Accepted Manuscript

Title: Saccharomyces cerevisiae: a novel and efficient

biological control agent for Colletotrichum acutatum during
pre-harvest

Author: Marcos Roberto Lopes Mariana Nadjara Klein

Luriany Pompeo Ferraz Aline Caroline da Silva Katia Cristina
Kupper

PII: S0944-5013(15)00057-9
DOI: http://dx.doi.org/doi:10.1016/j.micres.2015.04.003
Reference: MICRES 25770

To appear in:

Received date: 6-3-2015

Revised date: 1-4-2015
Accepted date: 4-4-2015

Please cite this article as: Lopes MR, Klein MN, Ferraz LP, Silva AC, Kupper
KC, Saccharomyces cerevisiae: a novel and efficient biological control agent
for Colletotrichum acutatum during pre-harvest, Microbiological Research (2015),
http://dx.doi.org/10.1016/j.micres.2015.04.003

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 1 Saccharomyces cerevisiae: a novel and efficient biological control agent for

2 Colletotrichum acutatum during pre-harvest

4 Marcos Roberto Lopesa, Mariana Nadjara Kleinb, Luriany Pompeo Ferrazb, Aline

t
ip
5 Caroline da Silvaa, Katia Cristina Kupper a,b,c*

cr
a
6 Universidade Federal de São Carlos, CEP 13600-970, Araras, SP, Brazil.
b
7 Universidade Estadual Paulista “Julio de Mesquita Filho”, CEP4884-900,

us
8 Jaboticabal/SP, Brazil.

an
c
9 Sylvio Moreira Citriculture Center/IAC, Laboratory Plant Pathology and Biological

10 Control - CEP 13490-970, Cordeirópolis/SP, Brazil.

M
11
d
p te
ce

12 ∗Corresponding author at: Laboratory Plant Pathology and Biological Control - Sylvio
Ac

13 Moreira Citriculture Center/IAC – Rodovia Anhanguera, Km 158 - CEP 13490-970,

14 Cordeirópolis/SP, Brazil. Tel.: +55 (19)3546-1399. E-mail address:

15 katia@centrodecitricultura.br (K.C. Kupper)

16

1
Page 1 of 32
17 Abstract

18 In this study, we evaluated the efficiency of six isolates of Saccharomyces cerevisiae in

19 controlling Colletotrichum acutatum, the causal agent of postbloom fruit drop that occur

20 in pre-harvest citrus. We analyzed the mechanisms of action involved in biological

t
ip
21 control such as: production of antifungal compounds, nutrient competition, detection of

22 killer activity, and production of hydrolytic enzymes of the isolates of S. cerevisiae on

cr
23 C. acutatum and their efficiency in controlling postbloom fruit drop on detached citrus

us
24 flowers. Our results showed that all six S. cerevisiae isolates produced antifungal

25 compounds, competed for nutrients, inhibited pathogen germination, and produced

an
26 killer activity and hydrolytic enzymes when in contact with the fungus wall. The

27 isolates were able to control the disease when detached flowers were artificially
M
28 inoculated, both preventively and curatively. In this work we identified a novel potential

29 biological control agent for C. acutatum during pre-harvest. This is the first report of
d

30 yeast efficiency for the biocontrol of postbloom fruit drop, which represents an
te

31 important contribution to the field of biocontrol of diseases affecting citrus populations

32 worldwide.
p
ce

33 Keywords: detached citrus flowers, antifungal compounds, Killer yeast, hydrolytic

34 enzymes, postbloom fruit drop.

Ac

35

36 1. Introduction

37 Postbloom Fruit Drop (PFD), caused by the fungus Colletotrichum acutatum

38 Simmonds, is one the most devastating diseases affecting citrus. Only flower tissues of citrus

39 are infected and develop postbloom fruit drop (Figure 1A). Natural citrus fruit

40 abscission usually occurs at the base of the peduncle at the attachment to the stem, but

2
Page 2 of 32
41 with postbloom fruit drop, abscission is at the base of the fruit leaving persistent calyces

42 and peduncles (Figure 1B) (Peres et al., 2005). When conditions are favorable during bloom

43 (i.e., rain), the fungus can disperse and cause greater damage to the crop (Silva-Junior et al.

44 2014). In general, this disease is controlled with protectant and systemic fungicide sprays (Goes

t
45 et al. 2008). Due to the numerous bloom cycles in certain commercial citrus varieties, the

ip
46 fungicidal method requires numerous sprays, which increases production costs and causes

47

cr
significant environmental risks. Therefore, new and effective control methods are needed that

48 satisfy consumer needs and are also safe for workers and the environment.

us
49 One alternative method studied under field conditions in recent years is the

50 bacterium Bacillus subtilis, which acts on a curative control mechanism (Klein et al.

51

52 an
2013; Kupper et al. 2012). In the current work, we were interested in identifying an

alternative method that could act preventively. Species of yeast, which act by competing
M
53 for nutrients and space, have been used as biocontrol agents against various pathogens,

54 but only those that occur in postharvest fruits (Droby et al. 2009; Kupper et al. 2013;
d

55 Moretto et al. 2014; Talibi et al. 2014). Furthermore, S. cerevisiae has been shown to
te

56 act as an important biocontrol agent against plant pathogens (Nally et al., 2012; Platania
p

57 et al., 2012; Gouvea et al. 2009).

ce

58 Yeast rely on antagonistic mechanisms of action to exert biological control. In

59 general, many yeast species compete for nutrients and space, induce resistance and
Ac

60 secrete lytic enzymes (Droby et al. 2009; Zhang et al. 2011). Johnson and Stockwell

61 (2000) reported that the antagonistic efficiency of microorganisms in flowers can be

62 attributed to their competition for nutrients, and in some cases, to the production of

63 compounds with an inhibitory action against pathogens. Fialho et al. (2010)

64 demonstrated that the production of volatile organic compounds by Saccharomyces

65 cerevisiae plays an essential role in the antagonistic activity on Guignardia citricarpa.

3
Page 3 of 32
66 Some yeast exhibit killer activity, which is defined as the ability to secrete

67 proteins or glycoprotein killer toxins that generally are lethal to other species of yeasts,

68 pathogenic bacteria and also to cells of filamentous fungi through different mechanisms,

69 including the hydrolysis of the major cell wall component β-1,3-glucans (Schmitt and

t
70 Breinig, 2006; Coelho et al. 2007; Hashem and Alamri, 2009; Muccilli et al., 2013).

ip
71 The objectives of this study were to evaluate six yeast isolates of S. cerevisiae

cr
72 for the possible biological control of PFD caused by C. acutatum in citrus, and to

73 determine the mechanisms of action that are involved in this biocontrol. In the case of

us
74 most citrus pathogens detected post-harvest, infection occurs pre-harvest. Therefore, a

yeast-based product that acts during the pre-harvest period would be particularly

an
75

76 important for diseases such as PFD, which occurs pre-harvest. Such a yeast would also
M
77 reduce the production of initial inoculum of pathogens, which would also indirectly help

78 control diseases that occur during the post-harvest period.

d

79
te

80 2. Material and methods

p

81 2.1. Microorganisms
ce

82 Six industrial yeast strains of Saccharomyces cerevisiae (ACB-CR1, ACB-KD1,

83 ACB-CAT1, ACB-BG1, ACB-K1 and ACB-PE2) isolated from the fermentation

Ac

84 process for ethanol production, characterized by electrophoretic karyotyping, and stored

85 in the collection of the Laboratory of Biochemistry and Plant Pathology at the

86 University if São Paulo (ESALQ), Piracicaba-São Paulo, Brazil (Fialho et al., 2010) and

87 later deposited in the microorganisms collection of the Apta Center Citrus “Sylvio

88 Moreira”/IAC, Cordeirópolis, São Paulo State, Brazil. The Colletotrichum acutatum

89 strain used in the present study was obtained from the same collection.

4
Page 4 of 32
 90

91 2.2. Evaluation of in vitro antagonism

92 C. acutatum was co-cultured on Petri dishes with each yeast isolate studied. Petri

93 dishes with PDA medium were inoculated with a 7-mm diameter disk of an actively

t
ip
94 growing fungal mycelium near the disk edge. Another disk of yeast (cultured for 48 h),

95 of the same size, was inoculated at the opposite edge. The control corresponded to the

cr
96 growth of the fungus without yeast.

us
97 The experiment was performed with a completely randomized design and five

98 replicates, incubating the cultures at 27°C for 18 days with a 12 h photoperiod. Mycelial

an
99 diameter of C. acutatum was measured in two perpendicular directions. The data were

100 statistically analyzed by analysis of variance (ANOVA) and Tukey’s test at 5%

M
101 significance. The experiment was repeated at least twice.

102
d

103 2.3. Production of antifungal compounds by yeast isolates

te

104

105 2.3.1. Production of volatile antifungal compounds

p

106 To assess the production of volatile compounds, each of the yeast isolates was
ce

107 simultaneously cultivated with C. acutatum on split plates, which prevented nonvolatile
Ac

108 compounds produced by the yeast from reaching the fungus. A C. acutatum culture disk

109 (7 mm in diameter) was placed on PDA medium on one side of the split plate. On the

110 other side, a disk of same size of the yeast strain (cultured for 48 h) was placed in the

111 medium, and the plates were hermetically sealed. After incubating the fungus and

112 different yeast strains at 25°C for 7 days, fungal growth was measured as the mycelial

113 diameter in the presence of yeast relative to the mycelial diameter in the absence of

114 yeast (control).

5
Page 5 of 32
115

116 2.3.2. Production of cell-free antifungal compounds in yeast

117 For each yeast isolate tested, a loop of inoculum from a 48 h culture was

118 transferred to a 250 mL Erlenmeyer flask containing 50 mL of potato dextrose broth

t
119 (PDB), followed by incubation at 150 rpm for 72 h in the dark.

ip
120 Using a protocol adapted from the technique described by Frighetto and Melo

cr
121 (1995), each yeast culture was filtered through Whatman Nº 4 filter paper and a

122 0.45-µm Millipore® membrane after the incubation period to remove the yeast cells. For

us
123 each yeast cell-free filtrate, a 10 mL aliquot was added to 90 mL of PDA medium and

poured into Petri dishes. After solidification, a 7 mm C. acutatum culture disk was

an
124

125 placed in the center of each Petri dish. For the control, C. acutatum was grown in PDA
M
126 medium containing sterile water instead of the yeast filtrate. The cultures were

127 incubated in a Biochemical Oxygen Demand (BOD) chamber at 25 °C for 9 days.

d

128 Mycelial diameter of C. acutatum was measured in two perpendicular directions.

te

129

130 2.3.3. Production of thermostable antifungal compounds

p

131 For each yeast isolate tested, a yeast culture disk was transferred to a 250 mL
ce

132 Erlenmeyer flask containing 50 mL PDB, followed by incubation at 150 rpm for 72 h in

133 the dark, as described before. Thereafter, 10 mL aliquots of each isolate were
Ac

134 transferred to vials containing 90 mL PDA medium and sterilized by autoclaving at 121

135 °C for 20 min. Each sterilized medium was poured into a Petri dish, and, following

136 solidification, a 7 mm C. acutatum culture disk was transferred to the center of each

137 plate. For the control, C. acutatum was grown on PDA medium containing sterile water

138 instead of the metabolites. The cultures were incubated in a BOD chamber at 25 °C for

139 7 days. Mycelial growth of C. acutatum was measured in two perpendicular directions.

6
Page 6 of 32
140

141 2.4. Effect of glucose on C. acutatum conidial germination at presence of yeast

142 isolates

143 For the assessment of nutrient competition between C. acutatum and S.

t
144 cerevisiae, agar-coated microscope slides were prepared with varying glucose

ip
145 concentrations (0.0%, 0.5%, 1.0%, 1.5%, 2.0%, and 2.5%), using a methodology

cr
146 adapted from Kupper et al. (2013). For each yeast isolate tested, 10 μL of a C. acutatum

147 suspension (1.0 x 104 conidia/mL) and 10 μL of a yeast suspension (1.0 x 106 cells/mL)

us
148 were spotted onto the slides, and the cultures were incubated in a BOD chamber at 25

°C for 15 h. Nutrient competition was assessed by counting the number of germinated

an
149

150 and ungerminated conidia among 100 randomly selected conidia. Conidia were
M
151 considered fully germinated when the length of the germ tube was at least the size of the

152 turgid conidia.

d

153
te

154 2.5. Detection of killer activity

155 A 1.0 x 105 cells/mL suspension of S. cerevisiae NCYC 1006 (killer-factor

p

156 sensitive) was prepared by culturing in YEPD medium (1% yeast extract, 2% peptone,
ce

157 2% glucose, 2% agar, 0.01% ampicillin, 0.01% nalidixic acid) at 28°C for 24 h. A 100

158 µL aliquot of the suspension was transferred to a Petri dish containing YEPD-methylene
Ac

159 blue medium buffered at pH 4.3-4.7, according to the methodology described by

160 Ceccato-Antonini et al. (2004). To evaluate the presence of killer factors, each yeast

161 isolate was spotted on separate Petri dishes with sterile toothpicks, and the cultures were

162 incubated at 28 °C for 3 days. The production of killer factors and the death of sensitive

163 cells were indicated by the presence of a growth inhibition zone and an adjacent blue

164 zone.

7
Page 7 of 32
165

166 2.6. Production of hydrolytic enzymes (β-1, 3-glucanase and chitinase)

167 The production and secretion of hydrolytic enzymes by the yeast isolates were

168 analyzed according to the methodology of Fialho et al. (2004). For each yeast tested, a

t
169 loopful of isolate was transferred to 20 mL liquid YEPD broth and incubated at 150 rpm

ip
170 for 72 h in the dark. Thereafter, 1 mL of each suspension was transferred to 15 mL

cr
171 Falcon tubes containing 10 mL YEPD or modified YEPD medium (containing C.

172 acutatum cell wall preparation in place of 1% glucose). The cultures were prepared in

us
173 triplicate and incubated at 150 rpm. After a 24 h incubation period, a 1.5-mL volume

was aliquoted and centrifuged at 3000 rpm for 10 min. The supernatant was recovered

an
174

175 and used for the quantification of β-1,3-glucanase and chitinase.

M
176 A protocol adapted from Bar-Simon et al. (2004) was used for the cell wall

177 preparation. Briefly, 1 mL of a C. acutatum suspension (1.0 x 105 spores/mL) was

d

178 cultured in 50 mL PDB medium at 150 rpm for 8 days. Mycelial contents were

recovered by filtration through Whatman No.1 filter paper then washed three times with
te

179

180 distilled water, homogenized in 0.1 M phosphate buffer (pH 7.2) for 2 min, and stored
p

181 at -20 °C overnight. After thawing and homogenization, approximately 20 mL of the

ce

182 fungal mycelia was transferred to a vial and macerated in liquid nitrogen, and the fungal

183 cell wall preparation was stored at -80 °C until use.

Ac

184

185 2.6.1. Quantification of reducing sugars

186 Reducing sugars released during enzymatic activity assays were quantified using

187 the 3,5-dinitrosalicylic acid (DNS) method described by Miller (1959). A 250 μL

188 volume of the reaction mixture was added to 250 μL DNS, and the solution was heated

189 in a boiling water bath for 10 min. After cooling on ice to a temperature of 25 °C, the

8
Page 8 of 32
190 solution was diluted with 2.5 mL distilled water and homogenized. Absorbance was

191 read at 540 nm using a reagent blank.

192

193 2.6.2. Quantification of β-1, 3-glucanase enzymatic activity

t
194 To assess the production of β-1,3-glucanases, a colorimetric assay was used to

ip
195 quantify glucose released from the laminarin substrate, together with the method for the

cr
196 quantification of reducing sugars. The reaction was performed using 200 μL McIlvaine

197 buffer (pH 6.0), 100 μL culture sample, and 100 μL laminarin (4 mg/mL). The reaction

us
198 was incubated at 50 °C for 1 h and then stopped using 200 μL DNS for the reducing

sugar quantification. Absorbance readings at 540 nm were subtracted from the

an
199

200 absorbance of the reaction mixture in the presence of a buffer solution (in place of the
M
201 culture medium). Additionally, the absorbance of a negative control (a buffer solution in

202 place of the substrate) was subtracted from each experimental reading. Absorbance
d

203 values were plotted on a standard curve of glucose, and enzymatic activity was
te

204 expressed in U/L, where a unit of activity (U) was defined as 1.0 g of reducing sugar

205 (glucose) released from laminarin under the assay conditions used.
p

206
ce

207 2.6.3. Quantification of chitinase enzymatic activity

208 Chitinase production was quantified as the amount of N-acetyl glucosamine

Ac

209 (NAG) released from glycol chitin substrate. Briefly, 100 μL of each yeast culture was

210 mixed with 200 μL McIlvaine buffer (pH 6.0) and 100 μL 0.01% glycol chitin (w/v) in

211 the same buffer. After incubation at 50 °C for 60 min, the reaction was stopped with

212 200 μL DNS, and reducing sugar quantification was performed as previously described

213 (section 2.9.1). A solution containing the reaction mixture combined with buffer

214 solution (in place of culture medium) was used as the reagent blank. Absorbance

9
Page 9 of 32
215 readings were also subtracted from a negative control reading (buffer solution in place

216 of the glycol chitin substrate). Enzymatic activity was expressed in U/L, where U was

217 defined as 1.0 g of reducing sugar (N-acetylglucosamine) enzymatically released from

218 glycol chitin under the assay conditions used.

t
219

ip
220

cr
221

222 2.7. Statistical analysis

us
223 To quantify antifungal compound production a completely randomized design

with five replicates was used. The data were subjected to analysis of variance

an
224

225 (ANOVA), and the mean values were compared using Tukey’s test at a 5% significance
M
226 level. All assays were performed in duplicate, and ASSISTAT software was used for the

227 statistical analysis. We showed the data from one experiment, because there were no
d

228 differences statistically significant between them.

te

229 To assess nutrient competition, a factorial design with eight replicates per

230 treatment was used. The mean values for each treatment were compared using Tukey’s
p

231 test at a 5% significance level, and the analyses were performed using ASSISTAT
ce

232 software.

233
Ac

234 2.8. In vivo efficiency of Saccharomyces cerevisiae in controlling Colletotrichum

235 acutatum on detached citrus flowers

236 Detached Valência sweet orange (Citrus sinensis) flowers were placed in plastic

237 boxes (Gerbox®), with the stems inserted in holes made into 0.5-cm thick synthetic

238 foam over filter paper that was soaked with sterile distilled water. The boxes were

10
Page 10 of 32
239 placed 70 cm below two germicidal lamps (Sankyo Denki G30T8, 30 watts) for 20

240 minutes before antagonist application and pathogen inoculation (Moretto et al. 2001).

241 Previously, colonies of each microorganism were cultivated in PDA. To apply

242 the biocontrol agents, an aliquot of 10 μL of suspension (1 x 107 cfu mL-1), was applied

t
ip
243 on each petal. The flowers were treated for 24 h before and 24 h after inoculation with

244 the pathogen (1 x 104 conidia mL-l). Flowers sprayed with conidial suspension of C.

cr
245 acutatum or with water were used as controls.

us
246 Each treatment was replicated three times, and each replication consisted of one

247 Gerbox® with 10 flowers. The boxes were maintained in a growth chamber at 22 ºC and

an
248 with a 12 h photoperiod. The evaluation was done 72 h after inoculation with the

249 pathogen by determining the percentage of healthy flowers. We considered as disease

M
250 flowers those that presented orange to peach colored lesions on the petals. The data

251 were analyzed by ANOVA using the statistical program ASSISTAT 7.6, and the means
d

252 were compared by Tukey’s test at 5% significance. The experiment was repeated at
te

253 least twice, but we showed the data from one experiment, because there were no
p

254 differences statistically significant between them.

ce

255

256 3. Results
Ac

257

258 3.1. Effect of Saccharomyces cerevisiae isolates on mycelial growth of C. acutatum

259 When we co-cultured C. acutatum in Petri dishes with each S. cerevisiae yeast

260 isolate, we observed that all yeast isolates inhibited mycelial growth of C. acutatum

261 relative to the control treatment (Figure 2). The ACB-CAT1 and ACB-CR1 isolates

11
Page 11 of 32
262 promoted the most inhibitions values correspond to 71% and 67%, respectively.

263 Moreover, the ACB-PE2 and ACB-K1 isolates were the least effective.

264

265 Fig. 2

t
ip
266

cr
267 3.2. Production of antifungal compounds by yeast isolates

268 The ACB-KD1, ACB-CAT1 and ACB-BG1 isolates produced volatile

us
269 compounds that inhibited the development of the pathogen colony, with colony

270 inhibition values ranging from 18-25% (Figure 3). All of the isolates produced cell-free

271

272 an
antifungal substances which inhibited the development of the pathogen colony, with

inhibition values that ranged from 11 to 37%, and with the ACB-CR1 isolate having the
M
273 highest inhibition rate. ACB-PE2 and ACB-BG1 were the only two isolates to produce

274 substances that endured high temperatures.

d
te

275
p

276 Fig. 3
ce

277

278 3.3. Effect of glucose on C. acutatum conidial germination at presence of yeas

Ac

279 isolates

280 When we assess the effect of glucose concentrations on C. acutatum conidial

281 germination in the presence of yeast isolates, our study showed that although all of the

282 isolates evaluated affected the germination of the pathogen, there were significant

283 interactions among the different glucose concentrations and S. cerevisiae isolates tested.

284 We observed that the addition 0.5% glucose favored the antagonistic activity, increasing

12
Page 12 of 32
285 the percentage of inhibition from 41 to 87%, depending on the yeast isolates (Tables 1

286 and 2).

287

288 Table 1

t
ip
289 Table 2

cr
290

us
291 3.4. Detection of killer activity

an
292 All six yeast isolates exhibited killer activity, they produced blue inhibition

293 zones, which are indicative of cell death, as showed in the Table 3 and illustrated in
M
294 Figure 4. The results showed considerable killer activity against S. cerevisiae NCYC

295 1006 (killer-factor sensitive), suggesting that these isolates could be potential as
d

296 biocontrol agents.

te

297

298 Table 3
p
ce

299 Fig. 4

300
Ac

301 3.5. Production of hydrolytic enzymes

302 To assess the production of hydrolytic enzymes produced by S. cerevisiae as a

303 possible mechanism of action against C. acutatum, colorimetric assays were performed

304 to quantify the reducing sugars (RS) released from the substrate using the pathogen cell

305 wall as a carbon source. All yeast isolates showed activity for β-1,3-glucanase and

306 chitinase in all conditions (Table 4).

13
Page 13 of 32
307

308 Table 4

309

310 3.6. In vivo assay of detached flowers

t
ip
311 Saccharomyces cerevisiae isolates were applied to the petals of detached citrus

312 flowers to evaluate their biocontrol activity against PFD. The isolates significantly

cr
313 decreased the incidence of infection by C. acutatum (Figure 5).

us
314 All yeast isolates had both curative and preventive biocontrol activity: 73 to 84%

315 of flowers were asymptomatic when the treatment was curative, and 50 to 86% were

an
316 asymptomatic when the treatment was preventive. When the flowers were untreated and

317 inoculated with the pathogen, the data showed 100% of diseased flowers.
M
318

319 Fig. 5
d

320
te

321 4. Discussion
p
ce

322 In this study, we evaluated the efficiency of six isolates of S. cerevisiae in

323 controlling C. acutuam, the causal agent of PFD in pre-harvest citrus, and analyzed the
Ac

324 mechanisms of action involved in the antagonistic activity. Our results showed that S.

325 cerevisiae was able to control the disease when detached flowers were artificially

326 inoculated, both preventively and curatively. The findings of this study converge with

327 work by other authors, who have reported the successful biological control of other

328 fungal diseases by using yeast species (Chanchaichaovivat et al. 2008; Weiler and

329 Schmitt 2003).

14
Page 14 of 32
330 All six S. cerevisiae isolates produced antifungal compounds, competed for

331 nutrients, inhibited pathogen germination, and produced killer activity and hydrolytic

332 enzymes.

333 While all isolates produced cell-free substances, only two (ACB-PE2 and ACB-

t
ip
334 BG1) produced metabolites that withstood the effects of high temperature, and three

335 other isolates (ACB-KD1, ACB-CAT1 and ACB-BG1) also produced volatile

cr
336 compounds (Figure 3). Metabolites from different yeasts species were found have had

us
337 no significant inhibitory effects on the pathogens. For example, Wang et al. (2010)

338 found that thermostable and cell-free metabolites produced by Rhodosporidium

an
339 paludigenum did not significantly inhibit germination of conidia of Botrytis cinerea,

340 and Zhang et al. (2011) found that cell-free metabolites produced by Pichia
M
341 guilliermondii did not inhibit B. cinerea.

342 Fialho et al. (2010) evaluated the same yeast isolates described in our work for
d

343 the control of the citrus fungal pathogen Guignardia citricicarpa and verified the
te

344 production of volatile organic compounds, reporting that ACB-CR1 reduced mycelial

345 growth of the pathogen by 87%. Although ACB-CR1 in our study produced volatile
p

346 compounds, these were not sufficient to significantly inhibit the colony size of C.
ce

347 acutatum, showing the difference of sensibility between plant pathogen microorganisms
Ac

348 to the produced volatile compounds by S. cerevisiae. Furthermore, the susceptibility of

349 a pathogen to yeasts may vary according to the chemical nature and mechanism of

350 action of the antifungal compounds produced (Walker et al. 1995).

351 The competition between microorganisms for essential environmental factors

352 such as nutrients presents itself as one of the fundamental mechanisms of biological

353 control in studies of yeast-plant pathogen interactions (Zhang et al. 2011; Kupper et al.

354 2013). In our study, we observed that the addition of 0.5% glucose favored the

15
Page 15 of 32
355 antagonistic activity of yeast, which further inhibited pathogen germination. In a

356 previous study from our own group (Kupper et al. 2013) working with the same S.

357 cerevisiae isolates, we observed that ACB-K1 and ACB-CR1 were more efficient at

358 inhibiting the germination of Penicillium digitatum, at rates of 78% and 85.7%,

t
359 respectively, and that the addition of 0.5% glucose further favored the inhibition of

ip
360 conidia germination. Similarly, Zhang et al. (2011) tested different sources of sugar

cr
361 (i.e., sucrose, fructose and glucose) and found that concentrations of 0.5 and 1%

362 increased the antagonistic activity of Pichia guilliermondii without benefiting the

us
363 pathogen, which was also reported in the current study. Finally, Chanchaichaovivat et

al. (2008) evaluated possible mechanisms of action by the yeast P. guilliermondii in the

an
364

365 control of anthracnose (Colletotrichum capsici) in pepper, and observed that increased
M
366 sugar concentrations reduced the germination of conidia of the fungus.

367 In our study, all S. cerevisiae isolates presented killer activity, inhibiting the
d

368 growth of a sensitive yeast strain. Similarly, Lutz et al. (2013) found that Cryptoccocus
te

369 albidus, Pichia membranifaciens and Cryptoccocus victoriae exhibited secretion of

370 killer toxins against Penicillium expansum and B. cinerea pathogens. In a study of 580
p

371 yeast isolates, Lima et al. (2013) reported that only 5% of these isolates showed killer
ce

372 activity against Colletotrichum gloeosporioides in papaya.

373 In this study, all S. cerevisiae isolates produced β-1,3-glucanase and chitinase
Ac

374 when they were cultured in medium supplemented with C. acutatum cell wall

375 fragments. Similar results were obtained by Zhang et al. (2011), who demonstrated the

376 high capacity of P. guilliermondii to control B. cinerea, reporting a high production of

377 lytic enzymes, especially β-1,3-glucanase and chitinase. Studies have shown that

378 extracellular lytic enzymes and B. subtilis antibiotics are important for the biocontrol of

379 plant pathogens, acting alone or together in the degradation of fungal cell walls

16
Page 16 of 32
380 (Leelasuphakul et al. 2006; Tweddell et al. 1993). In this work, we obtained similar

381 results for the biocontrol of C. acutatum with S. cerevisiae, which, in addition to

382 hydrolytic enzymes, produced several other metabolites that may act synergistically in

383 pathogen control. Possibly, these results might indicate that the bioactive compounds

t
384 produced by S. cerevisiae isolates could be applied in vivo as biofungicides for disease

ip
385 control.

cr
386 Due to harsh environmental conditions, one of the most critical steps to

387 implementing an efficient biological control is for the agent to establish itself on the

us
388 living plant tissue, e.g., the citrus flower surface. Agents with one or more mechanisms

of antagonistic action against the pathogen, including the production of antifungal

an
389

390 compounds in a short period of time (e.g., during citrus flowering), may harbor an
M
391 advantage over other types of agents.

392 However, for perennial crops such as citrus, where the inoculum (i.e., C.
d

393 acutatum) survives from one crop to the next, an effective biological control requires a
te

394 more specific and aggressive control strategy, one that can act throughout the entire

395 crop period in order to act preventively and not just curatively. Thus, yeast that compete
p

396 for space and nutrients may have more effective preventive control over the disease. In
ce

397 this study, all S. cerevisiae isolates were effective in controlling the disease both

398 preventively and curatively, and most isolates exhibited multiple modes of action.
Ac

399 Maybe the S. cerevisiae ACB-K1 isolate shouldn't be chosen, because it presented

400 lowest efficiency in vivo conditions as preventive control mechanism and no notable

401 results about its mode of action.

402 In future studies, we will further explore how yeast isolates interact on floral

403 surfaces and whether these microorganisms can be manipulated to enhance their

404 efficacy as controls. Further studies are needed to identify the best conditions for

17
Page 17 of 32
405 increasing isolate production of antifungal compounds and killer activity, with the aim

406 of developing future biocontrol products. These novel bioproducts could potentially

407 reduce the production of initial inoculum of pathogens, which would also indirectly help

408 control diseases that occur during the post-harvest period. This work represents an

t
409 important contribution to the field of biocontrol of pre-harvest diseases affecting citrus

ip
410 populations worldwide.

cr
411

us
an
M
d
p te
ce
Ac

18
Page 18 of 32
411 References

412 Bar-Shimon M, Yehuda H, Cohen L, Weiss B, Kobeshnikov A, Daus A, Goldway M,

413 Wisniewski M, Droby S. Characterization of extracellular lytic enzymes produced

414 by the yeast biocontrol agent Candida oleophila. Curr Genet 2004; 45: 140-148.

t
ip
415 Ceccato-Antonini SR, Tosta CD, Silva AC. Determination of yeast Killer activity in

416 fermentation sugarcane juice using selected ethanol-making strains. Braz Arch

cr
417 Biol Technol 2004; 47: 13-23.

us
418 Chanchaichaovivat A, Bhinyo P, Ruenwongsa P. Putative modes of action of Pichia

419 guilliermondii strain R13 in controlling chilli anthracnose after harvest. Biol

an
420 Control 2008; 47: p.207-215.

421 Coelho AR, Celli, MG, Ono EYS, Wosiacki G. Penicillium expansum versus antagonist
M
422 yeasts with perspectives of application in biocontrol and patulin degradation. Braz

423 Arch Biol Technol 2007; 50: 725–733.

d
te

424 Dennis C, Webster J. Antagonistic properties of species groups of Trichoderma III.

425 Hyphal interactions. Tr Brit Mycol Soc 1971; 57: 359-363.

p
ce

426 Droby S, Wisniewski M, Macarisin D, Wilson C. Twenty years of postharvest

427 biocontrol research: Is it time for a new paradigm? Postharvest Biol Technol 2009;
Ac

428 52: 137–145.

429 Fialho MB. Efeito in vitro de Saccharomyces cerevisiae sobre Guignardia citricarpa,

430 agente causal da pinta preta dos citros. Dissertation, Escola Superior de Agricultura.

431 ‘‘Luiz de Queiroz’’, Universidade de São Paulo, 2004.

432 Fialho MB, Toffano L, Pedroso MP, Augusto F, Pascholati SF. Volatile organic

433 compounds produced by Saccharomyces cerevisiae inhibit the in vitro development

19
Page 19 of 32
434 of Guignardia citricarpa, the causal agent of citrus black spot. World J Microbiol

435 Biotechnol 2010; 26: 925-932.

436 Frighetto RTS, Melo IS. Produção de antibióticos por microrganismos. In: Melo IS;

437 Sanhueza RMV. (Coords.) Métodos de seleção de microrganismos antagônicos a

t
ip
438 fitopatógenos. Jaguariúna: EMBRAPA-CNPMA, 1995; 40-46.

cr
439 Goes A, Garrido RBO, Reis RF, Baldassari RB, Soares MA. Evaluation of fungicide

440 applications to sweet orange at different flowering stages for control of postbloom

us
441 fruit drop caused by Colletotrichum acutatum. Crop Protect 2008; 27: 71-76.

an
442 Gouvea A, Kuhn OJ, Mazaro SM, Mio LLM, Deschamps C, Biasi LA, Fonseca VC.

443 Controle de doenças foliares e de flores e qualidade pós-colheita do morangueiro

M
444 tratado com Saccharomyces cerevisiae. Hortic Bras 2009; 27: 527-533.

445 Guetsky R, Shtienberg D, Elad Y, Dinoor A Combining biocontrol agents to reduce the
d

446 variability of biological control. Phytopathology 2001; 91:621-627.

te

447 Hashem M, Alamri S. The biocontrol of postharvest disease (Botryodiplodia

p

448 theobromae) of guava (Psidium guajava L.) by the application of yeast strains.
ce

449 Postharvest Biol Technol 2009; 53: 123–130.

450 Johnson KB, Stockwell VO. Biological control of fire blight. Fire Blight: the Disease
Ac

451 and its Causative Agent, Erwinia amylovora (ed. J.L. Vanneste) 2000; 319-337.

452 Klein MN, Silva AC, Lopes MR, Kupper KC. Application of microorganisms, alone or

453 in combination, to control postbloom fruit drop in citrus. Trop Plant Pathol, 2013;

454 38, 505-512.

20
Page 20 of 32
455 Kupper KC, Corrêa EF, Azevedo FA de, Silva AC. Bacillus subtilis to biological

456 control of postbloom fruit drop caused by Colletotrichum acutatum under field

457 conditions. Sci. Horticulturae 2012; 134: 139-143.

458 Kupper KC, Cervantes ALL, Klein MN, Silva AC. Avaliação de microrganismos

t
459 antagônicos, Saccharomyces cerevisiae e Bacillus subtilis para o controle de

ip
460 Penicillium digitatum. Rev Bras Frutic 2013; 35: 425-436.

cr
461 Leelasuphakul W, Sivanunsakul P. Phongpaichit S. Purification, characterization and

us
462 synergistic activity of β-1,3-glucanase and antibiotic extract from an antagonistic

463 Bacillus subtilis NSRS 89-24 against rice blast and sheath blight. Enzyme and

an
464 Microb Technol 2006; 38: 990–997.

465 Lima JR, Gonçalves, LRB, Brandão LR, Rosa CA, Viana FMP. Isolation, identification,
M
466 and activity in vitro of killer yeasts against Colletotrichum gloeosporioides isolated

467 from tropical fruits. J Basic Microbiol 2013; 53: 590–599.

d
te

468 Lutz MC, Lopes CA, Rodriguez, ME, Sosa MC, Sangorrín MP. Efficacy and putative

469 mode of action of native and commercial antagonistic yeasts against postharvest
p

470 pathogens of pear. Int J Food Microbiol 2013; 164: 166-172.

ce

471 Miller GH. Use of dinitrosalicylic acid reagent for determination of reducing sugar.
Ac

472 Anal Chem 1959; 31: 426-429.

473 Moretto CM, Cervantes ALL, Filho AB, Kupper KC. Integrated control of green mold

474 to reduce chemical treatment in post-harvest citrus fruits. Sci. Horticulturae 2014;

475 165: 433–438.

476 Moretto KCK, Gimenes-Fernandes N, Santos JM. Influence of Trichoderma spp. on

477 Colletotrichum acutatum mycelial growth and morphology and on infection of

478 “Tahiti” lime detached flowers Summa Phytopathol 2001; 27: 357-364.

21
Page 21 of 32
479 Muccilli S, Wemhoff S, Restuccia C, Meinhardt F. Exoglucanase-encoding genes from

480 three Wickerhamomyces anomalus killer strains isolated from olive brine Yeast

481 2013; 30: 33–43.

482 Nally MC, Pescea VM, Maturanoa YP, Mu˜noze CJ, Combinab M, Toroa ME,

t
483 Castellanos de Figueroa LI, Vazqueza F. Biocontrol of Botrytis cinerea in table

ip
484 grapes by non-pathogenic indigenous Saccharomyces cerevisiae yeasts isolated

cr
485 from viticultural environments in Argentina Postharvest Biol Technol 2012; 64:

486 40–48.

us
487 Peres NA, Timmer, LW, Adaskaveg JE, Correll JC. Lifestyles of Colletotrichum

an
488 acutatum Plant Dis 2005; 89:784-796.

489 Platania C, Restuccia C, Muccilli S, Cirvilleri G. Efficacy of killer yeasts in the

M
490 biological control of Penicillium digitatum on Tarocco orange fruits (Citrus

491 sinensis) Food Microbiol 2012; 30: 219-225.

d
te

492 Silva-Junior GJ, Spósito MB, Marin DR, Ribeiro-Junior PJ, Amorim L. Spatiotemporal

493 characterization of citrus postbloom fruit drop in Brazil and its relationship to
p

494 pathogen dispersal. Plant Pathol 2014; 63: 519–529.

ce

495 Schmitt MJ, Breinig F. Yeast viral killer toxins: lethality and self-protection. Nat. Rev.
Ac

496 Microbiol. 2006; 4: 212-221.

497 Talibi I, Boubaker H, Boudyach EH, Ait Ben Aoumar A. Alternative methods for the

498 control of postharvest citrus diseases. J Appl Microbiol 2014; 1-17.

499 Tweddell RJ, Jabaji-hare SH, Charest PM. Production of chitinases and β-1,3-

500 glucanases by Stachybotrys elegans, a mycoparasite of Rhizoctonia solani. Appl

501 Environ Microbiol 1993; 60: 489-495.

22
Page 22 of 32
502 Wang X, Yu T, Xia J, Yu D, Wang J, Zheng X. Biocontrol of postharvest gray mold of

503 cherry tomatoes with the marine yeast Rhodosporidium paludigenum. Biol Control

504 2010; 50: 164-171.

505 Weiler F, Schmitt MJ. Zygocin, a secreted antifungal toxin of yeast Zygosaccharomyces

t
ip
506 bailii, and its effect on sensitive fungal cells. FEMS Yeast Res 2003; 3: 69–76.

cr
507 Zhang D, Spadaro D, Garibaldi A, Gullino ML. Potential biocontrol activity of a strain

508 of Pichia guilliermondii against grey mold of apples and its possible modes of

us
509 action. Biol Control 2011; 57: 193–201.

an
510

511
M
d
p te
ce
Ac

23
Page 23 of 32
511 Table 1
512 Effect of glucose concentrations on Colletotrichum acutatum conidial germination after

513 15 hours of incubation at 25° C.

Concentrations of glucose Percentage of conidia germinated

48.93 a (1)

t
0.0

ip
0.5 33.48 c

cr
1.0 43.12 b

1.5 38.09 bc

us
2.0 41.62 b

2.5 40.33 b

an
(1)
514 Mean values marked with the same lowercase letter are not significantly different,

515 according to Tukey’s test. (p < 0.05).

M
516

517
d
p te
ce
Ac

24
Page 24 of 32
517 Table 2

518 Effect of Saccharomyces cerevisiae isolates on conidial germination of Colletotrichum

519 acutatum after 15 h of incubation at 25oC.

Isolates % of conidia germinated inhibition of conidial germination

t
ip
(%)

cr
Control 99.72 a(1) -

us
ACB-K1 59.03 b 40.69

ACB-CAT1 38.72 c 61.00

ACB-BG1 28.58 d
an 71.14
M
ACB-CR1 26.92 de 72.80

ACB-PE2 21.14 e 78.58

d
te

ACB-KD1 12.39 f 87.33

p

(1)
520 Mean values marked with the same lowercase letter are not significantly different,
ce

521 according to Tukey’s test. (p < 0.05).

522
Ac

25
Page 25 of 32
522 Table 3

523 Killer activity of yeast isolates cultured with the sensitive yeast Saccharomyces

524 cerevisiae NCYC 1006 in YEPD-methylene blue medium (pH 4.5) at 28 °C.

Yeast isolate(1) Killer activity

t
ACB-KD1 Blue zone /Inhibition zone(1)

ip
ACB-CAT1 Blue zone /Blue ring /Inhibition zone

cr
ACB-PE2 Clear zone/Blue zone/Blue ring/Inhibition zone

ACB-CR1 Blue zone /Blue ring /Inhibition zone

us
ACB-K1 Blue zone /Inhibition zone

an
ACB-BG1 Clear zone / Blue zone
(1)
525 Blue zone, Blue ring and Inhibition zone indicate cell death; Clear zone- indicates
M
526 inhibitory activity without cell death.

527
d
p te
ce
Ac

26
Page 26 of 32
527 Table 4

528 Production of chitinase and β-1,3-glucanase by Saccharomyces cerevisiae isolates as

529 measured by the amount of reducing sugars released.

Yeast isolate Chitinase RS(1) (g/L) β-1,3-glucanase RS (g/L)

t
ip
ACB-CAT1 0.052 0.036

cr
ACB-K1 0.047 0.017

ACB-BG1 0.035 0.160

us
ACB-PE2 0.012 0.069

an
ACB-KD1 0.050 0.011

ACB-CR1 0.015 0.044

M
(1)
530 RS: reducing sugar.

531
d
p te
ce
Ac

27
Page 27 of 32
 t
ip
cr
531

us
532 Fig. 1. Symptoms caused by Colletotrichum acutatum on citrus: A, infected flowers of

533 sweet orange; B, persistent calyces remaining following fruit drop.

534

an
M
d
p te
ce
Ac

28
Page 28 of 32
534

t
ip
cr
us
535

536

537 an
Fig. 2. Average diameter of Colletotrichum acutatum colony (cm) co-cultured with S.

cerevisiae isolates. Averages followed by the same letter were not statistically different
M
538 (Tukey’s test, p < 0.05).
d

539
te

540
p
ce
Ac

29
Page 29 of 32
540

t
ip
cr
us
541

542

543 an
Fig. 3. Production of volatile, thermostable and cell-free antifungal compounds by

Saccharomyces cerevisiae isolates, as measured by diameter of Colletotrichum

M
544 acutatum colony.
d

545
p te
ce
Ac

30
Page 30 of 32
 t
ip
cr
us
an
M
545

546 Fig. 4. Killer activity of the ACB-PE2 isolate as shown by a Blue Zone (BZ) and Blue

547 Ring (BR) - indicative of cell death of the Saccharomyces cerevisiae NCYC 1006 in
d

548 YEPD-methylene blue medium (pH 4.5) at 28 °C, and Clear Zone (CZ) - indicative of
te

549 inhibition without cell death.

p

550
ce
Ac

31
Page 31 of 32
 t
ip
cr
us
550
551

552

553 an
Fig. 5. Percentage of asymptomatic citrus flowers treated with S. cerevisiae isolates 24h

before (preventive) and 24h after (curative) inoculation with Colletotrichum acutatum.
M
554
555
d
p te
ce
Ac

32
Page 32 of 32