Putative modes of action of Pichia guilliermondii strain R13 in controlling chilli anthracnose after harvest

Biological Control 47 (2008) 207–215

Contents lists available at ScienceDirect

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

Putative modes of action of Pichia guilliermondii strain R13 in controlling chilli anthracnose after harvest Arun Chanchaichaovivat a, Bhinyo Panijpan a, Pintip Ruenwongsa a,b,* a b

Institute for Innovation and Development of Learning Process, Mahidol University, Rama VI Road, Bangkok 10400, Thailand Biochemistry Department, Faculty of Science, Mahidol University, Rama VI Road, Bangkok 10400, Thailand

a r t i c l e

i n f o

Article history: Received 3 April 2008 Accepted 28 July 2008 Available online 7 August 2008 Keywords: Biological control Chilli anthracnose Chitinase Colletotrichum capsici b-1,3-glucanase Pichia guilliermondii

a b s t r a c t The mode of action of an antifungal yeast, Pichia guilliermondii, strain R13, against the fungal pathogen of chilli anthracnose, Colletotrichum capsici, was examined both on agar plates and in chilli fruit wounds. Light microscopy revealed that strain R13 attached to the fungal pathogen, and this attachment apparently restricted the proliferation of C. capsici in the chilli fruit wounds. In chilli juice, strain R13 suppressed C. capsici spore germination and germ tube length, but the suppression was completely overcome by addition of 0.05% glucose, sucrose, or 1% of nitrate sources (NH4NO3, NaNO3, Ca(NO3)24H2O, Mg(NO3)26H2O, and KNO3), suggesting the yeast was competing with the fungus for these substrates. Strain R13 also produced hydrolytic enzymes, including b-1,3-glucanase, and chitinase, in both solid and liquid media. The activities of these enzymes were highest when the C. capsici hyphal cell walls, rather than laminarin or glucose, were the carbon source; the activities were approximately 2 and 15 times higher with hyphal cell wall than with laminarin or glucose. Unlike the other strains tested, strain R13 did not produce a lethal toxin when cultivated under similar conditions. This study provides evidence that attachment, competition for nutrients, and secretion of hydrolytic enzymes, at least partially, explain how P. guilliermondii strain R13 suppresses C. capsici. Ó 2008 Elsevier Inc. All rights reserved.

1. Introduction Biological control of postharvest diseases of fruits and vegetables by antagonistic microorganisms is now recognized as one of the most promising alternatives to the use of fungicides. Fungicides have important limitations, including risks to consumers and to the environment as well as possible induction of fungicide-resistant pathogens (Pal and Gardener, 2006; Scherm et al., 2003). Several yeast isolates from the naturally occurring microflora of fruits and vegetables suppress various postharvest fungal pathogens (Druvefors, 2004), and the yeast Pichia gulliermondii Wickerham strain R13 has recently been shown to control anthracnose disease caused by the fungus Colletotrichum capsici (Syd.) Butler & Bisby (Chanchaichaovivat et al., 2007). Some of these yeasts have been patented and registered as biological control agents and are available for postharvest application. The yeast Candida oleophila Montrocher (AspireÒ) has been used for controlling citrus fruit decay caused by Penicillium itallicum Wehmer and Penicillium digitatum Sacc. (Janisiewicz and Korsten, 2002). Other commercial products, i.e., Yield PlusÒ (Cryptococcus albidus (Saito) Skiner and ShemerÒ (Metschnikowia fructicola Kurtzman & Droby), are effective against a wide range of pathogens (e.g., Botrytis, Penicillium, * Corresponding author. Fax: +66 2 354 7345. E-mail address: [email protected] (P. Ruenwongsa). 1049-9644/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.biocontrol.2008.07.018

and Rhizopus) in apple, grape, and strawberry (Blachinsky et al., 2007; Elmer and Reglinski, 2006). Integration of the commercial yeast biocontrol agents with sub-lethal dose chemicals or heat treatment has also been effective (Janisiewicz and Korsten, 2002). Control of plant pathogenic fungi by these yeasts often involves several modes of action including competition for nutrients and space, the effect of a lethal toxin (termed ‘‘killer toxin”), production of hydrolytic enzymes, direct contact, and possibly induction of plant resistance (Janisiewicz et al., 2000). Killer toxins have been found in many yeast genera, including Candida, Cryptococcus, Debaromyces, Torulopsis, Williopsis (Magliani et al., 1977). Nutrient competition and production of exo-b-1,3-glucanase and chitinase are the main modes of action for the commercial yeast products AspireÒ, ShemerÒ, and Yield PlusÒ (Bar-Shimon et al., 2004; Chan and Tian, 2005; Droby, 2006). AspireÒ also induces pathogen resistance in plants by increasing ethylene biosynthesis, phytoalexin accumulation, and phenylalanine ammonia lyase activity (Droby et al., 2002). Other possible modes of action of the yeast antagonists have also been suggested. For example, in apple wounds infected by Botrytis cinerea Per. ex Fr. and Penicillium expansum Link ex Thom, yeast Cryptococcus laurentii (Kufferath) Skinner may out compete the wound pathogens by tolerating oxidative stress (Castoria et al., 2003). In peach fruit infected by P. expansum Link ex Thom, the yeast Pichia membranefaciens Hansen induces a number of proteins related to host defense mechanisms (Chan

208

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

et al., 2007). Similarly, the yeast C. laurentii (Kufferath) Skinner may control a postharvest disease of jujube fruit by producing b1,3-gucanase, a cell wall degrading enzyme involved in plant host defense (Tian et al., 2007). The antagonistic yeast P. guilliermondii has been extensively investigated as a possible biological control agent for the postharvest spoilage of grape fruit, apple, and soybean caused by the fungi P. digitatum Sacc., B. cinerea Per. ex Fr., P. expansum Link ex Thom, and Aspergillus flavus Link ex Fries, respectively (Droby et al., 1989; Paster et al., 1993; Wisniewski et al., 1990). Its adverse effects on fungal pathogens have been attributed to competition for nutrients and secretion of cell wall degrading enzymes (Droby et al., 1989, 1990). In the case of P. gulliermondii, strains 5A and US7, against P. digitatum Sacc. and P. italicum Wehmer, respectively, the ability of the yeast to attach to the pathogen’s hyphae may enhance nutrient competition or degradation of the fungal cell walls (Arras et al., 1999; Wisniewski et al., 1991). In pilot plant tests, P. gulliermondii, strain US7, performed as well as standard fungicide treatments in preventing postharvest fruit decay caused by fungi (Hofstein et al., 1990). Moreover, P. gulliermondii US7 and the other antagonistic yeasts do not secrete potentially hazardous antibiotics (Arras et al., 1999; Janisiewicz and Korsten, 2002), and treated fruit is safe for human consumption. The demand for chillies (Capsicum annuum L.) is increasing worldwide, and chillies for import and export must be free of diseases, fungal toxins, and chemical residues (Klieber, 2000). In several countries in Asia and in the United States, one of the major diseases of chilli is anthracnose caused by C. capsici (Syd.) Butler & Bisby (Kim et al., 1989; Hadden and Black, 1989). In nature, C. capsici can cause disease both in intact and in wounded chilli fruits, but the disease is more severe in wounded fruit. The appressorium of C. capsici adheres to the plant’s surface and then produces the infection peg, which penetrates the plant tissue leading to infection and disease (Mehrotra and Aggarwal, 2003). Control of anthracnose disease on chilli fruit still relies mainly on the use of synthetic fungicides (Jaffee, 2004). In our earlier report, P. guilliermondii, strain R13, isolated from Thai rambutan (Nephelium lappaceum L.) was more effective than chemical treatment in preserving chilli fruits (Chanchaichaovivat et al., 2007), but the mechanism by which P. guilliermondii strain R13 suppresses anthracnose disease in chilli is still unknown. In this study, we investigate the mode of action of the antagonistic yeast P. guilliermondii strain R13 against the anthracnose pathogen, C. capsici, in chilli fruits. We focused on the yeast’s attachment to the pathogen, competition for nutrients, production of cell wall degrading enzymes, and production of killer toxin. Clarifying the mechanism of biological control could help us to obtain better and more consistent control of anthracnose disease in chilli fruits by P. guilliermondii strain R13.

2. Materials and methods 2.1. Yeast and pathogen The yeast added, P. guilliermondii, strain R13, (Chanchaichaovivat et al., 2007) previously isolated from fruit of rambutan (N. lappaceum) in Thailand and identified by Collaborative Research Center for Bioscience and Biotechnology, Bangkok Thailand, was maintained on nutrient yeast dextrose agar (NYDA) slants containing 8 g l1 nutrient broth, 5 g l1 yeast extract, 10 g l1 glucose, and 20 g l1 agar. The yeast was grown at 28 °C for 24 h, stored at 4 °C and used within 7 days. The fungal pathogen C. capsici DOAC 1511 was obtained from the Mycological Laboratory of the Department of Agriculture (DAO), Bangkok,Thailand. The fungal pathogen was grown at 28 °C and maintained on potato dextrose agar (PDA)

slants at 4 °C. All culture media were from Difco Laboratories (Sparks, USA). 2.2. Effects of P. guilliermondii R13 on preventing infection of wounded chilli fruit inoculated with C. capsici Chilli fruits were wounded with a sterile cork borer (one 6-mm diameter wound per fruit) and treated with 20 ll of a P. guilliermondii cell suspension (5  108 cells ml1 as counted by hemocytometer) or sterile distilled water for the control. After 1–2 h (to allow penetration of the cell suspension into the wounds), the wounds were inoculated with 20 ll of a C. capsici spore suspension (5  104 spores ml1). After 5 days in a ventilated incubator in the dark at 28 °C, cross sections of chilli fruit tissues (8 mm in diameter) were removed from the center of the infected areas using a razor blade. Thin sections of the infected tissue were mounted on glass slides, stained with 1.0% toluidine blue-O for 10 min, blotted with tissue paper to remove excess dye, rinsed three times with a few drops of deionized water, and blotted with tissue paper. The stained tissue was covered with a coverslip and examined at 40 magnification with a light microscope equipped with a camera. Thirty replicate fruits were examined per treatment and the experiment was performed three times. 2.3. Adhesion between P. guilliermondii strain R13 and C. capsici on the surface of wounded chilli fruits In one experiment, chilli fruits were wounded, inoculated with strain R13, and then immediately inoculated with C. capsici as described in Section 2.2. The wounds were then covered with a sterile cellophane disk (10 mm in diameter, Bio-Rad 1650963). In another experiment, the C. capsici was added before P. guilliermondii; the added C. capsici spores were allowed to germinate and grow for 36 h on the chilli wound before yeast cells were added to the margin of the C. capsici hyphae and the wound was covered with a sterile cellophane disk. The cellophane disk method has been used to study yeast-mold interactions by several researchers e.g. El-Ghaouth et al. (1998), Saligkarias et al. (2002). In our experiment, both C. capsici and P. guilliermondii adhered tightly to the cellophane disks. After the disks were added in both experiments, fruits were incubated at 28 °C on enclosed plastic trays for 24 h in a ventilated cabinet (95% relative humidity) in the dark. Small sections of the cellophane disks (10 mm in diameter) were then washed according to the method of El-Ghaouth et al. (1998). They were then dipped two to three times in distilled water containing 1% (v/v) Tween 20 to remove yeast cells that did not adhere to hyphae. The sections were mounted on glass slides and examined at 100 magnification with a light microscope. Attachment of yeast cells to fungal hyphae was assessed. As a negative control for attachment, wounds were also treated with the yeast Issatchenkia orientalis Kudryavtsev strain ER1, which had been isolated from eggplant and which was not expected to adhere to C. capsici hyphae. Thirty replicate fruits were examined per treatment, and the experiment was performed three times. 2.4. Competition for nutrients The role of competition for nutrients on the biological control capability of P. guilliermondii strain R13 was investigated. Chilli juice was used as the liquid medium throughout this experiment. Chilli juice was prepared by homogenization with distilled water (15% w/v), and the homogenate was filtered through a Whatman No. 1 filter paper. Strain R13 (50 ll at 5  108 cells ml1) and C. capsici (50 ll at 5  104 spores ml1) were added to test tubes (18  150 mm) containing 5 ml of chilli juice supplemented or not with sugars or nitrates. The sugars, which were used at concen-

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

trations of 0.05–2.00% (w/v), were glucose, sucrose, and fructose. The nitrates, used at a concentration of 0.1%, included NH4NO3, NaNO3, Ca(NO3)24H2O, Mg(NO3)26H2O, KNO3, and Fe(NO3)39H2O. After 16 h at 28 °C, the test tubes were mixed, three drops of the sample from each tube were placed in a hemocytometer, and at least 100 C. capsici spores per treatment were observed at 100 magnification with a light microscope; germination and germ tube length were determined for each spore. There were three replicate tubes per treatment, and the experiment was performed three times. 2.5. Quantification of b-1,3-glucanase and chitinase activities in liquid culture Strain R13 was cultured in sterilized Vogel’s medium (Vogel, 1956) modified by adding 0.1 g l1 of sterilized yeast extract and one of the following sterilized carbon source (0.5 g l1): glucose, laminarin from Laminaria digitata (Hudson) Lamouroux (Sigma L-9634), colloidal chitin, or cell wall preparation (CWP) from C. capsici. Colloidal chitin was prepared according to the method of Roberts and Selitrennikoff (1988). Ten grams of chitin power (C9213, Sigma–Aldrich Co., USA) was added slowly into 180 ml of concentrated HCl and left at 4 °C overnight with vigorous stirring. The mixture was added to one liter of iced-cold 95% ethanol with rapid stirring, kept overnight at 35 °C, and centrifuged at 5000g for 20 min at 4 °C. The precipitate of colloidal chitin was washed with sterile distilled water until the pH was neutral (pH 7.0). CWP was obtained from C. capsici cultured in 15 ml of malt extract broth containing malt extract (17 g l1) and peptone (3 g l1) for 5 days at 28 °C. The cultures were centrifuged at 3000g to collect the mycelium. Pelleted mycelium was washed three times with distilled water. The washed mycelium was homogenized for 2 min in 30 ml of 50 mM Tris/HCl buffer (pH 7.2) and centrifuged for 10 min at 3000g, and the supernatant was discarded. The washed pellet (hyphae) was frozen at 70 °C and pulverized with a mortar and pestle. The pulverized mycelium was resuspended in an equal volume of 50 mM Tris/HCl buffer (pH 7.2) and centrifuged for 10 min at 3000g, and the supernatant was discarded. The pellet was subjected to three successive cycles of centrifugation and resuspension. The final pellet of CWP was stored at 20 °C until use. Erlenmeyer flasks (250 ml) containing 1% of P. guilliermondii strain R13 in 100 ml of modified Vogel’s medium plus 5% of one of the sterilized supplements (glucose, laminarin, colloidal chitin, or CWP) were incubated on a rotary shaker (200 rpm) at 28 °C. Three separate flasks were used for each supplement. After 120 h, cultures of strain R13 were centrifuged for 10 min at 3000g, and each supernatant was filtered through a 0.22-lm filter (Millex-GP, Millipore, USA). The culture filtrates were stored at 20 °C until assayed for enzymes. b-1,3-Glucanase activity from yeast strain R13 was assayed according to the method of Bergmeyer and Brent (1974) with laminarin as the substrate. One milliliter of laminarin (5 mg ml1 in 50 mM potassium acetate buffer, pH 5.0), was added to 1 ml of the strain R13 culture filtrates (see previous paragraph). The reaction mixture was incubated at 37 °C in water bath for 30 min, and the reaction was terminated by placing the mixture in boiling water for 10 min. Enzyme activity was determined by quantifying the glucose released from laminarin with the God-Perid glucose test kit (Boehringer Mannheim, GMBH, Germany). The kit consists of glucose oxidase, peroxidase, and 2,20 -azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) diammonium salt (ABTS). The glucose test kit reagent (3 ml at 1.5 mg ml1) was added to the reaction mixture, which was then incubated at 28 °C for 30 min before its optical density at 420 nm was measured. The amount of re-

209

leased glucose was determined using the glucose standard curve. Specific activity of b-1,3-glucanase was expressed as lmol glucose released/min/mg protein. Protein content of strain R13 was measured according to Bradford’s method (1976) using bovine serum albumin (Sigma A-9647, Sigma Chemical Co.) as the standard. Chitinase activity was determined according to the method of Wirth and Wolf (1990) with colloidal chitin azure (chitin covalently linked with Remazol Brilliant Violet 5R dye) as the substrate (Loewe Biochemica, Germany). Colloidal chitin azure (1 ml at 1 mg ml1) in 0.1 mM sodium acetate buffer, pH 5, was incubated with 1 ml of strain R13 culture filtrate for 1 h at 37 °C. The reaction was stopped by adding 50 ll of 1 M HCl. The reaction mixture was cooled on ice for 10 min and centrifuged at 15000g for 5 min. The optical density (550 nm) of the supernatant was measured (Beckman Coulter, DUÓ 800, USA). The unit of chitinase activity produced by strain R13 was determined, as described by Saligkarias et al. (2002), from a standard curve of chitinase from Streptomyces griseus (Krainsky) Waksman & Henrici (Sigma C-6137). One unit of chitinase was defined as the liberation of 1.0 mg of N-acetyl-D-glucosamine from chitin per hour at pH 5.0 at 37 °C. 2.6. Production of killer toxin Strain R13 was tested for killer toxin production by using the well test (Somers and Bevan, 1969). Strain R13 was first cultured in YEPD broth (1.0% yeast extract, 2.0% peptone, 0.1 M citrate phosphate buffer, pH 5.0) at 28 °C for 5 days. The culture was filtered through a 0.45-lm millipore filter (Millipore Corp., Bedford, Mass.). An 80-ll aliquot of the filtrate was added to a small well (8 mm in diameter) formed in a YEPD-MB agar plate (YEPD containing 0.01% methylene blue and 2.0% agar) previously seeded with 0.1 ml of a 24-h culture of a yeast (Saccharomyces cerevisiae Meyen ex Hansen strain 1006, National Collection of Yeast Cultures, Norwich, UK) that is sensitive to the killer toxin (Farris et al., 1991). As a positive control for killer toxin production, Kluveromyces lactis (Dombrowski) Johannsen and van der Walt killer strain CBS 2359 (Holland) was used (Aguiar and Lucas, 2000). Sterile distilled water was used as the negative control. After 5 days at 28 °C, all plates were examined for production of killer toxin, which was indicated by a zone of growth inhibition of the sensitive strain around the well. Each test was done in triplicate. 2.7. Statistical analysis Data for percentage of spore germination, spore germ tube length, and specific activity of hydrolytic enzymes were analyzed with the general linear model (GLM) procedure of SPSS software (version 10.0 for Windows, SPSS Inc., Chicago, IL, USA). The least significant difference (LSD) test at P < 0.05 was used for mean separation for all tests. 3. Results 3.1. Effects of P. guilliermondii strain R13 on infection of wounded chilli fruits by C. capsici Light microscopic examination of thin sections of the chilli wounds indicated a massive colonization by C. capsici. Proliferation of the pathogen in wounded tissue resulted in the collapse of cell walls and internal disorganization (Fig. 1a and b). Invading hyphae and acervuli that embedded in the parenchyma tissue beneath the wound caused extensive cellular degradation (Fig. 1c and d). Walls of the invaded and adjacent cells were severely damaged, and darkened spots appeared on chilli tissues. C. capsici cells also grew

210

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

Fig. 1. Light micrographs of chilli fruit inoculated with the pathogen Collectotrichum capsici alone (a–d), with the pathogen and the yeast antagonist Pichia guilliermondii strain R13 (e–g), and without pathogen or yeast (h). (a) Massive colonization of C. capsici in the chilli wounds appeared as the dark spots. (b) The pathogen proliferated (PP) in wounded tissue resulting in the collapse of the chilli tissue, CT, below the wound. (c) Invading hyphae (IH) and (d) Acervuli (AC) embedded in the parenchyma tissue beneath the wound and causing extensive cellular degradation. (e and f) P. guilliermondii strain R13 multiplied in the wound site and formed a matrix of yeast cells (MY) covering plant-cell layers. (g) The area of the wound (AW) preserved by P. guilliermondii cells. (h) Healthy chilli tissues that were not inoculated with the pathogen or yeast. Bars = 20 lm.

inter- and intracellularly through the cell layer immediately beneath the wound. Fungal colonization of the chilli wounds was reduced if the tissue was previously treated with strain R13 (Fig. 1e). Examination of 30 thin sections from thirty different fruits showed a decrease in numbers of invading hyphae in the treated tissues. Few fungal hyphae were detected in the wounded tissue layer treated with strain R13, and the invading hyphae were usually surrounded by the antagonistic yeast cells. Strain R13 multiplied in the wound and formed a matrix of yeast cells that covered layers of plant-cells that were ruptured during wounding (Fig. 1e and f), resulting in

preservation of the parenchyma tissue beneath the wound (Fig. 1f). Wounded tissues treated only with strain R13 showed no signs of cellular alterations (Fig. 1g). The healthy chilli tissues without yeast and fungal pathogen inoculation are shown in Fig. 1h. 3.2. Adhesion between P. guilliermondii strain R13 and C. capsici on the surface of wounded chilli fruits The cellophane disks covering the inoculated chilli wounds contained abundant yeast cells. After 12 h, yeast cells had adhered to C. capsici spores (Fig. 2a), and after 20 h, yeast cells formed aggre-

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

211

Fig. 2. Attachment of Pichia guilliermondii strain R13 to spores and hyphae of Colletotrichum capsici. (a) At 12 h after inoculation, C. capsici spore (S) is surrounded by yeast cells (1700). (b) At 20 h after incubation, a clump of yeast cells (CY) encircles the hyphal tip (HT) of a C. capsici germ tube (1000). (c and d) P. guilliermondii cells (Y) attached to C. capsici hypha (H) (1000). (e) Lower magnification (100) of C. capsici hypha clumped by P. guilliermondii cells. (f) Lack of attachment of yeast cells (Issatchenkia orientalis strain ER1) to C. capsici hypha (1000). Bars = 5 lm.

gates around hyphal tips of germinating C. capsici spores (Fig. 2b). When C. capsici was allowed to grow for 36 h before adding strain R13, the yeast cells attached to C. capsici hyphae (Fig. 2c–e). In the absence of strain R13, all the spores of C. capsici germinated and formed dense mycelia (data not shown).

3.3. Competition for nutrients by P. guilliermondii strain R13 The potential role of competition for nutrients in the biological control of C. capsici by strain R13 was examined by investigating the effects of increasing concentrations of three sugars and six

212

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

Table 1 Effects of increasing concentrations of glucose, sucrose, and fructose on inhibition of spore germination and germ tube elongation of Colletotrichum capsici by the antagonistic yeast Pichia guilliermondii strain R13 Treatment b

Sugar concentration (%)

Spore germinationa (%)

Germ tube lengtha (lm) 250 ± 10e

Juice + C. capsici

—

97 ± 5e

Juice + C. capsici + yeastb

—

7 ± 1a

76 ± 4a

Juice + C. capsici + glucosec

0.05 0.10 0.20

98 ± 4e 98 ± 3e 98 ± 5e

388 ± 10g 467 ± 12h 571 ± 9i

Juice + C. capsici + yeast + glucose

0.05 0.10 0.20

22 ± 3b 44 ± 5c 63 ± 4d

110 ± 7b 165 ± 9c 196 ± 10d

Juice + C. capsici + fructosec

0.05 0.10 0.20

97 ± 3e 97 ± 5e 98 ± 4e

258 ± 7e 287 ± 10f 292 ± 11f

Juice + C. capsici + yeast + fructose

0.05 0.10 0.20

27 ± 3b 28 ± 3b 25 ± 2b

123 ± 7b 126 ± 8b 119 ± 6b

Juice + C. capsici + sucrosec

0.05 0.10 0.20

97 ± 6e 98 ± 5e 99 ± 5e

298 ± 9f 373 ± 9g 450 ± 10h

Juice + C. capsici + yeast + sucrose

0.05 0.10 0.20

15 ± 2b 30 ± 2c 47 ± 4d

97 ± 4b 128 ± 8c 173 ± 10d

a Spore germination and germ tube length were determined after incubation for 16 h at 28 °C. Data are means ± standard errors of three independent experiments. Values in each column followed by a different letter indicate significant difference (P < 0.05) according to LSD test. b Control tubes were inoculated with C. capsici alone or C. capsici plus yeast strain R13 in chilli juice (15%) without supplemental sugar. c Control tubes were inoculated with C. capsici alone in chilli juice with supplemental sugars.

nitrates on spore germination and germ tube length (Tables 1 and 2). The data shows that increasing the concentrations of glucose and sucrose from 0.05 to 2.00% reduced the suppression of spore germination (Table 1), suggesting that the yeast competed with C. capsici for these sugars. However, increasing concentrations of fructose did not increase germination (Table 1). None of the sugars Table 2 Effects of various nitrates on biocontrol efficacy of antagonistic yeast Pichia guilliermondii strain R13 on spore germination and germ tube elongation of Colletotrichum capsici Treatment

Spore germinationa (%)

Germ tube lengtha (lm)

Juice + C. capsicib Juice + C. capsici + yeastb

97 ± 5f 7 ± 1a

250 ± 10e 76 ± 4a

Juice + C. capsici + Fe(NO3)39H2Oc Juice + C. capsici + yeast + Fe(NO3)39H2O

7 ± 3a 6 ± 1a

72 ± 6a 68 ± 4a

Juice + C. capsici + Ca(NO3)24H2Oc Juice + C. capsici + yeast + Ca(NO3)24H2O

97 ± 2f 36 ± 2b

261 ± 10ef 121 ± 9b

Juice + C. capsici + Mg(NO3)26H2Oc Juice + C. capsici + yeast + Mg(NO3)26H2O

97 ± 2f 44 ± 3bc

267 ± 11ef 142 ± 10bc

Juice + C. capsici + KNO3c Juice + C. capsici + yeast + KNO3

98 ± 2f 53 ± 4c

280 ± 12f 164 ± 11c

Juice + C. capsici + NH4NO3c Juice + C. capsici + yeast + NH4NO3

98 ± 3f 69 ± 5d

289 ± 12f 190 ± 6d

Juice + C. capsici + NaNO3c Juice + C. capsici + yeast + NaNO3

98 ± 3f 74 ± 5e

302 ± 9g 209 ± 7d

a Spore germination and germ tube length were determined after incubation in the presence or absence of 0.1% of various nitrate sources for 16 h at 28 °C. Data are means ± standard errors of three independent experiments. Value in each column followed by a different letter indicates significant difference (P < 0.05) according to LSD test. Values followed by shared letter are not significantly different (P < 0.05). b Control tubes were inoculated with C. capsici alone or C. capsici plus yeast strain R13 in chilli juice (15%) without supplemental nitrates. c Control tubes were inoculated with C. capsici alone in chilli juice with supplemental nitrates.

inhibited C. capsici germination. Similar germination data were obtained with nitrates except that Fe(NO3)39H2O appeared to inhibit C. capsici germination (Table 2). Although supplemental nutrients usually increased C. capsici germ tube length in the presence of the yeast, the nutrients also increased germ tube length in the absence of the yeast. We therefore could not infer that the suppression of germ tube length by the yeast was due to competition for these nutrients. 3.4. Quantification of b-1,3-glucanase and chitinase activities in liquid culture b-1,3-Glucanase and chitinase activities were first identified by examining agar cultures of P. guilliermondii strain R13. A clear zone of b-1,3-glucanase activity developed around the colony of strain R13 cultured on YEPD plate containing 0.2% laminarin azure as a substrate (unpublished data). The chitinase enzyme cleaves the b-1,4-glycosidic bonds of chitin into oligomers and monomers, resulting in a clear zone around the yeast strain R13 colony (unpublished data). Fig. 3 shows b-1,3-glucanase specific activity of strain R13 cultured with three different carbon sources. Cell wall preparation of C. capsici (CWP) induced the most b-1,3-glucanase specific activity, followed by laminarin and glucose, respectively (Fig. 3). CWP induced 2 times and 18 times higher b-1,3-glucanase activity than laminarin and glucose, respectively. Similar results were obtained for chitinase specific activity when strain R13 was cultured with the three carbon sources in a liquid medium (Fig. 4). CWP induced the most chitinase specific activity followed by colloidal chitin and glucose. CWP induced 1.3 and 11.9 times higher chitinase specific activity than colloidal chitin and glucose, respectively. 3.5. Production of killer toxin The well test for killer toxin production by P. guilliermondii strain R13 was performed by using the yeast S. cerevisiae, NCYC 1006, as the killer toxin-sensitive strain. K. lactis, CBS 2359, is

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

Fig. 3. b-1,3-glucanase activity in the culture filtrates of Pichia guilliermondii strain R13 grown for 120 h at 28 °C in Vogel’s medium amended to contain one of three carbon sources: glucose (0.5 gl1) ( ), laminarin (0.5 gl1) ( ), or cell wall preparation (0.5 gl1) (CWP) (j). b-1,3-Glucanase activity was measured at 37 °C for 30 min using laminarin as substrate. Bars represent the means ± standard errors of three independent experiments, with three replicates per experiment. Different letters above the bars indicate significant differences (P < 0.05) according to LSD test.

Fig. 4. Chitinase activity in the culture filtrates of Pichia guilliermondii strain R13 grown for 120 h at 28 °C in Vogel’s medium amended to contain one of three carbon sources: glucose (0.5 gl1) ( ), colloidal chitin (0.5 gl1) ( ), or cell wall preparation (0.5 gl1) (CWP) (j). Chitinase activity was measured at 37 °C for 1 h using chitin azure as substrate. One unit will liberate 1.0 mg of N-acetyl-D-glucosamine from chitin per hour at pH 5.0 at 37 °C. Each assay was performed in triplicate and the entire experiment was repeated three times. Bars represent the means ± standard errors of three independent experiments. Different letters above the bars indicate significant differences (P < 0.05) according to LSD test.

known to produce the killer toxin and was used as a positive control. After 5 days, strain R13 did not inhibit the growth of S. cerevisiae, but K. lactis did (data not shown). The results suggest that strain R13 does not produce killer toxin, at least under the conditions of the assay used. 4. Discussion Results from this study clearly indicate that P. guilliermondii strain R13 has multiple modes of action in controlling chilli fruit

213

anthracnose caused by C. capsici. Our findings conform to previous investigations with other antagonistic yeasts indicating that biological control agents with multiple modes of action appear superior to those with only one mode of action (Wszelaki and Mitcham, 2002). Information on modes of action is essential for enhancing the chances that this yeast will be successfully used for biological control. For example, understanding mode of action can facilitate registration for commercial use and can be useful for optimizing the formulation and delivery systems (Andrews, 1992). In general, biological control of postharvest diseases of fruits and vegetables is unique in that the antagonists are applied directly to the wounds (the point of entry for the pathogens) and the wound sites are often rich in nutrients. Most postharvest pathogens are necrotrophs, which require nutrients of the dying or dead tissues for germination and initiation of the pathogenic process (Janisiewicz et al., 2000). Our microscopic observations indicate that C. capsici grows vigorously in wounded chilli fruit, possibly because the fungus produces cell wall degrading enzymes. The fungal appressorium and infection peg also help produce physical pressure that may enhance entrance of the pathogen into the plant-cells (Lakshmesha et al., 2005; Mehrotra and Aggarwal, 2003). That P. guilliermondii strain R13 attaches to C. capsici hyphae and hyphal tips of germinated spores agrees with other reports concerning antagonistic yeasts; C. oleophila Montrocher strain I-182 and C. albidus (Saito) Skinner attached to the hyphal walls of B. cinerea Pers. ex Fr. and Monilinia fructicola (Winter) Honey, respectively (Chan and Tian, 2005; Saligkarias et al., 2002). Pichia guilliermondii, strains 5A and US7, have also been shown to attach to the hyphae of P. digitatum Sacc. and B. cinerea Pers. ex Fr, respectively (Arras et al., 1999; Saligkarias et al., 2002; Wisniewski et al., 1991). The adhesion process is influenced by microbial surface charges (Buck and Andrews, 1999a, b) and signal recognition (Chan and Tian, 2005). It is not surprising that P. guilliermondii, strain R13 competed with C. capsici for both sugars and nitrates because competition for sugars has been reported for several other strains. For example, P. guilliermondii, strain Y2 competed with B. cinerea for glucose and sucrose in infected strawberry leaves (Guetsky et al., 2002). Candida guilliermondii (Castellani) Berkhout strains 3C-1b and F1, and Metschnikowia pulcherrima Pitt & Miller strains 2.33 and 4.4 competed with P. expansum Link ex Thom and B. cinerea for several nitrates (Piano et al., 1997; Scherm et al., 2003). Researchers have suggested that attachment of yeast antagonists to the hyphae of plant pathogenic fungi may enhance nutrient competition as well as interfere with the ability of the pathogens to initiate infection (El-Ghaouth et al., 1998; Janisiewicz and Korsten, 2002). Like other antagonistic yeasts, P. guilliermondii, strain R13, can produce hydrolytic enzymes and our results demonstrated b-1,3glucanase and chitinase activity in both solid and liquid media. In addition, the specific activities of b-1,3-glucanase and chitinase from P. guilliermondii strain R13 cultured with cell wall preparation (CWP) from C. capsici were higher (2.0 times and 1.1 times, respectively) than those of P. guilliermondii, strain US7 grown on CWP from B. cinerea Pers. ex Fr. (Castoria et al., 1997; Wisniewski et al., 1991). However, the specific activities of b-1,3-glucanase and chitinase were lower (3.0 times and 1.5 times, respectively) than those of C. oleophila Montrocher, the commercial yeast antagonist (Saligkarias et al., 2002). Chitinase and b-1,3-glucanase hydrolyze fungal cell walls and inhibit the in vitro growth of several pathogenic fungi (Schlumbaum et al., 1986; Sela-Buurlage et al., 1993). It is possible that these hydrolytic enzymes play an important role in degradation of the C. capsici cell wall, especially when the yeast attaches to the pathogen hypha. Release of hydrolytic enzymes by yeast attached to hyphae has been discussed/documented in studies of other yeast antagonists including

214

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215

commercial ones (Chan and Tian, 2005; Droby, 2006; Saligkarias et al., 2002). In addition to degrading hyphae, chitinase and b1,3-glucanase may help induce resistance of plants to pathogenic fungi (van Loon et al., 1998). The results of this study clearly demonstrated that production of b-1,3-glucanase and chitinase depended on the carbon source, and that cell walls of the fungal pathogen induced the highest production of both enzymes. This agrees with other findings in which b-1,3-glucanase activities from P. guilliermondii, isolate 87, Trichoderma harzianum Rifai, and P. anomala (Hansen) Kurtzman were higher if the organisms were grown in media supplemented with fungal cell walls than in media supplemented with glucose (Elad et al., 1982; Jijakli and Lepoivre, 1998; Wisniewski et al., 1991). Low activities of b-1,3-glucanase and chitinase in glucose-supplemented medium may be due to catabolite repression because secretion of b-1,3-glucanase is down regulated in many fungi when glucose is the main source of carbon (Giczey et al., 2001). Although P. guilliermondii, strain R13, is similar to other antagonistic yeasts in that its mode of action involves nutrient competition and hydrolytic enzyme production, it differs from other antagonistic yeast in that it failed to produce killer toxin against the fungus in our experiments. Other species of Pichia (P. anomala (Hansen) Kurtzman, P. burtonii Boidin et al., P. farinose (Lindner) Hansen, and P. membranifacians Hansen) produce killer toxin against the molds, including Penicillium roqueforti Thom, Penicillium camembertii Thom, and Aspergillus nidulans (Eidam) Winter (Druvefors, 2004). Killer toxins are mainly produced and active under acidic conditions, and the activity decreases with increasing pH and temperature (Sawant et al., 1989). Because most killer toxins are only produced or are only active within a very narrow temperature and pH range (Druvefors, 2004; Suzuki and Nikkuni, 1989), antagonists that depend on killer toxin may often be ineffective biological control agents. Because antagonism by P. guilliermondii, strain R13, apparently does not involve killer toxin, biological control by this isolate may be less sensitive to variations in temperature and pH. On the other hand, strain R13 might produce killer toxin under environmental conditions that differed from those in our study. Multiple modes of action as documented in this study might explain why P. guilliermondii, strain R13, provided excellent control of postharvest disease of chilli fruit (Chanchaichaovivat et al., 2007). Additional modes of action, however, are possible and could include the yeast’s tolerance to oxidative stress and an ability to induce host plants to produce antioxidants and synthesis of pathogenesis-related proteins (Castoria et al., 2003; Chan et al., 2007; Tian et al., 2007). Although these yeast antagonists can reduce postharvest decay, they are applied directly to the fruit or vegetable and so must not be toxic or otherwise harmful to humans. The yeast P. guilliermondii has been shown to be non-toxic and has been accepted by many researchers as the biological control agent of postharvest fruit rots of apples, grapes, and tomatoes (Arras et al., 1999; Mari and Guizzardi, 1998; Wisniewski et al., 1991). Because P. guilliermondii strain R13 is one of the most promising biological control agents for anthracnose disease, large-scale studies in commercial storage facilities are now warranted. A future study on survival of P. guilliermondii on chilli fruit will augment knowledge on its application as a longer term biological control agent. Acknowledgments The authors thank Dr. Wantanalai Panbangred and Dr. Chuenchit Boonchird for helpful suggestions in microbiological techniques. We also thank Puritat Ratanabunlung for his helpful assistance during this work. We thank Mahidol Research Grant for funding.

References Aguiar, C., Lucas, C., 2000. Yeast killer/sensitivity phenotypes and halotolerance. Food Technology and Biotechnology 38, 39–46. Andrews, J.H., 1992. Biological control in the phyllosphere. Annual Review of Phytopathology 30, 603–635. Arras, G., Nicolussi, P., Ligios, C., 1999. Non-toxicity of some antifungal yeasts (Pichia guilliermondii, Rhodotorula glutinis and Candida oleophila) in laboratory animals. Annals of Microbiology And Enzymology 49, 125–131. Bar-Shimon, M., Yehuda, H., Cohen, L., Wiess, B., Kobeshnikov, A., Daus, A., Goldway, M., Wisniewski, M.E., Droby, S., 2004. Characterization of extracellular lytic enzymes produced by the yeast bicontrol agent Candida oleophila. Current Genetics 45, 140–148. Bergmeyer, H.U., Brent, E., 1974. Glucose determination with glucose oxidase and peroxidase. In: Bergmeyer, H.U. (Ed.), Methods of Enzymatic Analysis. Academic Press, New York, pp. 1205–1215. Blachinsky, D., Lazare, M., Shamai, I., Antonov, J., Husid, A., Bercovitz, A., Feldman, K., El-ad, B., Prengler, M., Keren-Zur, M., 2007. Commercial Applications of ‘Shemer’ for the Control of Pre- and Postharvest Diseases. The 28th Congress of the Israeli Phytopathological Society, Israel. pp. 194–215. Bradford, M.M., 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248–254. Buck, J.W., Andrews, J.H., 1999a. Attachment of the yeast Rhodosporidium toruloides is mediated by adhesives localized at sites of bud cell development. Applied and Environmental Microbiology 65, 465–471. Buck, J.W., Andrews, J.H., 1999b. Localized, positive charge mediates adhesion of Rhodosporidium toruloides to barley leaves and polystyrene. Applied and Environmental Microbiology 65, 2179–2183. Castoria, R., de Curtis, F., Lima, G., de Cicco, V., 1997. b-1,3-glucanase activity of two saprophytic yeasts and possible mode of action as biocontrol agents against postharvest diseases. Postharvest Biology and Technology 12, 293–300. Castoria, R., Caputo, L., de Curtis, F., de Cicco, V., 2003. Resistance of postharvest biocontrol yeasts to oxidative stress: a possible new mechanism of action. Phytopathology 93 (5), 564–572. Chan, Z., Tian, S., 2005. Interaction of antagonistic yeasts against postharvest pathogens of apple fruit and possible mode of action. Postharvest Biology and Technology 36, 215–223. Chan, Z., Qin, G., Xu, X., Li, B., Tian, S., 2007. Proteome approach to characterize proteins induced by antagonist yeast and salicylic acid in peach fruit. Journal of Proteome Research 6 (5), 1677–1688. Chanchaichaovivat, A., Ruenwongsa, P., Panijpan, B., 2007. Screening and identification of yeast strains from fruits and vegetables: potential for biological control of postharvest chilli anthracnose (Colletotrichum capsici). Biological Control 42, 326–335. Droby, S., Chalutz, E., Wilson, C.L., Wisniewski, M.E., 1989. Characterization of the biocontrol activity of Debaryomyces hansenii in the control of Penicillium digitatum on grapefruit. Canadian Journal of Microbiology 35, 794–800. Droby, S., Chalutz, E., Cohen, L., Weiss, B., Wilson, C., Wisniewski, M., 1990. Nutrient competition as a mode of action of postharvest biocontrol agents. In: Wilson, C.L., Chalutz, E. (Eds.), Biological Control of Postharvest Diseases of Fruits and Vegetables, Workshop Proceedings. Agricultural Research Service document ARS-92. US Department of Agriculture, Washington, DC, pp. 142–160. Droby, S., Vinokur, V., Weiss, B., Cohen, L., Daus, A., Goldschmidt, E.E., Porat, R., 2002. Induction of resistance to Penicillium digitatum in grapefruit by the yeast biocontrol agent Candida oleophila. Phytopathology 92, 393–399. Droby, S., 2006. Yeast Metschnikowia fructicola NRRL Y-30752 for inhibiting deleterious microorganisms on plants. US Patent 6994849. Druvefors, U., 2004. Yeast biocontrol of grain spoilage mold: mode of action of Pichia anomala (Doctoral dissertation, Swedish University of Agricultural Sciences). Retrieved January 17, 2005, from Epsilon Dissertations and Graduate Theses Archive Web site: http://diss-epsilon.slu.se/archive/00000552/. Elad, Y., Chet, I., Henis, Y., 1982. Degradation of plant pathogenic fungi by Trichoderma harzianum. Canadian Journal of Microbiology. 28, 719–725. El-Ghaouth, A., Wilson, C., Wisniewski, M., 1998. Ultrastructural and cytochemical aspect of the biocontrol activity of Candida saitoana in apple fruit. Phytopathology 88, 282–291. Elmer, P.A.G., Reglinski, T., 2006. Biosuppression of Botrytis cinerea in grapes. Plant Pathology 55, 155–177. Farris, G.A., Mannazzu, I., Budroni, M., 1991. Identification of killer factor in the yeast genus Metschnikowia. Biotechnology Letters 13, 297–298. Giczey, G., Kerényi, Z., Füölop, L., Hornok, L., 2001. Expression of cmgl, an exo-b-1,3glucanase gene from Coniothyrium minitans, increases during sclerotial parasitism. Applied and Environmental Microbiology 67, 865–871. Guetsky, R., Shtienberg, D., Elad, Y., Fischer, E., Dinoor, A., 2002. Improving biological control by combining biocontrol agents each with several mechanisms of disease suppression. Phytopathology 92 (9), 976–985. Hadden, J.F., Black, L.L., 1989. Anthracnose of pepper caused by Colletotrichum spp. In: Green, S.K., Griggs, T.D., Mclean, B.T. (Eds.), Tomato and Pepper Production in the Tropics. AVRDC, Shauhua, Taiwan, pp. 189–191. Hofstein, R., Droby, S., Chalutz, E., Wilson, C.L., Frilender, B., 1990. Scaling-up the production for application of an antagonist-from basic research to R & D. In: Wilson, C.L., Chalutz, E. (Eds.), Biological Control of Postharvest Diseases of Fruits and Vegetables, Workshop Proceedings. Agricultural Research Service document ARS-92 US Department of Agriculture, Washington, DC, pp. 197–210.

A. Chanchaichaovivat et al. / Biological Control 47 (2008) 207–215 Jaffee, S., 2004. Delivering and Taking the Heat: Indian Spices and Evolving Product and Process Standards. The World Bank, USA. Janisiewicz, W.J., Tworkoski, T.J., Sharer, C., 2000. Characterizing the mechanism of biological control of postharvest diseases on fruits with a simple method to study competition for nutrients. Phytopathology 90, 1196–1200. Janisiewicz, W.J., Korsten, L., 2002. Biological control of postharvest diseases of fruits. Annual Review of Phytopathology 40, 411–441. Jijakli, M.H., Lepoivre, P., 1998. Characterization of an exo-b-1,3-glucanase produced by Pichia anomala strain K, antagonist of Botrytis cinerea on apples. Phytopathology 88, 335–343. Kim, B.S., Park, H.K., Lee, W.S., 1989. Resistance to anthracnose (Colletotrichum spp.) in pepper. In: Green, S.K., Griggs, T.D., Mclean, B.T. (Eds.), Tomato and Pepper Production in the Tropics. AVRDC, Shauhua, Taiwan, pp. 184–188. Klieber, A., 2000. Chilli Spice Production in Australia. Rural Industry Research and Development Corporation, Australia. Lakshmesha, K.K., Lakshmidevi, D.N., Mallikarjuna, A.S., 2005. Changes in pectinase and cellulose activity of Colletotrichum capsici mutants and their effect on anthracnose disease on capsicum fruit. Archives of Phytopathology and Plant Protection 38 (4), 267–279. Magliani, W., Conti, S., Gerloni, M., Bertolotti, D., Polonelli, L., 1977. Yeast killer systems. Clinical Microbiology Reviews 10 (3), 369–400. Mari, M., Guizzardi, M., 1998. The postharvest phase: emerging technologies for the control of fungal diseases. Phytoparasitca 26 (1), 59–66. Mehrotra, R.S., Aggarwal, A., 2003. Plant Pathology, second ed. Tata McGraw-Hill, New Delhi. Pal, K.K., Gardener, B.M., 2006. Biological control of plant pathogens. The Plant Health Instructor. Doi:10.1094/PHI-A-2006-1117-02. Paster, N., Droby, S., Chalutz, E., Menasherov, M., Nitzan, R., Wilson, C.L., 1993. Evaluation of the potential of the yeast Pichia guilliermondii as a biocontrol agent against Aspergillus flavus and fungi of stored soya beans. Mycological Research 10, 1201–1206. Piano, S., Neyrotti, V., Migheli, Q., Gullino, M.L., 1997. Biological control of Metschnikowia pulcherrima against Botrytis postharvest rot of apple. Postharvest Biology and Technology 11, 131–140. Roberts, W.K., Selitrennikoff, V.C.P., 1988. Plant and bacterial chitinases differ in antifungal activity. Journal of General Microbiology 134, 169–176. Saligkarias, I.D., Gravanis, F.T., Epton, H.A.S., 2002. Biological control of Botrytis cinerea on tomato plants by the use of epiphytic yeasts Candida guilliermondii

215

strains 101 and US7 and Candida oleophila strain I-182: II. A studies on mode of action. Biological Control 25, 151–161. Sawant, A.D., Abdelal, A.T., Ahearn, D.G., 1989. Purification and characterization of the anti-Candida toxin of Pichia anomala Wc 65. Antimicrobial Agents and Chemotherapy 33, 48–52. Scherm, B., Ortu, G., Muzzu, A., Budroni, M., Arras, G., Migheli, Q., 2003. Biocontrol activity of antagonistic yeasts against Penicillium expansum on apple. Plant Pathology 85, 205–213. Schlumbaum, A., Mauch, F., Vogeli, U., Boller, T., 1986. Plant chitinases are potent inhibitors of fungal growth. Nature 324, 365–367. Sela-Buurlage, M.B., Ponstein, A.S., Bres-Vloemans, B., Melchers, L.O., van den Elzen, P., Cornelissen, B.J.C., 1993. Only specific tobacco chitinases and b-1,3 glucanases exhibit antifungal activity. Plant Physiology 101, 857–863. Somers, J.M., Bevan, E.A., 1969. The inheritance of the killer character in yeast. Genetical Research Communication 13, 71–83. Suzuki, C., Nikkuni, S., 1989. Purification and properties of the killer toxin produced by halotolerant yeast, Pichia farinisa. Agricultural and Biological Chemistry 53, 2599–2604. Tian, S., Yao, H., Deng, X., Xu, X., Qin, G., Chan, Z., 2007. Characterization and expression of b-1,3-glucanase genes in jujube fruit induced by the microbial biocontrol agent, Cryptococcus laurentii. Phytopathology 97 (3), 260–268. van Loon, C., Bakker, P.A.H.M., Pieterse, C.M.J., 1998. Systemic resistance induced by rhizosphere bacteria. Annual Review of Phytopathology 36, 453–483. Vogel, H.L., 1956. A convenient growth medium for Neurospora (Medium N). Microbiology Genetic Bulletin 13, 42–44. Wirth, S.J., Wolf, G.A., 1990. Dye-labelled substrates for the assay and detection of chitinase and lysozyme activity. Journal of Microbiological Methods 12, 197– 205. Wisniewski, M.E., Biles, C., Droby, S., 1990. The use of the yeast Pichia guilliermondii as a biocontrol agent: characterization of attachment to Botrytis cinerea. In: Wilson, C.L., Chalutz, E. (Eds.), Biological Control of Postharvest Diseases of Fruits and Vegetables, Workshop Proceedings. Agricultural Research Service document ARS-92. US Department of Agriculture, Washington, DC, pp. 167– 183. Wisniewski, M.E., Biles, C.L., Droby, S., McLaughlin, R., Wilson, C.L., Chalutz, E., 1991. Mode of action of postharvest biocontrol yeast, Pichia guilliermondii. I. Characterization of attachment to Botrytis cinerea. Physiological and Molecular Plant Pathology 39, 245–258.