Screening of Plant Epiphytic Yeasts for Biocontrol of Bacterial Fruit Blotch (Acidovorax Avenae Subsp

中国科技论文在线 http://www.paper.edu.cn

Biological Control 50 (2009) 164–171

Contents lists available at ScienceDirect

Biological Control

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

Screening of plant epiphytic yeasts for biocontrol of bacterial fruit blotch (Acidovorax avenae subsp. citrulli) of hami melon

Xiaodong Wang a,b,c, Guoqing Li a,b,*, Daohong Jiang a,b, Hung-Chang Huang d

a The State Key Laboratory of Agricultural Microbiology, Huazhong Agricultural University, No. 1 of the Lion Mountain Street, Hong Shan District, Wuhan, Hubei Province 430070, China b The Key Laboratory of Plant Pathology of Hubei Province, Huazhong Agricultural University, Wuhan 430070, China c Department of Plant Protection, Shihezi University, Shihezi, Xinjiang 832003, China d Agriculture and Agri-Food Canada, Research Centre, Lethbridge, Alta., Canada T1J 4B1

article info abstract

Article history: Bacterial fruit blotch (BFB) caused by Acidovorax avenae subsp. citrulli (Aac) is a serious disease of hami Received 1 December 2008 melon (Cucumis melo var. saccharinus) in Northern China. A study was conducted to screen plant epi- Accepted 23 March 2009 phytic yeasts for use as biocontrol agents of BFB. Results showed that 24 out of 463 yeast strains isolated Available online 31 March 2009 from leaves or flowers of plants collected from three provinces in China were antibiotic against Aac on agar medium and eight antagonistic yeast strains including strain 0732-1 formed inhibition zones larger Keywords: than 18 mm in diameter. Spray application of strain 0732-1 isolated from watermelon grown in Xinjiang Hami melon was effective in reducing incidence and severity of disease caused by Aac on leaves of hami melon. Treat- Cucumis melo var. saccharinus ment of hami melon seeds with cell-free cultural filtrates of the yeast strain 0732-1 resulted in a signif- Bacterial fruit blotch Acidovorax avenae subsp. citrulli icant reduction in severity of seedling blight caused by seedborne Aac, and the efficacy was not Pichia anomala significantly different (P > 0.05) from that of chemical seed treatments including streptomycin sulfate 0732-1 (0.1%, w/v) and hydrochloric acid (2%, v/v). Based on morphological and physiological characteristics Biocontrol and analysis of the DNA sequence of the internal transcribed spacer of ribosomal DNA, the yeast strain 0732-1 was identified as Pichia anomala Kurtzman. This study suggests that the yeast strain 0732-1 is an agent with potential for biocontrol of BFB of hami melon caused by Aac. Ó 2009 Elsevier Inc. All rights reserved.

1. Introduction watermelon caused by BFB in USA reached 5–50%, depending on the stage of infection and environmental conditions (Latin and Bacterial fruit blotch (BFB) caused by Acidovorax avenae subsp. Hopkins, 1995; Lessl et al., 2007). In China, BFB was reported on citrulli (Schaad et al.) Willems et al. (formerly Pseudomonas pseud- watermelon in 1998 in Hainan province (Zhang et al., 1998) and oalcaligenes subsp. citrulli)(Willems et al., 1992) is a serious dis- on hami melon in 2001 in Xinjiang Uygur Autonomous Region ease of cucurbits including hami melon (Cucumis melo L. var. (Zhao et al., 2001), also resulting in heavy losses to these two crops. saccharinus Naud.). The pathogen also causes BFB on watermelon Since BFB is a seedborne disease (Latin and Hopkins, 1995; Wal- (Citrullus lanatus L.) in many countries including the USA (Hopkins, cott et al., 2003), control strategies for this disease include preven- 1989; Latin and Rane, 1990; Somodi et al., 1991; Jacobs et al., 1992; tive measures such as use of pathogen-free seeds and seedlings, Black et al., 1994; Hamm et al., 1997; Langston et al., 1999), seed fermentation and seed treatment with peroxyacetic acid, Australia (Martin and Horlock, 2002), Brazil (Silveira et al., 2003), hydrochloric acid or copper-containing bactericides (Zhao et al., Canada (Walcott et al., 2004), China (Zhang et al., 1998; Zhao 2003; Hopkins et al., 2003; Feng et al., 2007). Although preventive et al., 2001), Israel (Burdman et al., 2005), Japan (Shirakawa control measures could provide some efﬁcacy against BFB (Fesse- et al., 2000), Mariana Islands (Wall et al., 1990), Thailand (Walcott haie and Walcott, 2005), seed fermentation and seed treatment et al., 2004) and Turkey (Demir, 1996). Besides fruit blotch, with chemicals sometimes could have negative impacts on seed A. avenae subsp. citrulli (Aac) also causes leaf blight, seedling blight quality, including reduction of seed germination and seedling and/or blossom rot of cucurbitaceous plants. Annual yield loss of growth (Hopkins et al., 2003; Zhao et al., 2003). Moreover, public concerns over the use of chemical pesticides suggest a need to uti- lize other environmentally safe measures for control of BFB on * Corresponding author. Address: The State Key Laboratory of Agricultural watermelon or hami melon. Microbiology, Huazhong Agricultural University, No. 1 of the Lion Mountain Street, Hong Shan District, Wuhan, Hubei Province 430070, China. Biological control is considered an ecologically sound and envi- E-mail address: [email protected] (G. Li). ronmentally friendly approach for management of plant diseases

X. Wang et al. / Biological Control 50 (2009) 164–171 165

including BFB on cucurbits (Fravel, 2005). Fessehaie and Walcott ensis Kuntze) grown in Xin Yang of Henan province, China. Each (2005) reported that seed treatment with either Acidovorax avenae plant sample was cut into pieces of ca 1 1 cm (width length), subsp. avenae AAA99-2 or Pseudomonas fluorescens A506 was effec- and 10 g of the sample was added to 100 ml of sterilized distilled tive in reducing incidence of BFB on seedlings of watermelon under water (SDW) in a 250 ml Erlenmeyer flask. After shaking growth chamber conditions. They also found that treatment of Aac- (150 rpm) on a rotary shaker (Model HQL150B, the Instrument contaminated blossoms of watermelon with each of these two Company of the Chinese Academy of Sciences, Wuhan, China) for antagonists significantly lowered the percentage of watermelon 30 min, the mixture in each flask was filtered through a single-lay- seeds infested with Aac under greenhouse conditions (Fessehaie ered cheesecloth. The resulting filtrate was designated as 101 and and Walcott, 2005). Other antagonistic microorganisms such as was serially diluted up to 102,103,104 and 105 with SDW. Ali- Bacillus spp. (Santos et al., 2006) and Streptomyces spp. (Yaeram quots of each diluted suspension were pipetted onto PDA medium et al., 2006), and unidentified endophytic bacteria isolated from amended with 0.1% (w/v) streptomycin sulfate in a Petri dish (9 cm hami melon (Lu and Luo, 2004) were reported as potential biocon- diameter), 100 ll per dish, and the suspension drop in each dish trol agents of BFB. However, information on control of Aac on was spread evenly using a flame-sterilized glass rod. The dishes cucurbits by yeasts remains unavailable. were incubated at 28 °C for 2 days and yeast colonies on the med- Yeasts are single-cell fungi, which are widely distributed in soil ium were individually identified by appearance, color and smell. and on the surface of leaves, stems, flower petals, fruits and seeds Yeast cells of selected colonies were transferred onto PDA slants of plants (Wilson and Wisniewski, 1989). Numerous studies indi- in glass tubes (18 2 cm, length diameter) using a sterilized cated that some yeast species or isolates are ideal biocontrol agents, inoculation loop. The stock cultures were labelled, incubated at as they are natural plant epiphytic colonizers, nonpathogenic to 28 °C for 4 days and stored at 4 °C until use. plants and human beings in most cases and can rapidly proliferate For preparation of yeast cultures, 100 ll of a cell suspension of (Wilson and Wisniewski, 1989). Commercial yeast products such each yeast strain (1 108 cells/ml) was inoculated into 100 ml of as Aspire based on Candida oleophila Montrocher (Ecogen Inc., Lang- PDB in an Erlenmeyer flask (250 ml) and the cultures were incu- horn, PA, USA), YieldPlus based on Cryptococcus albidus (Saito) Skin- bated at 28 °C for 2 days. The concentration of yeast cells in each ner (Anchor Yeast, Cape Town, South Africa), have been developed yeast culture was determined using a haemocytometer and a com- for control of postharvest diseases of fruits and vegetables caused pound light microscope. The two-day-old yeast cultures were used by a wide range of phytopathogenic fungi, including Botrytis cinerea, for in vitro and in vivo screenings of yeast strains. Colletotrichum spp., Penicillium spp. and Rhizopus spp. (Fravel, 2005). The objectives of this study were (i) to screen yeast strains 2.3. Antibiosis of yeast strains against Aac (in vitro screening) antagonistic to Aac; (ii) to evaluate the potential of the selected yeast strain 0732-1 for control of Aac on leaves and seedlings of Yeast strains were tested for antibiosis against Aac using the hami melon; and (iii) to characterize the identity of the yeast strain method described by Arun et al. (2007) with modifications. An ali- 0732-1. quot of 500 ll of a 2-day-old liquid culture of Aac grown in KB medium (1 109 cfu/ml) was pipetted onto 120 ml of KBA medium in a Petri dish (18 cm diameter) and spread evenly on the surface of the 2. Materials and methods agar medium. A sterilized cork borer (6 mm diameter) was used to remove agar pieces and create 20 wells in each dish, at 3 cm between 2.1. Bacterial strain, media and preparation of bacterial cultures wells, and used for testing 19 yeast strains. An aliquot of 100 llofa2- day-old culture (28 °C) of a yeast strain was pipetted into a well. Strain XJ05-1 of A. avenae subsp. citrulli (Aac) used in this study Wells filled with PDB at 100 ll/well were used as control. For each was isolated from a diseased plant of hami melon grown in Chang dish, there were 19 wells for 19 yeast strains and one well for the Ji of Xinjiang Uygur Autonomous Region, China. Cultural media control. There were three replicates for each yeast strain. The cul- used in this study included King’s medium B (KB), King’s medium tures were incubated at 28 °C for 2 days and the presence of a clear B agar (KBA), potato dextrose agar (PDA), potato dextrose broth zone surrounding a well in each dish was recorded. (PDB) and soybean sprouts extract broth (SSEB). KB was prepared To confirm the production of antibacterial substances by the according to the description by King et al. (1956), whereas KBA yeast strain 0732-1, an aliquot of 60 ll of a cell suspension was a modified KB containing 20 g of agar per liter. KB and KBA (1 108 cells/ml) of this strain was inoculated in a 150 ml flask were used for culturing Aac and KBA was also used for testing containing 60 ml of PDB and a total of 27 flasks were inoculated. the antibiotic activity of yeast strains against Aac. PDA and PDB The flasks were mounted on the shaker (Model HQL150B, the were made of fresh potato according to the procedure described Instrument Company of the Chinese Academy of Sciences, Wuhan, by Fang (1998) and used for isolating and culturing yeasts. SSEB China) and incubated at 28 °C and 150 rpm. Three flasks of the contained the same ingredients as PDB except for replacement of yeast cultures were randomly removed from the shaker at 12 h peeled potato with fresh soybean sprouts. intervals for a period of 96 h. The yeast suspension in each flask Aac cultures were prepared by adding an aliquot of 100 llofa was centrifuged for 15 min at 4 °C and 12,000 rpm. The superna- bacterial suspension of Aac (1 109 cfu/ml) into 50 ml of KB med- tant was pipetted out and filter-sterilized. The cell-free cultural fil- ium in an Erlenmeyer flask (150 ml). The culture was incubated at trate was tested for pH using a pH meter (Model pHS-3C, Shanghai 28 °C for 2 days and used for in vitro and in vivo screening of yeast Hongyi Instrument Company, Ltd., Shanghai, China) and for inhibi- strains. tion of Aac growth on Aac-containing KBA plates (9 cm diameter) using the method described previously. The precipitate (yeast 2.2. Isolation of yeast strains and preparation of yeast cultures cells) was dried at 60 °C overnight and weighed.

Yeast strains were isolated from leaves or ﬂowers of cucumber 2.4. Suppression of Aac infection of hami melon leaves by antagonistic (Cucumis sativus L.), celery (Apium graveloens L.), coriander (Corian- yeasts (in vivo screening) drum sativum L.) and lettuce (Lactuca sativa L.) grown in Wuhan of Hubei Province, China, from leaves of watermelon (Citrullus lanatus Seeds of hami melon (C. melo var. saccharinus cultivar 86-1) L.), soybean (Glycine max L.) and apricot (Prunus armenica L.) grown were sown in autoclaved loess-clay soil in plastic pots in Shi He Zi of Xinjiang, and from leaves of tea plants (Camellia sin- (10 15 cm, height diameter), 10 seeds/pot. The pots were kept 中国科技论文在线 http://www.paper.edu.cn

166 X. Wang et al. / Biological Control 50 (2009) 164–171 in a greenhouse (30–35 °C and 85–100% relative air humidity) and Aac suspension (1 106 cfu/ml) for 30 min and filtered through plants were watered as required. After emergence, the seedlings double-layered cheesecloth to remove the excess bacterial suspen- were thinned to 6 seedlings per pot. At the 2-true-leaf stage, the sion. The seeds were then air-dried and divided into four lots, 72 seedlings in each pot were sprayed with 10 ml of PDB cultures seeds/lot. The four seed lots were soaked in the cell-free cultural (1–5 108 cells/ml) of each selected yeast strain. The hami melon filtrate of the yeast strain 0732-1 (pH3.4), 2% (v/v) HCl (Shennong seedlings sprayed with SDW was used as controls. There were Chemical Reagent Co. Ltd., Yi Chang, China), 0.1% (w/v) streptomy- three pots (replicates) of hami melon seedlings for each selected cin sulfate (The Northern China Pharmaceutical Company Ltd., Shi yeast strain and six pots of hami melon seedlings for the control Jia Zhuang, China) or SDW (control), respectively. The seeds of the treatments. The hami melon seedlings pre-treated with yeasts for four treatments were separately filtered, air-dried and individually 24 h were then inoculated with Aac by spraying the bacterial bac- sown in autoclaved soil in plastic pots (10 15 cm, height diam- terial suspension (1 108 cfu/ml) of Aac on these seedlings at eter) at 8 seeds/pot and 9 pots/treatment with three pots of each 10 ml per pot. Three pots of hami melon seedlings pre-treated with treatment being treated as one replicate. The pots were kept under SDW were sprayed with the bacterial suspension of Aac as the po- outdoor conditions (30–33 °C) with treatments being arranged in a sitive control and the other three pots of hami melon seedlings randomized complete block design. Ten days later, seedlings in pre-treated with SDW were sprayed with SDW as the negative con- each pot were examined for blight symptoms on cotyledons, true trol. All the hami melon seedlings were incubated in the green- leaves and stems. The experiment was repeated once. house for 3 days and then examined for incidence and severity of disease on leaves. Disease incidence was defined as percentage of 2.6. Identification of the yeast strain 0732-1 diseased leaves in total leaves examined in each pot. Disease severity of each leaf was rated using a scale of 0–4, where 0 represented The yeast strain 0732-1 was identified by morphological and healthy, and 1, 2, 3 and 4 represented diseased with lesion area physiological features and molecular feature of the internal tran- <15%, 16–50%, 51–80% and >80%, respectively. Disease index (DI) scribed spacer (ITS) of the ribosomal gene. The cultures on PDA of hami melon leaf blight for each pot was calculated by the follow- at 28 °C for 2 days were used for observation of colony morphology ing formula (Fang, 1998): (size, color and texture), and the cultures in SSEB at 28 °C and , 150 rpm for 2 days were used for observation of morphology and X4 X4 proliferation of yeast cells, whereas the cultures on autoclaved car- DI ¼ 100 ðLi iÞ 4 ðLiÞ i¼0 i¼0 rot slices at 20 °C for 10 days were used for observation of ascospores (Barnett et al., 1983; Kurtzman and Fell, 1998; Liang and Chi, 2002). A minimum of 50 cells or ascospores were measured for size where i is the leaf blight severity rating, and Li is the number of of cells or ascospores of the yeast strain 0732-1. Physiological char- hami melon leaves in each pot corresponding to the disease severity acteristics of the yeast strain 0732-1 were determined by detecting rating i. The experiment was repeated three times. its ability to ferment sugars under semi-anaerobic conditions; and to assimilate a variety of carbon compounds as the major carbon 2.5. Suppression of seed transmission of Aac by cultural filtrates of the sources or to assimilate a variety of nitrogen compounds as the yeast strain 0732-1 major nitrogen sources under aerobic conditions (Barnett et al., 1983). The sugars used in the fermentation test included D-glucose, Seeds of hami melon (cv 86–1) were surface-sterilized with 2% D-galactose, maltose, sucrose and lactose. The carbon compounds sodium hypochlorite (v/v) for 15 min and rinsed twice in SDW. used in the assimilation test included D-glucose, D-galactose, malt- After air-drying for 2 days, seeds were soaked in 100 ml of an ose, lactose, D-xylose, starch D-mannose, inositol, L-arabinose, amri-

Fig. 1. A histogram showing inhibition of growth of Acidovorax avenae subsp. citrulli (Aac) by different yeast strains on King’s medium B agar (28 °C, 2 days). Inoculum of yeasts was from cultures in potato dextrose broth and Aac from cultures in King’s medium B. Vertical bars represent the standard errors of means (n = 3). Bars headed by the same letters are not signiﬁcantly different (P > 0.05) according to Duncan’s Multiple Range Test. 中国科技论文在线 http://www.paper.edu.cn

X. Wang et al. / Biological Control 50 (2009) 164–171 167

ta, glycerol, ethanol, methanol and sorbitol. The nitrogen com- ery (Hubei), coriander (Hubei), apricot (Xinjiang), cucumber (Hu- pounds used in the assimilation test included L-lysine, creatine, bei), lettuce (Hubei) and tea plants (Henan), respectively. In ammonium sulfate and potassium nitrate. Additional tests such contrast, the remaining 439 yeast strains were not antibiotic as production of acids, esters and starch, and growth at 42 °Cin against Aac, as no clear zones surrounding agar wells containing the presence of high concentration of D-glucose (50 or 60%, w/v) PDB cultures of these yeast strains appeared after incubation at were included (Barnett et al., 1983). Each test was repeated three 28 °C for 2 days. times. For analysis of the ITS sequence, genomic DNA was extracted from cells of the yeast strain 0732-1 collected from 2-day-old PDA cultures using the procedure described by Heras-Vazquez et al. (2003). Polymerase chain reaction (PCR) based method was used to amplify the ITS region (ITS1–5.8S rDNA–ITS2) with primers ITS1 (50-TCCGTAGGTGAACCTGCGG-30) and ITS4 (50-TCCTCCGCTTA TTGATATGC-30)(White et al., 1990). Amplification was performed in a PTC-100TM Peltier Thermal Cycler (Hercules, CA, USA) pro- grammed as follows: the initial denaturation at 95 °C for 5 min, followed by 35 cycles each beginning with denaturation at 94 °C for 30 s, annealing at 52 °C for 30 s and ending with DNA synthesis at 72 °C for 1 min, and the final extension at 72 °C for 5 min. The amplified DNA fragment was purified from agarose gel after elec- trophoresis using the Fermentas DNA Extraction Kit (Jingmei Bio- tech Co., Ltd., Shengzhen, China), ligated into the pMD18-T vector (TaKaRa Biotechnology Co., Ltd., Dalian, China) and transformed into competent cells of Escherichia coli JM109. Positive E. coli clones grown on Luria–Bertani agar medium containing ampicillin (50 lg/ml) were selected, individually tested for the size of the DNA insert by the PCR method and three E. coli clones harboring the plasmid containing the expected DNA insert were sent to Sin- oGenoMax Co. Ltd. (Beijing, China) for sequencing on the ABI PRISM 377–96 automated sequencer with the universal primer M13. The sequence were aligned using the CLUSTAL W program and compared with all available sequences in GenBank database through internet (http://www.ncbi.nlm.nih.gov) using Basic Local Alignment Search Tool (BLAST) (Altschul et al., 1990).

2.7. Data analyses

Data on diameter of inhibition zones, disease incidence and disease index for the antagonistic yeast strains used in this study were analyzed using analysis of variance (ANOVA) in SAS software (SAS Institute, Cary, NC, USA, Version 8.0, 1999). Data on the disease incidence (percentage) were arcsin-transformed to angular data prior to ANOVA. After each analysis, means were individually back-transformed to numerical values. Means for different yeast strains or different treatments in each experiment or trial were separated using Duncan’s Multiple Range Test at P= 0.05 level.

3. Results

3.1. Isolation and screening of antagonistic yeast strains (in vitro screening)

A total of 463 yeast strains were isolated from 283 plant sam- ples collected from three provinces (Henan, Hubei and Xinjiang) in China. Results of in vitro screening showed that 24 of the 463 yeast strains tested on KBA medium were antibiotic against A. avenae subsp. citrulli (Aac), as indicated by the appearance of clear zones surrounding agar wells containing PDB cultures of these yeast strains after incubation at 28 °C for 2 days. The average diameter of clear zones varied with yeast strains, ranging from 8.0 mm Fig. 2. Antibacterial activity of the yeast strain 0732-1 against Acidovorax avenae for strain 074101-2 to 19.7 mm for strain 0732-1 (Fig. 1). There subsp. citrulli (Aac). A, Inhibition of Aac growth by cell-free cultural ﬁltrates of the was no signiﬁcant difference (P > 0.05) in the average diameter of yeast strain 0732-1 on King’s medium B agar (28 °C, 2 days) showing inhibition zone (right well), whereas sterile distilled water (left well) did not form any clear zones among yeast strains 0732-1, 074111-5, 074112-1, inhibition zones. (B–D) Time-course of the yeast biomass (B), ambient pH (C) and 074102-2, 07457-1, 703030, 19-1 and xin12-6. These yeast strains the antibacterial activity of the liquid cultures of the yeast strain 0732-1 against Aac were isolated from watermelon (Xinjiang), soybean (Xinjiang), cel- (D). Vertical bars represent the standard errors of means (n = 3). 中国科技论文在线 http://www.paper.edu.cn

168 X. Wang et al. / Biological Control 50 (2009) 164–171

Fig. 3. Effect of spray-treatment of leaves of hami melon at the seedling stage with cultures of the yeast strain 0732-1 (A) or sterile distilled water (B) on severity of leaf blight caused by Acidovorax avenae subsp. citrulli.

Inhibition of Aac growth by the cell-free cultural filtrate of the yeast strain 0732-1 was observed (Fig. 2A). After incubation in PDB at 28 °C for 24–96 h, strain 0732-1 grew and the biomass Fig. 4. Effect of seed treatment with the cultural filtrates of the yeast strain 0732-1 (0732-1), streptomycin sulfate (Strep) or hydrochloric acid (HCl) on seedling blight reached 2.0–2.9 mg/ml (Fig. 2B) and the pH of the liquid culture of hami melon caused by Acidovorax avenae subsp. citrulli (Aac). (A) Seedlings of declined from the initial of 6.2 to the final of 3.2–4.4 (Fig. 2C). 10 days after inoculation showing severe seedling blight in the treatment of Aac The average diameter of inhibition zones caused by the cell-free alone (Aac), but slight seedling blight in the treatments of the yeast strain 0732-1 cultural filtrate of strain 0732-1 was consistently increased from (0732-1), streptomycin sulfate (Strep) and hydrochloric acid (HCl). (B) A histogram 4.0 mm after incubation for 24 h to 19.4 mm after incubation for showing difference in disease incidence of seedling blights of hami melon for these four treatments. Vertical bars represent the standard errors (n=3). 96 h in three replicates (Fig. 2D).

3.2. Effect of antagonistic yeasts on Aac infection of leaves of hami control (Aac) were severely diseased in all the three tests (Fig. 3 melon (in vivo screening) and Table 1). The yeast strain 0732-1 showed consistent suppression of leaf blight symptoms of hami melon caused by Aac with Eleven antagonistic strains of yeasts were evaluated for control reduction of disease incidence to 43.1–48.8% and disease index to of bacterial fruit blotch leaf symptoms on hami melon. Results 13.1–19.1, compared to the disease incidence of 75.3–90.0% and showed that while the hami melon leaves in the negative control the disease index of 40.0–75.3 for the Aac treatment (Fig. 3 and Ta- (SDW alone) were healthy, the hami melon leaves in the positive ble 1). In contrast, the yeast strains 079-1, 074102-2, 074112-1 and

Table 1 Efﬁcacy of different yeast strains in suppression of leaf blight of hami melon caused by Acidovorax avenae subsp. citrulli (Aac).

Yeast strain Test 1 Test 2 Test 3 Disease incidence (%) Disease index (0–100) Disease incidence (%) Disease index (0–100) Disease incidence (%) Disease index (0–100) CK11 0.0c3 0.0e3 0.0e3 0.0f3 0.0d3 0.0d3 CK22 90.0a 40.0a 90.0a 73.2a 82.0a 75.3a 0732-1 48.8b 14.2d 46.4d 19.1e 43.1c 13.3cd 07488-1 51.5b 17.0cd 64.4b 26.2de 54.8bc 34.5b 07463-4 54.4b 16.4cd 82.6a 72.6a ND ND 074111-5 55.0b 25.0b ND4 ND ND ND xin12-6 57.1b 19.2cd 54.8bcd 46.4bc 50.8bc 28.3b 0714-1 57.6b 17.5cd 52.0cd 39.4cd 58.5b 33.3b 19-1 58.5b 20.8bcd 59.0bc 51.4bc 62.2b 30.6b 074112-1 69.0b 37.5a ND ND ND ND 079-1 81.1a 37.5a 54.8bcd 34.5cde 50.0bc 25.8bc 074102-2 82.0a 30.8ab 90.0a 60.0ab ND ND 07114-1 90.0a 30.0ab 90.0a 46.7bc ND ND

1 CK1, negative control without inoculation of Aac or yeast cultures. 2 CK2, positive control with inoculation of Aac alone. 3 Means followed by the same letters within each column are not signiﬁcantly different (P > 0.05) according to Duncan’s Multiple Range Test. 4 ND, not determined. 中国科技论文在线 http://www.paper.edu.cn

X. Wang et al. / Biological Control 50 (2009) 164–171 169

Fig. 5. Pichia anomala strain 0732-1 showing formation of colonies (A) on potato dextrose agar at 28 °C for 2 days and pseudo-hyphae at the colony margin (B), formation of yeast cells and small budding propagules in soybean sprout extract broth at 28 °C for 2 days (C) and formation of hat-shaped ascospores (arrowheads) on autoclaved carrot slices at 20 °C for 10 days (D).

07114-1 were ineffective or slightly effective for suppression of Results of the physiological tests showed that under leaf blight of hami melon caused by Aac (Table 1). The remaining semi-anaerobic conditions, the yeast strain 0732-1 could ferment six yeast strains (07488-1, 07463-4, 074111-5, 0714-1, 19-1 and D-glucose, D-galactose, maltose and sucrose, but could not ferment xin12-6) were intermediate regarding efﬁcacy of reducing disease lactose (Table 2). Under aerobic conditions, it could assimilate D- incidence or disease index on leaves of hami melon (Table 1). glucose, D-galactose, maltose, lactose, D-xylose, starch, D-mannose, D-arabinose, mannitol, glycerol, ethanol and sorbitol, but could not 3.3. Suppression of Aac seedling blight of hami melon by the yeast assimilate methanol, lactose and inositol, as the major source car-

strain 0732-1 bon; and could assimilate L-lysine, KNO3 and (NH4)2SO4, but could not assimilate creatine, as the major source nitrogen (Table 2). Treatment of hami melon seeds with Aac alone resulted in Meanwhile, cultures of the yeast strain 0732-1 at 42 °C showed development of seedling blights with severe symptoms on cotyle- production of acids and esters, but not starch (Table 2). The phys- dons and leaves (Fig. 4A). The disease incidence was high, 97.6% in iological characteristics of the yeast strain 0732-1 were also similar the first trial and 83.3% in the second trial (Fig. 4B). In contrast, to those for P. anomala described by Barnett et al. (1983). treatment of hami melon seeds with the cultural filtrates of the The ITS and the flanking regions of the rDNA of the yeast strain yeast strain 0732-1, streptomycin sulfate or HCl resulted in devel- 0732-1 comprised 617 base pairs. Without the flanking regions, opment of seedling blights with mild symptoms on cotyledons and the sequence of ITS (ITS1+5.8S rDNA+ITS2) is composed of only leaves (Fig. 4A). The disease incidence was low, 2.8, 0 and 2.6% in 529 base pairs. This DNA sequence was submitted to GenBank un- the first trial and 7.0, 8.8 and 3.7% in the second trial, for the treat- der the Accession No. EU380207. BLAST analysis showed that the ments of strain 0732-1, streptomycin sulfate and HCl, respectively ITS sequence of the yeast strain 0732-1 was identical to that for (Fig. 4B). Pichia anomala strains WM 828, MTCC 237, MCCL 13, MCCL 1569, MCCL 209/2K and FY-102, which were assigned with GenBank 3.4. Identification of the yeast strain 0732-1 Accession Nos. DQ249195, AY231606, AY231609, AY231610, AY231611 and AY270936, respectively. The yeast strain 0732-1 was identified as Pichia anomala (Han- sen) Kurtzman on the basis of morphological and physiological 4. Discussion characteristics and analysis of the ITS sequence. Colonies of strain 0732-1 on PDA for 2 days were circular in shape, 0.6–1.1 cm in Numerous studies indicate that some yeast species, including diameter, smooth surface, milky white in color (Fig. 5A) and with Pichia anomala, are effective agents for control of post-harvest dis- pseudohyphae around the colony margin (Fig. 5B). Cells of the eases caused by fungal pathogens in crops such as fruits, vegeta- yeast strain 0732-1 were oval in shape, 5–7 3–5 lm in size bles (Wilson and Wisniewski, 1989; Fravel, 2005; Lassois et al., (length width) and multiplied by budding (Fig. 5C). Ascospores 2008) and cereal grains (Petersson and Schnürer, 1995, 1998; Pet- of strain 0732-1 were hat-shaped and 3.5–4.2 1.2–1.8 lm in size ersson et al., 1999; Druvefors et al., 2002). However, information (Fig. 5D). These morphological characteristics of strain 0732-1 on control of plant bacterial diseases by yeasts is rare. This study matched the descriptions for P. anomala (Barnett et al., 1983). demonstrates the effectiveness of P. anomala strain 0732-1 as a 中国科技论文在线 http://www.paper.edu.cn

170 X. Wang et al. / Biological Control 50 (2009) 164–171

Table 2 cantly enhanced the antibiotic activity and biocontrol efﬁcacy of Comparison of physiological characteristics between the yeast strain 0732-1 and Pantoea agglomerans strain E328 against Erwinia amylovora, the cau- Pichia anomala (Pa). sal agent of ﬁre blight of pear and apple. This phenomenon might oc- Test 0732-1 Pa1 cur in the interaction between P. anomala strain 0732-1 and Aac on Fermentation hami melon. Additional studies are required to characterize the nat- D-Glucose + + ure of acidic compounds produced by P. anomala strain 0732-1 and D-Galactose + V the role of the acidic compounds in biocontrol of Aac. Maltose + V Sucrose + + Lactose Acknowledgements

Assimilation This research was funded by the Natural Science Foundation of Carbon source D-Glucose + + China (Grant No. 30570079). We thank Mr. M.D. Wu of Huazhong D-Galactose + V Agricultural University for help of graphic preparations. Maltose + + Lactose References D-Xylose + V Starch + + D-Mannose + NL Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local Inositol NL alignment search tool. Journal of Molecular Biology 215, 403–410. Arun, C., Pintip, R., Bhinyo, P., 2007. Screening and identification of yeasts strains L-Arabinose + from fruits and vegetables: potential for biological control of postharvest chilli Mannitol + + anthracnose (Colletotrichum capsici). Biological Control 42, 326–335. Glycerol + + Black, M.C., Isakeit, T., Barnes, L.W., Kucharek, T.A., Hoover, R.J., Hodge, N.C., 1994. Ethanol + + First report of bacterial fruit blotch of watermelon in Texas. Plant Disease 78, Methanol 831. Sorbitol + + Burdman, S., Kopelowitz, J., Kritzman, G., Kots, N., 2005. Molecular, physiological, Nitrogen source and host-range characterization of Acidovorax avenae subsp. citrulli isolates from watermelon and melon in Israel. Plant Disease 89, 1339–1347. L-Lysine + + Barnett, J.A., Payne, R.W., Yarrow, D., 1983. Yeasts: characteristics and Creatine identification. Cambridge University Press. pp. 306–307. KNO ++ 3 Comitini, F., Ingeniis De, J., Pepe, L., Mannazzu, I., Ciani, M., 2004. Pichia anomala and (NH4)2SO4 ++Kluyveromyces wickerhamii killer toxins as new tools against Dekkera/ Brettanomyces spoilage yeasts. FEMS Microbiology Letters 238, 235–240. Additional tests Demir, G., 1996. A new bacterial disease of watermelon in Turkey: bacterial fruit Starch production blotch of watermelon (Acidovorax avenae subsp. citrulli (Schaad et al.) Willems Grew at 42 °C+NLet al.. Journal of Turkey Phytopathology 28, 43–49. In 50% D-Glucose + V Druvefors, U., Jonsson, N., Boysen, M.E., Schnürer, J., 2002. Efficacy of the biocontrol In 60% D-Glucose + V yeast Pichia anomala during long-term storage of moist feed grain under Acids production (PDB) + NL different oxygen and carbon dioxide regimens. FEMS Yeast Research 2, 389– Ester production + + 394. Feng, J.J., Chen, K.J., Kim, J.Y., Xu, Y., Zhou, X.Y., Li, J.Q., 2007. Effect of seed priming 1 Physiological characteristics of Pichia anomala (Pa) were listed in Barnett et al. on seedling growth and disinfection to Acidovorax avenae subsp. citrulli in (1983). ‘‘+”, ‘‘” and ‘‘V” represent positive reaction, negative reaction and variable triploid watermelon seeds. Acta Phytopathologica Sinica 37, 528–534. reaction, respectively. ‘‘NL” represents the characteristics not listed in Barnett et al. Fessehaie, A., Walcott, R.R., 2005. Biological control to protect watermelon blossoms (1983). and seed from infection by Acidovorax avenae subsp. citrulli. Phytopathology 95, 413–419. Fang, Z.D., 1998. Research Methodology for Plant Diseases (third ed.). Chinese biocontrol agent for control of leaf blight and seedling blight of Agriculture Press, Beijing. pp. 8–12. hami melon caused by A. avenae subsp. citrulli. Since A. avenae Fravel, D.R., 2005. Commercialization and implementation of biocontrol. Annual subsp. citrulli is an important bacterial pathogen of cucurbitaceous Review of Phytopathology 43, 337–359. Friel, D., Pessoa, N.M.G., Vandenbol, M., Jijakli, M.H., 2007. Separate and combined crops, including hami melon. Further investigation is warranted on disruptions of two exo-b-1, 3-glucanase genes decrease the efficiency of Pichia efficacy and practicality of P. anomala strain 0732-1 in control of anomala (strain K) biocontrol against Botrytis cinerea on apple. Molecular bacterial fruit blotch of hami melon under field conditions. Plant-Microbe Interaction 20, 371–379. Hamm, P.B., Spink, D.S., Clough, G.H., Mohan, K.S., 1997. First report of bacterial fruit Previous studies showed that P. anomala could suppress myce- blotch of watermelon in Oregon. Plant Disease 81, 113. lial growth or sporulation of Penicillium spp. and Aspergillus spp. in Heras-Vazquez, F.J.L., Mingorance-Cazorla, L., Clemente-Jimenez, J.M., Rodriguez- cereal grains stored under oxygen-limited conditions possibly Vic, F., 2003. Identification of yeast species from orange fruit and juice by RFLP and sequence analysis of the 5.8S rRNA gene and the two internal transcribed through competition (Petersson and Schnürer, 1995, 1998; Peters- spacers. FEMS Yeast Research 3, 1–3. son et al., 1999; Druvefors et al., 2002). Friel et al. (2007) reported Hopkins, D.L., 1989. Bacterial fruit blotch of watermelon: a new disease in the that production of b-1,3-glucanase by P. anomala strain K played an eastern USA. Cucurtaceae 89, 74–75. Hopkins, D.L., Thompson, C.M., Hilgren, J., Lovic, B., 2003. Wet seed treatment with important role in suppression of Botrytis cinerea, the causal agent of peroxyacetic acid for the control of bacterial fruit blotch and other seedborne grey mould diseases. Meanwhile, production of killer toxins by P. diseases of watermelon. Plant Disease 87, 1495–1499. anomala for suppression of the human pathogen Candida albicans Jacobs, J.L., Damicone, J.P., McCraw, B.D., 1992. First report of bacterial fruit blotch of watermelon in Oklahoma. Plant Disease 76, 1185. (Mathews et al., 1998) and spoilage yeasts Dekkera and Brettano- King, E.O., Ward, M.K., Raney, D.E., 1956. Two simple media for the demonstration myces (Comitini et al., 2004) was also reported. The present study of pyocyanin and fluorescin. Journal of Laboratory and Clinical Medicine 44, showed that P. anomala strain 0732-1 could produce anti-bacterial 301–307. substances (ABS), which are effective for suppression of Aac Kurtzman, C.P., Fell, J.W., 1998. The Yeasts: Taxonomic Study (fourth ed.). Elsevier Science, Amsterdam. growth on KBA medium. Production of ABS by the yeast strain Langston Jr, D.B., Walcott, R.R., Gitaitis, R.D., Sanders, F.H., 1999. First report of a 0732-1 might be responsible for control of seedling blight of hami fruit rot of pumpkin caused by Acidovorax avenae subsp. citrulli in Georgia. Plant melon caused by Aac, as treatment of hami melon seeds with cell- Disease 83, 100. Lassois, L., de Bellaire, L., Jijakli, M.H., 2008. Biological control of crown rot of free cultural filtrates of strain 0732-1 significantly reduced inci- bananas with Pichia anomala strain K and Candida oleophila strain O. Biological dence and severity of this disease. Control 45, 410–418. Production of acidic compounds by P. anomala strain 0732-1 was Latin, R., Rane, K., 1990. Bacterial fruit blotch of watermelon in Indiana. Plant Disease 74, 331. evidenced by lowering of the pH value in cultures observed in this Latin, R., Hopkins, D.L., 1995. Bacterial fruit blotch of watermelon: the hypothetical study. Pusey et al. (2008) reported that acidic ambient pH signifi- exam question becomes reality. Plant Disease 79, 761–765. 中国科技论文在线 http://www.paper.edu.cn

X. Wang et al. / Biological Control 50 (2009) 164–171 171

Lessl, J.T., Fessehaie, A., Walcott, R.R., 2007. Colonization of female watermelon Silveira, E.B., Mariano, R.L.R., Michereff, S.J., 2003.Variabilidade de isolados de blossoms by Acidovorax avenae ssp. citrulli and the relationship between Acidovorax avenae subsp. citrulli provenientes de melão produzido no estado do blossom inoculum dosage and seed infestation. Journal of Phytopathology 155, Rio Grande do Norte. Summa Phytopathologica 29, 285–261. 114–121. Wall, G.C., Santos, V.M., Cruz, F.J., Nelson, D.A., Cabrea, I., 1990. Outbreak of Liang, Q.F., Chi, Z.M., 2002. Study on screening of one strain of yeast with biological watermelon fruit blotch in the Mariana Islands. Plant Disease 74, 80. control effect and its physiological and biochemical characters. Journal of Walcott, R.R., Gitaitis, R.D., Castro, A.C., 2003. Role of blossoms in watermelon Shangdong Institute of Light Industry 16, 66–70. seed infestation by Acidovorax avenae subsp. citrulli. Phytopathology 93, 528– Lu, Y., Luo, M., 2004. Isolation of endophytic bacteria from hami melon and 534. screening of antagonistic bacteria. Journal of Shihezi University (Natural Walcott, R.R., Fessehaie, A., Castro, A.C., 2004. Differences in pathogenicity between Science Edition) 22 (Suppl.), 104–109. two genetically distinct groups of Acidovorax avenae subsp. citrulli on cucurbit Martin, H.L., Horlock, C.M., 2002. First report of Acidovorax avenae subsp. citrulli as a hosts. Journal of Phytopathology 152, 277–285. pathogen of gramma in Australia. Plant Disease 86, 1406. White, T.J., Bruns, T., Lee, S., Taylor, J., 1990. PCR protocols: a guide to methods and Mathews, H.L., Conti, S., Witek-Janusek, L., Polonelli, L., 1998. Effect of Pichia applications. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J. (Eds.), anomala killer toxin on Candida albicans. Medical Mycology 36, 199–204. Amplification and Direct Sequencing of Fungal Ribosomal RNA genes for Petersson, S., Schnürer, J., 1995. Biocontrol of mould in high-moisture wheat stored Phylogenetics. Academic Press, San Diego, pp. 315–322. under airtight conditions by Pichia anomala, Pichia guilliermondii and Willems, A., Goar, M., Thielemans, S., Gills, M., Kersters, K., Deley, J., 1992. Transfer Saccharomyces cerevisiae. Applied and Environmental Microbiology 61, 1027– of several phytopathogenic Pseudomonas species to Acidovorax as Acidovorax 1032. avenae subsp. citrulli, Acidovorax avenae subsp. cattleyae and Acidovorax konjaci. Petersson, S., Schnürer, J., 1998. Pichia anomala as a biocontrol agent of Penicillium International Journal of Systematic Bacteriology 42, 107–119. roqueforti in high-moisture wheat, rye, barley, and oats stored under airtight Wilson, C.L., Wisniewski, M.E., 1989. Biological control of postharvest diseases of conditions. Canadian Journal of Microbiology 44, 471–476. fruits and vegetables: an emerging technology. Annual Review of Petersson, S., Jonsson, N., Schnürer, J., 1999. Pichia anomala as a biocontrol agent Phytopathology 27, 428–441. during storage of high-moisture feed grain under airtight conditions. Yaeram, C., Thummabenjapone, P., Pachinburavan, A., 2006. Suitable medium for Postharvest Biology and Technology 15, 175–184. increasing biomass of antagonistic Streptomyces spp. against bacterial fruit Pusey, P.L., Stockwell, V.O., Rudell, D.R., 2008. Antibiosis and acidification by blotch disease of watermelon. Khon Kaen Agriculture 34, 12–19. Pantoea agglomerans strain E328 may contribute to suppression of Erwinia Zhang, R.Y., Tan, Z.Q., Wen, Y.T., Zhang, H.M., Jiang, R.M., 1998. Description and amylovora. Phytopathology 98, 1136–1143. identification of the causal organism of bacterial fruit blotch of watermelon. Santos, E.R., Gouveia, E.R., Mariano, R.L.R., Souto-Maior, A.M., 2006. Biocontrol of Chinese Journal of Tropical Crops 19, 70–76 (in Chinese with an English bacterial fruit blotch of melon by bioactive compounds produced by Bacillus abstract). spp.. Summa Phytopathologica 32, 376–378. Zhao, T.C., Sun, F.Z., Wang, J.R., Xie, Y.J., 2003. Control of bacterial fruit blotch of Somodi, G.C., Jones, J.B., Hopkins, D.L., Stall, R.E., Kucharek, T.A., Hodge, N.C., hami melon by seed treatment with bactericides. Plant Protection 29, 50–52 (in Watterson, J.C., 1991. Occurrence of a bacterial watermelon fruit blotch in Chinese with an English abstract). Florida. Plant Disease 75, 1053–1056. Zhao, T., Sun, F., Wang, B., Hui, W., 2001. Pathogen identification of hami melon Shirakawa, T., Kikuchi, S., Kato, T., Abiko, K., Kaiwa, A., 2000. Occurrence of bacterial fruit blotch. Acta Phytopathologica Sinica 31, 357–364 (in Chinese watermelon bacterial fruit blotch in Japan. Annals of the Phytopathological with English abstract). Society of Japan 66, 223–231.