksel et (HPAF2) at 20%, 40%, 60%, 80%, 100% (w/w)
al. 2001). In this work, we studied the effects of and 25%, 50%, 75%, 100% (w/w) addition
high protease activity flour (HPAF), caused by levels, respectively. AACC methods were used
feeding on developing kernels by Eurygaster spp. for production of bread (Method 10-11), cakes
and/or Aelia spp. on rheological properties and (Method 10-90) and cookies (Method 10-54).
quality characteristics of bakery products such as Texture analysis of bread, cake and cookie
bread, cookies and cakes. samples were determined by a TAPlus Texture
Analyzer (LLOYD Instruments, UK); color value
was determined by a Minolta Spectrophotometer
Materials and methods CM-3600 D, (Japan); and spread ratio of cookies
was determined using AACC Method 10.54.
Protein contents of flour samples were
Bread and cake volumes were determined by the
determined by the Dumas method (Velp
rapeseed displacement method.
Scientifica NDA-701, Italy). Ash and wet
Results 9
Chemical and physicochemical properties of
HPAF and the flours used for bread, cake and
cookie production are shown in Table 1.
Spread ratio
8
Cookies
The color, spread ratio and texture values of
cookies are shown in Fig. 1, 2 and 3; images are
shown in Fig. 4. No significant differences in
7
color values were observed, but cookies from 0 20 40 60 80 100
HPAF2 increased in hardness with increasing HPAF2 additon level (%)
levels of grain-damaged flour. The spread ratio Fig. 2 Spread ratio values of cookies.
increased as expected.
L a b 100
80
70
60
50
Texture
Color
40 50
30
20
10
0 0
0 20 40 60 80 100 0 20 40 60 80 100
HPAF2 additon level (%) Additon level (%)
Fig. 1 Color values of cookies (L, Brightness; a, Fig. 3 Effect of HPAF2 levels on texture values of
Redness; b, Yellowness). cookies.
Table 1. Some chemical and physicochemical properties of flours.
Flour 1 Flour 2 HPAF 1 HPAF 2

Protein (%) 10.52 7.98 8.80 8.4
Wet gluten (units) 38.8 17.6 -1 24
Zeleny sedimentation (ml) 40 20 28 18
Modified sedimentation (ml) 39 20 5 9
Ash (%) 0.45 0.65 0.61 0.71
Falling number (sec) 336 314 368 305
1 Could not be washed
0 20 40 60 80 100
HPAF2 addition level (%)
Fig. 4 Effects of HPAF2 on cookies.

Cakes in cake volumes depending on the HPAF2 addition
The color, volume and texture values of cakes level there were no significant differences between
are shown in Fig 5, 6 and 7; images of cakes are the samples in terms of textural properties such as
shown in Fig. 8. The qualities of cakes at various hardness, cohesiveness, springiness, gumminess,
added HPAF2 levels were comparable with chewiness, adhesiveness, stiffness values and crust
controls. Although a slight increase was observed color.
70 Bread
L a b
60 Color, volume and texture values of breads are
shown in Fig. 9, 10 and 11; images are shown in
50
Fig. 12. While there were no significant differences
40 in crust color, there were significant reductions
Color
30 in loaf volume with increasing HPAF1 addition

20 levels. Textural qualities (hardness, gumminess,
chewiness and stiffness) of breads at high HPAF1
10
supplementation (10% and 15% w/w) levels were
0 significantly increased, but there were no significant
0 25 50 75 100
differences in cohesiveness and springiness.
HPAF2 additon level (%)
Fig. 5 Cake crust color values (L, Brightness; a,
Redness; b,Yellowness). 80
L a b
200
60
Volume (cm3)
Color
40
20
0
0 25 50 75 100
150
0 25 50 75 100 HPAF1 additon level (%)
HPAF2 additon level (%) Fig. 9 Bread crust color values (L, Brightness; a,
Fig. 6 Cake volumes. Redness; b, Yellowness).
5 Hardness
Cohesiveness
4
Springiness
3 Gumminess
Texture
Chewiness
2 Stiffness
1
0
0 25 50 75 100
Fig. 7 Cake texture values.
0 25 50 75 100
HPAF2 addition level (%)
Fig. 8 Effects of HPAF2 on cakes.
200
190
Volume (cm3)
180
170
160
0 25 50 75 100
Fig. 10 Bread loaf volumes.
5 Hardness
Cohesiveness
4 Springiness
Gumminess
3
Texture
Chewiness
Stiffness
2
0
0 25 50 75 100
Fig. 11 Bread texture values.
0 2.5 5 10 15
Addition level (%)
Fig. 12 Effects of HPAF1 on bread.

In the commodity market, in addition to other AACC (1990) Approved methods of the AACC, 8th edn.
evaluation factors, the price of wheat is affected The Association of Cereal Chemists, St Paul, MN
by the level of bug damaged kernels. A better ICC (2012) Standard methods. Available at: http://www.
icc.or.at/standard_methods, June, 2012
understanding of the effects of bug damage
on wheat quality is essential for bread, cookie Paulian F, Popov C (1980) Wheat. In: Hafliger E (ed) Ciba-
Geigy, Basel, Switzerland, pp69-74
and cake manufacturers who might have to use
Köksel H, Sivri D, Özboy Ö, Başman A, Karacan H (2000),
damaged wheat. Therefore, the results of this
Cereal laboratory handbook, publications of Hacettepe
study might provide some practical information University Faculty of Engineering, Ankara, Turkey
for the cereals industry for decision making about Köksel H, Sivri D, Ng PKW, Steffe JF (2001) Effects of
the utilization of bug damaged wheat. transglutaminase enzyme on fundamental rheological
properties of sound and bug-damaged wheat flour
doughs. Cereal Chem 78:26-30
Acknowledgements Sivri D, Köksel H (2002) Wheat bug protease: A protease
enzyme with specific activity for gluten proteins.
This research was supported by The Scientific
In: Ng PKW, Wrigley CW (eds) Wheat quality
and Technological Research Council of Turkey elucidation: the Bushuk legacy. American Association
(TÜBİTAK) (Project No: 111 O 060). of Cereal Chemists, St Paul, MN, pp113-126
Sivri D, Köksel H, Bushuk W (1998) Effects of wheat
bug (Eurygaster maura) proteolytic enzymes on
electrophoretic properties of gluten proteins. New
Zealand J Crop and Hort Sci 26:117-125
Molecular cloning and characterization of four i-type
LMW glutenin subunit genes from T. zhukovskyi and
T. compactum
Q. Wang, Y. Li, X.Y. Li, X. Xiao, F.S. Sun, Y.S. Wang, G.Y. He, and G.X. Yang
The Genetic Engineering International Cooperation Base of Chinese Ministry of Science and Technology, Chinese
National Center of Plant Gene Research (Wuhan) HUST Part, College of Life Science and Technology, Huazhong
University of Science & Technology (HUST), Wuhan, 430074, China
Abstract
Four novel low-molecular-weight glutenin subunit (LMW-GS) genes, designated TzLMW-i1, TzLMW-i2,
TcLMW-i1 and TcLMW-i2, were cloned and characterized from the genomic DNA of Triticum zhukovskyi
and Triticum compactum. The coding regions of TzLMW-i1, TzLMW-i2, TcLMW-i1 and TcLMW-i2 were
759 bp, 837 bp, 756 bp and 1,170 bp in length, encoding peptides with 251, 277, 250 and 388 amino acid
residues, respectively. Analysis of deduced amino acid sequences showed that the four novel genes
were LMW-i types and the comparative results indicated that the four genes had similar structures
and higher homologies with LMW-i-type genes than with LMW-m and LMW-s type genes. Sequence
alignment results showed that TzLMW-i1, TcLMW-i1 and TcLMW-i2 have all eight cysteines at
conserved positions in the C-terminal domain, whereas TzLMW-i2 has an additional cysteine residue.
This additional cysteine residue was predicted to be involved in the formation of intermolecular
disulfide bonds. Therefore, TzLMW-i2 may act as a “chain brancher” and have the possibility of
increasing protein cross-links in the gluten network, forming larger glutenin polymers, which
contribute to dough elasticity and high bread-making quality. Moreover, TcLMW-i2 possessed a longer
repetitive domain, which was considered to be associated with good wheat quality. To investigate the
evolutionary relationship of the novel genes with other storage protein genes, a phylogenetic tree was
constructed. It showed that prolamin genes could be absolutely clustered into five groups, including
high-molecular-weight glutenin subunits (HMW-GS), LMW-GS, α/β-, γ- and ω-gliadins, and the four
LMW-GS genes cloned in the present study were grouped into the LMW-i subgroup.
Introduction Glu-A3, Glu-B3 and Glu-D3 loci on the short arms

The gluten proteins in wheat endosperm confer of chromosomes 1A, 1B and 1D (D’Ovidio and
the unique visco-elastic properties to dough and Masci 2004). Typical LMW-GS can be classified
make it suitable for processing into bread, pasta, into three groups: LMW-i, LMW-m, and LMW-s,
noodles and other food products. Glutenins based on the first amino acid residue in the
consist of high molecular weight glutenin mature protein (isoleucine, methionine and
subunits (HMW-GS) and low molecular weight serine, respectively) (D’Ovidio and Masci 2004).
glutenin subunits (LMW-GS), considered to Many LMW-GS genes have been cloned and
be major determinants of the unique dough characterized from hexaploid wheat varieties.
properties. However, the relationships of certain Some more recent studies have been focused
alleles of LMW-GS with dough qualities are still on the isolation of LMW-GS genes from related
not fully understood, at least in part because of wheat species, including Triticum monococcum,
the complexity of the LMW-GS gene family and Aegilops longissima, Triticum zhukovskyi and
extensive allelic diversity of LMW-GS in common Triticum dicoccoides. In this study, we report the
wheat. cloning and characterization of four novel LMW-
GS genes (designated TzLMW-i1, TzLMW-i2,
The LMW-GSs, which in common wheat account TcLMW-i1 and TcLMW-i2) from the genomic
for approximately 40% of the gluten proteins DNA of T. zhukovskyi and T. compactum, which
and 60% of total glutenins, are encoded by the

would provide new glutenin gene resources using MEGA 4.1 (http://www.megasoftware.
and insights into the evolution of Glu-3 loci in net/mega4/mega.html). Bootstrap tests were
Triticum species. performed with 1,000 replications.

The complete coding sequences of four LMW-
Plant materials
GS genes were successfully amplified from the
One T. zhukovskyi accession (ZH1) and one T. genomic DNA of T. zhukovskyi and T. compactum
compactum accession (TH2W), kindly provided by (Fig. 1 and Table 1). Two PCR products of
the Institute of Crop Sciences, Chinese Academy approximately 750 bp and 800 bp, from T.
of Agricultural Sciences, Beijing, were used to zhukovskyi were clearly detected using the primer
isolate LMW-GS genes. set pLMW-1 and pLMW-2; whereas two PCR
products of ca. 750 bp and 1,100 bp from T.
DNA cloning and sequencing compactum were observed
PCR products were separated in 1% agarose gels
and the expected fragments were purified from M 1 2
the gels using a gel extraction kit. The purified
DNA fragments were ligated into the pMD18-T 3, 000 bp
vector, and then transformed into competent cells
of E. coli strain TOP10 according to the standard 2, 000 bp
procedure. After identification of positive 1,400 bp
clones using colony PCR, DNA sequencing was
performed with three clones by a commercial
sequencing company. The sequences were 900 bp
analyzed with the help of Vector NTI Advance
10.0 software.
600 bp
Sequence comparison and phylogentic analysis
The relative molecular masses of the LMW- 400 bp
GS from T. zhukovskyi and T. compactum were
predicted using ProtParam tools (http://www.
expasy.org/tools/protparam.html). Complete 200 bp
amino acid sequences were used to perform
multiple alignments with known i-type LMW-
GS proteins using the Vector NTI Advance 10.0
Fig. 1 PCR amplification results from
program. The deduced amino acid sequences of
genomic DNA of ZH1 and TH2W. M, DNA
the cloned LMW-GS were used to construct a
marker; 1, 2, PCR amplification products of
phylogenetic tree by the neighbor-joining method
ZH1 and TH2W, respectively.
Table 1. Nucleotide compositions of deduced amino-acid sequences and molecular weights of four novel
LMW-i genes from Triticum zhukovskyi and Triticum compactum.
LMW-GS Nucleotide Residue no. No. of cysteine Predicted molecular

gene Species length (bp) of mature protein residues weight (kDa)
TzLMW-i1 T. zhukovskyi 759 231 8 26.4
TzLMW-i2 T. zhukovskyi 837 257 9 29.6
TcLMW-i1 T. compactum 756 230 8 26.4
TcLMW-i2 T. compactum 1170 368 8 42.7
Nucleotide composition, deduced amino-acid chromosome 1A than those on the chromosomes
sequences and molecular weights of four 1B and 1D (Table 2). This indicates that the
novel LMW-i genes from T. zhukovskyi and T. four i-type LMW-GS genes may be located on
compactum chromosome 1A.
After sequencing the PCR products, multiple As shown in Fig. 2, these four novel i-type
alignment of the sequences with known LMW-GS LMW-GS have a set of eight conserved Cys
genes (Fig. 2) showed that they belonged to i-type residues compared with other LMW-i subunits.
LMW-GS genes and therefore were designated Interestingly, TzLMW-i2 has an additional
TzLMW-i1, TzLMW-i2, TcLMW-i1 and TcLMW-i2. cysteine residue in its C-terminal cysteine-rich
The lengths of their coding sequences and region (Fig. 2), and this may allow additional
predicted relative molecular masses are shown inter-molecular disulfide bonding and may act as
in Table 1. Homology analysis demonstrated “chain brancher” in the gluten protein network
that LMW-GS genes cloned in this study show (Xu et al. 2006). Moreover, the TcLMW-i2 subunit
higher homology with the LMW-GS genes on has a longer N-terminal repetitive domain
Table 2. Identities of four LMW-GS genes identified in this study with homologous Glu-3 locus-encoded
proteins from the wheat A, B, and D genomes.
GenBank Identity (%)

Locus accession TzLMW-i1 TzLMW-i2 TcLMW-i1 TcLMW-i2
Glu-A3 AB062868 86% 88% 88% 88%
AB062871 86% 88% 88% 88%
AY453155 98% 95% 98% 98%
AY453154 98% 95% 98% 98%
AB062878 96% 92% 96% 96%
AB062877 96% 92% 96% 96%
AY453159 95% 99% 95% 95%
AY453158 97% 94% 99% 99%
AB062876 96% 93% 96% 96%
Average 94% 93% 95% 95%
Glu-B3 Y14104 79% 85% 80% 80%
AB062853 91% 85% 90% 90%
AB062860 90% 90% 90% 90%
Y17845 90% 84% 90% 90%
Y18159 91% 84% 90% 90%
AB062862 91% 85% 90% 90%
AB062861 91% 85% 90% 90%
AB062854 91% 85% 91% 91%
AB062855 91% 85% 91% 91%
Average 89% 85% 89% 89%
Glu-D3 AB062865 78% 86% 79% 79%

AB062866 78% 86% 79% 79%
AY585355 79% 86% 79% 79%
AY585350 81% 78% 81% 81%
AB062851 80% 89% 81% 81%
X13306 92% 92% 92% 92%
AB062867 78% 86% 78% 78%
AY299485 91% 92% 91% 91%
AY263369 91% 92% 91% 91%
Average 83% 87% 83% 83%

Fig. 2 Multiple alignments of the deduced amino acid sequences of the four i-type LMW-GS genes (TzLMW-i1, TzLMW-i2, TcLMW-i1and
TcLMW-i2) cloned in this study and five known i-type LMW-GS genes.
Session 3: Improvement of end use qualiƟes by geneƟc and alternaƟve approaches

151
than other i-type LMW-GS (Fig. 2). It was Fig. 3 Phylogenetic tree showing BAD12056-LMW-S
suggested that longer sequences of the relationships among LMW-GS, 88 EU189088-LMW-S
ABY58126-LMW-S
repetitive domain of LMW-GS possibly HMW-GS and gliadins from different 36
BAE96110-LMW-S
contribute to the formation of inter-chain Triticum species. The genes identified BAJO9388-LMW-S
52
22 AEH31546-LMW-S
hydrogen bonds, particularly with HMW- in this study are boxed. ACY08819-LMW-S
GS, and this is important in stabilizing EF437426-LMW-S
protein-protein interactions in gluten and in 67 ACY08822-LMW-S
36 EF437420-LMW-S
influencing the gluten structure and dough 70 EF437425-LMW-S
functionality (Lambourne et al. 2010). ACY08813-LMW-S
89 BAD12055-LMW-S
28 AAS66085-LMW-S
Phylogenetic analysis showed that all the
56 ABY58133-LMW-S
storage proteins examined were clustered FJ606796-LMW-M
into seven groups: s-type LMW-GS, m-type EF190885-LMW-M
26 EF190884-LMW-M
LMW-GS, i-type LMW-GS, HMW-GS,
69 AY299485-LMW-M
31
ω-gliadins, α/β-gliadins and γ-gliadins (Fig. AY263369-LMW-M
33 58
3). Four LMW-GS genes cloned in this study 19 AJ519835-LMW-M
X13306-LMW-M
were grouped together with other i-type FJ615309-LMW-M
LMW-GS from T. aestivum. 13 FJ615311-LMW-M
FJ615310-LMW-M
FJ972196-LMW-M
Y14104-LMW-M
Conclusions 99 EU369719-LMW-M
EU369720-LMW-M
The isolation of four i-type LMW-GS genes 69 FJ876823-LMW-M
from T. zhukovskyi and T. compactum are EU189089-LMW-M
63 EU369729-LMW-M
reported. TzLMW-i2 has an additional Cys TzLMW-i2
residue in the C-terminal cysteine-rich TzLMW-i1
66 CAA30570-LMW-i
region, whereas TcLMW-i2 has a longer AAS10192-LMW-i
N-terminal repetitive domain. These P10385-LMW-i
99 AE100698-LMW-i
characteristics suggest that these genes
27 ABG76006-LMW-i
may make positive contributions to dough ACZ51336-LMW-i
99 ACJ66627-LMW-i
functionality, but require further study.
Incorporation of these novel LMW-GS into 64 ACP27640-LMW-i
78 BAB78762-LMW-i
flour, or genetic transformation, may be ACP27638-LMW-i
employed to further evaluate their potential 87 ACY08811-LMW-i
56 BAB78764-LMW-i
for use in wheat quality improvement. TcLMW-i1
AAS10193-LMW-i
TcLMW-i2
AAV92075-LMW-i
References 85 AAV92079-LMW-i
AAS66083-LMW-i
D’Ovidio R, Masci S (2004) The low-molecular-
ABY58125-LMW-i
weight glutenin subunits of wheat gluten. J AAV92080-LMW-i
Cereal Sci 39:321-339 AY585355-LMW-M
ExPASy (2013) ProtParam Tool. Available at: http:// DQ457416-LMW-M
www.expasy.org/tools/protparam.html 89 M13713-gliadin
97 AJ133613-gliadin
Lambourne J, Tosi P, Marsh J, et al. (2010) M36999-gliadin
Characterisation of an s-type low molecular AJ937838-gliadin
weight glutenin subunit of wheat and its proline 95 U51303-gliadin
X02540-gliadin
and glutamine-rich repetitive domain. J. Cereal U08287-gliadin
99
Sci 51:96-104
88 U51307-gliadin
Molecular Evolutionary Genetics Analysis (2013) 83 D84341-gliadin
MEGA. Available at: http://www.megasoftware. 52 X12929-HMW
X61026-HMW
net/mega4/mega.html X61009-HMW
99
Xu H, Wang RJ, Shen X, et al. (2006) Functional 85 X12928-HMW
properties of a new low-molecular-weight 81 X13927-HMW
AF280605-gliadin
glutenin subunit gene from a bread wheat
AY591334-gliadin
cultivar. Theor Appl Genet 113:1295-1303
0.9 0.6 0.3 0.0

Effects of interactions between high molecular weight
glutenin subunits and waxy alleles on dough-mixing
properties in common wheat
Z.Y. Deng, S.N. Hu, F. Chen, F.F. Zheng, J.N. Chen, X.Y. Zhang, C.L. Sun, J.S. Chen, Y.X. Zhang, S.Y.
Wang and J.C. Tian
State Key Laboratory of Crop Biology, Key Laboratory of Crop Biology of Shandong Province, Group of
Shandong Agricultural University, Taian, Shandong, P.R. China, 271018, China
Abstract
The glutenin and waxy genes of wheat are important determinants of dough quality. This study was
conducted to evaluate the effects of high molecular weight glutenin subunits (HMW-GS) and waxy
alleles on dough-mixing properties over 3 years using a population (Nuomai 1 × Gaocheng 8901) of
290 recombinant inbred lines (RILs). The results indicated the following: 1) the Glu-A1 and Glu-D1
loci had greater impacts on Mixograph properties than the Wx-1 loci and the effects of Glu-D1d and
Glu-D1h on dough mixing were superior to those of Glu-D1f and Glu-D1new1 in this population; 2)
the interactions between the Glu-1 and Wx-1 loci affected some traits, especially midline peak value
(MPV), and lack of Wx-B1 or Wx-D1 led to increased MPV for all variants of Glu-1 loci.
Introduction 8901 (77546-2/Lingzhang, 1/7+8/5+10 and a/a/a)

The glutenin subunit composition and waxy has high gluten strength and good bread-making
subunit alleles are of particular interest to quality.
many wheat researchers and breeders (Zhai
et al. 2008; Zhang et al. 2009; Yamamori 2009), Field trials
but generally their effects on wheat quality are Field trials were conducted over three years
studied separately. There is no information utilizing a randomized complete block design.
on the effects of interaction between them on Two replicates were grown at the experimental
wheat quality, or how their contributions to fields of Shandong Agricultural University, Tai’an
quality and interactions affect quality in specific City (36°57′N, 116°36′E) in 2007-2008 and 2008-
populations. We therefore undertook research 2009 and at Suzhou (33°38′N, 116°58′E) City,
using a single recombinant inbred line (RIL) Anhui province, in 2010-2011.
population in three different environments,
to assess the effects of interactions between Methods
the Glu-1 and Wx-1 loci on dough-mixing Seed samples were milled using a Bühler
properties. experimental mill (Buhler, Buhler-Miag
Company, Germany) with a flour extraction of
approximately 70%. Dough-mixing properties
Materials and methods were analyzed for each sample with a 10 g
Mixogram at room temperature according to
Plant2 Materials
approved AACC method 54-40A (AACC 2004).
A population of 290 RILs was developed Mixograph characteristics were determined with
from the cross Nuomai 1 × Gaocheng 8901. the software program Mixsmart_version 3.80.
Nuomai 1 (Jiangsu Baihuomai/Guandong 107) These mean values were also analyzed with
(null/7+8/2.2+12 at Glu-l loci, b/b/b at Wx-1 loci) SAS (SAS Institute, NC, USA) using Fisher’s
has unique starch properties that are related to least significant difference (LSD) procedure and
high-quality white salted noodles. Gaocheng Pearson correlations.
Results and discussion too small for meaningful comparisons (Table 2).
There were significant differences in MPT between
Effects of HMW-GS and Wx allelic diversity on the Wx combinations b/a/a and the other six
mixing properties combinations, except a/a/a.
A high level of allelic diversity was observed at Effects of interactions between the Glu-1 and Wx
the Glu-A1 and Glu-D1 and the three Wx loci in loci on mixing properties
this population (Table 1). For the Glu-A1 loci,
significant differences between Glu-A1a and Interactions between the Glu-A1 and Wx loci had
Glu-A1c were observed for the Mixograph peak no significant influence on almost all Mixograph
time (MPT), Mixograph midline peak (MPV), properties except MPV (Table 3). Interaction
Mixograph midline peak integral (MPI) and between the Glu-A1 and Wx-B1 loci significantly
Mixograph band width at 8 minutes (8MW), but affected MPV. No significant interactions were
not the Mixograph band width at MPT (MPW). observed between Glu-D1 and Wx-A1 alleles
However, the MPT was significantly affected by for most Mixograph traits. However, significant
Glu-D1 allelic diversity. Significant differences interactions occurred between Glu-D1f and Wx-A1
were found between Glu-D1d and the other three for the MPV and 8MW. Significant interactions
alleles on MPT, MPI and 8MW. Only MPV was were also observed between Glu-D1f and Wx-B1
significantly affected by Wx-B1 alleles. Significant for MPV, MPW and 8MW. Only the interactions
differences in MPT and MPI were observed at the between Glu-D1new1 and Wx-B1 significantly
Wx-D1 locus. Therefore, the effects of the HMW- affected MPV. Thus, the interactions between
GS allelic diversity on the mixing characteristics different HMW-GS subunits and Wx proteins
were stronger than those of the Wx loci. affected some Mixograph traits, especially MPV.
Deletions of Wx-B1 and Wx-D1 increased MPV in
Effects of HMW-GS and Wx subunit all types of HMW-GS.
combinations on mixing characteristics
Interactions between HMW-GS subunits and
There were significant differences between Wx allele combinations (Table 4) indicated that
combinations 1/7+8/5+10 and the other three HMW-GS combinations significantly affected
combinations for the MPT and MPI (Table 2). the MPT, MPV and MPI, but Wx combinations
When Glu-A1 and Glu-B1 were the same, Glu-D1d only significantly influenced MPV. No significant
had better effect on MPT and MPI than Glu-D1f. effects were observed for HMW-GS × Wx alleles
Although four new combinations were found, on Mixograph traits.
the number of samples per combination was
Table 1. Effects of glutenin and waxy alleles on Mixograph properties of the recombinant inbred line
(RIL) population.
Mixograph properties1
Locus Allele2 N MPT3 MPV MPW MPI 8MW
Glu-A1 a 142 3.11a 62.74a 23.93a 158.10a 6.49a
c 125 2.75b 60.69b 23.14a 135.54b 5.68b
Glu-D1 d 162 3.28a 61.15b 23.10a 163.47a 6.72a
f 91 2.41b 63.06a 24.37a 122.66b 5.13b
h 8 2.67b 58.619b 22.69a 128.14b 5.36b
new1c 6 2.33b 63.49a 24.29a 121.74b 5.46b
Wx-A1 a 139 2.97a 61.23a 23.68a 147.82a 6.28a
b (null) 128 2.91a 62.38a 24.98a 147.33a 5.93a
Wx-B1 a 146 3.00a 60.65b 23.26a 147.71a 6.09a
b (null) 121 2.87a 63.14a 23.89a 147.39a 6.12a
Wx-D1 a 132 3.05a 61.25a 23.44a 151.07a 6.16a
b (null) 135 2.84b 62.29a 23.65a 144.14b 6.03a
1 Means followed by different letters indicate significant differences within columns at P < 0.05.
2 MPT, Mixograph peak time; MPW, Mixograph band width at MPT; MPV, Mixograph midline peak value; MPI, Mixograph
midline peak integral; 8MW, Mixograph band width at 8 minutes.
3 New allele.

Table 2. The effects of HMW-GS subunit and Wx allele combinations on Mixograph properties in the
recombinant inbred line (RIL) population.
Mixograph properties1
Glu-A1 Glu-B1 Glu-D1 N MPT2 MPV MPW MPI 8MW
1 7+8 5+10 88 3.44a 62.25b 23.52b 173.78a 7.17a
1 7+8 2.2+12 46 2.56c 64.29a 25.05a 132.62c 5.32bc
null 7+8 5+10 74 3.09b 59.84c 22.59b 151.22b 6.18ab
null 7+8 2.2+12 45 2.26d 61.81b 23.68b 112.48d 4.94c
1 7+8 5+12 6 2.83 58.02 21.41 134.64 5.8
1 7+8 2.2+10 2 2.45 62.87 23.26 124.58 5.47
null 7+8 5+12 2 2.2 60.41 26.55 108.65 4.01
null 7+8 2.2+10 4 2.27 63.8 24.81 120.32 5.46
Wx-A1 Wx-B1 Wx-D1
a a A 42 3.12ab 60.13c 23.59ab 152.08ab 6.53ab
a a B 38 2.84b 60.28c 23.40ab 138.85b 5.86ab
a b A 30 2.94b 62.1bc 23.33ab 148.03ab 6.32ab
b a A 25 3.35a 60.46c 22.60b 165.28a 6.85a
a b B 29 2.95b 63.16ab 24.53a 150.86ab 6.41ab
b a B 41 2.82b 61.64bc 23.20ab 140.754b 5.4b
b b A 35 2.84b 62.44bc 23.97ab 142.32b 5.32b
b b B 27 2.74b 65.18a 23.72ab 149.52ab 6.65ab
1 Means followed by different letters within columns indicate significant differences at P < 0.05
2 MPT, Mixograph peak time; MPW, Mixograph band width at MPT; MPV, Mixograph midline peak value; MPI, Mixograph
midline peak integral; 8MW, Mixograph band width at 8 minutes.
Table 3. Mean values of Mixograph properties for combinations of alleles at the Glu-1 and Wx loci in the
recombinant inbred line (RIL) population.
Mixograph characteristics1
Locus 1 Locus 2 MPT2 MPV MPW MPI 8MW
Glu-A1a Wx-A1a 3.16a 62.08a 23.78a 157.84a 6.62a
Wx-A1b 3.06a 63.46a 24.09a 158.38a 6.35a
Glu-A1c Wx-A1a 2.75a 60.26a 23.56a 135.37a 5.89a
Wx-A1b 2.74a 61.15a 22.63a 135.86a 5.45a
Glu-A1a Wx-B1a 3.18a 61.68b 23.78a 158.20a 6.59a
Wx-B1b 3.03a 64.07a 24.11a 157.97a 6.36a
Glu-A1c Wx-B1a 2.80a 59.44b 22.65a 135.35a 5.51a
Wx-B1b 2.69a 62.13a 23.65a 135.89a 5.87a
Glu-A1a Wx-D1a 3.23a 62.46a 23.66a 162.61a 6.65a
Wx-D1b 2.99a 63.03a 24.20a 153.32a 6.32a
Glu-A1c Wx-D1a 2.83a 59.76b 23.17a 136.79a 5.70a
Wx-D1b 2.68a 61.52a 23.07a 134.54a 5.65a
Glu-D1d Wx-A1a 3.32a 61.03a 23.56a 164.50a 7.17a
Wx-A1b 3.26a 60.49a 22.28b 161.91a 6.23a
Glu-D1f Wx-A1a 2.47a 61.55b 24.30a 122.39a 4.85b
Wx-A1b 2.40a 65.04a 24.86a 126.52a 5.59a
Glu-D1h Wx-A1a 3.24a 60.68a 23.28a 160.20a 7.03a
Wx-A1b 3.05a 62.88a 23.66a 156.86a 5.96a
Glu-D1new1 Wx-A1a 2.14a 63.01a 22.70a 108.70a 4.66a
Wx-A1b 2.36a 62.96a 24.27a 121.10a 5.30a
(cont’d on next page)
Table 3. Mean values of Mixograph properties for combinations of alleles cont’d...
Mixograph characteristics1
Locus 1 Locus 2 MPT2 MPV MPW MPI 8MW
Glu-D1d Wx-B1a 3.44a 59.85b 23.00a 168.19a 7.23a
Wx-B1b 3.07b 62.06a 22.78a 156.02a 5.93a
Glu-D1f Wx-B1a 2.46a 61.79b 23.80b 122.39a 4.81b
Wx-B1b 2.41a 64.93a 25.57a 126.76a 5.69a
Glu-D1h Wx-B1a 3.00a 60.95a 22.98a 148.17a 5.62a
Wx-B1b 3.28a 62.16a 23.80a 166.87a 7.29a
Glu-D1new1 Wx-B1a 2.36a 60.35b 23.26a 116.19a 5.05a
Wx-B1b 2.18a 65.35a 23.93a 115.60a 5.01a
Glu-D1d Wx-D1a 3.46a 60.36a 22.45a 170.50a 7.08a
Wx-D1b 3.10b 61.20a 23.42a 155.24b 6.27a
Glu-D1f Wx-D1a 2.40a 62.47a 25.29a 119.69a 4.78a
Wx-D1b 2.47a 63.57a 24.04a 127.41a 5.46a
Glu-D1h Wx-D1a 3.17a 60.59a 23.29a 156.61a 6.46a
Wx-D1b 3.14a 62.89a 23.63a 161.35a 6.70a
Glu-D1new1 Wx-D1a 2.26a 64.76a 24.05a 116.91a 4.92a
Wx-D1b 2.27a 61.00a 23.13a 114.73a 5.14a
1 Means followed by different letters indicate significant differences within columns at P <0.05
2 MPT, Mixograph peak time; MPW, Mixograph band width at MPT; MPV, Mixograph midline peak value; MPI, Mixograph midline peak
integral; 8MW, Mixograph band width at 8 minutes.
Table 4. The effects of interactions between HMW-GS subunits and Wx alleles on mixing properties in the
recombinant inbred line (RIL) population using the Generalized Linear Model (GLM) in SAS.
Trait1 HMW-GS Wx HMW-GS×Wx

DF 7 7 44
MPT Type III SS 52.97 4.58 20.74
Mean square 7.57 0.65 0.47
F value 16.89 1.46 1.05
P <0.0001 0.18 0.39
MPV Type III SS 661.86 644.43 1087.32
Mean square 94.55 92.06 24.71
F value 4.61 4.49 1.21
P <0.0001 <0.0001 0.19
MPW Type III SS 213.51 58.26 475.92
Mean square 30.50 8.32 10.82
F value 2.89 0.79 1.02
P 0.01 0.60 0.44
MPI Type III SS 83752.95 4891.83 61492.40
Mean square 11964.71 698.83 1397.55
F value 10.44 0.61 1.22
P <0.0001 0.75 0.18
8MW Type III SS 90.00 37.77 459.42
Mean square 12.86 5.40 10.44
F value 1.47 0.62 1.2
P 0.18 0.74 0.20
1 MPT, Mixograph peak time; MPW, Mixograph band width at MPT; MPV, Mixograph midline peak value; MPI, Mixograph midline peak
integral; 8MW, Mixograph band width at 8 minutes; HMW-GS, high molecular weight glutenin subunits.

Interactions between HMW-GS and waxy AACC Chemists (2004) Approved methods of the
proteins play important roles in determining American Association of Cereal Chemists, 10th ed.
AACC, St. Paul, MN
dough-mixing properties at the protein subunit
Yamamori M (2009) Amylose content and starch
and molecular levels. Among them, MPV was
properties generated by five variant Wx alleles for
particularly affected by interactions between the granule-bound starch synthase in common wheat
Glu-1 and Wx-1 loci, and its QTL was located (Triticum aestivum L.). Euphytica 165:607-614
around Wx-B1. Variation at the Wx-1 loci played Zhai H, Tian J, Sun C (2008) Comparison of grain traits
a minor role in variation in dough-mixing and rheological characteristics of dough for wheat
properties, but the Glu-A1 and Glu-D1 loci were with different waxy protein deficiency. Journal of the
major contributors to dough-mixing properties. Chinese Cereals and Oils Association 23:17-21
Interactions between different HMW glutenin Zhang Y, Tang J, Yan J, Zhang Y, Zhang Y, Xian X, He Z
and waxy alleles affected the different mixing (2009) The gluten protein and interactions between
components determine mixograph properties in an F6
characteristics of the Mixograph test. recombinant inbred line population in bread wheat. J
Application of molecular markers for HMW-GS in wheat
cultivar improvement
X. M. Chen1, Y. Zhang1, X. C. Xia1, Z. H. He1,2, D. S. Wang1 and Y. Zhang1
1Institute of Crop Sciences, Chinese Academy of Agricultural Sciences (CAAS), 12 Zhongguancun South
Street, Beijing 100081, China; 2CIMMYT-China Office, C/O CAAS , 12 Zhongguancun South Street, Beijing
100081, China
Abstract
To improve both yield and bread-making quality simultaneously, 12 BC2F5 breeding lines with
high molecular weight glutenin subunits (HMW-GS) 7+8 and 5+10 derived from the cross Yumai
34/3*Lunxuan 987 and selected by markers were used to investigate the yield and quality performance.
Cultivar Yumai 34, with excellent bread-making quality, was used as the donor of HMW-GS 7+8 and
5+10, and Lunxuan 987 as the high yielding backcross parent. Nine of twelve lines had higher values
for major quality parameters than Lunxuan 987, including dough development time and farinogragh
stability, maximum extensograph resistance, and superior loaf volume and total score. Three of four
lines with the 1BL/1RS translocation showed negative quality effects. The yields of 11 lines were
increased by 3.3 to 19.5% compared with Lunxuan 987; among them, three named CA1063, CA1062 and
CA1061 were significantly higher yielding than Lunxuan 987, and higher than the Zhongmai 175 check
by 4.1%, 3.5% and 1.9%, respectively. Eight lines, that were not significantly different in yield from
Zhongmai 175, exhibited better quality than both Lunxuan 987 and Zhongmai 175. The results indicated
that yield and quality could be improved at the same time by molecular marker assisted selection.
Introduction through selection for superior HMW-GS should

Yield increases and quality improvement are be a major objective in Chinese wheat breeding.
the most important wheat breeding objectives. Traditionally, SDS-PAGE was used to identify
Generally, it is very hard to improve both yield HMW-GS, but the method needs wheat seeds as
and quality simultaneously because of a negative the material for analysis and can only be carried
correlation between them. However, it may be out after harvesting. Hence, it is not efficient to
possible to overcome this constraint by marker apply in marker assisted selection in breeding.
assisted breeding. Although high molecular Molecular markers based on DNA are not
weight glutenin subunits (HMW-GS) account for affected by environment or growth stage and are
only about 10% of the wheat storage proteins, therefore much more applicable and efficient.
they play a key role in determining bread- Gene functional markers (Bagge et al. 2007)
making quality (Payne et al. 1981, 1983, 1984). are the ideal molecular marker for breeding.
Most studies have shown significant correlations Functional markers of HMW-GS have been
between HMW-GS and bread-making quality developed to distinguish glutenin subunits 1, 2*,
(Lukow et al. 1989; Payne et al. 1988; Rogers Null, 5, 10, 2, 12, 7, 8 (Bustos et al. 2000; D’Ovidio
et al.1989). Some studies have pointed out et al. 1995; Lei et al. 2006; Smith et al. 2001), and
limitations of HMW-GS in predicting end-use can be used in marker assisted breeding.
quality (Graybosch et al. 1990). However, all
The facultative wheat cultivar Yumai 34, a high
agree that certain HMW-GS, such as 5+10, are
quality bread-making, low yielding cultivar with
more important than others in determining
HMW-GS 1, 7+8, 5+10, was grown in the Yellow
bread-making quality. Some results show that
and Huai River Winter Wheat Region from 1996
low frequencies of HMW-GS associated with
to 2008. Lunxuan 987, on the other hand, has
good bread-making quality, such as 2*, 7+8, 5+10,
high yield, but poor bread-making quality with
in Chinese wheat cultivars is a major reason for
HMW-GS 1, 20, 2+12. It was grown in the North
poor bread-making quality (Liu et al. 2005; Song
Winter Wheat Region as a leading variety from
et al. 2003). Therefore, improving gluten quality

2004 to present. In year 2004, Yumai 34, as a Quality analysis
donor of good quality, and Lunxuan 987, as a Kernel protein (14% moisture basis) and moisture
backcross parent, were crossed and molecular contents were determined by NIR analysis using
markers for HMW-GS 7+8 and 5+10 were used a Foss 1241 NIR instrument (Sweden). Hard
during backcrossing and subsequent selection. and soft grain sample were tempered to 16.5%
In this paper we report the results of quality and and 14.5% moisture content, respectively. Grain
yield improvement on that material through samples were milled into flour using a Buhler
molecular marker assisted selection. experimental mill. Farinograph and extensograph
values were determined using methods AACC
56-60, 54-21, and 54-10, respectively. Pan bread-
Materials and methods making and quality evaluation were done
Plant materials according to He et al. (2004).
Twelve BC2F5 advanced lines from Yumai Molecular marker detection
34/3*Lunxuan 987 with HMW-GS 7+8 and
5+10 were used in yield trials. The lines were HMW-GS 7+8 and 5+10 were detected according
selected by molecular marker assisted and to the methods of Lei et al. (2006) and Ma et al.
agronomic trials from 56 F5 lines. DNA was (2003). 1BL/1RS translocation was identified
extracted from individual plants of the BC1F1 based on the method of Francis et al. (1995).
and molecular markers for HMW-GS 7+8 and
5+10 were screened. The plants with HMW-GS
7+8 and 5+10 were backcrossed with Lunxuan Results and discussion
987 with at least 15 plants being crossed each Molecular marker tests on the 12 final lines
time. The BC2F1 plants were self-pollinated and
those combining good agronomic characters Eleven lines had HMW-GS 7+8 and 5+10, and
and desirable HMW-GS were selected. Self- CA1066 contained only 7+8, and four lines
pollination and selection were continued, and possessed the 1BL/1RS translocation (Table 1).
12 BC2F5 lines were finally obtained. The check
cultivar Zhongmai 175 and recurrent parent Quality improvement
Lunxuan 987 were also used as controls. There were no differences in protein content
between the recurrent parent Lunxuan 987
Field trials and the 12 advanced lines, except for CA1067,
All the materials were tested in the 2010- and the extensibilities of all lines were lower
2011 cropping season at CAAS Changping than that for Lunxuan 987 (Table 1). Nine lines
Experimental Station, Beijing. Field trials were had higher values than Lunxuan 987 for most
conducted in random complete blocks with two quality parameters, such as dough development
replications. Plots consisted of six 4.12 m rows time and farinogragh stability, maximum
with 20 cm spacing and 40 cm between plots. extensogragh resistance, loaf volume, and
The seeding rate was about 270 million seedlings total bread-making score. These nine lines,
per ha. Basal fertilizers, including 2250 kg/ha CA1063, CA1062, CA1061, CA0998, CA1064,
of organic manure, 375 kg/ha of (NH4)2HPO4, CA0996, CA1060, CA1059, and CA1058, had
and 225 kg of carbamide, were applied pre- HMW-GS 7+8 and 5+10, and only CA1058 had
sowing. Winter irrigation was carried out to the 1BL/1RS translocation. This indicated that
promote plant survival. Irrigation and 150 kg/ quality was improved by the use of molecular
ha of carbamide were applied at plant stem marker selection. Lines CA1065 and CA1067,
elongation in spring. The final irrigation was which contained HMW-GS 7+8 and 5+10 as
applied during grain filling, and aphids were well as the 1BL/1RS translocation, had lower
controlled with chemicals. values of stability and maximum resistance
than the eight non-1BL/1RS lines, showing
that the 1BL/1RS translocation had negative
effects on the two parameters. The exceptional
1BL/1RS translocation line CA1058 had stability
and maximum resistance values equal to the
Table 1. Yield and quality parameters of tested lines derived from cross Yumai 34/3*Lunxuan 987. MR,
maximum resistance.
Farniograph Extensograph
Yield1 ±CK12 ±CK2 HMW-GS3 1BL/ protein Stability Exten- MR Bread
Line (kg/ha) % % 7+8 5+10 1RS (14%MB) (min) sibility (BU) score
CA1063 8433.0 a 4.1 19.5 + + - 12.0 7.6 146.2 256.2 62.5
CA1062 8386.5 a 3.5 18.9 + + - 11.9 8.3 135.2 208.6 58.5
CA1061 8259.0 a 1.9 17.0 + + - 11.8 9.6 138.0 269.9 64.0
Zhongmai175 8103.0 ab 0.0 14.8 - - - 12.1 2.6 187.1 100.9 49.0
CA0998 8059.5 ab -0.5 14.2 + + - 12.1 10.2 138.0 330.5 53.5
CA1064 7764.0 abc -4.2 10.0 + + - 12.6 4.7 139.8 197.0 64.0
CA0996 7746.0 abc -4.4 9.8 + + - 12.5 8.5 143.0 244.0 65.0
CA1065 7738.5 abc -4.5 9.7 + + + 12.9 3.5 134.1 135.5 64.5
CA1059 7726.5 abc -4.6 9.5 + + - 12.3 10.5 134.5 393.9 60.5
CA1066 7714.5 abc -4.8 9.3 + - + 12.7 1.9 134.8 53.5 48.0
CA1058 7696.5 abc -5.0 9.1 + + + 12.3 9.2 144.4 318.3 63.0
CA1067 7290.0 bcd -10.0 3.3 + + + 13.7 2.6 131.6 112.0 60.0
Lunxuan987 7056.0 cd -12.9 0.0 - - + 12.3 2.1 151.2 78.6 41.0
CA1060 6700.5 d -17.3 -5.0 + + - 12.6 14.5 134.5 391.2 71.0
1 Values with same letter are not significantly different at P=0.05
2 CK1, Zhongmai 175 ; CK2, Lunxuan 987; HMW-GS, High molecular weight glutenin subunit
3 +, presence, -, absence
other eight non-1BL/1RS lines. CA1058, Conclusion

CA1059, and CA1060 that had similar genetic Yield and quality were improved simultaneously
background, being derived from the same by marker assisted selection of HMW-GS 7+8 and
BC2F4 plant, showed higher values in most 5+10 combined with yield trials. The 1BL/1RS
quality parameters and very similar values were translocation generally, but not always, had a
found between CA1058 and CA1059. Thus the negative effect on quality.
effect of the 1BL/1RS translocation on quality
was different in different lines. CA1066 only
with HMW-GS 7+8 showed similar quality to References
Lunxuan 987, implicating the importance of Bagge M, Xia XC, Lübberstedt T (2007) Functional markers
HMW-GS 5+10 in bread-making quality. in wheat. Current Opinion in Plant Biol 10:211-216
Bustos AD, Rubio P, Jouve N (2000) Molecular
characterization of the inactive allele of the gene
Yield improvement Glu-A1 and the development of a set of AS-PCR
markers for HMW glutenins of wheat. Theor Appl
The yields of 11 lines were increased by 3.3 Genet 100:1085-1094
to 19.5% compared to Lunxuan 987 (Table 1). D’Ovidio R, Masci S, Porceddu E (1995) Development of
Among them, CA1063, CA1062 and CA1061 a set of oligonucleotide primers specific for genes at
had significantly higher yields than Lunxuan the Glu-1 complex loci of wheat. Theor Appl Genet
987, and their yields were also higher than 91:189-194
the check cultivar Zhongmai 175 by 4.1, 3.5, Francis HA, Leitch AR, Koebner RMD (1995) Conversion
and 1.9%, respectively. Eight lines that were of a RAPD-generated PCR product, containing a novel
statistically similar in yield to Zhongmai 175, dispersed repetitive element, into a fast and robust
assay for the presence of rye chromatin in wheat.
exhibited better quality than Lunxuan 987 and Theor Appl Genet 90:636-642
Zhongmai 175, thus indicating that yield and
Graybosch RA, Peterson CJ, Hansen LE, Mattern PJ (1990)
quality were simultaneously improved by Relationship between protein solubility characteristics,
molecular marker assisted selection combined 1BL/1RS, high molecular weight glutenin composition,
with yield trials. and end-use quality in winter germplasm. Cereal
Chem 67:342-349

He ZH, Yang J, Zhang Y, Quail KJ (2004) Pan bread and Payne PI, Holt LM, Jackson EA, Law CN. (1984) Wheat
dry white Chinese noodle quality in Chinese winter storage proteins: their genetics, and their potential
wheats. Euphytica 139:257-267 for manipulation by plant breeding. Phil Trans Roy
Lei ZS, Gale KR, He ZH, Gianibelli C, Larroque O, Butow Soc B 304:359-371
BJ, Morell M, Ma W (2006) Y-type gene specific Payne PI, Lawrence GJ (1983) Catalogue of alleles for
markers for enhanced discrimination of high- the complex gene loci Glu-A1, Glu-B1, and Glu-D1
molecular weight glutenin alleles at the Glu-B1 locus which code for high molecular weight subunits of
in hexaploid wheat. J Cereal Sci 43:94-101 glutenin in hexaploid wheat. Cereal Res Commun
Liu L, Yan J, Zhang Y, He ZH, Peña RJ, Zhang LP (2005) 11:29-35
Allelic variation at the Glu-1 and Glu-3 loci and Payne PI, Holt LM, Krattiger Carillo JM (1988)
presence of 1B/1R translocation, and their effects on Relationship between seed quality and HMW
processing quality in cultivars and advanced lines glutenin subunit composition determined using
from autumn-sown wheat regions in China. Scientia wheat grown in Spain. J Cereal Sci 7:229-235
Agric Sin 38:1944-1950 Rogers WJ, Payne PI, Harinder K (1989) The HMW
Lukow OM, Payne PI, Tkachuk R (1989) The HMW subunit and gladin composition of German-grown
glutenin subunit composition of Canada wheat wheat varieties and their relationship with bread-
cultivars and their association with bread-making making quality. Plant Breeding 103:89-100
quality. J Sci Food & Agric 46:451-460 Smith HA, Bariana HS, Ogbonnaya FC (2001)
Ma W, Zhang W, Gale KR (2003) Multiplex-PCR typing Implementation of markers in Australian wheat
of high molecular weight glutenin alleles in wheat. breeding. Austr J Agric Res 52:1349-1356
Euphytica 134:51-60 Song JM, Liu AF, Wu XY, Liu JJ, Zhao ZD, Liu GT
Payne PL, Corfield KG, Blackman JA (1981) Correlation (2003) The quality relationship between the content
between the inheritance of certain high-molecular- and composing of HMW-GS. Scientia Agric Sin
weight subunits of glutenin and bread-making quality 36(2):128-133
in progenies of six crosses of bread wheat. J Sci Food &
Agric 32:51-60
Complex quality characterization of Hungarian
wheat cultivars
G. Balázs1, A. Harasztos1, Sz. Szendi1, A. Bagdi1, M. Rakszegi2, L. Láng2, Z. Bedő2, F. Békés3,
and S. Tömösközi1
1Budapest University of Technology and Economics (BUTE), Department of Applied Biotechnology and Food
Science, Budapest, Hungary; 2Agricultural Institute, Center for Agricultural Research, Hungarian Academy of
Sciences, Martonvásár, Hungary; 3FBFD PTY LTD, Beecroft, NSW 2119, Australia
Abstract
Old landraces and new Hungarian wheat cultivars bred at the Agricultural Institute of the Hungarian
Academy of Sciences (Martonvásár, Hungary) along with international standard cultivars (Glenlea
and Bezostaja-1) were characterized, covering qualitative and quantitative analysis of gluten and
non-gluten protein as well as the carbohydrate composition of the flours, and rheological analysis.
The aims were to: i) evaluate the genetic potential of these lines with novel and conventional methods;
ii) look for correlations between parameters obtained from chemical composition and from different
conventional and novel analytical methods; iii) develop predictive models for quality attributes as a
function of chemical data, and iv) obtain an improved understanding about rheological parameters
and biochemical background. A subset of results is highlighted to illustrate the usefulness and
outcomes of the study. The results presented were evaluated with regard to different genetic
backgrounds and protein compositions. The chemical parameters were also used in correlation studies
and to develop mathematical modeling for estimation of complex wheat quality.
Introduction aim of incorporating the collected knowledge

The current meaning of wheat quality – in into selection procedures in breeding programs
terms of both economics-related parameters aimed at quality improvement.
such as milling yield and water absorption,
and consumers’ increased demand for healthy, Materials and methods
nutritionally more valuable products –
Twenty-five Hungarian landraces and elite
requires a more complex outlook and scientific
cultivars and two internationally well-known
background than before. Knowledge about the
reference varieties (Glenlea and Bezostaja-1)
relationships among gluten proteins, rheological
grown at the same site at the Agricultural
characteristics of dough and quality of products
Institute of Hungarian Academy of Sciences
offers possibilities not only for understanding,
(Martonvásár) were characterized. They covered
but also for predicting, technological behaviour.
a wide range of variation in qualitative and
However, poor correlations in the case of some
quantitative aspects of gluten and non-gluten
parameters, like water absorption and viscous
proteins as well as starchy and non-starchy
properties, draw attention to the roles of further
carbohydrates.
quality-related wheat components, including
starch, non-starch carbohydrates and non- For profiling of gluten proteins and the albumin
gluten proteins. The spread of more healthy and globulin fractions, a Lab-on-a-Chip (LOC),
cereal products like whole grain products Bioanalyzer 2100 from Agilent was applied.
and enrichment in dietary fiber gives an extra Possible improvements in sample preparation
dimension to understanding the detailed effects for this technology were investigated, and the
of different non-gluten components. results were compared with corresponding
reference methods. SE-HPLC was also applied
This integrated approach was applied in our
to characterize the protein polymer profile by
work on a sample population containing old
UPP% (unsoluble protein polymer percentage).
and new Hungarian wheat cultivars with the
The amylase/amylopectin ratio was determined

by colorimetric methods, and starch damage was using the LOC technique and compared with
measured with SDmatic (Chopin Technologies). those obtained from MALDI-TOF analysis
Quantitative variations in extractable water (Table 1). As discussed in detail in recent
(Weax), and total arabinoxylan (Totax) content published related works (Baracskai et al.
were determined by GC-FID, with the hydrolysis 2011; Kovács et al. 2013; Balázs et al. 2012)
and derivation of sugars. These chemical both the Hungarian landraces and elite lines
parameters were related to different rheological from Martonvásár used in the breeding
measurements, such as MixoLab (Chopin program represent a reasonably wide range
Technologies), RVA (Rapid Visco Analyser, Perten of biodiversity in glutenin alleles. This subset
Instruments.), and the automated, micro sized contains 3, 4 and 2 HMW-GS and 6, 8 and
version of the Zeleny sedimentation test (Sedicom, 5 LMW-GS alleles at the A, B and D loci,
BME-Labintern Ltd, Hungary). theoretically representing 5,760 combinations.
Elution sequences of prolamin proteins often
Results and discussion show unexpected results. Using LOC for
Allelic composition and variability in chemical separating HMW-GS the order of the largest
composition subunits was Ax2*<Ax1<Dx2 compared to the
expected Ax1<Dx2<Ax2*. Similarly, the elution
High molecular weight (HMW) and low molecular sequence Dy9<Dy8<Dy10 was found instead of
weight (LMW) glutenin subunit (GS) allelic the expected By8<By9<Dy10.
compositions of the samples were determined
Table 1. Glutenin allelic composition of samples.
Variety name Glu-A1 Glu-B1 GluD-1 Glu-A3 Glu-B3 Glu-D3

Landraces
Bánkuti-1201 b al a f i c
Bánkuti-1205 b c a a i c
Diószegi-N12 a u a a f m
Fertődi-293-24-5 a c a c c d
Fleischmann-481 b u d b b b
Szekacs1242 a c d b a d
TF-Gyulavári c u a c c c
TF-Homokszentgyörgy a u a a a a
TF-Komádi b c a b c c
TF-Ravazd a c a b c m
TF-Rétság b c d b c m
Lovászpatonai-407 b c d b b b
Elite varieties
MV-Csárdás b c a c j b
MV-Emese b u d c b c
MV-Karizma a al d a b c
MV-Ködmön b c d c h b
MV-Kolo b c d f g b
MV-Kolompos b c a c f d
MV-Lucilla b c d d b b
MV-Magdaléna b c a f j b
MV-Marsall b c d b j c
MV-Mazurka b f a d b b
MV-Suba a c d c b b
MV-Toldi b c d f b b
Control varieties
Bezostaja-1 b c d c c b
Glenlea b al d g g c
About 80% of the LMW-GS allelic combinations a single measurement it is possible to analyze
were differentiated by LOC; however it is not the conventional, mainly protein-related dough
proven to be a decisive method (Balázs et al. properties (like dough strength and stability),
2012). HMW/LMW ratio can also be determined and with a temperature program as a continuous
by LOC. In this case the main problem is the variable, it is possible to characterize the mostly
occasional fusion of the last HMW-GS and the carbohydrate-related viscous parameters.
ladder protein used as an inner standard for the
In our study, it was an acceptable method for
applied LOC system. However, using a correction
differentiation of samples. The dough properties
with the average area of the inner standard,
correlated well with “classic” rheological
meaningful ratios in the interval with acceptable
measurements, like stability with MixoLab
reproducibility (RSD below 10%) can be achieved.
and Farinograph (r=0.74), (Fig. 1); however,
A large level of variation was observed among the carbohydrate-related parameters showed
the quantitative protein composition of samples: less significant correlations with conventional
the Glu/Gli ratio varied between 0.84 and 1.68 methods like RVA parameters.
(average 1.38), whereas the UPP% values (one of
the most important protein parameters in relation
Table 2. Variation in arabinoxylan content and
to bread-making quality) covered an interval of
Zeleny test values in the sample population.
32.2-54.3 (average 44.6). Similarly, large differences
have been observed in both water soluble and total Weax Totax
arabinoxylan content as well as in the values of A/X A/X Zeleny
Zeleny sedimentation test (Table 2). Weax ratio Totax ratio test
Novel methods in quality measurement Min 0.26 0.55 1.10 0.52 2.85
Average 0.65 0.75 1.63 0.62 4.68
Mixolab is a relatively new complex rheological Max 1.09 1.70 2.57 0.74 7.13
instrument from Chopin Technologies. During
Stability - Mixolab (min)

N = 78
Scatterplot: Stability - Mixolab (min) vs. Farino_Stability_sec (Casewise MD ?? Mean = 8, 538974
Farino _Stability_sec = -1,408 + 1,6669 * Stability - Mixolab (min) Std.Dv. = 2, 101229
Max = 11, 670000
Correlation: r = ,73838 Min. = 3, 950000
40 Farino_Stability _sec
N = 78
20 Mean = 12, 825641
Std.Dv. = 4, 743428
Max = 18, 700000
0 Min. = 4, 200000
22
20
18
16
Farino _Stability_sec
14
12
10
8
6
4
2
2 4 6 8 10 12 14 0 20 40
Stability - Mixolab (min) 0,95 Conf. Int.
Fig. 1 Relationship between stability values determined by MixoLab (x) and Farinogrph (y).
The correlation graph was made with Statistica 10.0

Application of the results to water absorption Conclusions
prediction as an example New and old wheat cultivars were characterized
In contrast to dough strength and extensibility, using conventional and novel methods together
where the PSS model of Békés et al. (2006) with reliable predictive models. The large
we successfully applied, prediction of Rmax variations observed in glutenin allelic composition
and EXT from glutenin allelic composition, at the 6 loci, in Glu/Gli ratio, UPP%, amount of
water absorption cannot be estimated using arabinoxlyans and the different rheological quality
only protein parameters. Analysis of soluble parameters measured with Mixolab, and variation
components (both proteins and non-starch in Zeleny number indicated that the population
carbohydrates) together with the level of starch had a high potential for use in breeding new high
damage in the flour provided meaningful quality varieties for different end products. This
relationships. A multiple regression model that level of characterization of samples provides
could be useful in breeding to select suitable a solid basis for classification of wheat and
parents for crossing to improve water absorption helps to understand the specific characteristics
was developed. Measured and predicted water of a superior wheat class, the so called Pannon
absorption values are compared in Fig. 2. Premium.
Acknowledgements
This research work was supported by the
64 Hungarian National Research Fund (OTKA 80292
and OTKA 80334) and the Hungarian National
Project (TECH-09-A3-2009-0221) for development
62
of breeding, agricultural production and food
Water absorption (predicted)
industrial processing system of Pannon wheat

60 varieties.
58 References
Balázs G, Tömösközi S, Harasztos A, Németh T, Tamás A,
Morgounov A, Ma W, Békés F (2012) Advantages and
56 limitation of lab-on-a-chip technique in the analysis
of wheat proteins. Cereal Res Comm DOI: 10.1556/
CRC.2012.0015
54 Baracskai I, Balázs G, Liu L, Ma W, Oszvald M, Newberry
52 56 60 64 68 M, Tömösközi S, Láng L, Bedő Y, Békés F (2011) A
Water absorption (measured) retrospective analysis of HMW and LMW glutenin
alleles of cultivars bred in Martonvásár, Hungary.
Fig. 2 Relationship between measured and predicted
Cereal Res Comm 39:226-237
water absorption.
Békés F, Kemény S, Morell M (2006) An integrated
approach to predicting end-product quality of wheat.
European J Agron 25:155-162
Kovács A, Rakszegi M, Láng L, Ma W, Békés F, Bedő
Z (2013) Application of a rapid electrophoresis
technique analysing the glutenin subunit composition
of Hungarian wheat varieties in breeding selection
practices. Cereal Res Comm. In press
Bread-making qualities of a blend of protein-removed rice
flour and extra-strong wheat flour
W. Maruyama-Funatsuki1, 2, T. Noda2, K. Nagasawa2, Z. Nishio2, M. Ito2, T. Umemoto2, and H. Yamauchi3
1NARO Western Region Agricultural Research Center, Fukuyakma, Japan; 2NARO Hokkaido Agricultural
Research Center, Sapporo, Japan;3Obihiro University of Agriculture and Veterinary Medicine, Obihiro, Japan
Abstract
Rice (Oryza sativa subsp. japonica) flour with the protein removed has good bread-making quality when
it is blended with flour of the Japanese wheat cultivar ‘Yume-Chikara’, which has extra-strong dough
properties. The concentration of NaOH used as a soaking solution to remove the rice protein was
0.10–0.20%. This is useful information allowing the rice/wheat bread market to utilize rice for bread
and to increase the level of self-sufficiency of cereals in Japan.
Introduction for two nights at 4°C in a laboratory. Washed and

The amount of domestically produced wheat (0.8 dried pastes were ground in a mortar and sieved
million tonnes per year) in Japan accounts for (75 μm mesh). Medium particle size (μm), starch
only 14% of gross consumption, whereas almost damage (%), protein content (%), and amylose
95% of rice is domestically produced. However, content (%) of each flour sample were determined
gross consumption of rice has been decreasing (Table 1). The protein-removed rice flour was
in Japan. Almost all (>95%) of the rice in Japan examined by an SEM (TM-1000, HITACHI, Ltd.,
is consumed as boiled rice, with very little being Tokyo), and images were compared with images
utilized as rice flour. Utilization of rice flour to of rice flour milled by a cross jet mill.
make bread, noodles and confectionery could
therefore be effective for increasing production Processing qualities of blended protein-
and the level of self-sufficiency of Japanese removed rice flour and extra-strong wheat flour
cereals. Although bread is the main product A blend consisting of 20% rice flour and 80%
of flour in Japan, bread made from rice/wheat wheat flour from the Japanese extra-strong
flour has smaller specific loaf volumes (SLVs), wheat ‘Yume-Chikara’ was used to evaluate
and SLVs generally decline with increasing bread-making qualities of protein-removed rice
amounts of rice flour. Starch damage in rice flour flour. Bread-making was performed by a no-
is thought to be a major determinant of bread- time method. SLVs of bread were determined. A
making quality (Araki et al. 2009). To explore panel test was performed by the method of the
rice and wheat flour blends suitable for rice/flour Japan Yeast Industry Association in which the
bread with good bread-making quality, SLVs of maximum score is 70.
bread made from a mixture of protein-removed
‘japonica’ rice flour and an extra-strong wheat
flour (20:80) were determined. Results and discussion
Properties of protein-removed rice flour
Materials and methods More protein was removed with higher
concentrations of NaOH (Table 1). The amylose
Properties of protein-removed rice flour content and starch damage for each flour sample
Water-soaked ‘japonica’ rice grains were finely were similar. The medium particle size of
milled by a stone mill and then spray-dried in protein-removed rice flour was the same as that
a commercial factory (Joetsu Starch Co., Ltd., of a single starch granule. Rice starch granules
Nagaoka). The rice flour was then soaked in seemed to be individually detached from the
different concentrations of NaOH overnight or amyloplast by the alkaline treatment (Fig. 1).

Table 1. Basic properties and bread-making qualities of protein-removed rice flour1.
Bread-making
Flour properties qualities of blend3
NaOH Medium particle Starch Protein Amylose SLV
concentration size damage content content ratio3 Panel
Treatment2 (%) μm % % % % score
I - 23.2 2.1 4.7 19.3 82 48.3
II 0.01 7.7 5.6 3.8 20.2 79 50.8
III 0.03 7.4 4.7 3.0 19.8 80 50.3
IV 0.05 6.3 5.5 1.1 20.2 91* 51.3
V 0.10 6.4 5.8 0.2 20.8 88* 52.5
VI 0.15 6.0 5.8 0.2 20.6 92* 52.8
VII 0.20 5.4 8.3 0.2 20.6 90* 52.3
VIII 0.20 6.4 5.7 0.2 19.8 85* 52.3
1 Mean values of more than two determinations.
2 Rice grains were soaked in different concentrations of NaOH overnight at 4°C except for treatment VIII, for which the rice flour was
soaked for two nights.
3 Ratio of specific loaf volumes (SLV; ml/g) of each loaf was calculated against the SLV of 100% Yume-Chikara loaves. SLV ratios with an
asterisk are significantly different (P <0.05) from the SLV of 100% for Yume-Chikara.
A) B)
Fig. 1 SEM images of rice flour. Rice flour milled by a cross jet mill in semi-dry conditions
(A) and protein-removed rice flour treated with 0.2% NaOH (B). Magnification, 500x.
Processing qualities of the blend of protein- With regard to good bread-making qualities
removed rice flour and extra-strong wheat flour of rice flour, the appropriate concentration of
NaOH as a soaking solution to remove rice
The SLV ratios of the rice/wheat blend loaves
protein is thought to be 0.10–0.20%. It is also
were higher at lower protein contents of rice
thought that low protein content, rather than
flour, viz. IV, V, VI, VII and VIII (Fig. 2), but all
low starch damage, contributes more to good
showed higher loaf scores than those made from
bread-making quality of rice/wheat flour blends.
rice flour without alkaline treatment (Table 1, Fig.
2, 3). It seems that storage proteins in ‘japonica’
rice flour prevent polymerization of wheat
glutenin in blended flour and eventually reduce
the SLVs of bread.
100% Rice flour: Rice flour: Rice flour: Rice flour: Rice flour:
Yume-Chikara I IV V VI VII
SLV ratio 100 82 91 88 92 90

Panel score 59.0 48.3 51.3 52.5 52.8 52.3
Fig. 2 Rice/wheat blended loaves. SLV = specific loaf volume.
94
Conclusions
SLVs of rice/wheat bread tended to be small in
SLV ratio of rice/wheat loaf (%
92
100: Yume - Chikara loaf)
proportion to the protein content of the rice flour.

90 A panel test demonstrated that bread using rice
88 flour with a protein content of less than 0.2%
protein had high scores. Protein was removed by
86
treatment of the rice flour with 0.1–0.2% NaOH.
84 R = -0.859*
82
Acknowledgements
80
The authors are grateful to Ms. K. Shimizu for
78
technical assistance.
52 56 60 64 68
Protein content of rice flour (%)
Fig. 3 Correlation between rice flour protein
Reference
content and specific loaf volume (SLV) of rice/
Araki E, Ikeda TM, Ashida K, Takata K, Yanaka M, Iida S
wheat blended loaf. (2009) Effects of rice flour properties on specific loaf
volume of one-loaf bread made from rice flour with
*, Significant at the P = 0.05
wheat vital gluten. Food Sci Technol Res 15:439-448

Effect of cultivar and roller milling on the levels of
phenolic compounds in Italian durum wheat
A. Pasqualone1, L.N. Delvecchio1, G. Mangini2, and A. Blanco2
Department of Soil, Plant and Food Sciences (Di.S.S.P.A.), University of Bari, Via Amendola, 165/a, 70126,
Bari, Italy;1Food Science and Technology Unit; 2Plant Breeding Unit
Abstract
Various studies have analyzed the polyphenol contents of the soft, medium and hard wheat classes
of bread wheat (Triticum aestivum L.). Marginal attention has been devoted to durum wheat (Triticum
turgidum L. spp. turgidum var. durum), the essential raw material for producing high quality pasta, one
of the basic foods in the Italian diet. The aim of this work was to evaluate the effect of roller milling
and cultivars on soluble phenolic contents (SPC) in durum wheat. A set of 20 Italian durum cultivars
was considered. SPC ranged from 1.69 to 2.26 mg ferulic acid equivalent (FAE)/g of wholemeal.
Various milling by-products derived from the same grain lot were analyzed. The SPC of different
milling fractions differed significantly, with the highest levels in bran and “cruschello” middling (about
3.8 mg FAE/g). The decrease in SPC was progressive as the contribution of inner parts of kernels
increased. The results showed that bran and bran-rich middlings from selected Italian durum cultivars
could be potential sources of phenolic substances to be used as functional food ingredients per se, or as
base materials to prepare antioxidant extracts.
Introduction Phenolic acids are the most abundant

Functional bioactive compounds either occur polyphenols in wheat, with ferulic acid present
naturally in food, or are formed during in highest amounts (47 mg/100 g in whole-
the processing of foods; they may have grain wheat flour, measured after hydrolysis),
physiological and biochemical functions when principally in an insoluble-bound form. Free,
consumed by humans (Biesalski et al. 2009). soluble-conjugated, and insoluble-bound
Among them, phenolic substances have been forms of ferulic acid are estimated to occur in
widely studied for their functional value. They a ratio of 0.1:1:100 (Adom and Liu 2002). Other
may act in a variety of ways, such as protecting phenolic acids detected in wheat are sinapic
DNA from oxidative damage, deactivating acid, p-coumaric acid, vanillic acid, caffeic
carcinogens, inhibiting the expression of acid, and diferulic acid, and in minor amounts,
mutated genes, and promoting the activity of syringic acid, o-coumaric acid, gentisic acid and
enzymes that cause carcinogenesis, as well as 2-hydroxybenzoic acid (Zieli et al. 2000).
detoxification of xenobiotics (Yang et al. 2001). Various studies have analyzed the polyphenol
Phenolic compounds include a diverse group contents of soft, medium and hard bread wheats
of substances in the plant kingdom, present (Triticum aestivum L.) and showed that they
in virtually all plant foods, but their dietary varied greatly according to the grain portion
intakes greatly vary depending on the type and being considered. Milling and refining processes
quantity of vegetable foods consumed. Fruits, reduce the polyphenol contents (Adom et al.
vegetables and beverages, such as fruit juices, 2005). Environmental factors, such as soil pH,
tea and wine, are important sources of phenolic rainfall and temperature, also affect the total
compounds in the human diet. Flavonoids, polyphenol contents (Yu and Zhou 2004; Gelinas
phenolic acids and tannins are considered the and McKinnon 2006), and varietal effects are
main dietary phenolic compounds; flavonoids also significant (Adom et al. 2005; Verma et al.
and phenolic acids account for 30% and 60% 2008; Revanappa and Salimath 2011). Finally, the
of total polyphenols in the Mediterranean diet, antioxidant activity of polyphenols is affected
respectively (King and Young 1999). by the action of polyphenol oxidase (PPO) (E.C.
1.14.18.1).
Marginal attention (Lempereur et al. 1997; Menga in 50:50 (v/v) acetone-water. The results were
et al. 2010) has been devoted to the polyphenol expressed as ferulic acid equivalents (FAE).
content of durum wheat (Triticum turgidum
L. ssp. turgidum var. durum), the essential raw Determination of PPO activity
material for producing high quality pasta, one PPO activity was determined as described in
of the basic foods in the Italian diet. The aim of Taranto et al. (2012); 30 seeds from three samples
this work was to evaluate the effect of milling of each plot were incubated in 5 mL of 0.01 M
processes and cultivars on the levels of phenolic disodium tyrosinate solution (pH 10.7) at 37°C for
compounds in Italian durum. The relationship 19 h. After removal of the seeds, the absorbance of
between phenolic levels and flour color with PPO the solution was measured at 405 nm.
activity were also investigated.
Color measures
Materials and methods The brown index (BI) of dough, calculated as
100-L*, was determined using a reflectance
Samples colorimeter (Chroma Meter CR-300, Minolta,
Twenty Italian durum cultivars (Appio, Appulo, Osaka, Japan) equipped with a pulsed xenon arc
Arcangelo, Arcobaleno, Ariosto, Ciccio, Cirillo, lamp. Dough was prepared by hand-kneading
Colosseo, Dauno, Duilio, Fiore, Giotto, Iride, Isa, 1 g of wholemeal from each sample with 700
Messapia, Normanno, Parsifal, Produra, Russello, μL distilled water in a clean porcelain capsule,
Simeto) were grown in the experimental field until water was completely absorbed and dough
of the University of Bari at Valenzano (Italy) in appeared smooth and elastic. The dough was then
2010, in a randomized complete block design compressed between two sterile Petri plates to
with three replications. The plots consisted of 1 m obtain uniformly thick discs. Color measurements
rows, 30 cm apart, with 50 germinating seeds per were performed immediately after kneading
plot. Three sets of different milling by-products and compression. Brown indices of milling
derived from the same grain lots were supplied by-products were determined by placing them
by local durum wheat mills. These included: into the granular materials attachment (CR-A50,
debranned wheat, bran and various middlings, Minolta, Osaka, Japan) of the colorimeter to obtain
called “cruschello”, “farinaccio” and “farinetta” a smooth surface suitable for color readings.
containing, in different proportions, both starchy
endosperm and fragments of outer grain layers Quality determinations on milling by-products
(Table 1). Levels of ash, total dietary fiber and starch were
determined according to AACC Methods 32-45.01,
Determination of phenolic substances 08-01.01, and 76-13.01, respectively.
Grains of each cultivar were milled on a
laboratory mill (Cyclotec Sample Mill, Tecator Statistical analyses
Foss, Hillerød, Denmark). Following McCallum One-way analysis of variance was carried out to
and Walker (1990), 0.1 g of wholemeal flour compare the analytical data.
was incubated at room temperature for 2 h with
1 mL acetone-water 50:50 (v/v) on an orbital
shaker set at 200 rev/min. This was followed Results and discussion
by centrifugation at 8,000 g for 5 min. The SPC The distribution of SPC in the durum cultivars
of supernatant extract was determined with is shown in Fig. 1. SPC ranged from 1.69 to 2.26
Folin-Ciocalteu reagent. The reaction mixture mg FAE/g wholemeal. The highest values were
contained 100 μL extract, 500 μL Folin-Ciocalteu observed in Ariosto, followed by Colosseo,
reagent as described by Yu and Zhou (2004) and Russello and Appio, whereas the lowest level
2 ml of 15% sodium carbonate. The final volume was in Normanno, followed by Appulo and
was made up to 10 mL with distilled water. After Arcangelo. The majority of cultivars showed levels
2 h, and then centrifugation at 12,000 g for 3 min, ranging from 1.81 to 1.90 mg FAE/g wholemeal.
the absorbance at 765 nm was read. The phenolic Four cultivars, namely Normanno, Ciccio, Simeto
content was estimated on the basis of a standard and Duilio, were recently analyzed by Menga et
curve prepared using ferulic acid solutions al. (2010). Although the absolute values of SPC

obtained in our determinations were all slightly PPO activities of wholemeal of the Italian durum
higher than the levels detected by Menga et al. cultivars (Fig. 2) ranged from 0.39 to 0.78, with
(2010) (probably due to both environmental the lowest value in Normanno, and the highest in
effects and different extraction procedures), Parsifal. The majority of cultivars showed values
the trend was the same: SPC increased from in the range of 0.41–0.50. Despite the fact that
Normanno, to Ciccio, Simeto, and Duilio. the antioxidant activity of polyphenols is known
to be affected by the action of PPO, in the set
The results in Italian durum cultivars were higher
of samples considered a significant correlation
than those reported for bread wheat, although
between SPC and PPO activity (R = 0.247) was
the methods used to extract and evaluate the
not observed. However, further investigation is
phenolic contents varied considerably among
needed on a wider number of samples to better
reports, making comparisons of data rather
evaluate the significance of this correlation.
difficult. Finally, an environmental effect cannot
be excluded, since as reported Mediterranean The brown index of doughs from the examined
climatic factors, such as drought stress and cultivars ranged from 30.12 to 32.46 (Fig. 3), with
high temperatures during grain filling (typical the majority of cultivars showing low values
of the area where the samples for the current (below 31.5). High brown index values are
research were grown) lead to higher phenolic
contents (Menga et al. 2010). In fact, among
their physiological functions, these secondary 12
metabolites stabilize cell walls and act as screens
against UV radiation (Abdel-Aal et al. 2001; 10
Daniel et al. 1999).
Number of samples
8
In bread wheat, Revanappa and Salimath (2011)
assessed phenolic contents of 0.216–0.246 mg 6
of gallic acid equivalent (GAE)/g in wholemeal
from Indian cultivars, Adom et al. (2003) detected 4
levels of 709.8–860.0 μmoL of GAE/100 g in US
cultivars, Vaher et al. (2011) observed levels of 2
0.168–0.459 mg GAE/g in 15 Estonian cultivars,
and Menga et al. (2010) recorded contents 0
0.30-0.40 0.41-0.5 0.51-0.6 0.61-0.7 >0.71
between 0.783 and 1.068 mg FAE/g in wholegrain
PPO activity (Abs 405 nm)
from 25 cultivars. The last measurements were
similar to those from 30 durum samples grown Fig. 2 Distribution of polyphenol oxidase activities
in the same conditions (0.780–0.946 mg FAE/g) in wholemeal from 20 Italian durum cultivars.
(Menga et al. 2010). PPO = polyphenol oxidase.
7 6
6 5
5
Number of samples
Number of samples
4
4
3
3
2
2
1 1
0 0
1.6-1.7 1.71-1.8 1.81-1.9 1.91-2 >2.1 30-30-50 30.51-31 31.01-31.5 3151-32 >32
Phenolic content (mg ferulic acid equivalent/g whole meal) Brown Index of dough
Fig. 1 Distribution of soluble phenolic contents in Fig. 3 Distribution of brown index of the dough
wholemeal from 20 Italian durum cultivars. obtained from wholemeal of 20 Italian durum cultivars.
usually considered undesirable. In this cultivar Adom et al. (2005) detected 0.487–0.530 mg GAE/g
set no significant correlation was observed in bran/germ fractions of US bread wheat. The
between SPC and the brown index of dough (R = brown index of the various milling by-products
0.132), but again, further investigation is needed paralleled the trend observed for SPC, i.e.,
on a wider diversity of samples, that should increased with the proportion of the outer layer
include landraces and wild tetrapoloid accessions components.
from germplasm collections as modern cultivars
may have a restricted range of variability.
Conclusions
To assess the effect of roller milling on SPC
content, debranned wheat, bran and various As for bread wheat, bran and bran-rich middlings
middlings derived from the same grain lot from selected Italian durum cultivars could be a
(referred to, in Italian, as “cruschello”, “farinaccio” potential source of phenolic substances to be used
and “farinetta”) were analyzed. These middlings as functional food ingredients per se, or as a source
contained different proportions of both starchy material to prepare enriched antioxidant extracts.
grain and fragments of the outer grain layers.
The contribution of the external layers of
Acknowledgement
the caryopsis decreased from “cruschello”, to
“farinaccio” and “farinetta”, as indicated by This research was funded by MIUR, within the
decreasing levels of total dietary fiber and ash, PON ISCOCEM Project no. 01_01145.
and by increasing values of starch (Table 1).
The SPC of the examined by-products differed References
significantly, with the highest values in bran Abdel-Aal ESM, Hucl P, Sosulski FW, Graf R, Gillott C,
and “cruschello” (about 3.8 mg FAE/g). The SPC Pietrzak L (2001) Screening spring wheat for midge
decrease was progressive as the contribution resistance in relation to ferulic acid content. J Agric
of inner parts of the kernel increased. These Food Chem 49:3559-3566
findings were similar to those reported for bread Adom KK, Liu RH (2002) Antioxidant activity of grains. J
wheat, where the SPC of bran fractions are about Agric Food Chem 50:6182-6187
15-fold higher than those of endosperm (Adom et Adom KK, Sorrells ME, Li RH (2003) Phytochemical
al. 2005; Gallardo et al, 2006; Mattila et al. 2005). profiles and antioxidant activity of wheat varieties. J
Agric Food Chem 51:7825-7834
Among 51 cultivars of bread wheat of different
Adom KK, Sorrells ME, Liu RH (2005) Phytochemicals
classes and some cultivars of durum, Verma
and antioxidant activity of milled fractions of different
et al. (2008) detected SPC levels ranging from wheat varieties. J Agric Food Chem 53:2297-2306
3.40 to 6.70 mg GAE/g bran. Mattila et al. (2005) Biesalski HK, Dragsted LO, Elmadfa I, Grossklaus R,
detected phenolic content levels of 4.52 mg/g Müller M, Schrenk D, Walter P, Weber P (2009)
in bran, Velioglu et al. (1998) reported 0.349 mg Bioactive compounds: definition and assessment of
FAE/g in commercial bread wheat germ, and activity. Nutrition 11:1202-1205
Table 1. Mean values of soluble phenolic contents (SPC) and brown index (BI) of different commercial
roller milling by-products (n = 3).
SPC BI Total dietary fiber Starch Ash

Sample (mg FAE/g d.m.) (100 – L*) (g/100 g w.m.) (g/100 g w.m.) (g/100 g w.m.)
Durum bran 3.83±0.14a 34.53±0.04a 12.2±0.3 - 6.80±0.03
Durum middling “cruschello” 3.76±0.34a 28.52±0.09b 11.5±0.2 - 6.00±0.02
Durum middling “farinaccio” 3.20±0.03c 22.52±0.03d 5.1±0.1 31.1±0.9 4.19±0.01
Durum middling “farinetta” 1.66±0.09b 17.01±0.03c 3.0±0.1 42.3±1.1 2.60±0.01
Debranned durum 1.60±0.05b 25.19±0.02e n.d. n.d. n.d.
a Different letters indicate significant differences at P<0.05; n.d., not determined; d.m., dry matter; w.m., wet matter (moisture content
<15.50 g/100 g)

Daniel O, Meier MS, Schlatter J, Frischknecht P (1999) Revanappa SB, Salimath PV (2011) Phenolic acid and
Selected phenolic compounds in cultivated plants: antioxidant activities of different wheat (Triticum
Ecologic functions, health implications, and aestivum L.) varieties. J Food Biochem 35:759-775
modulation by pesticides. Environ Health Perspect Taranto F, Delvecchio LN, Mangini G, Del Faro L, Blanco
107:109-114 A, Pasqualone A (2012) Molecular and physico-
Gallardo C, Jimenez L, Garcia-Conesa MT (2006) chemical evaluation of enzymatic browning of whole
Hydroxycinnamic acid composition and in vitro meal and dough in a collection of tetraploid wheats. J
antioxidant activity of selected grain fractions. Food Cereal Sci 55:405-414
Chem 99:455-463 Vaher M, Matso K, Levandi T, Helmja K, Kaljurand M
Gelinas P, McKinnon CM (2006) Effect of wheat variety, (2011) Phenolic compounds and the antioxidant
farming site, and bread-baking on total phenolics. Int J activity of the bran, flour and whole grain of different
Food Sci Technol 41:329-332 wheat varieties. Procedia Chem 2:76-82
King A, Young G (1999) Characteristics and occurrence Velioglu YS, Mazza G, Gao L, Oomah BD (1998).
of phenolic phytochemicals. J Am Dietetic Assoc Antioxidant activity and total phenolics in selected
99:213-218 fruits, vegetables, and grain products. J Agr Food
Lempereur I, Rouau X, Abecassis J (1997) Genetic and Chem 46:4113-4117
agronomic variation in arabinoxylan and ferulic acid Verma B, Hucl P, Chibbar RN (2008) Phenolic content and
contents of durum wheat (Triticum durum L.) grain antioxidant properties of bran in 51 wheat cultivars.
and its milling fractions. J Cereal Sci 25:103-110 Cereal Chem 85:544-549
Mattila P, Pihlava JM, Hellstrom J (2005) Contents of Yang CS, Landau JM, Huang M-T, Newmark HL (2001)
phenolic acids, alkyl- and alkenylresorcinols, and Inhibition of carcinogenesis by dietary polyphenolic
avenanthramides in commercial grain products. J compounds. Annu Rev Nutr 21:381-406
Agric Food Chem 53:8290-8295 Yu L, Zhou K (2004) Antioxidant properties of bran
McCallum JA, Walker JR (1990) o-diphenol oxidase extracts from ‘Platte’ wheat grown at different
activity, phenolic content and colour of New Zealand locations. Food Chem 90:311-316
wheats, flours and milling streams. J Cereal Sci 12:83-96 Zieli Å, Ki H, Kozaowska H (2000) Antioxidant activity
Menga V, Fares C, Troccoli A, Cattivelli L, Baiano A and total phenolics in selected cereal grains and their
(2010) Effects of genotype, location and baking on the different morphological fractions. J Agric Food Chem
phenolic content and some antioxidant properties of 48:2008-2016
cereal species. Int J Food Sci Technol 45:7-16
The effect of bug (Eurygaster spp. and Aelia spp.)
damaged flours on formation of acrylamide and
hydroxmethylfurfural (HMF) in cakes, cookies and breads
S. Özçandir, D. Sivri Özay, and V. Gökmen
Hacettepe University, Department of Food Engineering, Beytepe, Ankara, Turkey
Abstract
Grain pests like Pentatomid insects named as wheat bugs (Eurygaster spp., Aelia spp. and Nysius
huttoni) or “suni bug” (Eurygaster spp.) have detrimental effects on grains, especially Gramineae.
While the insects absorb their nutrients they leave their digestive secretions in the grain before
harvest. The proteolitic enzymes in their salivary secretion result in processing problems in dough
and low-quality end products. It has been shown that wheat bug damage causes high protease
activity in flour, resulting in hydrolysis of gluten and release of some amino acids. Acrylamide and
hydroxymethylfurfural (HMF) are mainly formed through the Maillard reaction and dehydration
of certain sugars. They can be regarded as the most important heat-induced contaminants occurring
in bakery products. Formation of acrylamide, free amino acids, especially asparagine, and reducing
sugars in food are important precursors. In this study, the effects of high protease activity flour
damaged by Eurygaster spp. and/or Aelia spp. on formation of acrylamide and HMF in bakery
products such as bread, cakes and cookies were determined by liquid chromatography coupled with
tandem mass spectrometry (LC-MS/MS). High protease activity flour due to Eurygaster spp. and/
or Aelia spp. damage was added at different levels in cookies (0%, 20%, 40%, 60%, 80% and 100%),
cakes (0%, 25%, 50%, 75% and 100%) and bread produced by modified AACC (2.5%, 5%, 10% and
15%) formulations. No variations were observed in acrylamide and HMF contents of bread and cake
samples with added bug-damaged wheat flour. Although no change was observed in the HMF content
of cookies, formation of acrylamide increased up to 30% (96 μg/kg) at the highest addition level (100%)
as compared to the control (77 μg/kg). These results suggested that bug-damaged flour causes no
health risk for baking products such as bread and cakes due to acrylamide formation, but there is an
increased potential health risk in cookies. Variations in free amino acids in high protease activity flours
during hydrolysis (0, 30, 45, 60 and 120 min) were also determined by liquid chromatography coupled
with high resolution mass spectrometry (LC-HRMS). The amounts of asparagine in bug-damaged
flour (HPAF1) used in cookie and cake production were increased by approximately 23%.
Introduction been shown that bug-damaged wheat causes

Grain pests like Pentatomid insects named as high protease activity in flour (HPAF), and then
“suni bug” (Eurygaster spp.) and “wheat bugs” hydrolysis of gluten and release of some amino
(Eurygaster spp., Aelia spp. and Nysius huttoni) acids due to increased solubility (Sivri et al. 1998;
have detrimental effects on grains, especially Sivri and Köksel 2000).
those of the Gramineae, in some Mediterranean, Acrylamide and hydroxymethylfurfural (HMF)
Middle Eastern, East European, and Near Eastern are mainly formed through the Maillard Reaction
countries, and in New Zealand. While the insects and dehydration of certain sugars. They are
absorb their nutrients during the nymph and the most important heat-induced contaminants
adult periods, they leave digestive secretions in occurring in bread and bakery products. In the
the grain before harvest (Paulianian and Popov formation of acrylamide in food, free amino
1980). The proteolytic enzymes in their salivary acids, especially asparagine and reducing sugars
secretions cause processing problems in dough are important precursors (Fig. 1).
resulting in low-quality end products. It has

In this study, we investigated the effects of high AACC methods (AACC 1990) were used for
protease activity flour damaged by Eurygaster production of bread (Method 10-11), cakes
spp. and/or Aelia spp. on formation of acrylamide (Method 10-90) and cookies (Method 10-
and HMF in bakery products such as breads, 54). Acrylamide contents of bread, cake and
cakes and cookies. In addition, asparagine levels cookie samples were determined by liquid
were determined during hydrolysis of gluten of chromatography coupled to tandem mass
HPAF. spectrometry (LC-MS/MS) (Gökmen et al. 2009).
High performance liquid chromatography
(HPLC, Agilent Technologies, Waldbronn,
Materials and methods Germany) was used for determination of HMF
High protease activity flour (HPAF1) due to contents in samples (Gökmen and Şenyuva 2006).
Eurygaster spp. and/or Aelia spp. damage was Variation of asparagine content in bug damaged
used for bread production at 2.5%, 5%, 10%, flours was monitored by liquid chromatography
15% (w/w) addition levels. HPAF2 was added coupled with high resolution mass spectrometry
at 20%, 40%, 60%, 80%, 100% (w/w) and 25%, (LC-HRMS, Thermo Fisher Scientific, San Jose,
50%, 75%, 100% (w/w) levels in cookie and cake CA, USA) during gluten hydrolysis at 0, 30, 45, 60
formulations, respectively. and 120 min (Gökmen et al. 2012).
NH2 NH2
O O
O CH2 CH2
+ HO H
R1 CH OH CH OH
R2 NH2 R1 N
O R2 O
carbonyl source asparagine
H2O
NH2
O NH2
CH2 O
R1 CH OH CH2
N
R1 CH2
R2 O CO2 N
decarboxylation R2
Schiff base
H2O
R1 NH2 NH2
R1
O O O + NH
+ CH2 CH
R2 R2
CH2 CH2
H2N ACRYLAMIDE
(Beta alanine amide)
NH2
O + NH3
CH
CH2
ACRYLAMIDE
Fig. 1 Acrylamide formation in the Maillard Reaction (Zyzak et al. 2003).
Results kg) (Fig. 2, 3). Acrylamide contents in cookies
No variations were observed in acrylamide and at different HPAF2 addition levels are shown as
HMF contents of both bread and cake samples chromatographic peaks in Fig. 4. The amount
(results not shown). Although no change in of asparagine in bug-damaged flour (HPAF2)
the HMF content of cookies was observed, used in cookie production was increased by
formation of acrylamide was increased by up to approximately 26% (Fig. 5). No change was
30% (100 μg/kg) at the highest HPAF2 addition observed in the asparagine content of bug-
level (100%) compared to the control (77 μg/ damaged flour (HPAF1) used for bread.
Acrylamide
120
1.07 100%
100
80%
60%
80 40%
20%
0%
g/kg
60
40
20 0.92 0.99
0 0
0 20 40 60 80 100 0.95 1.00 1.05 1.10 1.15 1.20 1.25
HPAF2 addition level (%) Time
Fig. 2 Effect of HPAF2 on formation of acrylamide Fig. 4 Chromatograms of acrylamide contents of
in cookies. cookies at different HPAF2 addition levels.
HMF Asparagine
1.6 120
HPAF2 HPAF1
1.4
100
1.2
80
1
mg/kg
0.8 60
0.6
40
0.4
20
0.2
0 0
0 20 40 60 80 100 0 30 45 60 120
HPAF2 addition level (%) Time (min)
Fig. 3 Effect of HPAF2 on formation of acrylamide Fig. 5 Variation in asparagine during hydrolysis.
in cookies.

Conclusions Sivri D, Köksel H (2000) Characterization and partial
purification of a gluten hydrolyzing proteinase from
These results suggest that bug-damaged flour bug (Eurygaster spp.) damaged wheat. In: Shewry
causes increased health risks due to acrylamide DR, Tatham AS (eds). Wheat Gluten, Bristol, UK,
formation in cookies. pp287-290
Gökmen V, Ataç, B, Serpen A, Morales FJ (2009)
Multiple stage extraction strategy for the
Acknowledgements determination of acrylamide in foods. J Food Comp
& Analysis 22:142-147
This research was supported by The Scientific
Gökmen V, Serpen A, Moğol BA (2012) Rapid
and Technological Research Council of Turkey determination of amino acids in foods by
(TÜBİTAK) (Project No: 111 O 060). hydrophilic interaction liquid chromatography
coupled to high-resolution mass spectrometry. Anal
& Bioanal Chem 403:2915-2922
References Gökmen V, Şenyuva HZ (2006) Improved method for
AACC (1990) Approved methods of the AACC, 8th the determination of hydroxymethylfurfural in
Edition. AACC, St. Paul, MN baby foods using liquid chromatography-mass
spectrometry. J Agric & Food Chem 54:2485-2489
Paulian F, Popov C (1980) Sunn pest or cereal bug.
Wheat Technical Monograph, Ciba-Geigy Ltd, Basel, Zyzak DV, Sanders RA, Stojanovic M, Tallmadge DH,
Switzerland, 69-74 Eberhardt BL, Ewald DK, Gruber DC, Morsch
TR, Strothers MA, Rizzi GP, Villagran MD (2003)
Sivri D, Köksel H, Bushuk W (1998) Effects of wheat Acrylamide formation mechanism in heated foods. J
bug (Eurygaster maura) proteolytic enzymes on Agric & Food Chem 51:4782-4787
electrophoretic properties of gluten proteins. NZ J
Crop & Hort Sci 26:117-125
Session 4: Starch and health attributes of the
wheat grain
Starch biosynthesis in rice grains: Natural variation and

genetic improvement
Qiaoquan Liu1, Minghong Gu1, Jiayang Li2, and Yongcheng Shi3
1Collegeof Agriculture, Yangzhou University, Yangzhou 225009, Jiangsu, China; 2Institute of Genetics and
Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China; 3Department of Grain Science
and Industry, Kansas State University, Manhattan, KS 66506, USA
Starch, the reserve carbohydrate in cereal grains, makes up approximately 80-90% of the dry weight
of the mature cereal grain, and contains two distinct polymers, amylose and amylopectin. The ratio of
amylose and amylopectin as well as their fine structure varies with the botanical source of starch and
affects its end use quality. Rice, rich in starch, is used as a major diet for more than half of the world’s
population. In Asia, over 2,000 million people obtain 60% to 70% of their daily calories from rice and
its processed products. Compared with other crops, rice starch contains tiny starch granules with
a narrow size distribution, making rice starch ideally suited as a cosmetic dusting power, a textile
stiffening agent, and a fat mimetic in foods.
The grain quality in rice was greatly affected by starch composition and structure. Therefore, starch
biosynthesis might play a crucial role in the formation of rice quality, especially the cooking and
eating quality. In our studies, the natural variation of 18 genes involved in starch synthesis, as well
as the diversity of the grain starch quality, was carefully investigated among many representative
germplasms. The effects of the variation on grain quality were firstly studied through association
analysis, and then confirmed by using near-isogenic lines and/or transformants. In this presentation,
the genetic variation of these starch synthesis related genes (SSRGs) and their effects on grain quality
will be presented. The strategies and application of molecular techniques for quality improvement in
rice will be also presented and discussed.
Session 4: Starch and health aƩributes of the wheat grain 179

Bio-fortification for high micronutrients in wheat breeding
programs in China
Y. Zhang1, L. Wang1, and Z.H. He1, 2
1Instituteof Crop Science, Chinese Academy of Agricultural Sciences (CAAS), 12 Zhongguancun South Street,
Beijing 100081, China; 2CIMMYT China Office, C/O CAAS, 12 Zhongguancun South Street, Beijing 100081, China
Abstract
Two hundred and sixty-five Chinese wheat cultivars were sown at Anyang, Henan province, in 2005-2006
to evaluate genetic variation in major mineral elements in the grain. They displayed a wide variation
for all elements investigated, ranging from 28.0 to 65.4 mg kg-1 for Fe and 21.4 to 58.2 mg kg-1 for Zn.
Cultivars Jimai 26, Henong 326, and Jingdong 8 displayed high Fe and Zn concentrations, for which
Jingdong 8 was the most promising for increasing Fe and Zn. Thirty-seven cultivars grown at Anyang
over two seasons were used to determine phenolic acids, and the fractions of free and bound types were
analyzed using HPLC with measurements of individual phenolic acids in each fraction. Most of the
parameters were significantly influenced by cultivars, seasons, and their interactions, with cultivars being
predominant. The average concentration of bound phenolics was 661 μg g-1 of dm (dry matter), making
up 97.5% of the total phenolic acids, with ferulic acid accounting for 70.7% of it. Free phenolics made
up only 2.5%, with syringic acid accounting for 44.7% of that. Bound phenolics predominated. Cultivars
Liangxing 66 and Zhongmai 895 were stable with high values in phenolic acids across seasons, indicating
they could be used as parents in breeding for health beneficial phenolic acid content.
Introduction the risk of chronic diseases. Bio-fortification

Nutritional problems related to diets of cereals through plant breeding has multiplicative
such as wheat have been observed throughout advantages (Bouis et al. 2000), in reaching
the world including China, and the consequences the total population in both urban and rural
of malnutrition create immense economic and areas without a need for ongoing investment
social costs (Bouis et al. 2000; Chen 2004). More once the bio-fortified cultivars have been
than half of the world’s population suffers developed, and not disappearing after initial
from micronutrient under-nourishment, and successful investment and research, as long as a
20% have metabolic problems (Ward et al. domestic agricultural research infrastructure is
2008). It is estimated that two billion people maintained (Bouis et al. 2000).
worldwide suffer from iron deficiency, Common wheat (Triticum aestivum L.) is one
especially children and women, and mineral of the most important staple food crops and
deficiencies in vulnerable groups have led to the cheapest source of calories and protein for
concern by the Chinese Government (Chen local inhabitants in China. The Northern China
2004). The prevalence of anemia in China is Winter Wheat Region plays the most important
about 25% on average, but can be as high as role in producing and contributing about 70%
60% in children and 35% in pregnant women in of the total national wheat production, and
rural areas, largely due to dietary Fe deficiency. 75% or more of the commercial grain (Zhuang
Cardiovascular incidence in urban China is 2003). Grain is traded across the nation, adding
about 18% and has almost doubled in the last 10 to local consumption in the form of various
years (Chen 2004). These nutritional problems kinds of noodles and steamed bread as the main
have led to two bio-fortification programs, viz. end products, especially in rural areas. The
HavestPlus, with the objective of developing objectives of this study were to: 1) determine the
high micronutrient levels in staple crops through levels of mineral elements Fe and Zn in grain
breeding, and the HEALTHGRAIN program, of current leading cultivars from the Northern
with the objective of increasing the protective China Winter Wheat region, and 2) characterize
roles of phytochemicals in crops, to address the phenolic acid profiles and concentrations in

grain of leading Chinese cultivars and promising Free phenolic acids in the wheat flour were
advanced lines. The information generated from extracted following a previously reported
the study could be important for Chinese wheat method (Adom et al. 2003, 2005). Bound
breeding programs, with potential application phenolic acids were extracted according to the
in other wheat producing countries, including published method of Mattila et al. (2005) with
India, Pakistan, Turkey and Europe, who are also minor modifications. Individual phenolic acids
participants in the Bio-fortification program. in the wheat meal extracts were analyzed by
an Agilent 1100 Series high-performance liquid
chromatogram equipped with UV detector and
Materials and methods an Agilent TC-C18 (250×4.60mm, 5μm) column.
Two hundred and sixty-five cultivars Identification and quantification of phenolic
recommended by major breeding programs in acids in samples were performed by comparison
the Northern China Winter Wheat Region were with chromatographic retention times and areas
collected and grown as 1 m single row plots in of external standards. Calibration curves of
irrigated field trials with 2 replications at the phenolic acid standards were constructed using
CAAS Institute of Crop Science Anyang wheat authentic standards that had undergone the same
breeding station (lat. 36°06′N, long. 114°21′E, extraction procedure to ensure that losses due to
61 masl) in Henan province, in 2005-2006, to the extraction were accounted for. All chemicals
investigate the current status of Fe and Zn in used in this study were of analytical grade, and
those cultivars. Each trial was planted on the samples were prepared and analyzed at least in
locally recommended date at a seeding rate of 130 duplicate.
kg ha-1. Cultivars were sown in a latinized alpha
lattice design (Zhang et al. 2007), and one local
check was also included in every 10 cultivars. Results and discussion
Grain samples were harvested and cleaned by
Fe and Zn
hand to avoid contamination with minerals.
There was a wide range of variation among
Thirty-seven high-yielding commercial cultivars genotypes for all mineral elements in the grain,
recommended from the Yellow and Huai River with the highest concentrations of all elements
Valleys Winter Wheat Region, were sown at almost double the lowest. Cultivars showed
Anyang in 2 m x 2 row plots in irrigated field almost the same mean value and range of Fe and
trials in 2008-2009 and 2009-2010 in a completely Zn concentrations as those reported by Zhang et
randomized block design in two replications al. (2007), ranging from 28.0 to 60.2 mg kg-1 for
for phenolic acids profile characterization. Fe with mean of 39.2 mg kg-1, and from 21.4 to
Field management and timing of management 58.2 mg kg-1 for Zn with mean of 32.3 mg kg-1.
practices including fertilization matched local All the samples were sound, and the 1,000 kernel
commercial practices. weights and protein contents ranged from 28.0 to
Thousand kernel weights were obtained as 48.7 g and from 10.1 to 15.6%, with means of 34.7
averages of three samples of 200 seeds. Grain g and 12.6%.
protein content was obtained with a near- Jimai 26, Henong 326, Han 3475, and Jingdong 8
infrared (NIR) analyzer (Instalab 600, Newport displayed high Fe concentration, all with values
Scientific Sales and Services Ltd., Australia) more than 50 mg kg-1; Jimai 26, Henong 326,
following AACC approved method 39-10 (AACC Nongda 3197, and Jingdong 8 displayed high Zn
2000). Each sample was milled into fine powder concentrations, with values of more than 46.1
with a 60-mesh screen, thoroughly mixed, and mg kg-1, among which Jimai 26 and Henong 326
cooled immediately and stored at −20°C until also displayed high values of Cu, Mg, K, P, and
analyzed. Mineral element concentrations of protein content. These cultivars could be used
grain were measured using inductively coupled as parents in breeding programs targeting high
plasma atomic emission spectrometry (ICP- nutrient concentrations. The high concentrations
AES, OPTIMA 3300 DV) to quantify aqueous of Fe and Zn for Jingdong 8, the leading cultivar
constituents following microwave digestion with used as a local check in the northern part of
HNO3-H2O2 solution. the Northern China Winter Wheat Region
with high yield potential, were combined with

good resistance to powdery mildew, and good of dm, and bound phenolic acids ranged from 530
tolerance to high temperature during grain filling μg g-1 of dm to 854 μg g-1 of dm. Contributions
(Zhuang 2003). This made it the most promising from the free types ranged from 1.5% in Xingtai
cultivar, since lines selected for high Fe and Zn 456 to 3.6% in Jishi 02-1, whereas those of the
must have good agronomic performance and bound type ranged from 96.4% in Jishi 02-1 to
broad adaptation. Zhongyou 9507, Zhongyou 98.5% in Xingtai 456, in agreement with the large
16, and Liken 2, which were all reselected from variations observed in previous studies (Adom
Zhongzou 8131-1 (Zhuang 2003), had high et al. 2005; Li et al. 2008). This also indicates that
concentrations of Fe and Zn. genetic improvement could play a significant role
in further improvement of the health benefits of
Based on personal daily requirements of Fe
wheat. Ferulic acid, representing up to 90% of
and Zn, and the information of bioavailability
total polyphenols, is the predominant phenolic
provided by the WHO (2006), the proposed
acid in wheat, contributing from 62.4% in Pumai
concentrations for Fe and Zn was 57 and 41
9 to 79.2% in Zhengmai 366, in agreement with
mg kg-1, respectively. However, only Jimai 26
previous reports (Adom et al. 2005). Ferulic acid
had a Fe concentration of more than 60 mg
is a precursor for synthesis of other aromatic
kg-1, whereas tens of cultivars in this study had
compounds in plants, and comprises a significant
Zn concentrations of more than 40 mg kg-1.
proportion of each of the phenolic acid fractions,
Therefore, there is sufficient genetic variability
with concentrations ranging from 2 to 8 μg g-1 of
to develop wheat cultivars with increased grain
dm for the free form to 353-628 μg g-1 of dm for
Zn levels and promising genetic variability for
the bound form, consistent with the study of Li
Fe, but due to the lower bioavailability of Fe
et al. (2008). The average percentage contribution
when compared with Zn, target levels for Fe are
to the grain ferulic acid by the bound type was
significantly higher and meeting them will be
99%, much more than reported previously
more challenging.
(Adom et al. 2003, 2005; Li et al. 2008). This may
Phenolics largely be due to the undetermined soluble
conjugated phenolic acids in this study. Li et al.
The mean concentration of free phenolic acids (2008) indicated that free phenolic acids make
in flour samples from 37 cultivars was 17 μg the smallest (typically less than 1%) contribution
g-1 of dm, making up only 2.5% of the total to the total phenolic acid concentrations in
phenolic acids determined. Syringic, caffeic, wheats, with the average total free type being of
and ferulic acids accounted for 44.7, 30.8, and 11 μg g-1 of dm; whereas undetermined soluble
24.5% of it. There were no detectable amounts conjugated type in this study made up a greater
of free phenolic acids chlorogenic, vanillic, proportion (about 22%) of the total phenolic acid
p-coumaric, and gentisic. The concentration concentrations, containing around 162 μg g-1
of bound phenolic acids was 661 μg g-1 of dm, of dm on average; while bound phenolic acids
making up 97.5% of the total phenolic acids contributed the highest proportion (around
determined, among which ferulic accounted for 77%) of the total phenolic acids, with mean
70.7% of it, followed by caffeic and p-coumaric. concentrations of 492 μg g-1 of dm.
This is in agreement with Ward et al. (2008).
Bound phenolics are more likely to survive upper The highest values over the two seasons for
gastrointestinal digestion and can be released concentrations of free and bound phenolic acids
from the colon through microflora digestion in 12 cultivars are shown in Table 1. There were
activity (Adom et al. 2003). Thus, wheat phenolics highly significant effects of season and cultivar
are more likely to exert health benefits in the × season interactions. In 2008-2009, Jishi 02-1
colon where they are released. This may partly and Taishan 23 had the highest free phenolic
explain the reduced incidence of colon cancers acids concentrations, and Kaimai 18, Linyou
and other chronic diseases associated with the 145, Xingtai 456, and Zhoumai 22 had the
consumption of wheat grain products (Jacobs highest bound phenolic acids concentrations.
et al. 1998). However, there were considerable In 2009-2010, Jimai 19, Liangxing 66, and
differences between cultivars in both free and Yanzhan 4110 had the highest free phenolic acids
bound forms, as well as their compositions. Free concentrations, and Kaimai 18, Jimai 19, and
phenolics ranged from 9 μg g-1 of dm to 27 μg g-1 Zhengmai 366 had the highest bound phenolic

Table 1. The highest free and bound phenolic acids contents in grain of 12 selected wheat cultivars
(μg g-1 dry matter).
Season 2008-2009 2009-2010

Cultivar Free Cultivar Bound Cultivar Free Cultivar Bound
Jishi 02-1 45 Kaimai 18 756 Jimai 19 25 Kaimai 18 951
Taishan 23 39 Linyou 145 725 Liangxing 66 25 Jimai 19 923
Zhoumai 18 28 Xingtai 456 711 Yanzhan 4110 25 Zhengmai 366 902
Zhongmai 349 22 Zhoumai 22 705 Zhoumai 16 24 Jishi 02-1 851
Shi 4185 22 Hengguan 35 694 Zhengmai 366 23 Shaanyou 225 846
Yumai 47 21 Han 6172 693 Shaanyou 225 23 Xiaoyan 6 844
Xiaoyan 6 21 Zhongmai 895 684 Jinhe 9123 20 Jimai 22 839
Liangxing 66 20 Shaanyou 225 683 Zhongmai 895 19 Zhongmai 155 815
Xinong 979 20 Liangxing 66 674 Zhoumai 22 16 Han 6172 809
Zhongmai 895 19 Zhongmai 349 666 Zhongyu 5 16 Zhongmai 895 783
Yannong 19 19 Zhengmai 366 651 Zhongmai 349 16 Liangxing 66 777
Jimai 20 19 Yanzhan 4110 645 Taishan 23 16 Zhoumai 22 757
acids concentrations. Liangxing 66, Zhongmai Conclusions

349, and Zhongmai 895 had high concentrations Jimai 26, Henong 326, and Jingdong 8 displayed
of free and bound phenolic acids in 2008-2009, high Fe and Zn concentrations, among which
and Jimai 19, Liangxing 66, Shaanyou 225, Jingdong 8 is the most promising leading
Zhengmai 366, Zhongmai 895, and Zhoumai cultivar with high yield potential. Liangxing 66
22 had high concentrations of free and bound and Zhongmai 895 were stable across seasons
phenolic acids in 2009-2010. Liangxing 66, with high values in free and bound phenolics,
Taishan 23, Zhongmai 349, and Zhongmai 895 and could be used as parents in breeding for
had high concentrations of free phenolic acids in higher concentrations of health-beneficial
both seasons, whereas Liangxing 66, Han 6172, phenolic acids.
Shaanyou 225, Kaimai 18, Zhoumai 22, Zhongmai
895, and Zhengmai 366 had high levels of bound
phenolic acids concentrations in both seasons. References
It is likely that Liangxing 66 and Zhongmai 895 AACC (2000) Approved methods of the AACC. 10th ed.
display more consistent performance than the American Association of Cereal Chemists, St. Paul,
others for both free and bound phenolic acids. MN
This suggests that these cultivars with higher Adom KK, Sorrells ME, Liu RH (2003) Phytochemical
and more stable concentrations of phenolic acids profiles and antioxidant activity of wheat varieties. J
could be selected for production and for use as Agric & Food Chem 51:7825-7834
parents in breeding Adom KK, Sorrells ME, Liu RH (2005) Phytochemicals
and antioxidant activity of milled fractions of
different wheat varieties. J Agric & Food Chem
53:2297-2306
Bouis HE, Graham RD, Welch RM (2000) The
Consultative Group on International Agriculture
Research (CGIAR) Micronutrients Project: justification
and objectives. UNU Food & Nutrition Bulletin
21:374-381
Chen CM (2004) Ten Year Tracking of Nutritional Status
in China. People’s Medical Publishing House, Beijing
(in Chinese)

Jacobs DR, Meyer KA, Kushi LH, Folsom AR (1998) Ward JL, Poutanen K, Gebruers K, Piironen V, Lampi
Whole grain intake may reduce risk of coronary AM, Nyström L, Andersson AAM, Åman P, Boros
heart disease death in postmenopausal women: D, Rakszegie M, Bedõ Z, Shewry PR (2008) The
The Iowa Women’s Health Study. Am J Clin Nutr HEALTHGRAIN cereal diversity screen: concept,
68:248-257 results and prospects. J Agric & Food Chem 56:9699-
Li L, Shewry PR, Ward JL (2008) Phenolic acids in 9709
wheat varieties in the HEALTHGRAIN diversity World Health Organization (2006) Guidelines on food
screen. J Agri & Food Chem 56:9732-9739 fortification with micronutrients. World Health
Mattila P, Pihlava JM, Hellström J (2005) Contents of Organization, Geneva, Switzerland
phenolic acids, alkyl- and alkenylresorcinols, and Zhang Y, Wang DS, Zhang Y, He ZH (2007) Variation
avenanthramides in commercial grain products. J of major mineral elements concentration and their
Agric & Food Chem 53:8290-8295 relationships in grain of Chinese wheats. Scientia
Agric Sin 40:1871-1876 (in Chinese)
Zhuang QS (2003) Chinese Wheat Improvement and
Pedigree Analysis. China Agricultural Press, Beijing
(in Chinese)

Allergen potential of non-prolamin seed proteins in
Brachypodium distachyon
A. Juhász1, Gy. Gell1, E. Sebestyén1,2 , R. Haraszi3 , L. Tamás4, and E. Balázs1
1Applied Genomics Department, Agricultural Institute, Centre for Agricultural Research, HAS, Martonvásár,
Hungary; 2Universitat Pompeu Fabra, Dr. Aiguader 88, E08003 Barcelona, Spain; 3Institute for Reference
Materials and Measurements, European Commission Directorate General Joint Research Center, Geel, Belgium;
4Department of Plant Physiology and Molecular Plant Biology, Eötvös Loránd University, Budapest, Hungary
Abstract
Prolamin proteins are considered key players in the development of wheat-related health problems,
such as celiac disease or wheat allergy. However, there are further protein families that cause allergic
responses. The relatively small number of expressed wheat seed proteins reported in the protein
databases makes it difficult to get a complete overview of the allergome of wheat seed. Whole genome
sequencing projects carried out on cereals such as wheat and barley will facilitate an understanding
of the complex genetic effects of cereal allergies. Brachypodium distachyon serves as the evolutionary
closest sequenced genome for the Triticeae. To better understand the potential detrimental effects of
different seed proteins, an in-silico analysis was carried out using publicly available epitope databases
and Brachypodium distachyon genome sequence data. This study summarizes our results in identifying
potentially harmful non-prolamin proteins that may cause either celiac disease or wheat allergy-
related symptoms.
Introduction that have been proven to act as allergens

Wheat prolamins are considered to be key include wheat germ agglutinin, wheatwins
players in wheat-related health problems, (pathogenesis-related proteins of the PR-4
such as the autoimmune response in patients family, (Altenbach et al. 2007), triosephosphate
suffering from celiac disease (CD) or different isomerase, glycerinaldehyde-3-phosphate
categories of wheat allergies (WA) (e.g., Mittag dehydrogenases, acyl-CoA oxidase, fructose
et al. 2004.). Alpha and gamma gliadins are the bisphosphate aldolase, and triosephosphate
main protein families responsible for CD. These isomerase (Tatham and Shewry 2008).
proteins contain a number of T cell stimulatory The diversity and number of these proteins,
epitopes either in their repetitive regions (Shan as well as their effects compared to prolamin
et al. 2002) or elsewhere (Van de Wal et al. 1998). proteins, are not clear. The relatively small
Wheat allergies are an extremely diverse group number of expressed wheat seed proteins
of sensitivities with variable symptoms caused by found in the protein databases makes it difficult
different allergens. Omega gliadins are the main to get a complete overview of the allergen
contributors to symptoms of wheat dependent characteristics of wheat seed. Whole genome
exercise induced anaphylaxis. Sulphur rich sequencing projects carried out on cereals, such
prolamins such as alpha gliadins and LMW as wheat and barley will facilitate efforts to
glutenins also contribute to wheat allergies. Other reveal the complexity of genetic effects of cereal
members of the wheat prolamin superfamily, allergies. So far, Brachypodium distachyon serves
such as alpha-amylase inhibitors, or non-specific as the evolutionary closest sequenced genome
lipid transfer proteins, may contain epitopes to Triticeae (Vogel et al. 2010). Due to its low
that are related to allergy symptoms, including prolamin content, Brachypodium distachyon serves
baker’s asthma (Amano et al. 1998). The effects as a good platform to study allergen potential of
of allergens are not exclusively related to the non-prolamin proteins. In the present work, the
prolamin superfamily, although most of the Brachypodium distachyon genome sequence was
methods and clinical tests focus on identification utilized as a model to map allergen potential.
of proteins in this family. Wheat non-prolamins

Materials and methods and frequency of epitopes, possible antibody
The Immune Epitope Database (IEDB - www. responses and induced diseases were analyzed
immunoepitope.org) was used to retrieve and on both digested and non-digested proteins.
analyze epitope entries of the Pooideae. This
subfamily includes all species relevant in terms
of cereal-related food allergies and also the genus Results and discussion
Brachypodium. The epitope hits were filtered Five hundred and seventy-three linear epitopes
against Homo sapiens as host organism and related to non-pollen type wheat allergies and
epitopes reported to be pollen allergens were CD were collected from IEDB; mainly originating
excluded from the analyses. Only linear epitopes from Triticum aestivum. As a result of the
causing all kinds of wheat allergies and CD were conservancy analysis, 632 Brachypodium proteins
included in the study. The epitopes obtained were were found that contained at least one epitope
mapped to the Brachypodium protein sequence with 100% sequence homology. Sequences were
database v1.2 using the Epitope Conservancy compared to Brachypodium developing seed
Analysis tool at IEDB (Bui et al. 2007). Potential ESTs and rice homologs. From the 632 proteins,
seed allergens were identified using the tBlastN 206 potential allergens were identified as being
algorithm (Altschul et al. 1990) and Brachypodium expressed in seed tissues. The distribution of
developing seed ESTs and rice homologs. Trypsin, potential allergen proteins causing symptoms
pepsin (pH 1.3) and chymotrypsin were included related to CD and wheat allergies among the
simultaneously in the in-silico digestion analysis Brachypodium chromosomes showed that 81%
and epitopes that were resistant to the enzymatic of the identified allergen proteins possessed CD
cleavage were identified. Gene Ontology (GO) epitopes in their sequences. The majority of toxic
terms were assigned to individual proteins proteins, 33.5% of the possible CD proteins, were
using the Biological Networks Gene Ontology located on chromosome 1, followed by 23.4%
tool (BiNGO) and v1.0 Jan 2009 annotations of on chromosome 3 and 21% on chromosome 2.
the Brachypodium genome from the Gramene Potential allergen seed proteins were digested
database (http://www.brachypodium.org). The with trypsin, pepsin and chymotrypsin using
hypergeometric test and Benjamini & Hochberg the PeptideCutter tool in Expasy, and 82.6% of
False Discovery Rate (FDR) correction was used to the CD-related proteins were resistant to the
identify over-represented GO terms at the p = 0.05 digestion. Similarly, 63.83% of the wheat allergy-
significance level (Benjamini and Yekutieli 2001). related Brachypodium proteins remained intact
The Cytoscape platform was used to visualize the after enzyme digestion (Table 1). There were
results (Shannon et al. 2003). Variability in number six CD-related epitopes present in this set of
Table 1. Distribution of harmful proteins and epitopes encoded on different Brachypodium distachyon
chromosomes.
Chromosome 1 2 3 4 5
Celiac disease
No. of proteins with intact 49 21 33 24 11
epitopes when digested
Intact epitopes in B. distachyon QQQP PQQLPQ QQQP QQQP QQQP
seed proteins IPEQ QQQP IPEQ IPEQ
QQQQQQQQLQ IPEQ LQQQQQQQQQ IPEQ
LQQQQQQQQQ
Wheat allergy
No. of proteins with intact 5 12 6 4 3
epitopes when digested
Intact epitopes in B. distachyon AASVPE LQQQQQQQQQ GSQVPE LQQQQQQQQQ QQQPP
seed proteins QQQQQQQQLQ AASVPE QQQPP QQQPP QQPPQQ
QQQPP ADINNE QQPPQQ QQPGQ
QQPPQQ QQQPP
QQPPQQ

proteins; viz. FFQQ, IPEQ, LQQQQQQQQQ, alpha-amylase inhibitors, omega gliadins, high
PQQLPQ, QQQP, and QQQQQQQQLQ. The molecular weight (HMW) glutenins and low
epitopes were identified either in alpha-gliadins molecular weight (LMW) glutenins. Epitopes that
or annotated as peptides originating from gluten were resistant to digestion by the three enzymes
proteins. are presented in Table 1.
Forty-seven proteins that have wheat allergen To group the potential allergen proteins into
epitopes were identified in the analyzed protein biologically informative groups, their molecular
pool of B. distachyon. The frequencies of potential functions were characterized using sequence
allergens were highest for chromosomes 2 annotations and GO terms. The Biological
(42.6%) and 1 (23.4%). These proteins possessed Networks Gene Ontology tool (BiNGO) was
14 WA epitopes in their sequences. The epitopes used to find proteins with significantly over-
were originally identified in alpha gliadins, represented GO terms both in CD- and wheat
5.00E-2 <5.00E-7
A)
acyl-alkyI
acyl convertase
nucleic transferase
acid UDP-glycosyl
binding transferase transferase
protein glycogenin
binding glycosyl
binding glycosyl transferase
catalytic transferase
protein molecular activity
hexosyl glycosyl
dimerization function
transferase transferase
transcription nutrient
protein
regulator reservoir
heterodimerization
activity activity
B) transcription transcription
cofactor factor
binding protein
activity dimerization
protein
heterodimerization
transcription
initiation transcription protein
regulator binding unfolded
factor
activity protein
activity
heat binding
shock
molecular protein
function binding
nucleic
acid
binding DNA
binding
Fig. 1 Molecular functions of significantly over-represented proteins in potential allergens in Brachypodium

distachyon extracted without digestion. A, map of predominant molecular function terms in proteins
containing celiac disease-related epitopes; B, map of predominant molecular function terms in proteins
possessing wheat allergy-related epitopes. Node and font sizes are proportional to the number of genes
assigned to the same gene ontology (GO) term. Significantly over-represented GO terms are labeled with
gray nodes. Significant p-values are presented on a color scale.

allergy-related proteins (Fig. 1). Proteins identical to an epitope originally found in
containing CD epitopes fulfill molecular omega gliadins. This protein shared about 40%
functions such as binding (both protein sequence homology with several alpha gliadins.
and nucleic acid binding), catalytic activity, Another group of proteins is involved in protein
transcription regulator activity, nutrient reservoir dimerization and hetero-dimerization activities,
activity and enzyme regulator activity. Among or heat shock protein binding and unfolded
proteins with known molecular functions, only protein binding. The majority of potential wheat
6% of the CD-related proteins were assigned to allergens identified here have a role in nucleic
nutrient reservoir functions (Fig. 1). The number acid binding, transcription factor binding and
of potential wheat allergen proteins was 47, and transcription regulation, and many of them
10 proteins could not be assigned to GO terms. possess zinc–finger binding domains. There are
Like CD-related proteins, some of the proteins no studies yet reported on the immunoallergic
also have primary functions in nucleic acid characteristics of wheat proteins involved in
and protein binding and also in regulation of transcriptional and translational regulations.
transcription. This might be due to the relatively low number
of related proteins present in mature seeds
When potential allergens were assigned to gene
compared to some of the major allergens, such
ontology terms, one of the most remarkable
as gliadins. The common extraction techniques
results was the low level nutrient reservoir
used for two dimensional gel electrophoresis
activity. About 6% of the CD-related proteins
will likely fail when aiming for the detection of
belong to storage proteins whereas none of the
these proteins since they are low in abundance
WA proteins have storage activity. This is due
and often require specific extraction techniques.
to the low levels of contribution by prolamin
However, a predicted transcription factor with
proteins to allergen proteins found in this study.
zinc-ion binding, APFI, has been reported as
Apparently, prolamins are not major nutrient
IgE binding protein (De Angelis et al. 2007).
reserves in Brachypodium seed. The proteins
Similarly, proteins with zinc ion binding capacity,
associated with nutrient reserve functions in B.
such as plant homeodomain zinc-finger proteins
distachyon belong to avenin-like prolamins, or
have been reported as major allergens for asthma
predicted proteins, and no gliadin- or glutenin-
and eczema (Rahman et al. 2010).
like prolamins were identified. Most of the
proteins with known molecular functions Transcription factors with zinc-fingers, heat-
belong to binding proteins, either nucleic shock related proteins, and different signal-
acid binding proteins or proteins involved in recognition receptors and transporters were
protein binding. Proteins with functions such shown to be predominant in the soybean pollen
as protein dimerization, heterodimerization transcriptome (Haerizadeh et al. 2009). This
and transcription factor binding also belong in suggests that transcription factors possess some
this group of binding proteins. Proteins with peptides causing similar IgE responses to some
transcription regulator activity include bZIP prolamin proteins. It is not known whether these
transcription factors, transcriptional initiation transcription regulation-related proteins are
factors, transcriptional co-repressors, and present in mature Brachypodium seed.
proteins with zinc-finger domains. All these
proteins contain several 4-10 amino acid long
polyglutamine stretches in their sequences, and Conclusions
these might serve as toxic epitopes, in a similar High throughput genome sequencing is now one
way to prolamin epitopes. Enzyme regulators of the major methods in cereal allergy research to
include protein phosphatases and GTPases. determine potential allergen epitopes. Although a
Proteins with catalytic activity involve mainly number of epitopes may be false positives due to
acyl- and glycosyl-group transferases. the complexity of human responses to food intake,
The most striking difference between CD and the study presented here has several relevant
WA related proteins analyzed in this study outcomes. Our results emphasize the importance
was that there were no proteins assigned with of considering the allergen characteristics of
storage protein function among the WA proteins. non-prolamin seed proteins for identifying new
There was only one protein containing a peptide allergen protein families. The detected number of

non-prolamin type allergen proteins is expected to De Angelis M, Rizzello CG, Scala E, De Simone C, Farris
increase when extensive seed proteomic studies, GA, Turrini F, Gobbetti M (2007) Probiotic preparation
has the capacity to hydrolyze proteins responsible for
high throughput bioinformatics and computational
wheat allergy. J Food Prot 70:135-44
tools are used simultaneously. The pleiotropic
Haerizadeh F, Wong CHE, Bhalla PL, Gresshoff PM, Singh
effects of utilizing technologies such as RNAi for MB (2009) Genomic expression profiling of mature
tissue specific allergenic-gene knock outs could soybean (Glycine max) pollen. BMC Plant Biol 9:25
lead to compensation and over-expression of other Juhász A, Gell Gy, Sebestyén, E, Haraszi R, Tamás L,
proteins that may be high in glutamine and proline Balász E (2012) Brachypodium distachyon as a model
residues, and result in increased levels of allergic for defining the allergen potential of non-prolamin
response. These responses might be caused by proteins. Funct and Integr Genomics 12:439-446
protein families that normally do not show harmful Mittag D, Niggeman B, Sander I, Reese I, Fiedler EM,
effects. A more detailed analysis of the results Worm M, Vieths S, Reese G (2004) Immunoglobulin
E-reactivity of wheat-allergic subjects (baker`s asthma,
presented here is found in Juhász et al. (2012).
food allergy, wheat-dependent, exercise-induced
anaphylaxis) to wheat protein fractions with different
solubility and digestibility. Mol Nutr Food Res 48:380-
Acknowledgement 389
This project was supported by the Hungarian Rahman N, Stewart G, Jones G (2010) A role for the
National Research Fund, project number K100881. atopy-associated gene PHF11 in T-cell activation and
viability. Immunol Cell Biol 88:817-24
Shan L, Molberg O, Parrot I, Hausch F, Filiz F, Gray GM,
Sollid LM, Khosla C (2002) Structural basis for gluten
References intolerance in celiac sprue. Science 297:2275-2279
Altenbach SB, Kothari KM, Tanaka CK, Hurkman WJ Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT,
(2007) Genes encoding the PR-4 protein wheatwin Ramage D, Amin N (2003) Cytoscape: a software
are developmentally regulated in wheat grains and environment for integrated models of biomolecular
respond to high temperatures during grainfill. Plant Sci interaction networks. Genome Res 13:2498-2504
173:135-143
Tatham AS, Shewry PR (2008) Allergens in wheat and
Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ related cereals. Clin Exp Allergy 38:1712-1726
(1990) Basic local alignment search tool. J Mol Biol
215:403-10 Van de Wal Y, Kooy YMC, Van Veelen PA, Pena SA,
Mearin LM, Molberg O, Lundin KEA, Sollid LM,
Amano M, Ogawa H, Kojima K, Kamidaira T, Suetsugu Mutis T, Benckhuijsen WE, Drijfhout JW, Koning F
S, Yoshihama M, Satoh T, Samejima T, Matsumoto I (1998) Small intestinal T cells of celiac disease patients
(1998) Identification of the major allergens in wheat recognize a natural pepsin fragment of gliadin. Proc
flour responsible for baker’s asthma. Biochem J Natl Acad Sci USA 95:10050-10054
330:1229 -1234
Vogel JP, Garvin DF, Mockler TC, Schmutz J, Rokhsar D,
Benjamini Y, Yekutieli D (2001) The control of the false Bevan MW, et al. (The International Brachypodium
discovery rate in multiple testing under dependency. Initiative) (2010) Genome sequencing and analysis
Ann Statist 29:1165-1188 of the model grass Brachypodium distachyon. Nature
Bui HH, Sidney J, Li W, Fusseder N, Sette A (2007) 463:763-768
Development of an epitope conservancy analysis tool
to facilitate the design of epitope-based diagnostics and
vaccines. BMC Bioinformatics 26:8(1):361

Blood pressure lowering peptides from wheat gluten
X. Huang, P. Kanerva, H. Salovaara, T. Sontag-Strohm, and J. L. Oponen
Department of Food and Environmental Sciences, P.O. Box 66, FI-00014, University of Helsinki, Finland
Abstract
High blood pressure is an important health problem worldwide and it has associated risks of
cardiovascular and kidney disease. The renin-angiotensin system (RAS) is the most important system
that regulates blood pressure. Angiotensin-I converting enzyme (ACE) can stimulate high blood
pressure via the RAS. Obtaining ACE-inhibitory peptides from food sources is a cheap and efficient
approach to correct the problem. This work studied liberation of ACE-inhibitory peptides from wheat
gluten by enzymatic hydrolysis of thermolysin or/and prolyl endoprotease from Aspergillus niger
(Clarex), followed by 3 kDa cut-off ultrafiltration. The strongest ACE inhibition activity was obtained
when the sample was first hydrolyzed by thermolysin and then Clarex (IC50 value 0.016 mg protein/
ml). The ACE-inhibitory tripeptide leucine-glutamine-proline (LQP) was detected by a reverse-phase
liquid chromatography-mass spectrometry (RPLC-MS) system. Liberation of ACE-inhibitory peptides
by specific enzymes and isolation by membrane filtration techniques appeared a promising approach
to produce blood pressure lowering peptides from wheat gluten.
Introduction pressure. ACE-inhibitory peptides can be isolated

Hypertension, or high blood pressure, is from both food and non-food proteins. Most
one of the most significant chronic health ACE-inhibitory peptides studies are from milk
problems. It is highly frequent and associated proteins. But ACE-inhibitory peptides are also
with cardiovascular and kidney diseases found in animal sources, fish sources and plant
as well as brain health. Prolonged intake of sources.
antihypertensive drugs also causes adverse ACE-inhibitory peptides are usually very short
effects, including hypotension, cough, sequences with low molecular weight. Certain
fatigue and taste disturbances. Moreover, common structural characteristics for small ACE-
hypertension is one of the most expensive inhibitory peptides have been revealed. Firstly,
diseases to treat, due to the cost of drugs and proline, tyrosine or tryptophan commonly
treatment. Therefore, obtaining ACE-inhibitory exist at the C-terminals of peptides; secondly,
peptides from food protein would be a cheap hydrophobic amino acids with aliphatic sides,
and natural solution, and the development like isoleucine, leucine and valine, often exist
of food products containing Angiotensin-I at the amino-ends of ACE-inhibitory peptides;
converting enzyme (ACE)-inhibitory peptides thirdly, ACE-inhibitory tripeptides obeying the
are convenient for intake. ACE regulates blood two previous requirements frequently have a
pressure in the renin-angiotensin system (RAS). basic amino acid (lysine, arginine), or proline in
ACE is a peptidyl-dipeptidase hydrolyzing the middle positions. (De Leo et al. 2009).
Angiotensin I to Angiotensin II by removing
The aim of this work was to hydrolyze Vital
dipeptide histidine-leucine from C-terminal.
wheat gluten, a side product from the wheat
Angiotensin II is a potent vasoconstrictor
starch industry, by enzymes thermolysin and
and controls blood pressure. ACE hydrolyses
proline endopeptidase (Clarex), to analyze
bradykinin, a vasodilator, by cleaving dipeptide
hydrolysis characteristics and their in vitro ACE-
phenylalanine-arginine from the C-terminal to
inhibitory activities, and furthermore, to purify
inactive fragments. Inactive kinins are no longer
and identify targeted ACE-inhibitory tripeptides
able to cause blood vessels to enlarge; therefore,
from hydrolysate mixtures.
bradykinin has no blood pressure lowering
functions. Inhibition of ACE can lower the blood

Materials and methods IC50 value was determined by plotting ACE
inhibition activity percentage against logarithm
Enzyme hydrolysis of protein concentration of diluted hydrolysates.
Hydrolysate thermolysin (T): Vital wheat gluten IC50 value is the concentration of the sample
(Raisio Oy, Finland) was incubated in HEPES needed to inhibit 50% of ACE activity.
buffer (pH 8) with thermolysin (T7902 Sigma-
Aldrich). Reverse-phase liquid chromatography-mass
spectrometry (RPLC-MS)
Hydrolysate Clarex (C): Vital wheat gluten was
incubated in sodium acetate buffer (pH 4) with Identification of peptides was carried out by a
prolyl endopeptidase (Brewer’s Clarex, DSM Waters HPLC system and a photodiode array
Food Specialities). detector connected to an electrospray positive
mode Micromass ESCi system. Hydolysate T+C
Hydrolysate Clarex and thermolysin (C+T): were separated by C18 column (Waters XBridge
Clarex hydrolysis first and then thermolysin in Prep BEH130) and fractions were collected. For
conditions described above, pH was adjusted further separation, fractions were separated by
by 2M NaOH. C8 column (Discovery BIO wide pore, Sigma).
Hydrolysate thermolysin and Clarex (T+C): Solvent B was 0.1% formic acid in acetonitrile.
thermolysin hydrolysis first and then Clarex in Samples were eluted at a flow rate 3.5 ml/
conditions described above, pH was adjusted min with gradient 0-35% B in 125 min in a C18
by 2M HCl. column. For the C8 column, flow rate was 1.0
ml/min and gradient was 0-35% B in 64 min.
Size-exclusion chromatography (SEC) External standard tripeptide leucine-glutamine-
proline (LQP) was purchased from LifeTein
All hydrolysates were extracted with 1.5% SDS,
LLC, USA.
pH 6.9, 50 mM phosphate buffer at 50 °C for 1
hour. After centrifugation, the supernatant was
collected and mixed with running buffer (0.5%
SDS, 20% acetonitrile, pH 6.9, 50 mM phosphate
buffer). SEC columns Superdex Peptide Hydrolysis of Vital wheat gluten by thermolysin
100/300 GL and Superdex 200 10/300 GL (GE and/or Clarex caused breakdown of protein
Healthcare Biosciences AB, Uppsala, Sweden) (Fig. 1). Hydrolysate thermolysin contained
were coupled in series in a model 1200 HPLC more polypeptides than other hydrolysates.
system equipped with multiple wavelengths. Combination of two enzymes resulted in further
The injection volume was 100 μl, flow rate was degradation of gluten. After ultrafiltration
0.5 ml/min, and stopping time was 120 min. with a 3 kDa cut-off, hydrolysate C+T
contained the highest protein content (Table
Ultrafiltration 1), and hydrolysate T the least among all the
hydrolysates. However, hydrolysate T+C
Ultrafiltration was applied by Amicon Ultra-4
contained the strongest ACE inhibition activity;
Centrifugal Filter 3K (Millipore, Carrigtwohill,
followed by hydrolysate T. Thermolysin has
Co. Cork, Ireland). Maximum 4 ml was filled
relatively more cleavage sites than Clarex. The
and centrifuged at 20°C, 3,220 g for 30 min.
specificity of thermolysin is to cleave leucine,
In vitro ACE inhibition activity assay valine, alanine and phenylalanine residues at
the N-terminal. The resulting peptides after
Inhibition activity was analyzed using substrate thermolysin may have hydrophobic residues
FAPGG (N-[3-(2-Furyl)acryloyl]-Phe-Gly-Gly, at the N-terminals. When hydrolyzed by
F7131 Sigma). Diluted hydrolysate was pre- Clarex, the peptides contain proline residue at
incubated with ACE (A6778, Sigma) for 2 min the C-terminal. In hydrolysate C+T, peptides
at 37°C. The reaction was started by adding having proline residues were formed first.
FAPGG to the solution. A temperature of 37°C These peptides have 4-12 amino acids, and
was maintained for 10 min, and absorbance the glutamine residue is more often at the
was monitored during the last 5 min against N-terminal (Stepniak et al. 2006). After further
wavelength 340 nm (UV-1800, SHIMADZU). hydrolysis of thermolysin, some ACE-inhibitory

peptides can be formed, but not as efficiently selected because the sequence is highly repeated
as in T+C hydrolysis. Ultrafiltration is very in gluten, and LQP was found in wheat HMW,
important in this process, since bioactive peptides wheat LMW, α/β, γ, ω-gliadins (Loponen 2004).
are usually very short, containing 2-20 amino From the main peak of fraction 9, the mass
acids. The 3 kDa cut-off is around 20 amino
acids in size. Unhydrolyzed protein and big
polypeptides are excluded from the hydrolysate Table 1. Protein content of ultrafiltrated hydrolysates
mixture, while only small peptides remain. by Dumas method and Angiotensin-I converting
enzyme (ACE)-inhibition activity expressed by IC50
Because hydrolysate T+C showed the strongest value. Protein content was measured in triplicate. IC50
ACE inhibition activity, T+C was analyzed by value was calculated by software GraphPad Prism. C:
RPLC-MS (Fig. 2). The hydrolysate is a mixture Clarex hydrolysis, C+T: first Clarex then thermolysin
of different peptides, and it was difficult to hydrolysis, T: thermolysin hydrolysis, T+C: first
detect any hydrolysis products. LQP was used thermolysin followed by Clarex hydrolysis.
as an external standard in a C18 column, and
it eluted at fraction 9, which was collected for Protein content IC50 value
further separation (data not shown). LQP is a Hydrolysate (mg/ml) (mg/ml)
known ACE-inhibitory peptide with an IC50 C 4.98±0.27 0.070
value 1.9 μM (Miyoshi et al. 1991). LQP was T 4.11±0.24 0.029
C+T 6.50±0.10 0.040
T+C 4.42±0.13 0.016
Polymeric Monomeric
Proteins Proteins Polypeptides Small peptides and free amino acids
1,740
Control
1,540 C
C+t
1,340
Absorbnce 210 nm (mAU)
T
1,140 T+C
940
740
540
340
140
10 15 20 25 30 35 40 45 50
Elution Volumn (mL)
100 kd 15kd 6.5kd 1.4kd
Fig. 1 Size-Exclusion Chromatography (SEC) analysis of each hydrolysate. Control: Vital wheat
gluten. C: Clarex hydrolysis. C+T: first Clarex then thermolysin hydrolysis. T: thermolysin
hydrolysis. T+C: first thermolysin followed by Clarex hydrolysis. Polymeric and high molecular
weight proteins ≥100 kDa, monomeric protein 100 kDa - 15 kDa, polypeptides 15 kDa - 1.4 kDa,
peptides and amino acids <1.4 kDa. Molecular mass standards: 0.5 mM aprotintin (6.5 kDa), 6.5
mM cyanocobalamin (1.4 kDa), and 1.5 M glycine (75 kDa). (Molecular mass standards from
Loponen et al. 2009).

to charge ratio 357.2 was detected, which is peptides may be liberated after hydrolysis; for
LQP with positive charge (Fig. 3). Leucine and example, LPP, IAP, PYP, FP, and YP. These
isoleucine have the same molecular mass, and sequences are found in gluten, and it is possible
this detected peptide could be IQP. However, to liberate these sequences by hydrolysis of
IQP sequence is much less found in gluten; it is thermolysin and Clarex.
only in γ-gliadin (BIOPEP). Other ACE-inhibitory
f2 f3
6,0E+08
f5
Total current (TIC) intensity
f6
f1
f8
f7
4,0E+08 f11
f4 f12
f9 f10
2,0E+08
10 15 20 25 30 35 40 45 50 55 60 65 70 75 80
Retention time (min)
Fig. 2 Mass chromatogram of hydrolysate T+C (first thermolysin followed by Clarex) after C18
column separation.
242.1
100
[M+H]+
%
367.2
[2M+H]+
243.2
158.0
225.1 368.2
214.0 244.0
0
150 200 250 300 350 400 450 500 550 600 650 700 750 800 880 900 950 1,000
Fig. 3 Mass spectrum of the main peak in fraction 9 separated by C8 column from hydrolysate T+C
(first thermolysin followed by Clarex).

Vital wheat gluten degraded by enzymatic De Leo F, Panarese S, Gallerani R, Ceci LR (2009)
hydrolysis was shown to produce ACE- Angiotensin converting enzyme (ACE) inhibitory
peptides: production and implementation of
inhibitory peptides. Combined with functional food. Curr Pharm Des 15:3622-3643
ultrafiltration, enzyme hydrolysis by
Loponen J (2004) Angiotensin converting enzyme
thermolysin followed by Clarex showed the inhibitory peptides in Finnish cereals: a database
most potent ACE-inhibitory activity in in vitro survey. Agric & Food Sci in Finland 13:39-45
measurement. An ACE-inhibitory tripeptide Loponen J, Kanerva P, Zhang C, Sontag-Strohm T,
leucine-glutamine-proline that is frequently Salovaara H, Gänzle MG (2009) Prolamin hydrolysis
encrypted in gluten proteins was identified. and pentosan solubilization in germinated-rye
Targeted hydrolysis of gluten can liberate sourdoughs determined by chromatographic and
immunological methods. J Agric & Food Chem 57:746-
known tripeptides with ACE-inhibitory activity.
753
Miyoshi S, Ishikawa H, Kaneko T, Fukui F, Tanaka
H, Maruyama S (1991) Structures and activity of
angiotensin-converting enyzme inhibitors in an alpha-
zein hydrolysate. Agri Biol Chem 55:1313-8
Stepniak D, Spaenij-Dekking L, Mitea C, Moester M, De
Ru A, Baak-Pablo R, Van Veelen P, Edens L, Koning
F (2006) Highly efficient gluten degradation with a
newly identified prolyl endoprotease: implications
for celiac disease. Amer J Physiol - Gastrointestinal &
Liver Physiology 291:G621-G629

How reliable are the measurements of residual gluten
in gluten-free foods?
P. Kanerva, T. Sontag-Strohm, and H. Salovaara
Department of Food and Environmental Sciences, University of Helsinki, Finland
Abstract
Enzyme-linked immunosorbent assays (ELISA) are used to quantify proteins that are harmful
for people with coeliac disease, i.e., wheat, barley and rye prolamins, and to control the safety of
gluten-free foods. Prolamins consist of a complex mixture of proteins that set high requirements for
quantification. Modification of proteins further increases their diversity, and risks the reliability of
analysis results. The reactivity of prolamins with different prolamin-specific antibodies was examined
and clear differences between the recognition of antibodies of prolamin subgroups were shown
by western blotting. The results of ELISA demonstrated that barley prolamins are not accurately
quantified by R5 antibody. However, with a barley prolamin standard, more precise results were
obtained. Further studies were carried out to study how enzymatic or chemical modification affects
the measurement of prolamin contents. Both sandwich and competitive R5 ELISAs detected residual
rye prolamins after enzymatic hydrolysis in sourdough systems with comparable results. However,
after deamidation by acid treatment ELISAs based on R5 and ω-gliadin antibodies failed to measure
accurately wheat prolamins. The results demonstrate that current gluten analysis methods cannot
accurately quantify prolamins in all food matrices and more work is needed to improve the reliability
of gluten quantification.
Introduction is based on an R5 antibody, which was raised

Coeliac disease is one of the most common food against the prolamin fraction of rye (Valdés et
intolerances in the world. Current estimates of al. 2003). The antibody recognizes wheat, rye
its prevalence are between 1 and 2% (Mustalahti and barley prolamins reacting mainly with an
et al. 2010). Prolamins of wheat, barley and rye epitope QQPFP (glutamine-glutamine-proline-
are known to be an environmental trigger of phenylalanine-proline). The antibody against
coeliac disease, and when combined with certain ω-gliadin was widely used before the R5
genetic predispositions the disease may develop. antibody, and is still used in many commercial
Ingestion of these proteins by susceptible people assays (Skerritt and Hill 1990). The ω-gliadin
leads to an inflammation in the small intestine antibody especially recognizes heat-stable
and eventually to destruction of the villi. This ω-type prolamins from wheat and rye; however,
causes abdominal pain and discomfort, and it has a low reactivity with barley prolamins,
disrupts absorption of nutrients. Currently, the which limits its use. More recently, antibodies
only treatment for coeliac disease is a diet free have been raised against specific sequences in
from prolamins, i.e., a gluten-free diet. prolamins that were shown to elicit symptoms
in coeliac patients, or to activate T-cells. These
Immunological ELISA methods are used to include an antibody raised against T-cell
quantify the harmful proteins in gluten-free activating epitope present in α-gliadin (Spaenij-
foods. Methods are based on prolamin-detecting Dekking et al. 2004), an antibody PN3 raised
antibodies. Several antibodies against different against in vivo toxic 19mer (Ellis et al. 1998),
prolamin subfractions have been developed, and antibodies A1 and G12 raised against
and some of them are used in commercially 33mer, which contains three overlapping T-cell
available prolamin analysis assays. Currently activating epitopes (Morón et al. 2008). All
Codex Alimentarius has endorsed the method of these antibodies are used in commercially
of Mendez (R5 ELISA) for quantitation of available assays for prolamin quantification.
prolamin content in gluten-free foods. The assay

Modification of prolamins in order to increase Materials and methods
their applicability in foods further increases
Preparation of oat samples contaminated by
their diversity. Modification often changes the
barley flour
protein surface, which also affects the antibody
recognition and, therefore, puts the reliability of Samples with known amounts of contamination
prolamin analysis at risk. Proteins are modified were prepared by mixing barley flour with
to increase their solubility or foam and emulsion pure oat flour (Kanerva et al. 2006). The amount
forming ability. Wheat gluten is a good candidate of hordein was decreased by mixing more
as it is a vegetable protein with a relatively low pure oat flour to the samples. The samples
price. Modification of prolamins may, however, were extracted with 60% v/v ethanol or with
turn out to be problematic for people on a gluten- a cocktail solution (Gluten Unity, Spain). The
free diet, since gluten proteins can be found ethanol extraction was performed at 40˚C for 1
in unexpected sources, such as meat, fish or h, and the samples were centrifuged for 10 min
milk products. Numerous techniques such as at 2500 g. Analysis was performed within 24 h
deamidation and hydrolysis have been developed of extraction. The cocktail solution containing
to modify proteins such that they have more 250 mM 2-mercaptoethanol and 2 M guanidine
desired characteristics. Deamidation increases hydrochloride was used for the extraction of
the solubility by increasing the negative charge prolamins according to the manufacturer’s
of proteins when an amide group of glutamine recommendations.
or asparagine is replaced by a carboxyl group.
The hordein standard was extracted from
Deamidation also changes the immunological
barley flour with 60% v/v ethanol following
behavior of proteins, and the deamidation of
the same protocol as with oat samples. The
prolamin peptides by a tissue transglutaminase
protein content of the extract was analyzed by
is described to be an important part of the
the DC Protein Assay Kit (BioRad Laboratories,
pathogenesis of coeliac disease (Molberg et
USA), which is based on the Lowry method. A
al. 1998). It is not precisely known whether
γ-globulin was used as a standard.
industrial deamidation increases or decreases the
harmfulness of prolamins to people with gluten Preparation of deamidated gluten
intolerance, but either way, deamidation influences
their detection by antibody-based immunological Deamidated wheat gluten was prepared using
assays. Enzymatic hydrolysis, on the other hand, a modified procedure of Ma et al. (1986).
results in peptides with smaller size and hence Preparation was started by adding 200 ml of
improved solubility. Enzymatic hydrolysis also 0.1 M HCl to 5 g of the vital gluten followed by
has the potential to decrease the harmfulness of heating the suspension at 100°C for 2 h. After
prolamins if the process is carried far enough and the treatment, the suspension was neutralized
proteins are degraded into small peptides that no with NaOH, dialyzed against distilled water
longer have immunological activity. Because of the (MWCO 12-14 kD) and the contents were
tight structure of prolamins caused by their high lyophilized.
proline content, the enzymes of the gastrointestinal
system are incapable of efficiently hydrolyzing Electrophoresis and western blotting
prolamins (Shan et al. 2005). However, cereal The prolamin compositions and cross-reactivity
grains themselves contain enzymes that are able of antibodies with prolamin groups were
to degrade prolamins under optimal conditions evaluated with sodium dodecyl sulphate –
(Loponen et al. 2004). The use of these enzymes polyacrylamide gel electrophoresis (SDS-PAGE)
was suggested as an oral therapy, or they could and western blotting. Sequential extraction
offer a tool for developing new gluten-free cereal- following a modified Osborne fractionation was
based products. performed. Wheat, barley and rye flour samples
were first extracted three times with distilled
This study concentrated on the behavior of
water and then three times with 0.5 M NaCl at
prolamins in immunological gluten assays and
room temperature. Then the precipitates were
with different prolamin-specific antibodies.
washed briefly with water before extraction
Further studies were carried out to determine
three times with 40% 1-propanol at 50ºC, three
how enzymatic or chemical modification affects
times with 40% 1-propanol and 1% dithiothreitol
measurement of prolamin contents.

(DTT) at 50ºC and three times with SDS sample Enzyme-linked immunosorbent assays (ELISA)
buffer (4% (w/v) SDS, 20% (v/v) glycerol, 125 mM Sandwich ELISA based on the R5 antibody
tris(hydroxymethyl)aminomethane, 1% (w/v) (Diffchamb, France) was used to detect barley
dithiothreitol, bromophenol blue, pH 8.5). contamination in oat samples. The method was
The first extracts of 1-propanol, 1-propanol with performed according to the manufacturer with a
DTT and SDS sample buffer, were concentrated small modification in ethanol extraction.
by allowing 100 μl of extract to evaporate at room Competitive and sandwich ELISAs based on R5
temperature. The dried samples were solubilized antibody (R-Biopharm, Germany) were used
in 40 μl of the SDS sample buffer. Proteins were to detect residual rye prolamins in sourdough
separated in NuPage gradient Bis-Tris gel (4-12%) systems after enzymatic hydrolysis. ELISAs
(Invitrogen, USA). NP MES SDS running buffer based on R5 (R-Biopharm, Germany) and
was used with NuPAGE antioxidant (Invitrogen, ω-gliadin antibody (Tepnel Biosystems Ltd.,
USA). The run was done at 150 V and the running England) were used to measure wheat prolamins
time was 70 min. A Novex Sharp Pre-stained after deamidation by acid. The samples were
Standard was used as a molecular marker extracted and the method performed according
(Invitrogen, USA). After electrophoretic separation, to the manufacturer’s instructions.
the proteins were transferred to polyvinylidene
fluoride (PVDF) membranes (BioRad
Laboratories, USA) in 25 mM tris(hydroxymethyl) Results and discussion
aminomethane, 192 mM glycine and 20% (v/v) The ω-gliadin antibody and the R5 antibody
methanol. The transfer was performed at 30 V showed clear differences in reactivity with
for 1 h. After blocking with 1% (w/v) bovine prolamin subfractions of wheat, barley and rye
serum albumin, the membrane was incubated in (Fig. 1). The ω-gliadin antibody reacted mainly
solution of either a conjugated ω-gliadin antibody with prolamins of higher molecular weight than
(Diffchamb, France) or a conjugated antibody R5 the R5 antibody. Western blot analyses showed
(R-Biopharm, Germany). The antibodies were that the R5 antibody was able to react with low
diluted to 1/250 and 1/22, respectively, with the molecular weight (LMW) glutenin subunits
dilution buffers provided by the assays. Sodium (lane 2) in addition to the gliadins (lane 1). The
azide and bovine serum albumin were added to R5 antibody also recognized all rye prolamin
the buffers. After incubation, the membranes were subfractions. These results partly explain the high
washed and stained with tetramethylbenzidine variance in prolamin measurements. Differences
(Promega, USA).
WHEAT BARLEY RYE
1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
Fig. 1 Western blot (ω, R5) and SDS-PAGE (gel) analysis of 40% 1-propanol (1), 40% 1-propanol with 1%
dithiotreitol (2) and the SDS sample buffer (3) extracts of wheat, barley and rye. ω and R5 are western
blots by the ω-gliadin antibody and the R5 antibody, respectively. The prolamin subgroups are marked
on the right side of the gel (reproduced from Kanerva et al. 2011b).

in relative proportions of prolamin subgroups in been shown that barley is as common as wheat
samples will evidently lead to different results as a contaminant in oat based products (Gelinas
with different antibodies. et al. 2008), thus also requiring accuracy in the
quantification of barley prolamins.
The results with oat samples deliberately
contaminated by barley demonstrate that the The results showed substantial degradation
currently used gluten analysis methods are not of rye prolamins in an enzyme-active rye-
able to quantify barley prolamins accurately malt sourdough system. The prolamin levels
when mixed in oats. The gliadin standard used were lowered by 99.5% from the original
in R5 ELISA gave much higher levels for gluten levels (Loponen et al. 2009). Such extensive
content than estimated based on the amount degradation of rye prolamins suggest the use
of added barley flour. However, more precise of sourdough as part of gluten-free baking.
results were obtained when the hordein standard Although, the competitive technique is
was used (Fig. 2). This shows that wheat and recommended for samples containing small
barley prolamins are so different from each other hydrolyzed peptides, consistent results were
that they cannot be quantified based on the same obtained with the sandwich assay.
reference material. Better results were obtained
According to our results, ELISA based on the
when each was quantified by its own standard,
ω-gliadin antibody was not able to quantify
although this approach can be quite impractical
wheat gluten after deamidation in acidic
with varying food samples.
conditions. In addition, competitive and
Wheat is a more common cereal than barley sandwich R5 ELISAs failed to detect deamidated
or rye, and most of the ELISAs that are used gluten proteins except at much higher
today for prolamin quantification are optimized concentrations (Fig. 3). In practice, this means
on antibody recognition of wheat gliadin. that the results obtained by these ELISA assays
However, since prolamins of barley and rye are would be 125 to 600 times smaller than the true
also considered harmful for people with coeliac gluten contents. Consequently, deamidated
disease, they should get further attention to gluten peptides can exist in food products and
provide more accurate results for gluten-free remain undetected, and thus cause a risk for
products that contain barley contamination. It has people with coeliac disease.
1,800 600
Gliadin standard Hordein standard
1,500
500
1,200
400
ppm of hordein
ppm of gluten
900
308
300
500
400 200
300
200 100 77
100 32 63 38
8 16 10
0 0
0 0.025 0.05 0.1 0.2 0 0.03 0.13 0.26 1
Amount of barley (%) Amount of barley (%)
Fig. 2 Gluten contents of barley-contaminated oat samples when detected by the sandwich R5 ELISA
method using either gliadin or hordein standards. The dark lines show the level of hordein added to the
samples. Error bars indicate standard deviations (reproduced from Kanerva et al. 2006).

R5 competitive ELISA R5 sandwich ELISA
2.4 2
2.2 1.8
2.0 1.6
1.8 DA gluten
1.4
1.6 X 600
1.4 X 125 1.2
OD
OD
1.2 1
1.0 Standar
Vital gluten 0.8
0.8 Vital gluten DA gluten
0.6
0.6 Gliadin standar
0.4 0.4
0.2 0.2
0.0 0
0.00 0.01 0.10 1.00 10.0 100.0 1 10 100 1000 10000 100000
Protein content (g/ml) Protein content (ng/ml)
Fig. 3 Reactivity of gluten before (vital gluten) and after acid deamidation (DA gluten) by competitive and
sandwich R5 ELISAs (reproduced from Kanerva et al. 2011a).
Conclusions Loponen J, Kanerva P, Zhang C, Sontag-Strohm T, Salovaara

H, Gänzle M (2009) Prolamin hydrolysis and pentosan
The results demonstrate that current gluten solubilization in germinated-rye sourdoughs determined
analysis methods cannot accurately quantify by chromatographic and immunological methods. J
prolamins in all food matrices. The reliability of Agric Food Chem 57:746-753
the measurements is questioned if the complexity Ma C-Y, Oomah BD, Holme J (1986) Effect of deamidation
of prolamins and changes in their structures and succinylation on some physicochemical and baking
properties of gluten. J Food Sci 51:99-103
are not carefully considered. Reliability can be
improved substantially by selecting the right Molberg Ø, Macadam SN, Körner R, Quarsten H,
Kristiansen C, Madsen L, Fugger L, Scott H, Norén O,
reference materials and taking differences Roepstorff P, Lundin KEA, Sjöström H, Sollid LM (1998)
between the reactivity of antibodies into account. Tissue transglutaminase selectively modifies gliadin
In the future, it is essential to improve gluten peptides that are recognized by gut-derived T cells in
analysis methods so that they can be applied celiac disease. Nat Med 4:713-717
reliably to measure all harmful prolamins, Morón B, Cebolla A, Manyani H, Álvarez-Maqueda M,
including modified ones, in gluten-free foods. Megías M, del Carmen TM, López MC, Sousa C (2008b)
Sensitive detection of cereal fractions that are toxic to
celiac disease patients by using monoclonal antibodies
to a main immunogenic wheat peptide. Am J Clin Nutr
References 87:405-414
Ellis HJ, Rosen-Bronson S, O’Reilly N, Ciclitira PJ (1998).
Mustalahti K, Catassi C, Reunanen A, Fabiani E, Heier M,
Measurement of gluten using a monoclonal antibody McMillan S, Murray L, Metzger M-H, Gasparin M, Bravi
to a coeliac toxic peptide of A gliadin. Gut 43:190-195 E, Mäki M and the members of The Coeliac EU Cluster,
Gélinas P, McKinnon CM, Mena MC, Méndez E (2008) Epidemiology (2010) The prevalence of celiac disease
Gluten contamination of cereal foods in Canada. Int J in Europe: Results of a centralized, international mass
Food Sci Tech 43:1245-1252 screening project. Ann Med 42:587-595
Kanerva P, Sontag-Strohm T, Ryöppy P, Alho-Lehto P, Shan L, Qiao S-W, Arentz-Hansen H, Molberg Ø, Gray GM,
Salovaara H (2006) Analysis of barley contamination Sollid LM, Khosla C (2005) Identification and analysis
in oats using R5 and ω-gliadin antibodies. J Cereal Sci of multivalent proteolytically resistant peptides from
44:347-352 gluten: implications for celiac sprue. J Proteome Res
Kanerva P, Brinck O, Sontag-Strohm, T, Salovaara H, 4:1732-1741
Loponen J (2011a) Deamidation of gluten proteins Skerritt JH, Hill AS (1990) Monoclonal antibody sandwich
and peptides decreases the antibody affinity in gluten enzyme immunoassays for determination of gluten in
analysis assays. J Cereal Sci 53:335-339 foods. J Agric Food Chem 38:1771-1778
Kanerva P, Sontag-Strohm T, Brinck O, Salovaara H Spaenij-Dekking EHA, Kooy-Winkelaar EMC, Nieuwenhuizen
(2011b) Improved extraction of the prolamins for WF, Drijfhout JW, Koning F (2004) A novel and sensitive
gluten detection from processed foods. Agric Food Sci assay for the detection of T cell stimulatory epitopes of
20:206-216 α/β- and γ-gliadin. Gut 53:1267-1273
Loponen J, Mikola M, Katina K, Sontag-Strohm T, Valdés I, García E, Llorente M, Méndez E (2003) Innovative
Salovaara H (2004) Degradation of HMW glutenins approach to low-level gluten determination in foods
during wheat sourdough fermentations. Cereal Chem using a novel sandwich enzyme-linked immunosorbent
81:87-93 assay protocol. Eur J Gastroenterol Hepatol 15:465-474

Effects of HMW- & LMW-glutenins and grain hardness on
size of gluten polymers
V. S. Lesage1,2, L. Rhazi3, T. Aussenac3, B. Meleard4, and G. Branlard1,2
1INRA, UMR 1095 GDEC, F-63039 Clermont-Ferrand, France; 2Université Blaise Pascal, UMR 1095 GDEC,
F-63177 Aubière, France; 3Institut Polytechnique LaSalle Beauvais, F-60026 Beauvais, France; 4ARVALIS -
Institut du Végétal, F- 91720 Boigneville, France
Abstract
Bread quality depends on various traits of wheat flour including storage protein composition, grain
hardness and size of gluten polymers. To better understand relationships between these traits, a
statistical study was performed on a set of French varieties cultivated in 3 locations during 2 years.
Variation in grain hardness was primarily associated with SNPs within the PinB gene and not in
PinA (no PinA variants were detected), and also with the HMW-Glu-D1 locus. The mass of storage
protein polymers varied from 5 to 49 million Daltons, as evaluated by Asymmetric Flow Field Flow
Fractionation. Polymer mass and its heterogeneity were mainly influenced by the LMW-Glu-D3
locus. Percentages of storage protein fractions, evaluated by SE-HPLC, were significantly affected
by the LMW-Glu-B3, -A3 and HMW-Glu-D1 loci. Interactions between glutenins and puroindoline-B
alleles significantly influenced polymer characteristics and percentages of ω-gliadins. These effects
were variable depending on glutenin and puroindoline alleles. Environmental effects on flour quality
traits were also studied. The sum of mean June and July temperatures, the two last months of grain
development, was significantly correlated with polymer characteristics but not with storage protein
proportions. Moreover, temperatures during the last 2 months of grain development had variable effects
on the characteristics of gluten polymers depending upon grain hardness. This study is the first report
of interactions between glutenins and puroindoline alleles affecting storage protein polymer traits.

Previous studies on wheat near-isogenic lines A total of 68 bread wheat varieties were
showed that kernel hardness affects the molecular cultivated two successive years (2009, 2010) in 3
mass of storage protein polymers (Lesage et al. locations in France. In each location, 40 varieties
2011). The amount of unextracted polymeric were grown in two replicates in conventional
proteins (UPP) is well known to have a major conditions with full fungicide protection.
influence on dough properties. Many studies have
Grain hardness (GH) was assessed by Near-
developed approaches to assess the polymeric size
Infrared Spectrometry (NIRS) on whole meal
of storage proteins using SE-HPLC (Dachkevitch
flour according to the AACC 39-70A method.
and Autran 1989). The Asymmetrical Flow
Three classes of grain hardness were recorded
Field-Flow Fractionation (AFFFF) technique is
according to NIRS scores: Soft varieties below
a tool able to assess the size diversity of gluten
35, Medium varieties between 35 and 75, and
polymers. Only a few studies have reported
Hard varieties above 75. Protein content (PC)
variation in wheat storage proteins polymers
was measured by NIRS according to the AACC
using AFFFF (Stevenson and Preston 1996)
39-11.01 method. High molecular weight
and their influence on rheological properties of
(HMW)- and low molecular weight (LMW)-
dough (strength, elasticity and extensibility). To
glutenin alleles were determined by SDS-PAGE
assess effects of both grain hardness and glutenin
electrophoresis. Puroindoline A and B genes
alleles on storage protein polymers, genetic
(PinA and PinB) were sequenced by the Sanger
diversity of various traits (grain hardness, protein
method. Storage protein polymers characteristics
content, glutenin alleles, polymer characteristics
(mass (Mw2), polydispersity index (Mw/Mn2),
and percentages of five protein fractions) was
polymer radius (Rw2)) were evaluated by
investigated in a set of French cultivars.

Asymmetrical Flow Field-Flow Fractionation radius, Rw2) were mainly influenced by the
(AFFFF) described by Chiaramonte and Glu-D3 locus, and to a lesser extent by Glu-B1
colleagues (Chiaramonte et al. 2012). Percentage (Table 1).
of five protein fractions (%F1 to %F5) was
As shown by R² values, the percentages of SE-
obtained by Size Exclusion-HPLC (SE-HPLC) on
HPLC protein fractions reflected the diversity
a TSK G4000-SW column (Tosohaas). Statistical
of glutenin loci much more than polymer
analyses (General Linear Model and Partial
characteristics. It is known that fractions F1 to
Least-Squares procedures) were performed using
F5 contain different protein classes (F1: HMW
Statgraphics software (Statpoint Technologies,
glutenin polymers, F2: LMW glutenin polymers,
Inc., VA, USA). These analyses allowed
F3: ω-gliadins, F4: γ and β gliadins, F5:
assessment of the influence of glutenin loci and
α-gliadins and albumins-globulins). The Glu-B3
glutenin alleles, as well as puroindoline-B alleles,
locus had a significant effect on SE-HPLC
on polymer characteristics and on SE-HPLC
percentages of all protein fractions. The Glu-A3
fractions. Environmental effects on polymer size
locus impacted %F1, %F3 and %F4, whereas the
were also studied.
Glu-D1 locus significantly influenced %F1, %F2
and, to a lesser extent, %F3.
Results and discussion Glu-D1 also had a significant effect on grain
1. Large variations in polymer characteristics hardness (Table 1), which was significantly
were measured using AFFFF higher in varieties carrying Glu-D1 5+10 alleles
than in those carrying Glu-D1 2+12 (Fig. 1).
Over the 240 samples of bread wheat, a very large
variation of polymer characteristics was observed.
For example, the mass (Mw2) varied from 5.4 84
million to 48.8 million Daltons. Polymer radius
(Rw2) varied from 36.5 to 116.2 nm. On the same
samples, polymer mass obtained by SE-HPLC 74
varied from 0.75 million to 2 million Daltons.
G.H.
Taking into account that polymer mass measured 64

by AFFFF and SE-HPLC F1 and F2 fractions were
not correlated, these results indicate that SE-HPLC 54
results do not represent the entire polymer mass
variation in storage proteins.
44
2. Glutenin loci affect differently the polymer 2+12 3+12 4+12 5+10
characteristics and SE-HPLC fractions Glu-D1
Polymer characteristics (molecular mass, Mw2; Fig. 1 Relationship between Glu-D1 alleles and
polydispersity index, Mw/Mn2; and polymer grain hardness.
Table 1. Effects of glutenin loci on some quality traits.
Protein Grain
content hardness Mw2 Mw/Mr2 Rw2 %F1 &F2 &F3 &F4 %F5
Glu-A1 ns ns ns ns ns ns ns ns ns ns
HMW-GS Glu-b1 ns ns * ** * * * ns ns ns
Glu-D1 ns *** ns ns ns *** *** * ns ns
Glu-A3 ns * ns ** ns *** ns ** *** ns
LMW-GS Glu-B3 ns ns ns ns ns *** *** *** *** **
Glu-D3 ns ns *** *** *** ns ns ns ns *
R2 % 0.2 19.9 9 15.9 11.2 47.1 58.7 16.9 45.3 10.5
HMW-GS = high molecular weight glutenin subunits; LMW-GS = low molecular weight glutenin subunits; ns = not significant;
MW2 = molecular mass; Mw/Mn2 = polydispersity index; Rw2 = polymer radius.

This observation indicated a strong relationship fifth allele, PinB-D1c, contained a SNP (T266C)
between Glu-D1 alleles and grain hardness, both leading to a L89P substitution from the starting
of which are important characteristics for wheat methionine. PinB-D1b, -D1d and -D1c were
end use quality. already described in the literature, whereas the
PinB-D1b+d allele is new.
3. Effects of glutenin alleles on polymer mass
5. Interaction between glutenin and
Partial Least Square analysis revealed that
puroindoline B loci
glutenin alleles had different effects on polymer
mass, some being positive, and others negative Interaction effects were calculated for glutenin
(Fig. 2). Glu-B1-7, usually associated with poor loci and 3 alleles of PinB (PinB-D1a, -D1b
bread quality, had a much more positive effect and –D1d); the frequencies of alleles PinB-
on final mass of polymer than other alleles. The D1c and -D1b+d were too low for meaningful
two Glu-D3 alleles showed opposite effects on comparisons.
polymer mass, Glu-D3-b being positive and Glu-
Grain hardness was obviously highly
D3-c negative. Glu-B3-b also had a negative effect
explained by the interaction of the glutenin and
on polymer mass Mw2.
puroindoline-B loci, except, strikingly, for the
4. Characterization of puroindoline-B alleles interaction between Glu-A3 and PinB (Table 3).
Nevertheless, explanation of the trait by the
PinA and PinB are major genes of the Ha locus
involved in grain hardness. They were sequenced
in all varieties. The PinA-D1a allele was present Table 2. Amino acid sequence variation of
in all varieties and no SNP or deletions were puroindoline-B in 68 French bread wheat varieties.
found within its sequence. PinB sequences
comprised five alleles, the first one, Pinb-D1a, Puroindoline variant Partial amino acid sequence
being the PinB allele of Soft varieties. Three Position: 73-75--------------------89
alleles, PinB-D1b, -D1d and -D1b+d, carried SNPs PINB-D1a WPTKWWKGGC ----------- L
at positions T217A, G223A, T217A + G223A from PINB-D1b WPTKWWKSGC ----------- L
the start codon, respectively. These SNPs led to PINB-D1d WPTKWRKGGC ----------- L
the following modifications in the sequence of PINB-D1c WPTKWWKGGC ----------- P
puroindoline B: W73R, G75S and W73R + G75S PINB-D1b + d WPTKWRKSGC ----------- L
from the starting methionine (Table 2). The
Mw2
Glu-A1 Glu-B1 Glu-D1 Glu-A3 Glu-B3 Glu-D3
0.24
B1-7
0.14 D3-b
Standardized coef.
0.04
-0.06
-0.16
GluA1-n
GluA1-1
GluA1-2*
GluB1-68
GluB1-7
GluB1-78
GluB1-79
GluB1-1316
GluB1-1415
GluB1-1718
GluB1-6.1-22
GluD1-212
GluD1-312
GluD1-412
GluD1-510
GluA3-a
GluA3-ef
GluA3-d
GluB3-b
GluB3-bp
GluB3-c
GluB3-cp
GluB3-d
GluB3-f
GluB3-g
GluB3-j
GluD3-b
GluD3-c
Fig. 2 Effects of individual glutenin alleles on polymer mass Mw2.

interaction was always lower or at least equal Glu-B3 had significant interactions with the
(R² = 69.4% for Glu-B3 x PinB) than that of all puroindoline B locus, increasing the explained
loci Glu + PinB without interaction. effect on the %F3 fraction from 19.4% to 22.8%
and 20.2%, respectively. No effects on F1 and F2
Some HMW-GS loci and LMW-GS loci also
percentages were evident.
exhibited significant interaction with the
PinB locus for polymer characteristics and 6. Effects of interaction between glutenin alleles
%F3 (Table 3). Glu-B1 and Glu-A3 loci effects and puroindoline-B alleles
on polymer mass (Mw2) increased when
In regard to interactions between glutenins and
combined with effects of PinB alleles from 10%
puroindoline-B alleles, we observed different
to 13.6% and 13.9%, respectively. The effect
effects depending on alleles. For instance, Glu-B1-b
of interaction on polymer dispersity index
encoding subunits 7+8 showed a favorable effect
(Mw/Mn2) increased from 16.8% to 17.8% and
on polymer mass (Mw2) in Soft varieties carrying
20.5% for PinB x Glu-B1 and PinB x Glu-A3,
PinB-D1a, whereas their effect was unfavorable in
respectively. In regard to the polymer radius
Medium and Hard genotypes carrying PinB-D1b
trait (Rw2), the interaction was not significant
and -D1d (Fig. 3A). In contrast, the interaction
for LMW-Glutenin loci x PinB, whereas a
between Glu-A3 alleles and the puroindoline B soft
marginally significant interaction effect was
allele -D1a resulted in a much lower polymer mass
revealed for PinB x HMW-Glu-B1, increasing
than the mass observed for the interaction between
the R² from 12.7% to 14.4%. Concerning
Glu-A3 alleles and PinB-D1b alleles (Fig. 3B).
percentages of protein fractions, Glu-B1 and
Table 3. Effects of interaction between the glutenin and puroindoline-B loci.
Grain hardness Mw2 Mw/Mn2 Rw2 %F3

Glu Glu x PinB Glu Glu x PinB Glu Glu x PinB Glu Glu x PinB Glu Glu x PinB
effect effect R2% effect effect R2% effect effect R2% effect effect R2% effect effect R2%
A1 ns *** 25.8 ns ns 11.3 ns ns 16.9 ns ns 14.3 ns ns 19
HMW-GS b1 ns *** 56.6 * ** 13.6 ** ** 17.8 * ** 14.4 ns * 22.8
D1 *** *** 27.7 ns ns 10.9 ns ns 14.6 ns ns 12.4 * ns 16.4
A3 * ns 17.7 ns ** 13.9 ** *** 20.5 ns ns 15.2 ** ns 15.6
LMW-GS B3 ns *** 69.4 ns ns 10.1 ns ns 17 ns ns 13.2 *** *** 20.2
D3 ns *** 45.4 *** ns 4.9 *** ns 7.4 *** ns 6.8 ns ns 21.7
2
R % 19.9 9 15.9 11.2 16.8
Glu + PinB R2 % 69.4 10 16.8 12.7 19.4
HMW-GS = high molecular weight glutenin subunits; LMW-GS = low molecular weight glutenin subunits; ns = not significant;
MW2 = molecular mass; Mw/Mn2 = polydispersity index; R2 = polymer radius.
Mw2 Mw2
29 PinB 21 PinB
D1a D1a
24 D1b 19 D1b
D1d D1d
x 106 Da
x 106 Da
19 17
14 15
9 13
1 6+8 7+8 7+9 13+16 17+18 6.1+22 a d ef
Glu-B1 Glu-A3
Fig. 3 Interaction values found for PinB x Glu-B1 alleles (A) and PinB x Glu-A3 alleles (B). MW2 = polymer mass.

7. Temperature during grain development SE-HPLC and that AFFFF gives more reliable
was a major factor influencing polymer data for native polymer characterization than
characteristics SE-HPLC.
Variance analysis revealed that both year and It should be noted that the present results,
location significantly influenced polymer indicating that Hard varieties resulting from
characteristics (R²: 44.9%, 51.0%, and 56.4% SNPs of PinB had lower polymer mass than Soft
for Mw2, Mw/Mn2 and Rw2, respectively) varieties, are contrary to those obtained for Hard
confirming that environment had a major and Soft Falcon NILs, where the Hard phenotype
influence on gluten polymers. resulted from deletion of the PinA gene (Lesage
et al. 2012). In this previous study, we observed
We further analyzed this environmental effect,
that 1) stress-related and folding proteins were
using the sum of mean temperatures for June and
more abundant; 2) endosperm development
July, the two last months of grain development to
reached completion earlier; and 3) polymer mass
ripeness. As revealed through variance analysis,
was higher in the developing Hard kernels than
the summed mean temperatures for June and
in the Soft ones. This discrepancy between effects
July had a similar statistical effect to year and
of PinA and PinB variants on polymer mass could
location effects on polymer characteristics only.
be due to different roles played by these two
The other traits were not significantly influenced.
proteins in the endosperm, and particularly in
Moreover, the summed mean temperatures for the ER compartment that controls cellular stress
June and July differentially affected the molecular response. Study of the interactions between
weights of polymers in the three hardness classes storage proteins, puroindolines and environment
(Fig. 4). A difference of one hundred degrees could be important for understanding protein
(recorded between 6 experimental locations) matrix formation in wheat grain.
increased the final mass of polymers by an order
of 2 in Hard varieties (from 10 to 20 million Da),
Mw2
and by an order of nearly 3 in Soft varieties (from
13 to 33 million Da). Considering these huge 35
Soft ns
variations, it becomes clear that they may account 30 Medium
for the strong impact of genotype x environment Hard ns ** p<0.01
interactions on the dough quality often observed 25
and rarely explained.
x 106 Da
20
Interestingly, no significant effect was observed 15
on percentages of protein fractions, indicating
10
that storage protein composition was not affected
by temperature during the two last months of 5
grain development. The fact that even the F1
0
fraction was not impacted by temperature clearly 1080 1100 1120 1140 1160 1180 1200 1220
indicates that the F1 fraction does not contain the
Sum mean temp. (°C)
largest polymers. These results provide evidence
that the largest polymers were not retrieved by Fig. 4 Differential effect of sum of mean
temperatures on polymer mass (MW2) depending
on grain hardness classes.

We highlight that AFFFF provides more reliable Chiaramonte E, Rhazi L, Aussenac T, White Jr DR (2012)
data than SE-HPLC on polymer size. SE-HPLC Amylose and amylopectin in starch by asymmetric
flow field-flow fractionation with multi-angle light
reflects more clearly the composition of storage scattering and refractive index detection (AF4–MALS–
proteins. Interactions between glutenins and RI). J Cereal Sci 56:457-463
puroindoline-B alleles had significant effects Dachkevitch T, Autran J-C (1989) Prediction of baking
on polymer characteristics. These effects varied quality of bread wheats in breeding programs by size-
depending on the particular glutenin allele. In exclusion high-performance liquid chromatography.
addition, temperatures during the last two months Cereal Chem 66:448-456
of grain development differentially impacted Lesage VS, Bouchet B, Rhazi L, Elmorjani K, Branlard
polymer characteristics according to grain G, Marion D (2011) New insight into puroindoline
function inferred from their subcellular localization
hardness. Further molecular analysis of these
in developing hard and soft near-isogenic endosperm
results needs to be carried out on grain produced and their relationship with polymer size of storage
under controlled environmental conditions. proteins. J Cereal Sci 53:231-238
Lesage VS, Merlino M, Chambon C, Bouchet B, Marion D,
Branlard G (2012) Proteomes of hard and soft near-
isogenic wheat lines reveal that kernel hardness is
related to the amplification of a stress response during
endosperm development. J Exp Bot 63:1001-1011
Stevenson SG, Preston KR (1996) Flow Field-Flow
Fractionation of wheat proteins. J Cereal Sc 23:121-131

Structural and mechanical properties of compression-
molded wheat gluten, gliadin and glutenin enriched films
F. Rasheed1, W.R. Newson1, R. Kuktaite1, M. Hedenqvist2, M. Gälstedt3, and E. Johansson1
1Department of Plant Breeding, The Swedish University of Agricultural Sciences, Box 104, SE-23053 Alnarp,
Sweden; 2Department of Fiber and Polymer Technology, Royal Institute of Technology, SE-10044 Stockholm,
Sweden; 3 Innventia, Box 5604, SE-11486 Stockholm, Sweden
Abstract
Within the future bioeconomy there is a need to replace petroleum-based plastics with renewable
materials derived from plant polymers, e.g., starch and proteins. The wheat gluten proteins gliadin
and glutenin offer a unique and interesting combination of strength and elasticity for utilization in
the bio-based material industry. In addition, wheat gluten proteins have good oxygen barrier and
film forming properties with high biodegradability potential. The aim of the present study was to
examine the structural properties and protein-protein interactions in compression-molded plasticized
wheat gluten, gliadin and glutenin enriched films. Differences in structural and mechanical properties
of plasticized wheat gluten, gliadin and glutenin enriched films were also evaluated. Size exclusion
high performance liquid chromatography (SE-HPLC) and reverse phase high performance liquid
chromatography (RP-HPLC) were used to study the structural and polymerization behavior of
compression-molded bioplastic films. Tensile testing was used to measure the mechanical properties.
Results obtained by SE-HPLC showed that the protein solubility of compression molded wheat
gluten, gliadin and glutenin enriched bioplastic films (with 30% glycerol) in SDS extraction buffer
was extremely low compared to raw gluten, gliadin and glutenin-enriched proteins. The solubility
of proteins remained low even after repeated sonication. The low protein solubility is due to a high
degree of protein aggregation during compression-molding at high temperature (130°C) and pressure
(10 MPa). The RP-HPLC results also showed that not all the proteins were extracted when using DTT
and 6 M urea, indicating the presence of covalent types of bonding other than disulphide cross-linking.
Varying mechanical properties of plasticized wheat gluten, gliadin and glutenin enriched films were
determined in the study.
Introduction et al. 2001). Wheat gluten protein, a byproduct

Plastics are ideal raw materials for many from the bio-ethanol industry, is a suitable raw
industrial and household applications due to material for use in bio-based material industry
certain advantageous properties, including (Ye et al. 2006). Moreover, gluten is a relatively
low cost, recyclability, rigidity, and elasticity. inexpensive raw material (less than $1 per kg) and
Due to the limitations of petroleum resources, commercially available in large quantities (Reddy
rising oil prices, waste management and and Yang 2007). Gluten proteins are cohesive,
environmental pollution issues there is a need to viscoelastic, non-toxic, largely biodegradable, and
find alternatives to petro-based plastics (Momani have good oxygen barrier properties ideally suited
2009). Therefore, significant recent attention for production of biodegradable plastics (Domenek
has been given to research activities aimed at et al. 2004; Yuan et al. 2010). Gluten proteins
the synthesis of biodegradable plastics from are classified into two subgroups, gliadin and
renewable resources (Sun et al. 2007). glutenin. Both the gliadin and glutenin proteins
impart their effects on the viscoelastic behavior of
Plant polymers, e.g., starch and protein, gluten proteins. Gliadins are monomeric proteins
offer interesting characteristics such as high and contribute viscosity and extensability to the
biodegradability, abundant resources and good gluten. Glutenins are polymeric proteins that
oxygen barrier properties that could be utilized in contribute strength and elasticity to the gluten
the bio-based materials industry (Irissin-Mangata (Orth and Bushuk 1972).

Wheat gluten films can be produced using the Results and discussion
solution casting method or through thermo-
molding at high temperatures (130-160°C). Bio- Tensile testing
based films and sheets, made from wheat gluten Tensile results showed that glutenin enriched
without plasticizers are fragile and difficult to films had higher stiffness than gliadin-enriched
work with (Irissin-Mangata et al. 2001). Glycerol films. However gliadin-enriched films had
is frequently used as a plasticizer for protein- higher elongation compared to glutenin-
based bioplastics (Sun et al. 2008). It increases enriched films.
the polypeptide chain mobility of proteins by
reducing intermolecular forces and thus increases Size-exclusion high performance liquid
flexibility of wheat gluten-based films (Jerez et al. chromatography
2005; Gao et al. 2006; Sun et al. 2008). Results of size exclusion high performance
The aim of the study was to investigate the liquid chromatography are shown in Fig. 1.
structural properties of films produced from All protein fractions in raw powder form show
whole wheat gluten and individual glutenin- higher solubility in SDS-buffer and significant
enriched and gliadin-enriched fractions. We amounts of proteins were extracted even with
also studied protein-protein interaction in films two steps of sonication (Fig. 1). This shows
produced from individual protein fractions and that gluten proteins in their native form are
evaluated the mechanical properties of the films. less aggregated, but do have some disulphide
bonding. Protein extractability of plasticized
WG, gliadin and glutenin-enriched films
Materials and methods compared to raw WG powder was very low.
Materials All plasticized protein films showed very low
solubility in SDS buffer even after sonication.
The commercial wheat gluten powder used to This indicates that hot compression molding of
make gliadin and glutenin enriched fractions was gluten, gliadin and glutenin proteins at 130oC
supplied by Lantmännen Reppe AB, Sweden. in the presence of 30% glycerol induces strong
covalent bonding in protein films.
Preparation of gluten, gliadin and glutenin-
enriched bioplastic films
Bioplastic films of 0.5 mm thickness were
120
compression molded at 130°C for 10 min at 200 bar.
100
Tensile testing
Protein solubility V/sec
Dumbbell-shaped specimens (ISO 37, type 3) 80

were tested using an Instron 5566 universal test
machine at 23°C and 50% RH using a crosshead 60
speed 100 mm/min, 30 mm initial clamp
separation and 100 N load cell, and analyzed 40
using Bluehill software (Instron AB, Danderyd
Sweden). 20
Size exclusion high performance liquid 0

chromatography 1Ex 2Ex 3Ex 1Ex 2Ex 3Ex 1Ex 2Ex 3Ex
The gluten, gliadin and glutenin enriched films Gluten powder Gliadin Powder Glutenin powder
were chopped into very small pieces. Protein
solubility was determined using three extraction Fig. 1 Solubility of gluten, gliadin and glutenin-
steps according to Johansson et al. (2001). enriched fractions determined by size exclusion
high performance liquid chromatography (SE-
Reverse phase high performance liquid HPLC); 1 Ex, SDS buffer; 2 Ex, SDS buffer+30S
chromatography sonication; 3Ex, SDS buffer+60S+60S+30S
Reverse phase chromatography followed the sonication.
method used by Olabarieta et al. (2006).

Reverse Phase high performance liquid References
chromatography Domenek S, Feuilloley P, Gratraud J, Morel MH, Guilbert
Protein solubility revealed by RP-HPLC showed S (2004) Biodegradability of wheat gluten based
bioplastics. Chemosphere 54:551-559
that glutenin proteins are more polymerized
Gao C, Stading M, Wellner N, Parker ML, Noel TR, Mills
compared to whole gluten or gliadins. At the
ENC, Belton PS (2006) Plasticization of a protein-based
first step of extraction (with 70% ethanol) of film by glycerol: a spectroscopic, mechanical, and
glutenin fractions, some of the glutenin proteins thermal study. J Agric Food Chem 54:4611-4616
are extracted; these proteins could be unbound Irissin-Mangata J, Bauduin G, Boutevin B, Gontard N
monomeric proteins. Solubility significantly (2001) New plasticizers for wheat gluten films. Eur
decreased in subsequent steps of extraction until Polym J 37:1533-1541
the maximum proteins were extracted in the last Jerez AA, Partal P, Martinez AI, Gallegos C, Guerrero
step with the addition of DTT and urea. This A (2005) Rheology and processing of gluten based
could be due to the breakage of disulphide bonds bioplastics. Biochem Engineering J 26:131-138
by the addition of DTT and urea with 100oC Johansson E, Prieto-Linde ML, Jonsson JO (2001) Effects
of wheat cultivar and nitrogen application on storage
heating.
protein composition and breadmaking quality. Cereal
In contrast to glutenin proteins the maximum Chem 78:19-25
extraction of gliadins was reached in the Momani B (2009) Assessment of the Impacts of Bioplastics:
initial extraction steps due to the absence of Energy Usage, Fossil Fuel Usage, Pollution, Health
Effects, Effects on the Food Supply, and Economic
intermolecular disulphide bonds and a lower
Effects Compared to Petroleum Based Plastics.
degree of polymerization. Worcester Polytechnic Institute, MA, 58pp
Olabarrieta I, Gällstedt M, Ispizua I, Sarasua J-R,
Hedenqvist MS (2006) Properties of aged
Conclusion montmorillonite-wheat gluten composite films. J Agric
This study gave a better insight into the Food Chem 54:1283-1288
polymerization, and structural and mechanical Orth RA, Bushuk W (1972) Comparative study of the
proteins of wheats of diverse baking qualities. Cereal
properties of bioplastic films prepared from
Chem 49:268-275
different types of gluten proteins. It indicated
Reddy N, Yang Y (2007) Novel protein fibers from wheat
that compression molding of gluten, gliadin and gluten. Biomacromolecules. 8:638-643
glutenin proteins in the presence of 30% glycerol
Sun S, Song Y, Zheng Q (2007) Morphologies and
most likely induces strong covalent bonding properties of thermo-molded biodegradable plastics
in protein films. Glutenin proteins are more based on glycerol-plasticized wheat gluten. Food
aggregated than those from whole wheat gluten Hydrocolloids 21:1005-1013
or gliadins. Sun, S, Song Y, Zheng Q (2008) Morphology and
mechanical properties of thermo-molded bioplastics
based on glycerol-plasticized wheat gliadin. J Cereal
Sci 48:613-618
Ye P, Reitz L, Horan C, Parnas R (2006) Manufacture and
biodegradation of wheat gluten/basalt composite
material. J Polym Environ 14:1-7
Yuan Q, Lu W, Pan Y (2010) Structure and properties of
biodegradable wheat gluten/attapulgite nanocomposite
sheets. Polym Degrad Stabil 95:1581-1587

DNA sequence variations of four Wx alleles generating
polymorphic Wx-A1 protein for decreased amylose content
in wheat
M.Yamamori1, and C. Guzman2
1NARO Institute of Crop Science, Tsukuba, 305-8518, Japan; 2University of Córdoba, Córdoba,14071, Spain
Abstract
Waxy (Wx) genes encode granule-bound starch synthase I (also called Wx protein) responsible for
amylose synthesis of cereal grain starch. In this study, four wheat Wx-A1 alleles, viz. Wx-A1c, -A1e,
-A1i and -A1j producing polymorphic Wx proteins were studied at the DNA sequence level. It had
been suggested that apparent amylose contents conferred by the alleles increase in the order of Wx-
A1e (= waxy phenotype) <-A1i<-A1c<-A1j (= amylose level of wild type Wx-A1a, or about 20%). DNA
sequencing of these alleles detected SNPs and insertion/deletion variations, of which a particular
SNP causing amino acid changes in Wx-A1e and -A1c was identified as the likely factor responsible
for decreased amylose. On the other hand, a transposable-like element of 376 bp present in the 3’UTR
of Wx-A1i most probably lowered the levels of Wx protein and amylose. The fourth allele, Wx-A1j,
possessed four SNPs, of which two altered amino acids of the Wx-A1j protein and should cause the
protein polymorphism. Based on the DNA sequence determination, functional markers for Wx-A1c,
-A1e and -A1i were developed.

The major component of wheat flour is starch, Four variant Wx-A1 alleles, Wx-A1c, -A1e, -A1i
which normally consists of two glucose polymers, and -A1j (Table 1) were examined. Wx-A1c derives
amylose and amylopectin. The amylose/ from six Pakistani common wheat cultivars, Wx-
amylopectin ratio, i.e., amylose content, is known A1e derives from two durum wheat strains, Wx-
to affect the quality of food products made from A1i derives from one durum strain, and Wx-A1j
wheat flour. The waxy (Wx) gene, which encodes is from common wheat cv. MI. As controls for
granule-bound starch synthase I (also called Wx the variant, Wx-A1 alleles, Wx-A1a of common
protein), is responsible for amylose synthesis wheat cv. Chinese Spring and -A1a of durum cv.
of cereal grain starch. Due to allohexaploidy, Langdon were used. Wx-A1 genes were cloned
common wheat (Triticum aestivum L.) has and sequenced from exon1 to exon12 (most of
three Wx proteins, Wx-A1, -B1 and -D1. Some the transcribed region), except for Wx-A1c in four
Wx alleles produce polymorphic Wx proteins Pakistani cultivars. In these, a part (994 bp) of Wx-
and affect amylose content (Yamamori 2009; A1c was directly sequenced.
Yamamori and Yamamoto 2011). For example,
For amylose measurement by the auto-analyzer
a variant Wx-A1 protein in some Pakistani
method (Yamamori and Yamamoto 2011), wheat
wheat cultivars shows a lower molecular weight
materials carrying Wx-A1 protein but lacking both
and a little more basic isoelectric point in two-
Wx-B1 and -D1 proteins were used. Materials were
dimensional electrophoretic gels and produces
harvested from the field in 2010 and 2011.
less amylose. This variant Wx-A1 allele is named
Wx-A1c (Yamamori et al. 1994).
In this study, four Wx-A1 alleles producing Results and discussion
polymorphic Wx protein, or generating lower Average apparent amylose contents of the 2010 and
levels of amylose, were investigated to elucidate 2011 harvests (Table1) were in the order: Wx-A1b
the causes of these variations at the DNA (2.1%, waxy) = -A1e(1.9%) <-A1i(7.5%) <-A1c(16.8%)
sequence level. This enabled the development of <-A1j (22.8%) = -A1a(23.2%, wild type). These results
DNA markers for these alleles. were consistent with those of Yamamori (2009).

Allele Wx-A1c originating from PZ QT105 and The latter SNP exists in both wild type alleles
WB357 had two SNPs, compared to Wx-A1a Wx-B1a and -D1a in Chinese Spring, and the
of Chinese Spring (Fig. 1). One was present in corresponding positions of Wx-B1 and -D1
the 8th exon, changing glutamic acid in Wx- proteins are arginine. These two Wx proteins
A1a into glycine in -A1c. The other was in the showed greater capacities for amylose synthesis
9th exon, changing tryptophan in Wx–A1a to than wildtype Wx-A1 (Miura et al. 1999). Thus,
arginine in -A1c. These amino acid changes the SNP in the 9th exon should not be responsible
should be responsible for the Wx-A1c protein for the decreased amylose of Wx-A1c. To explore
polymorphism (Table 1). Two SNPs were also the conservation of the amino acid (glutamic
present in Wx-A1c which was directly sequenced. acid) changed by the SNP in the 8th exon, amino
Table 1. Wx alleles examined, origins in wheat, phenotype of Wx proteins and apparent amylose contents.
Apparent amylose(%)***
Wx allele Origin* Phenotype of Wx protein** 2010 harvest 2011 harvest Average
PZ QT105, PZ WB357
Wx-A1c (PZ WB6, PZ WB27, Greater mobility on SDS-PAGE 16.8 ± 0.9c (4) 16.8 ± 1.1c (4) 16.8
PZ 49P70-27, PZ 49P70-9 more basic PI
-A1e KU 3659, KU 3655 More basic PI 0.9 ± 0.5a (4) 2.9 ± 0.0a (4) 1.9
-A1i KU 9259 Decreased amount 7.0 ± 0.1b (6) 8 ± 0.2b (4) 7.5
-A1j MI More basic pI than Wx-A1c 23.0 ± 0.3d (3) 22.6 ± 1.4d (4) 22.8
Wx-A1a Bai Huo - 22.5 ± 1.5d (6) 23.8 ± 1.5d (4) 23.2
(wild type) Langdon - - - -
-AIb (waxy) Kanto 107 Null 0.8 ± 0.4a (3) 3.3 ± 0.4a (3) 2.1
* Wheat strains carrying the variant Wx-A1 protein but lacking both Wx-B1 and -D1 proteins were developed (Yamamori 2009,
Yamamori and Yamamoto2011) and used, except for PZ 49P70-9, KU 8937B and KU 3655. PZ is an abbreviation of Pakistani cv
Zairaishu. For the four wheat strains in parentheses, the Wx-A1c gene region containing SNPs was directly sequenced.
** Yamamori et al. (1994, 1995), Yamamori and Yamamato (2011).
* Values with the same letter in the same harvest year were not significantly different (p<0.05 Turkey’s studentized range test).
Numbers in parentheses show numbers of strains used for measurement.
GCC GCTA1j TGG AGGA1c, j CGC CGTA1j

500bp
(Arg) (Arg) GAG AGGA1j (Trp) (Arg) (Arg) (Arg)
(Glu) (Lys) GAG GGGA1c
TATA 㻔 (Glu) (Gly)* TGA AATAAA
ATG
CAAT
5’ 1 2 3 4 5 6 7 8 9 10 111 12 3’
G AA1e
C AA1e Microsatellite
GCC GCTA1e, i CCG CCTAA1e, i GAG AAGA1e
(Ala) (Ala) (Pro) (Pro) (Glu) (Lys)*
A1i: insertion A1e, i: deletion CCC CCGA1e,i
of 3 bp (AAG) of 40 bp (Pro) (Pro)
Fig. 1 Structure of the Wx-A1 gene showing nucleotide and amino acid sequence changes in Wx-A1c, -A1e,
-A1i and -A1j alleles. Amino acids are shown in parentheses. Asterisks indicate possible factors decreasing
the amylose content conferred byWx-A1c, -A1e and -A1i. TATA and CAAT signals are from AY050175 and
the putative polyadenylation signal AATAAA is from Clark et al. (1991). Exons 1–12 are shown by boxes;
the boundaries of exon 1 were determined by EST clones BQ605981 and AW448831. The horizontal thin line
indicates the promoter region, which was not sequenced in Wx-A1a of Chinese Spring, or-A1c, -A1d and -A1j
in this study. The 3’ UTR region (47 bp) shown by the dotted line was not examined in genomic DNA.

acid sequences of starch (glycogen) synthases Wx-A1j had four SNPs (Fig. 1), of which, two in
from nine other organisms were compared using the 3rd and 9th exons changed amino acids. These
ClustalW (Fig. 2). Seven higher plants, including should result in the variant Wx-A1j protein (Table
wheat, had glutamic acid at this position, but this 1). The SNP in the 9th exon was identical to the
was not the case in three microorganisms. This one inWx-A1c. However, results from amylose
amino acid position exists in a conserved region measurement (Table 1) suggested that two amino
of starch synthases (Li et al. 1999; Leterrier et al. acid changes would not affect amylose production.
2008). Therefore, amino acid changes caused by
Compared to Wx-A1a of Langdon, Wx-A1i had
SNPs in the 9th exon are responsible for lower
a 3 bp (AAG) insertion and a 40 bp deletion
levels of amylose.
in the 1st intron (Fig. 1). Since an insertion of
Wx-A1e from two durum wheat accessions had CAGAAGAAG happened in this region of
five SNPs in five exons (Fig. 1), of which one Chinese Spring Wx-A1a and a deletion of 40 bp
in the 9th exon caused an amino acid change; was present in Wx-A1e, these variations would
glutamic acid in Wx-A1a of Langdon was lysine not result in decreased Wx-A1i protein or lowered
in Wx-A1e. This amino acid change should cause amylose content (Table 1). On the other hand,
a slightly more basic pI of the Wx-A1e protein an insertion sequence of 376 bp was found in an
(Table 1). Glutamic acid at this position is present AT-repeat microsatellite present in the 3’UTR
in the conserved region of Wx proteins or starch of the 12th exon (Fig. 1). This insertion sequence
synthases from plants and E. coli (Ainsworth et showed characteristics of a transposable element;
al. 1993, Fig. 2). Nichols et al. (2000) and Yep et it had a target site duplication (TATATAATA) and
al. (2004) reported that the negatively-charged incomplete terminal inverted repeats; it generated
glutamic acid at this position is essential for a possible complex secondary structure by
catalysis of starch synthases. Furthermore, lysine inverted repeats, and was AT-rich (69%). Searching
is positively charged. Thus, this amino acid would the transposable-like element (TLE) sequence as a
be responsible for lack of amylose production query in the NCBI database (http://www.ncbi.nlm.
(waxy phenotype) of Wx-A1e protein. nih.gov/) and CerealsDB (http://www.ncbi.nlm.
Position of amino acid change

134k (-A1j) 405G(-A1c) 480K (-A1e)
Species
Ta (AB019622) KDAWDTSVISEIKVVDRYERV RLEEQKGPD SRFEPCGLIQL
Hv (CAA30755) KDAWDTSVISEIKVADEYERV RLEEQKGPD SRFEPCGLIQL
Os (CAA46294) KDAWDTSVVAEIKVADRYERV RLEEQKGSD SRFEPCGLIQL
Zm (P04713) KDAWDTSVVSEIKMGDGYETV RLEEQKGPD SRFEPCGLIQL
St (CAA41359) KDAWDTSVAVEVKVGDSIEIV RLEEQKGSD SRFEPCGLIQL
Ath (NP174566) KDAWDTCVVVQIKVGDKVENV RLEEQKGSD SRFEPCGLIQL
Ps (AAB26591) KDAWDTNVLVEVKVGDKIETV RLEEQKGSD SRFEPCGLVQL
Ec (AA23870) RRGVTD-AQVVSRRDTSAGHI RLTSQKGLD SRFEPCGLTQL
Atu (P0A3F2) KAAVTDPVKCFEFTDLLGEKA RLTWQKG I D SRFEPCGLTQL
Cr (XP001696617) PGVFDTGLLAPAAQPTRVTYA RLEVQKGVD SRWEPCGLVAL
** *** * ** ***** *
Fig. 2 Amino acid sequence comparison between plant GBSSs and micro-organism glycogen synthases.
Ta, Triticum aestivum; Hv, Hordeum vulgare; Os, Oryza sativa; Zm, Zea mays; St, Solanum tuberosum;
Ath, Arabidopsis thaliana; Ps, Pisum sativum; Ec, Escherichia coli; Atu, Agrobacterium tumefaciens; Cr,
Chlamydomonas reinhardtii. Numbers 134, 405 and 480 with arrows indicate positions in the Wx-A1
protein at which amino acid changes (K, G, M, K) have occurred in Wx-A1j, -A1c and -A1e, respectively.
Asterisks show identical amino acids among the ten starch synthases.

nih.gov/) produced matching sequences from the References
wheat genome in which the flanking sequences Ainsworth C, Clark J, Balsdon J (1993) Expression,
had mostly TA or AT repeats. organization and structure of the genes encoding
the waxy protein (granule-bound starch synthase) in
RT-PCR and 3’RACE analyses of Wx-A1i using wheat. Plant Mol Biol 22:67-82
RNA extracted from immature endosperm Clark JR, Robertson M, Ainsworth CC (1991) Nucleotide
showed that the first part of the 376 bp sequence of a wheat (Triticum aestivum L.) cDNA clone
insertion sequence was present after ATrepeat encoding the waxy protein. Plant Mol Biol 16:1099-
microsatellite. This finding suggested that the 1101
376 bp insertion sequence generated alternative Leterrier M, Holappa LD, Broglie KE, Beckles DM (2008)
polyadenylation signals in Wx-A1i. Abnormal Cloning, characterisation and comparative analysis
of a starch synthase IV gene in wheat: functional and
mRNA caused by the insertion would be the most
evolutionary implications. BMC Plant Biol 8:98
likely cause of the decreased Wx-A1i protein and
Li Z, Chu X, Mouille G, Yan L, Kosar-Hashemi B, Hey S,
the lowered amylose content (Table 1). Napier, J, Shewry P, Clarke B, Appels R, Morell MK,
CAPS (cleaved amplified polymorphic sequence) Rahman S (1999) The localization and expression of
the class II starch synthases of wheat. Plant Physiol
markers were developed for Wx-A1c and -A1e,
120:1147-1155
using primers 7A-F2 and 7A-R2a (Huang and
Miura H, Araki E, Tarui S(1999) Amylose synthesis
Brûlé-Babel 2010). These primers
capacity of the three Wx genes of wheat cv. Chinese
In the Wx-A1 gene amplicon, 1038 bp which Spring. Euphytica 108:91-95
included target SNPs of Wx-A1c (in the 8th exon) Nichols DJ, Keeling PL, Spalding M, Guan H (2000)
and -A1e (in the 9th exon) were produced. In the Involvement of conserved aspartate and glutamate
residues in the catalysis and substrate binding of
amplicon of Wx-A1a, four recognition sites of maize starch synthase. Biochemistry 39:7820-7825
BseRI (GAGGAG) are present. One of these was
Yamamori M (2009)Amylose content and starch properties
eliminated by a SNP in Wx-A1c (GAGGGG). generated by five variant Wx alleles for granule-bound
Thus, after digestion of the Wx-A1a amplicon by starch synthase in common wheat (Triticum aestivum L.).
BseRI, five DNA bands appeared in an agarose Euphytica 165:607-614
gel, whereas four bands were detected in -A1c. Yamamori M, Nakamura T, Endo TR, Nagamine T (1994)
In the amplicon of Wx-A1a, seven recognition Waxy protein deficiency and chromosomal location
sites of TaqI (TCGA) are present. A SNP in Wx- of coding genes in common wheat. Theor Appl Genet
89:179-184
A1e (TCAA) eliminated one of the sites. For
Wx-A1i, the size difference of the PCR amplicon Yamamori M, Nakamura T, Nagamine T (1995)
Polymorphism of two waxy proteins in the emmer
was usable as a dominant marker. Primers 3’FR2 group of tetraploid wheat, Triticum dicoccoides, T.
(GCGGGCTCGTCGACACTATC) and 3’FR1 dicoccum, and T. durum. Plant Breeding 114:215-218
(TACCACACCAAAATGCAAGCAC) were Yamamori M, Yamamoto K (2011) Effects of two novel Wx-
developed and produced a major amplicon of over A1 alleles of common wheat (Triticum aestivum L.) on
1,000 bp in agarose gels of genotypes with Wx-A1i. amylose and starch properties. J Cereal Sci 54:229-235
However, the control wheat cultivar with Wx-A1a Yep A, Ballicora MA, Sivak MN, Preiss J (2004)
produced a major band of about 850 bp. Identification and characterization of a critical region
in the glycogen synthase from Escherichia coli. J Biol
Chem 279:8359-8367
Conclusions
DNA sequencing of four Wx-A1 alleles identified
SNPs and deletion/insertion of base sequences.
These were mostly identified as base variations
responsible for Wx-A1 protein polymorphism
or decreased amylose content. These findings
enabled development of DNA markers for the
three variant Wx-A1 alleles.

ISBN: 978-607-8263-30-1
E-mail: [email protected]
Web site: www.cimmyt.org

Glutenworkshop

Uploaded by

Copyright:

Available Formats

Glutenworkshop

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

Glutenworkshop

Uploaded by

Copyright:

Available Formats

Proceedings

11th International Gluten Workshop

Zhonghu He and Daowen Wang (eds.)

Ministry of Agriculture (MOA), International Collaboration Program (2011-G3)

National Natural Science Foundation of China (NSFC)

AGRIS Category Codes: Q04 Food Composition

Dewey decimal classification: 572.651 HE

v Zhensheng Li: Welcome Speech

1 Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics

iv Proceedings 11th InternaƟonal Gluten Workshop

Ladies and gentlemen, dear colleagues:

Zhengsheng Li: Welcome Speech 11th InternaƟonal Gluten Workshop v

Dear colleagues and friends:

vi Proceedings 11th InternaƟonal Gluten Workshop

Dear colleagues and friends:

viii Proceedings 11th InternaƟonal Gluten Workshop

Wheat A genome sequencing and its application for

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 1

Introduction in dryland/rain-fed environments. Understanding

2 Proceedings 11th InternaƟonal Gluten Workshop

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 3

Table 1. Sample alignments of peptides to International Wheat Genome Sequencing Consortium

IWGSC_6DS_v1_2096398 11638 11680 + Peptide=LQCVGSQVPEAVLR

4 Proceedings 11th InternaƟonal Gluten Workshop

triticin [Triticum aestivum]

Alignment statistics for match #1

Score Expect Identities Positives Gaps

Score Expect Identities Positives Gaps

Score Expect Identities Positives Gaps

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 5

6 Proceedings 11th InternaƟonal Gluten Workshop

USDA-ARS Western Regional Research Center, Albany, CA 94710, U.S.A.

Introduction individual environmental factors aﬀect the

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 7

Quality Proteome Transgenic Plants

Materials and methods Results and discussion

8 Proceedings 11th InternaƟonal Gluten Workshop

Total % peptides % peptides % peptides

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 9

10 Proceedings 11th InternaƟonal Gluten Workshop

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 11

12 Proceedings 11th InternaƟonal Gluten Workshop

UMR1095 GDEC, F-631 77 Aubière, France

Introduction al. 2007). Only a few proteomics studies have

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 13

14 Proceedings 11th InternaƟonal Gluten Workshop

The subunits encoded by the three LMW alleles

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 15

Some results from heat treatment

16 Proceedings 11th InternaƟonal Gluten Workshop

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 17

Introduction the complex wheat-Fusarium interaction, one

18 Proceedings 11th InternaƟonal Gluten Workshop

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 19

20 Proceedings 11th InternaƟonal Gluten Workshop

Session 1: Capturing discoveries from genomics, proteomics, and transcriptomics 21

Introduction We previously profiled the gene expression of

22 Proceedings 11th InternaƟonal Gluten Workshop

BAC3 (6A): 94,535 bp BAC11 (6A): 114,651 bp BAC2 (6D): 210,855 bp

BAC5 (6B): 170, 743 bp