See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/263428448

Isolation and selection of Hemileia Vastatrix antagonists

Article in European Journal of Plant Pathology · August 2014

DOI: 10.1007/s10658-014-0430-9

CITATIONS READS

19 2,075

5 authors, including:

Fernando Haddad Eduardo S.G. Mizubuti

Brazilian Agricultural Research Corporation (EMBRAPA) Universidade Federal de Viçosa (UFV)
31 PUBLICATIONS 453 CITATIONS 211 PUBLICATIONS 4,146 CITATIONS

SEE PROFILE SEE PROFILE

Luiz antonio Maffia

Universidade Federal de Viçosa (UFV)
93 PUBLICATIONS 1,954 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Characterization of Mycospaherella fijiensis isolates View project

genetic and alternative control of Xanthomonas axonopodis pv. manihotis the causal agent of cassava bacterial blight View project

All content following this page was uploaded by Fernando Haddad on 09 January 2015.

The user has requested enhancement of the downloaded file.

 Biological Control 49 (2009) 114–119

Contents lists available at ScienceDirect

Biological Control
journal homepage: www.elsevier.com/locate/ybcon

Biological control of coffee rust by antagonistic bacteria under field conditions

in Brazil
Fernando Haddad, Luiz A. Maffia *, Eduardo S.G. Mizubuti, Hudson Teixeira
Departamento de Fitopatologia, Universidade Federal de Viçosa, Av. P. H. Rolfs s/n, 36570-000 Viçosa, MG, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Rust (Hemileia vastatrix) is the most important coffee disease in Brazil. Organic coffee production has
Received 13 March 2008 increased in the country and a research program aimed to develop alternatives to chemicals for disease
Accepted 4 February 2009 control was required. Seven bacterial isolates, isolated from organic coffee plantings and selected in
Available online 13 February 2009
greenhouse tests, were evaluated under commercial organic crop conditions in 2005 (Experiment 1)
and 2005/2006 (Experiment 2), in Machado, MG, Brazil. Ten treatments consisting of the seven bacterial
Keywords: isolates, copper hydroxide, calcium silicate and water were applied as three or four monthly sprays in
Coffea arabica
Experiment 1 or 2, respectively. Rust severity and incidence were evaluated monthly. In Experiment 1,
Hemileia vastatrix
Pseudomonas
the sprays started in January when rust incidence was 23.8%, and none of the treatments reduced rust
Bacillus progress significantly. In Experiment 2, the sprays began in November 2005, when rust incidence was
Biocontrol approximately 7.5%. There were significant differences (P < 0.0001) between treatments regarding max-
Disease management imum incidence and severity (as assessed in June, 2006), the rate of increase of the incidence between
November 2005 and June 2006 and for the areas under disease progress curves for both rust incidence
and severity. Lower values for these treatments were obtained in the plots treated with copper hydroxide
or Bacillus sp. isolate B157, and intermediate values with the Pseudomonas sp. isolate P286. In a third
experiment conducted in 2007 in Ervália, MG, isolates B157 and P286 were also evaluated; isolate
B157 reduced rust intensity as effectively as copper hydroxide. Isolate B157 is considered a potential bio-
control agent for coffee rust for organic crop systems in Brazil.
Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction Rust control by resistant varieties is promising, but physiologi-

cal and genetic variability of the pathogen reduces the durability of
Brazil is the largest coffee (Coffea arabica L.) producer and ex- resistance (Van der Vossen, 2005). Considering that most coffee
porter worldwide (MAPA, 2007). There is an increasing demand, areas are planted with susceptible cultivars, the predominance of
especially from Japan and the European and North American coun- a highly variable pathogen population (Nunes et al., in press),
tries, for the so-called ‘‘specialty coffees”, including organic grown and the increasingly restrictions on copper based fungicides, new
coffee, which have a high added-value and can attain a price twice alternatives for rust management are necessary. The organic pro-
that of the traditional coffee (Caixeta and Pedini, 2002). duction system is a feasible system for the development and com-
The organic coffee cropping system has official standards which mercialization of biological products (Fravel, 2005; Harman, 2000;
prohibit the use of pesticides; even the usage of copper-based fun- Pedini, 2000). In this scenario, biological control of coffee leaf rust
gicides is being restricted (Carvalho et al., 2002). Although the or- is a viable alternative.
ganic cropping systems favor human health and the environment, Rust biocontrol has been investigated previously (Shiomi et al.,
the intensity of plant diseases is expected to be higher in these sys- 2006; Bettiol et al., 1994; Bettiol and Várzea, 1992). However, no
tems because of the restrictions on pesticide usage. Leaf rust study was specifically addressed to the organic system and was
(caused by Hemileia vastatrix Berk. and Br) is the most important conducted under field conditions. Considering the need for alterna-
coffee disease and can cause yield losses up to 50% (Zambolim tives to manage coffee rust in organic systems, we initiated a re-
et al., 1997). When organic and conventional growing systems search program in 2003 aimed at finding biological agents for
were compared regarding progress of diseases, higher levels of cof- disease control. A total of 393 microorganisms (154 bacteria and
fee rust intensity were recorded in the organic systems (Botelho, 239 fungi) were isolated from leaves, leaf residues (crop debris)
2006). and soil taken from organic coffee systems in three localities of
Minas Gerais state. Seventeen isolates reduced the infection fre-
quency and the sporulation of H. vastatrix in leaves of 6-month-
* Corresponding author. Fax: +55 31 3899 1091.
E-mail address: lamaffi[email protected] (L.A. Maffia). old plants of variety Catuaí by more than 70% (Haddad et al.,

1049-9644/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.biocontrol.2009.02.004
 F. Haddad et al. / Biological Control 49 (2009) 114–119 115

2004a). These isolates were sprayed on coffee plants up to 16 days growth in MD1-agar, and endospore formation, according to
before H. vastatrix inoculation and seven bacterial isolates reduced Schaad et al. (2001). Genus identity was confirmed based in the
rust severity as efficiently as copper sulfate (Haddad et al., 2004b). fatty acid pattern through the Midi System and compared with
Through fatty acids profile, six isolates were identified as Bacillus the TSBA 4.0 Library, in the Laboratory of Biotechnology and Biodi-
spp. and one as Pseudomonas sp. (Unpublished data). versity of the Departmento de Microbiologia, at the Universidade
The objective of this study was to determine if the seven iso- Federal de Viçosa.
lates of bacteria selected under greenhouse conditions would effi- The seven bacterial isolates (six of Bacillus sp.—B10, B25, B157,
ciently control coffee rust under field conditions. Preliminary B175, B205, and B281—and one of Pseudomonas sp.—P286) were
results of this research were previously presented by Haddad preserved at 4 °C on 523 culture medium (Kado and Heskett,
et al. (2007). 1970) in tubes and deposited in the culture collection of the Labo-
ratório de Epidemiologia do Departamento de Fitopatologia at the
Universidade Federal de Viçosa, Brazil, under the same code num-
2. Materials and methods ber provided in this article.
To prepare cell suspensions used as inoculum, each isolate was
2.1. Bacterial isolates and inoculum preparation transferred to 523 medium in Petri dishes, that were then incu-
bated at 30 °C in the dark. After 24–36 h incubation, the bacterial
A total of 393 microorganisms were collected and isolated from cells were washed off the culture medium using sterile saline solu-
samples of leaves, crop debris (mostly leaves), and soil from organ- tion 0.85%. The concentration of the bacterial suspensions was ad-
ic coffee farms in the Minas Gerais State. From leaves, 1-cm discs justed to O.D.540 = 2.0 in the laboratory. The suspensions were
were cut with a cork borer and transferred to 100 ml of sterile dis- transported to the field on ice in an insulated box and were diluted
tilled water (SDW) containing 0.05% Tween 20 in 250 ml-Erlen- tenfold in saline solution (0.85%) with Tween 20 (0.05%) to
meyers. Flasks were shaked at 130 rpm for 60 min (Swadling and O.D.540 = 0.2 just before spraying. The volume of saline/Tween
Jeffries, 1996). Serial dilutions of 101 to 103 in SDW were done, solution was pre-determined in the laboratory, and flasks with
and 0.2 ml of each dilution was plated onto potato dextrose agar the exact amounts for dilutions were taken to the field.
medium (PDA). The Petri dishes were kept at 25 °C, under 12 h
day length (for now on, unless otherwise stated, all cultures/plants
2.2. Location of the field experiments and comparisons of treatments
were kept under 12 h day length). Other leaf discs were washed,
plated on paraquat-chloramphenicol-agar medium (PCA) (Peng
Three field experiments were conducted over three years in
and Sutton, 1991), and kept at 25 °C. Similar procedure was fol-
commercial organic coffee plantings with 20-year-old coffee plants
lowed with leaf debris, but serial dilutions were from 101 to
in the Minas Gerais state. Two experiments were conducted near
105. Each dilution was plated on PDA amended with antibiotics
Machado (876 m altitude, 2°360 0900 S and 45°480 1500 W), with variety
(0.05 g of streptomycin sulfate and 0.05 g of chloramphenicol)
Mundo Novo, and one experiment was conducted near Ervália
(PDA + antibiotics), MS medium (Park et al., 1992), or 523 medium
(839 m altitude, 20°540 5600 S and 42°370 2000 W), with variety Catuaí.
(Kado and Heskett, 1970). SDW (90 ml) was added to each 10 g-soil
The experiments in Machado were conducted from January to Sep-
sample, the suspension was agitated for 10 min in a magnetic stir-
tember 2005 (Experiment 1) and from November 2005 to Septem-
rer, serially diluted from 101 to 105 in SDW, 0.2 ml aliquots were
ber 2006 (Experiment 2). The third experiment was conducted
taken from the original suspension and each dilution, and plated
near Ervália from November 2006 to September 2007.
on the same media described above. The Petri dishes were kept
In Experiments 1 and 2, 10 treatments were compared: treat-
at 25 °C. Endospore-forming bacteria were isolated by keeping ali-
ments 1–7 consisted of each bacterial isolate; treatment 8 was cop-
quots of leaf, crop debris, and soil suspensions for 20 min in a stir-
per hydroxide (0.88 g/plant); treatment 9 was calcium silicate
red water bath at 80 °C, serial dilutions (101 to 105) were made,
(0.14 g Si/plant); and treatment 10 was water (control). The cal-
0.2 ml of each dilution was plated on 523 medium, and kept at
cium silicate treatment was included because it was commonly
27 °C. All Petri dishes were checked daily, and individualized colo-
used by some Machado growers for rust control. Based on the re-
nies were transferred to PDA in culture tubes that were maintained
sults of Experiments 1 and 2, four treatments consisting of isolate
at 10 °C, 12 h day length.
B157, isolate P286, copper hydroxide, and water were compared in
Suspension of each fungal (106 spores/ml) or bacterial
Experiment 3.
(O.D.540 = 0.2) isolate was sprayed on the underside of leaves of
For all experiments, each treatment was sprayed (400 ml per
6-month-old plants of variety Catuaí. After 48 h in moist chamber
plant) through a backpack sprayer with a hollow cone-type nozzle
at 25 °C, H. vastatrix was inoculated; the plants remained 48 h in
(600 or 800 l/ha in Machado or Ervália, respectively). Following
moist chamber at 25 °C in the dark, and were taken to a growth
growers’ spray schedules, the treatments were monthly sprayed
chamber at 22 °C, 12 h day length. Four variables were assessed:
three times in Experiment 1, beginning in January 2005, when rust
latent period (LP), infection frequency (IF), number of urediniosp-
incidence on leaves was above 23.8%; four times in Experiment 2,
ores produced per leaf (UPL), and urediniospore germination
beginning in November 2005, when incidence averaged 7.6%; and
(UG). For all variables, treatment efficacy (TE) was estimated in
four times in Experiment 3, beginning in January 2006, when inci-
relation to control plants sprayed with water and inoculated with
dence averaged 5%.
H. vastatrix. In a previous study, 17 isolates reduced more than 70%
both the IF and the UPL in 6-month-old Catuaí coffee plants (Had-
dad et al., 2004a). The efficacy of these isolates was further evalu- 2.3. Design and evaluation of experiments and data analysis
ated, and seven isolates were the most efficient in reducing rust
severity, mainly if applied before H. vastatrix inoculation (Haddad All three experiments were conducted in a randomized com-
et al., 2004a). These seven isolates were chosen to be assessed in plete block design with four replications of each treatment. One
the present study. experimental unit consisted of eight plants in Experiments 1 and
To identify the genus of the selected bacteria, the following 2, and six plants in Experiment 3. Each block followed a planting
tests were conducted: Gram reaction (KOH method and staining), row (planted at 4.5  1.5 m in Experiments 1 and 2 or at
anaerobic growth, growth on YDC at 30 °C, pigment production 2.5  2 m in Experiment 3). As borders, one row of plants between
in King B medium, growth at 40 °C, urease test, oxidase test, blocks and one plant at each end of a plot were left unsprayed.
116 F. Haddad et al. / Biological Control 49 (2009) 114–119

At the day of treatment applications and at monthly intervals, to September. The increase in SEV was from January to May in
10 leaves from branches in the middle third of each of six (Exper- the control and in the plots treated with isolate P286, or July in
iments 1 and 2) or four (Experiment 3) plants of a plot were ran- the plots treated with copper hydroxide and isolate B157, and
domly sampled. As the first and the last plants of a plot were not dropped until September (Fig. 1).
sampled, 60 leaves in Experiments 1 and 2 or 40 leaves in Experi- The treatments did not differ regarding y0INC and y0SEV
ment 3 were assessed. The sampled leaves were transported to the (P = 0.90 and P = 0.91, respectively), but differed (P < 0.0001)
laboratory in paper bags in an insulated box. Leaves were assessed regarding the other variables. For all assessed variables, plots trea-
and data on disease incidence (INC), and severity (SEV) were col- ted with isolate B157 did not differ from those treated with chem-
lected. For INC, the percentage of diseased leaves was calculated. ical control. Isolate P286 was less effective than copper hydroxide.
For SEV, images of all leaves collected for each replication of each Overall, disease intensity in plots treated with isolate B157 was
treatment were obtained and digitalized using a digital camera lower than in the control plots (Table 2).
and the percentage of the leaf area rusted was estimated using As in Experiment 2, lowest values of the variables were attained
the Quant software (Vale et al., 2003). In each experimental unit, with copper hydroxide and B157. Values of y0INC and y0SEV for
incidence value was calculated based on the total number of leaves P286 were intermediate, and highest values were recorded in the
assessed whereas severity was averaged over the assessed leaves. control plots (Table 2).
For each experiment, values for INC and SEV were used to cal-
culate the area under the disease progress curve (AUDPC) (Madden
et al., 2007), the initial disease (y0), the maximum disease (ymax), 4. Discussion
and the increment on disease (D) as ymax  y0. All variables were
subjected to analysis of variance, using the SAS System v. 9.1. Con- The biocontrol agents selected were efficient in reducing the
sidering the high number of treatments in Experiments 1 and 2, intensity of leaf rust under field conditions. The incidence was
their means were compared through Scott–Knott test (a = 0.01), greatly reduced, and management strategies of leaf rust with bio-
using the SAEG system. For Experiment 3, with fewer treatments, control agents, particularly B157 and P286, can be based on the ef-
the means were compared by the Tukey test (a = 0.01), using the fects of treatments in this component of disease intensity. The
SAS System v. 9.1. significant correlation between rust incidence and severity has
been previously demonstrated in field experiments in Brazil (Sil-
va-Acuña et al., 1999). Incidence is easier to estimate, less influ-
3. Results enced by subjectivities, and its use can increase the rate in which
evaluations can be made. For this reason, rust incidence in coffee
3.1. Experiment 1 has been commonly used to assess intensity, and this discussion
will be based on incidence values.
Disease progress curves for INC and SEV followed the same In the first experiment, none of the treatments reduced both
trend: both rose from January to April (beginning of harvest time), rust incidence and severity. Although some isolates reduced dis-
when ymax was achieved, and decreased during the period between ease incidence when compared to the control, all values of max-
April and September, 2005 (Fig. 1). imum incidence were above 37%, which is unacceptable for a
The treatments did not significantly affect y0INC, y0SEV, ymax- successful rust control (Garçon et al., 2004; Zambolim et al.,
SEV, DINC, DSEV and AUDPC–SEV (Table 1). However, the values 1997). When the sprays started in January 2005, incidence was
for both ymaxINC and AUDPC–INC in plots treated with isolates already 23.8%, well above 5%, the established limit to start the
B10, B157, B175, B205, B281, and P286 were significantly lower sprays of protectant fungicides (Carvalho et al., 2002; Zambolim
than the control (sprayed with water). The value of ymaxINC for et al., 1997). Given that the isolates protected leaves from infec-
copper hydroxide was significantly lower than the control (Table tion of H. vastatrix (Haddad et al., 2004b) by reducing uredini-
1). ospores germination (unpublished data), the sprays of
treatments must have been started earlier. Therefore, their low
3.2. Experiment 2 efficacy, including that of copper oxychloride, was related to
the delay in starting the sprays. In the second and third experi-
Progress curves of INC and SEV followed a trend similar to ments, the treatment sprays started when disease incidence ran-
Experiment 1: they increased from November 2005 to June 2006 ged between 5 and 8%. Therefore, the isolate B157 of Bacillus sp.
(harvest time), when the maximum was achieved, and dropped was as efficient as the copper fungicide in controlling rust. It
from June to September (Fig. 1). seems that the epidemiological effect of this isolate on rust con-
The treatments did not differ regarding y0INC and y0SEV, but trol is similar to the effect of fungicides on polycyclic diseases, in
differed (P < 0.0001) regarding ymaxINC, ymaxSEV, DINC, DSEV, which the reduction of progress rate has the most remarkable
AUDPC–INC, and AUDPC–SEV (Table 1). Considering the last six epidemiological effect (Zadoks and Schein, 1979). This effect
variables, lowest values were always attained with copper hydrox- was clearly seen in the third experiment, in which the progress
ide and B157. These two treatments did not differ from P286 for curves of rust in plants sprayed with either B157 or copper
ymaxSEV and DSEV. Isolate P286 effect was intermediate between hydroxide were similar.
the two best treatments above and the group formed by B25, Significant reductions in rust incidence was expected to be
B10, B281, B175, B205, calcium silicate, and the control that gener- achieved with the use of all bacterial isolates, given that they pro-
ated the largest values, especially of ymaxINC, ymaxSEV, DINC and tected coffee plants against H. vastatrix for at least 16 days under
DSEV (Table 1). greenhouse conditions (Haddad et al., 2004a). It is possible that
A subset of the best treatments was selected to be tested in the inability of these isolates to control coffee rust in the field
Ervália, MG, a municipality located in a different geographical area. was due to a lack of adaptability to unfavorable conditions. Besides
the antagonistic effect, effective biocontrol agents must compete
3.3. Experiment 3 with the phylloplane residents, as well as reproduce and survive
under unfavorable conditions (Fokkema, 1990). Four Bacillus iso-
Progress curves for INC increased from January to June (harvest lates efficient in controlling pumpkin powdery mildew (Podosphae-
time), when the maximum was achieved, and dropped from June ra fusca) remained viable on the phylloplane for at least 16 days
 F. Haddad et al. / Biological Control 49 (2009) 114–119 117

A 70 Control B
CuOH 2
60 B157 3
P286
50 B10
B25

Severity (%)
B175
Incidence (%)

40 2
B205
B281
30 CaSi

20 1

10

0 0
Jan Feb Mar Apr May Jun Jul Aug Sep Jan Feb Mar Apr May Jun Jul Aug Sep
Months Months

C 70 D
60 3

50
Incidence (%)

Severity (%)

40 2

30

20 1

10

0 0
Nov Dec Jan Mar Apr May Jun Jul Aug Sep Nov Dec Jan Mar Apr May Jun Jul Aug Sep

Months Months

E 70
F
60 3

50
Incidence (%)

Severity (%)

40 2

30

20 1

10

0 0
Jan Feb Mar Apr May Jun Jul Aug Sep Jan Feb Mar Apr May Jun Jul Aug Sep
Months Months

Fig. 1. Progress of coffee leaf rust on organically grown coffee plants treated with chemical and biological agents in three separate field experiments conducted in Machado
and Ervália, Minas Gerais state, Brazil. In all experiments, disease intensity was assessed as incidence (A, C and E) and severity (B, D and F). In two experiments, conducted in
Machado, from January to September 2005 (A and B) and from November 2005 to September 2006 (C and D), 10 treatments consisting of water (Control), copper hydroxide
(CuOH2), each of six isolates of Bacillus sp. (B10 to B281), one isolate of Pseudomonas sp. (P286), and calcium silicate (CaSi) were compared. In a third experiment, conducted in
Ervália, from January to September 2007 (E and F), four treatments consisting of water, copper hydroxide, and isolates B157 and P286 were compared.

(Romero et al., 2004), whereas the population of a B. subtilis antag- strategies to manage diseases that affect the aerial plant parts (Kin-
onist to Cercospora beticola declined sharply in the first 4 days of kel, 1997). These studies are being pursued.
application (Collins et al., 2003). Therefore, studies related to the Species of both Pseudomonas and Bacillus are effective biocontrol
dynamics of leaf populations of microorganisms as well as ecolog- agents of several rusts (Teixeira et al., 2005; Flaishman et al., 1996),
ical studies of presumed biocontrol agents are essential to direct including coffee rust under controlled conditions (Shiomi et al.,
118 F. Haddad et al. / Biological Control 49 (2009) 114–119

Table 1
Coffee leaf rust incidence (INC) and severity (SEV) values: initial (y0), maximum (ymax), increment (D) and area under disease progress curve (AUDPC). Ten treatments (copper
hydroxide, seven bacterial isolates, calcium silicate and control) were compared in two experiments conducted in an organic coffee plantation in Machado, MG, Brazil, from
January to November 2005 and from November 2005 to September 2006.

Treatment January to November 2005 November 2005 to September 2006

y0 (%) ymax (%) D (%) AUDPC (% days) y0 (%) ymax (%) D (%) AUDPC (% days)
INC SEV INC SEV INC SEV INC SEV INC SEV INC SEV INC SEV INC SEV
Cu(OH)21 19a2 0.48a 46b 1.50a 27a 1.02a 7323a 205a 7.50a2 0.14a 21c 0.78b 14c 0.64b 4135d 121d
B1573 24a 0.52a 48b 1.44a 24a 0.92a 6791b 187a 6.67a 0.15a 28c 1.25b 21c 1.10b 4830d 158d
P286 27a 0.63a 37b 1.38a 10a 0.74a 6210b 189a 8.33a 0.20a 39b 1.64b 31b 1.45b 6154c 217c
B25 26a 0.54a 52a 1.92a 26a 1.38a 7591a 222a 7.50a 0.16a 57a 3.19a 49a 3.03a 8567a 359a
B10 28a 0.63a 46b 1.71a 18a 1.08a 7146b 205a 7.50a 0.17a 50a 2.50a 42a 2.33a 7622b 296b
B281 22a 0.57a 38b 1.40a 16a 0.82a 6275b 189a 8.33a 0.12a 53a 2.87a 45a 2.75a 8116b 326b
B175 31a 0.64a 41b 1.60a 10a 0.96a 6634b 191a 7.50a 0.16a 50a 2.83a 43a 2.67a 7979b 327b
B205 24a 0.59a 40b 1.71a 16a 1.12a 6293b 200a 6.67a 0.19a 54a 2.92a 48a 2.73a 7936b 320b
SiCa4 17a 0.43a 57a 2.22a 40a 1.79a 8120a 237a 7.92a 0.15a 53a 2.93a 45a 2.79a 8895a 363a
Control 20a 0.44a 65a 2.65a 44a 2.21a 8624a 255a 7.92a 0.15a 53a 3.02a 45a 2.87a 9042a 385a

Each treatment was monthly sprayed (400 ml/plant) three times in Experiment 1 (January to March 2005) and four times in Experiment 2 (November 2005 to February 2006).
1
Copper hydroxide (0.88 g/plant).
2
Means with the same letter in each column are not different (Scott–Knott test, a = 0.01).
3
B or P, isolates of Bacillus or Pseudomonas (O.D.540 = 0.2 just before being sprayed), respectively.
4
Calcium silicate (0.14 g Si/plant).

Table 2 is assumed that this isolate has a good perspective of commercial

Coffee leaf rust incidence (INC) and severity (SEV) values: initial (y0), maximum use in coffee plantings. Further studies, including toxicological
(ymax), increment (D) and area under disease progress curve (AUDPC). Sprays of four
ones, are required to define strategies for coffee rust management
treatments were compared in one experiment conducted in an organic coffee
plantation in Ervália, MG, from January to November 2007. with isolate B157.
Organic coffee in Brazil is mostly produced by small growers
Treatment y0 (%) ymax (%) D (%) AUDPC (% days)
with family labor. The growers that allowed us to conduct the
INC SEV INC SEV INC SEV INC SEV experiments on their farms were very concerned about the results
Cu(OH)21 4.38a2 0.08a 23c 0.60c 18c 0.52c 3564c 88c as no fungicides were applied to their crops. Nevertheless, at the
B1573 3.75a 0.05a 25c 0.60c 21c 0.55c 3350c 74c end of the experiments, the growers were surprised with the re-
P286 5.00a 0.05a 34b 1.01b 29b 0.96b 4886b 156b
sults and convinced that biocontrol was an effective alternative
Control 5.00a 0.05a 50a 1.80a 45a 1.75a 7114a 259a
to manage coffee rust. Given the peculiarities of organic coffee pro-
1
Copper hydroxide (0.88 g/plant). duction and the farms’ interest in the experiments, it is expected
2
Means with the same letter in each column are not different (Tukey test,
that biological control will be readily accepted by the organic
a = 0.01).
3
B or P, isolates of Bacillus or Pseudomonas (O.D.540 = 0.2 just before being
growers. It is also foreseen that the results gained will not be re-
sprayed), respectively. Each treatment was sprayed (400 ml/plant) monthly (Janu- stricted to organic coffee production and may be used in general
ary to April 2006). coffee crop systems, to support a more sustainable and ecologically
sound crop production.

2006; Bettiol et al., 1994; Bettiol and Várzea, 1992; Porras et al.,
Acknowledgments
1999). In the greenhouse, B. subtilis suspensions sprayed 24 h before
inoculation of plants with H. vastatrix significantly reduced rust
This research has been sponsored by the Conselho Nacional de
severity on coffee plants (Bettiol and Várzea, 1992). Extracts of
Desenvolvimento Científico e Tecnológico (CNPq) and Fundação de
metabolites produced by B. subtilis in growth medium reduced rust
Apoio à Pesquisa do Estado de Minas Gerais (FAPEMIG).
severity by almost 100% when sprayed 24 h before pathogen inocu-
lation (Bettiol et al., 1994). However, there are no published reports
References
of biological control of rust under field conditions. Therefore, our
study is new and opens an opportunity to successful biocontrol of Bettiol, W., Saito, M.L., Brandão, M.S.B., 1994. Control of coffee leaf rust with
coffee leaf rust, mostly for organic producers. Isolates B157 and products based on Bacillus subtilis. Summa Phytopathologica 20, 119–122.
P286 were noticeable in controlling the disease and their perfor- Bettiol, W., Várzea, V.M.P., 1992. Controle biológico da ferrugem (Hemileia vastatrix)
do cafeeiro com Bacillus subtilis em condições controladas. Fitopatologia
mance is likely to improve when properly formulated. Brasileira 17, 91–95.
At the present time, coffee rust control is strongly based on fun- Botelho, A.O., 2006. Estudo do progresso e manejo de doenças foliares (Hemileia
gicides. Under organic systems, only copper based fungicides are vastatrix e Cercospora coffeicola) em cafeeiros durante a transição do sistema de
cultivo convencional para o orgânico. UFLA, MSc Thesis, Lavras.
allowed. This may lead to an accumulation of copper in the soils Caixeta, G.Z.T., Pedini, S., 2002. Comercialização de café orgânico. Informe
(Chaignon et al., 2003) and may be detrimental to human health Agropecuário 23, 149–152.
and the environment (Flemming and Trevors, 1989). Considering Carvalho, V.L., Cunha, R.L., Chalfoun, S.M., 2002. Manejo ecológico das principais
doenças do cafeeiro. Informe Agropecuário 23, 101–114.
these harmful effects, the use of biological control with the B157 Chaignon, V., Sanches-Neira, I., Herrmann, P., Jaillard, B., Hinsinger, P., 2003. Copper
isolate of Bacillus sp. may become a viable alternative to coffee rust bioavailability and extractability as related to chemical properties of
management, given that the control achieved with this organism contaminated soils from a vine-growing area. Environmental Pollution 123,
229–238.
was as effective as with copper hydroxide. It is known that Bacillus
Collins, D.P., Jacobsen, B.J., Maxwell, B., 2003. Spatial and temporal population
spp. are commercially used to control other plant diseases (Fravel, dynamics of a phyllosphere colonizing Bacillus subtilis biological control agent
2005; Raaijmakers et al., 2002). Isolate B157 was obtained from an of sugar beet cercospora leaf spot. Biological Control 26, 224–232.
organic coffee planting, it has been thoroughly evaluated under Flaishman, M.A., Eyal, Z., Zilberstein, A., Voisard, C., Haas, D., 1996. Suppression of
Septoria tritici blotch and leaf rust of wheat by recombinant cyanide-producing
controlled conditions in the past five years, and it performed well strains of Pseudomonas putida. Molecular Plant–Microbe Interactions 9, 642–
in the three-year experiments under field conditions. Therefore, it 645.
 F. Haddad et al. / Biological Control 49 (2009) 114–119 119

Flemming, C.A., Trevors, J.T., 1989. Copper toxicity and chemistry in the Peng, G., Sutton, J.C., 1991. Evaluation of microorganisms for biocontrol of Botrytis
environment: a review. Water, Air, and Soil Pollution 44, 143–158. cinerea in strawberry. Canadian Journal of Plant Pathology 13, 247–257.
Fokkema, N.J.E., 1990. Biological control of aerial plant surfaces. In: Komada, H., Porras, N.C., Leguizamon, C.J.E., Martinez, M.M., 1999. Inducción de resistencia por
Kiritani, K., Bay-Petersen, J. (Eds.), The Biological Control of Plant Diseases. Food Pseudomonas spp., en plántulas de cafe contra la roya del cafeto Hemileia
& Fertilizer Technology Center for the Asian and Pacific Region, Taipei, Taiwan, vastatrix Berk. y Br. Ascolfi-Informa 25, 12–14.
pp. 41–48. Raaijmakers, J.M., Vlami, M., Souza, J.T., 2002. Antibiotic production by bacterial
Fravel, D.R., 2005. Commercialization and implementation of biocontrol. Annual biocontrol agents. Antonie van Leeuwenhoek 81, 537–547.
Review of Phytopathology 43, 337–359. Romero, D., Pérez-García, A., Rivera, M.E., Cazorla, F.M., de Vicente, A., 2004.
Garçon, C.L.P., Zambolim, L., Mizubuti, E.S.G., Vale, F.X.R., Costa, H., 2004. Controle Isolation and evaluation of antagonistic bacteria towards the cucurbit powdery
da ferrugem do cafeeiro com base no valor de severidade. Fitopatologia mildew fungus Podosphaera fusca. Applied Microbiology and Biotechnology 64,
Brasileira 29, 486–491. 263–269.
Haddad, F., Maffia, L.A., Mizubuti, E.S.G., Romeiro, R.S., 2004a. Isolating and Schaad, N.W., Jones, J.B., Chun, W., 2001. Laboratory Guide for Identification of Plant
screening antagonists to Hemileia vastatrix. In: Annual Meeting of the Pathogenic Bacterial. APS Press, St. Paul, Minnesota.
American Phytopathological Society, Anaheim, CA. Phytopathology 94, S37. Shiomi, H.F., Silva, H.S.A., Melo, I.S., Nunes, F.V., Bettiol, W., 2006. Bioprospecting
Haddad, F., Maffia, L.A., Mizubuti, E.S.G., Romeiro, R.S., 2004b. Biocontrol of coffee endophytic bacteria for biological control of coffee leaf rust. Scientia Agricola
leaf rust with antagonists isolated from organic crops. In: Annual Meeting of the 63, 32–39.
American Phytopathological Society, Anaheim, CA. Phytopathology 94, S37. Silva-Acuña, R., Maffia, L.A., Zambolim, L., Berger, R.D., 1999. Incidence-severity
Haddad, F., Maffia, L.A., Mizubuti, E.S.G., Teixeira, H., 2007. Biocontrol as an relationships in the pathosystem Coffea arabica–Hemileia vastatrix. Plant
alternative for leaf rust management in organically-grown coffee. IOBC/WPRS Disease 83, 186–188.
Bulletin 30, 17. Swadling, I.R., Jeffries, P., 1996. Isolation of microbial antagonists for biocontrol of
Harman, E., 2000. Myths and dogmas of biocontrol: changes in perceptions derived grey mould disease of strawberries. Biocontrol Science and Technology 6, 125–
from research on Trichoderma harzianum T-22. Plant Disease 84, 377–393. 136.
Kado, C.I., Heskett, M.G., 1970. Selective media for isolation of Agrobacterium, Teixeira, D.A., Alfenas, A.C., Mafia, R.G., Maffia, L.A., Ferreira, E.M., 2005. Evidências
Corynebacterium, Erwinia, Pseudomonas, and Xanthomonas. Phytopathology 60, de indução de resistência sistêmica à ferrugem do eucalipto mediada por
965–975. rizobactérias promotoras de crescimento de plantas. Fitopatologia Brasileira 30,
Kinkel, L.L., 1997. Microbial population dynamics on leaves. Annual Review of 350–356.
Phytopathology 35, 327–347. Vale, F.X.R., Fernandes Filho, E.I., Liberato, J.R., 2003. QUANT. A software for plant
Madden, L.V., Hughes, G., van den Bosch, F., 2007. The Study of Plant Disease disease severity assessment. In: 8th International Congress of Plant Pathology,
Epidemics. The American Phytopathological Society, St. Paul, USA. Christchurch, New Zealand, p. 105.
MAPA, 2007. Ministério da Agricultura, Pecuária e Abastecimento. Van der Vossen, H.A.M., 2005. State-of-the-art of developing durable resistance to
<www.agricultura.gov.br>. biotrophic pathogens in crop plants, such as coffee leaf rust. In: Zambolim, L.,
Nunes, C.C., Maffia, L.A., Mizubuti, E.S.G., Brommonschenkel, S.H., Silva, J.C., in press. Zambolim, E.M., Várzea, V.M.P. (Eds.), Durable Resistance to Coffee Leaf Rust.
Genetic diversity of populations of Hemileia vastatrix from organic and Suprema Gráfica e Editora, Visconde do Rio Branco, Brasil, pp. 1–29.
conventional coffee plantations in Brazil. Australasian Plant Pathology. Zadoks, J.C., Schein, R.D., 1979. Epidemiology and Plant Disease Management.
Park, Y.-H., Stack, J.P., Kenerley, C.M., 1992. Selective isolation and enumeration of Oxford University Press, New York, USA.
Gliocladium virens and G. roseum from soil. Plant Disease 76, 230–235. Zambolim, L., Vale, F.X.R., Pereira, A.A., Chaves, G.M., 1997. Café (Coffea arabica L.),
Pedini, S., 2000. Produção e certificação de café orgânico. In: Zambolim, L. (Ed.), controle de doenças. In: Vale, F.X.R., Zambolim, L. (Eds.), Controle de doenças de
Café: Produtivdade, qualidade e sustentabilidade. Editora Suprema Gráfica e plantas: grandes culturas, vol. 1. Suprema Gráfica e Editora, Visconde do Rio
Editora, Visconde do Rio Branco, Brasil, pp. 333–360. Branco, Brasil, pp. 83–140.

View publication stats