761

Two Selection Strategies of Epiphytic Native Yeasts with Potential Biocontrol Capacity against Postharvest Pear Pathogens in Patagonia M.C. Lutz, A. Robiglio, M.C. Sosa, C.A. Lopes and M.P. Sangorrín Laboratorio de Microbiología y Biotecnología, Facultad de Ingeniería Laboratorio de Fitopatología, Facultad de Ciencias Agrarias and Instituto Multidisciplinario de Investigación y Desarrollo de la Patagonia Norte Universidad Nacional del Comahue Argentina Keywords: postharvest diseases, antagonistic microorganisms, Penicillum expansum,

Botrytis cinerea Abstract

To reduce the use of fungicides, biological control with yeasts has been proposed worldwide in recent years. In order to find antagonistic yeasts adapted to pear storage conditions (-1/0°C for 7 months), two isolation strategies were explored. In 2007 (strategy A), the yeasts were isolated from the surface of healthy fruits. Aliquots of the obtained suspensions were seeded on GPY agar and incubated at 26°C. In 2008 (strategy B), washes from healthy wounds after 150 days at 0°C were used to inoculate fresh pear wounds with Penicillium expansum. Yeasts were isolated from healthy wounds after 50 days of incubation in cold. From both A and B isolation strategies, one isolate from each yeast species was tested for antagonistic activity against P. expansum and Botrytis cinerea by in vivo (pear wounds at 0°C) and in vitro (dual cultures at 20°C) assays. By means of strategy A, six yeast species were identified. Among them, the best antagonists were A. pullulans and R. mucilaginosa, which reduced only P. expansum disease incidence (33%). From strategy B, five of six species obtained, Cryptococcus weringae, C. victoriae, Cystofilobasidium infirmominiatum, Rhodotorula laryngis and A. pullulans, showed the highest antagonistic activity against P. expansum; they completely controlled disease incidence at 100 days. Only Cryptococcus weringae and C. victoriae reduced incidence of B. cinerea (80%) at 100 days. Differences between in vivo and in vitro biocontrol assays were observed. In in vitro assays, all yeasts produced a greater growth inhibition of P. expansum than of B. cinerea. Strategy B was the most effective strategy for the selection of antagonistic yeasts for postharvest disease control. INTRODUCTION

Argentina is the largest pear (Pyrus communis) producing country and the major exporter in the Southern Hemisphere. The main pear-growing area is situated in the provinces of Rio Negro and Neuquén, Patagonia. Postharvest decays are one of the main factors that determine losses and compromise the quality of pear fruit in cold storage. Gray mold caused by Botrytis cinerea and blue mold, caused by Penicillium expansum are the most important storage diseases of pears in Argentina (Dobra et al., 2008).

Public concern about human health and environmental safety has limited the use of fungicides in postharvest. Due to this restriction, the interest in finding alternative approaches to control postharvest diseases has greatly increased. Biological control by means of microorganisms has been one of the most extensively studied alternatives and appears to be a viable postharvest technology (Droby et al., 2009; Janisiewicz and Korsten, 2002; Wisniewski et al., 2007). In our region, a postharvest biocontrol system for reducing decays of pears caused by P. expansum and B. cinerea is promising. These pathogens are mostly associated with wounds, mainly produced during harvest, transportation and packaging of fruit.

Several microorganisms present naturally on the surface of fruits and vegetables

Proc. 11th International Pear Symposium Eds.: E. Sánchez et al. Acta Hort. 909, ISHS 2011

762

have been shown to have an antagonistic effect against postharvest fungal diseases (Filonow et al., 1996). Among the microorganisms, yeasts may be good biocontrol agents (BCA) for protection of fruits due to i) their adaptation to the immediate environment and to the nutritional conditions prevailing at the wound site, ii) their capacity to grow at low temperatures and iii) their ability to colonize wounds (Sharma et al., 2009).

To obtain potential antagonistic microorganisms it is necessary to optimize isolation and selection regarding both the environmental conditions and the fruit substrate on which they act. Pear fruit are stored in cold chambers; thus, for controlling their postharvest diseases to a satisfactory level, a microbial antagonist should have the ability to survive under cold storage conditions. On the other hand, isolation of antagonists can be improved by using fruit from unmanaged orchards where natural antagonist populations have not been disturbed by the use of chemical products (Falconi and Mendgen, 1994; Janisiewicz and Korsten, 2002). Considering these factors, a variety of strategies to select microbial antagonistic yeasts have been proposed in many countries. In this sense, Wilson et al. (1993) proposed the isolation of potential antagonistic yeasts from wounds without decay symptoms, with excellent results.

The main objectives of this work were: (1) to compare two strategies of isolation and selection of pear epiphytic native yeasts naturally adapted at fruit storage conditions and (2) to evaluate the biocontrol efficacy of these yeasts against the main postharvest pear fruit pathogens, P. expansum and B. cinerea. MATERIAL AND METHODS Isolation, Selection and Identification of Yeasts

In 2007 and 2008, two different strategies of isolation and selection of epiphytic yeasts were explored. In each year, pear fruit coming from a packinghouse in northern Patagonia in organic transition were used after seven months of storage at -1/0°C. 2007

In the first strategy (A), the epiphytic yeasts were isolated from the surface of 20 healthy pear fruit. Two blocks (2×2 mm) were removed from each fruit by using a sharp knife and immediately immersed in 9 ml of sterile distilled water with 0.05% w/v of Tween. Tissue blocks were sonicated, centrifuged and resuspended in 100 µl of distilled water. Aliquots of the suspensions were seeded on GPY agar (w/v: 0.5% yeast extract, 0.5% peptone, 4% glucose, 1.5% agar) supplemented with ampicillin (0.5 µg/ml) and incubated at 26°C. In order to select the yeasts with the highest growth capability at storage temperature (-1/0°C), aqueous suspensions of the yeasts (106 cells/ml) were prepared for each isolate and 5 µl were seeded on GPY-agar plate surface. Plates were incubated at 0/-1°C for 7 days and growth was evaluated daily. Assays were performed in triplicate. 2008

In the second strategy (B), a more selective method than strategy A was studied. The native yeasts were isolated from fruit wound washes according to Wilson et al. (1993) with modifications. This method included several steps: (i) pear fruit were wounded and stored 150 days at 0°C. (ii) Pear fruit blocks (5×5 mm), including healthy wounds, were washed with sterile water. (iii) Each wash was used to inoculate fresh pear wounds against 10 µl of Penicillium expansum (104 cells/ml). (iv) Washes that controlled P. expansum incidence (I) by 50% and provided decay reduction (RD) by 40% after 50 days at 4°C were selected for yeast isolation. Suspensions from these wounds, which did not show decay were seeded on pear juice agar plates and incubated at 0°C for 48 h.

Yeasts were identified by ITS1-5.8S rDNA-ITS2 PCR-RFLP (Esteve-Zarzoso et al., 1999). Patterns obtained for each isolate after digestion with the restriction enzymes Cfo I, Hae III and Hinf l were compared with those of reference strains available in the yeast identification database (www.yeast-id.com). The identifications were confirmed by

763

sequencing of D1/D2 domains of the 26S rRNA gene. The yeast isolates were stored at -20 and -80°C using glycerol 20% v/v.

In Vivo Biocontrol Assays

Biocontrol assays were carried out on ‘Packham’s Triumph’ pears. One yeast isolate from each species was evaluated against postharvest pathogens in wounds of pear fruit at -1/0°C. Regional isolates of B. cinerea and P. expansum obtained from infected pear fruit were used. Pathogen cultures were maintained on potato-dextrose agar medium (PDA) at 8°C until use. Conidial suspensions were prepared from 7-old-day fungus cultures on PDA and adjusted to the minimal conidial concentration (MCC) determined for each pathogen (1×103 conidia/ml for P. expansum and 1×104 conidia/ml for B. cinerea). Yeast cells were obtained from 48 h cultures grown on GPY agar. Aqueous suspensions of the yeasts were prepared for each isolate and adjusted to 106 cells/ml.

In 2008, six yeasts isolated by means of strategy A and selected by their growth in cold were evaluated in the bioassay. Pear fruit selected according to size, uniformity and absence of injuries were surface-disinfected with 70% ethanol. One wound (3 mm deep and 3 mm wide) was made at the equator of each fruit with a disinfected tool and 20 μl of the corresponding yeast suspension (106 cells/ml) was inoculated into each wound. After 1 h, the treated wounds were inoculated with 20 μl of either P. expansum or B. cinerea conidia suspension at their respective MCC. Wounds only inoculated with pathogens were used as controls. After inoculation, pears were placed in polyethylene bags and stored in boxes under standard conditions (0/-1°C and 95% RH) for 2 months.

In 2009, six yeasts isolated by strategy B were evaluated in a bioassay according to previously described conditions and methods. Fruit were incubated during 6 months and wounds were examined for decay and lesion diameters (mm) every 15 days. Incidence (I) calculated as number of decayed wounds over total number of wounds, and percentage of decay reduction (DR) calculated as [(mean lesion diameter in control - mean lesion diameter in treatment)×100/mean lesion diameter in control] were used as yeast selection criteria. Each treatment was replicated three times with three fruit per treatment.

In Vitro Antagonism

To evaluate in vitro antagonistic capacity of each yeast isolate, dual inoculation of the pathogen and antagonist was carried out. Yeast and pathogen suspensions were prepared as described above. Briefly, 10 μl of the suspension of P. expansum or B. cinerea (104 conidia/ml) was seeded on PDA plates at 50 mm from the test yeast line. Control plates were simultaneously seeded with the pathogen and antagonists in separate plates. Plates were incubated at 20°C for 7 days. Inhibition of pathogen growth in dual cultures was assessed by the percentage of inhibition of radial growth after incubation (Royse and Ries, 1978). Three replications per treatment were set up for each pathogen-antagonist combination.

Statistical Analysis

Data were subjected to analysis of variance (ANOVA) and means were separated according to the Fisher least significant difference test (P=0.05) using STATISTICA data analysis software system, version 6 (Stat-Soft, 2001, France). The treatments were arranged in a completely randomized design.

RESULTS Isolation, Selection and Identification of Yeasts

A total of 24 yeast isolates were obtained on GPY during 2007 from the surface of healthy pear fruit (Strategy A, Table 1). In in vitro assays on GPY not all yeast isolates grew at -1/0°C (data not shown). In 2008, 23 washes from healthy fruit wounds were obtained after 150 days at 0°C (Strategy B). 56% of washes reduced the P. expansum

764

incidence by 50% and decay diameter by 40%. A total of 18 yeast isolates resulted from the seeding of these washes on pear juice agar (Table 1). Only two species were isolated by both strategies, A. pullulans and C. albidus, while four species were only isolated from strategy A and four from strategy B (Table 1). A large number of A. pullulans isolates were obtained. Four different species belonging to the genera Cryptoccocus were identified.

Biocontrol Assays 2008

A. pullulans and R. mucilaginosa were the best antagonists among the six yeasts isolated from strategy A and selected by growth in cold conditions. The results presented in Table 2 indicate that the application of A. pullulans and R. mucilaginosa followed by storage at -1 /0°C for 60 days resulted in low averages of P. expansum decay incidence in pears (33%, compared with 95% in the control fruit). With regard to the lesion diameter, 88% decay reduction was observed in pear wounds treated with either A. pullulans or R. mucilaginosa. Although a decrease in B. cinerea decay incidence was not obtained with any of the yeast isolates evaluated in this assay, the yeast R. mucilaginosa controlled the lesion diameter by 30% in fruit wounds after 30 days of incubation. 2009

Five yeast isolates belonging to Cryptococcus weringae, C. victoriae, Cystofilobasidium infirmominiatum, Rhodotorula laryngis and A. pullulans, inhibited completely the P. expansum decay after 100 days of conservation at -1/0°C (Table 2). Among them, Cr. victoriae and A. pullulans stood out by maintaining this control level after 200 days (data not shown). Cryptococcus victoriae and Cryptococcus weringae were the best yeasts against B. cinerea, reducing disease incidence by 80% after 100 days. The remaining yeast species did not reduce incidence, whereas they showed high percentages of decay reduction (65 and 80%) (Table 2).

In Vitro Antagonism

All yeast isolates selected from strategy A showed good antagonistic activity on agar plates at 20°C against both P. expansum and B. cinerea. Among the six species evaluated, C. difluens and C. albidus inhibited growth of P. expansum by more than 50%. A. pullulans and R. mucilaginosa (the best antagonists in biocontrol assay against P. expansum) demonstrated 44-47% and 52-46% growth inhibition of P. expansum and B. cinerea, respectively (Table 2). On the other hand, the six species isolated by strategy B evidenced low in vitro antagonism against P. expansum (Table 2). Only one isolate of A. pullulans inhibited the radial growth on GPY agar of both pathogens in dual cultures (25 and 36%, respectively) (Table 2).

DISCUSSION AND CONCLUSIONS

Pear fruit from unmanaged orchards (without use of chemical compounds) and conserved for a long time at -1/0°C provided a promising source for isolation of antagonists to be used against the main postharvest fruit pathogens, P. expansum and B. cinerea. Six different yeast species were recovered using two different methods of isolation and selection (strategy A and B). Two yeast species were common to both strategies (Table 1). The most frequently isolated genera were Aureobasidium, Cryptococcus and Rhodotorula. These genera have been previously reported as effective biocontrol agents against a number of postharvest diseases of various fruits (Calvente et al., 1999; Roberts, 1990; Vero et al., 2009).

Strategy B was highly effective for the selection of antagonistic microorganisms in pears, and it was in agreement with data reported by Wilson et al. (1993) for apple, tomato, and orange fruits. In other words, the best antagonistic yeast strains against both pathogens, particularly for P. expansum, were obtained in our work by the selective

765

strategy B (Table 2). This method, based in the recovery of yeasts from healthy wound washes, considers both the substrate (pear) and the environmental conditions (-1/0°C) in which yeasts have to act.

Although a high number of A. pullulans isolates was obtained from both strategies, the isolate that showed the highest effectivity as a biocontrol agent was isolated from a fruit wound wash. This isolate completely controlled P. expansum decay after 200 days of fruit conservation. The antagonistic capability of Aureobasidium pullulans has been already reported for the control of B. cinerea and P. expansum decays in apple fruit (Ippolito et al., 2000; Vero et al., 2009). In our work, both A. pullulans and C. victoriae isolated from wounds completely controlled P. expansum decay after 200 days of incubation. With respect to antagonistic activity against B. cinerea, only two strains isolated from washes (C. victoriae and C. weringae) diminished decay incidence (to 80%). Different species belonging to the genus Cryptococcus have demonstrated biocontrol activity at cold temperatures. C. laurentii isolated from apple leaves and fruit was associated with the reduction of gray mold in apples stored at low temperature (Roberts, 1990). Chand-Goyal and Spotts (1997) also observed the capacity of Cryptococcus infirmo-miniatus and C. laurentii for controlling decay of apples and pears caused by B. cinerea and P. expansum.

In relation to the in vitro assays, we did not observe a correlation with biocontrol assays in pear fruit. As has been previously reported, dual cultures are only useful to describe action mechanisms of biocontrol microorganisms, but they are not adequate to select effective biocontrol agents. Elucidation of the mechanisms by which antagonists inhibit postharvest pathogens is important for both the development of a more reliable procedure for effective antagonist application and the elucidation of rationale for the selection of more effective antagonists (Droby et al., 2009). The mechanisms of action involved in the biocontrol of the antagonistic yeasts selected in this work must be explored.

In conclusion, the best selection strategy for good antagonist yeasts is based on in vivo assays in fruit wounds and under real environment conditions. Different yeast species can be recovered depending on the selection strategy used. A. pullulans and C. victoriae yeast isolates appear as promising tools in the biocontrol of the main pathogens of pear fruit stored in cold chambers.

Literature Cited Calvente, V., Orellano, M.E., Benuzzi, D. and Sanz de Tosetti, M.I. 1999. Antagonistic

action of siderophores from Rhodotorula glutinis upon the postharvest pathogen Penicillium expansum. Int. Biodeter. Biodegrad. 43:167-172.

Chand-Goyal, T. and Spotts, R.A. 1997. Control of postharvest pear diseases using natural saprophytic yeast colonists and their combination with a low dosage of thiabendazol. Postharvest Biol. Technol. 7:51-64.

Dobra, A.C., Sosa, M.C. and Dussi, M.C. 2008. Low incidence of fungal and bacterial diseases in the pear production of north Patagonia. Arg. Acta Hort. 800:907-912.

Droby, S., Wisniewski, M., Macarisin, D. and Wilson, C.M. 2009. Twenty years of postharvest biocontrol research: is it time for a new paradigm? Postharvest Biol. Technol. 52:137-145.

Esteve-Zarzoso, B., Belloch, C., Uruburu, F. and Querol, A. 1999. Identification of yeasts by RFLP analysis of the 5.8S rRNA gene and two ribosomal internal transcribed spacers. Int. J. Syst. Bacteriol. 49:329-337.

Falconi, C. and Mendgen, K. 1994. Epiphytic fungi on apple leaves and their value for control of the postharvest pathogens Botrytis cinerea, Monilinia fructigena and Penicillium expansum. Zeitschrift für Pflanzenkrankheiten und Pflanz. 101:38-47.

Filonow, A.B., Vishniac, H.S., Anderson, J.A. and Janisiewicz, W.J. 1996. Biological control of Botrytis cinerea in apple by yeasts from various habitats and their putative mechanisms of antagonism. Biol. Cont. 7:212-220.

Ippolito, A., El-Ghaouth, A., Wilson, C.L. and Wisniewski, M. 2000. Control of

766

postharvest decay of apple fruit by Aureobasidium pullulans and induction of defense responses. Postharvest Biol. Technol. 19:265-272.

Janisiewicz, W.J. and Korsten, L. 2002. Biological control of postharvest diseases of fruits. Annu. Rev. Phytopathol. 40:411-441.

Sharma, R.R., Singh, D. and Singh, R. 2009. Biological control of postharvest diseases of fruits and vegetables by microbial antagonists: a review. Biol. Cont. 50:205-221.

Roberts, R.G. 1990. Postharvest biological control of gray mold of apple by Cryptococcus laurentii. Phytopathol. 80:526-530.

Royse, D.J. and Ries, S.M. 1978. The influence of fungi isolated from peach twigs on the pathogenicity of Cytospora cincta. Phytopathol. 68:603-607.

Vero, S., Garmendia, G., Gonzalez, M.B., Garat, M.F. and Wisniewski, M. 2009. Aureobasidium pullulans as a biocontrol agent of postharvest pathogens of apples in Uruguay. Biocontrol Sci. Technol. 19:1033-1049.

Wilson, C., Wisniewski, M., Droby, S. and Chalutz, E. 1993. A selection strategy for microbiall antagonists to control postharvest diseases of fruits and vegetables. Scientia Hort. 53:183-189.

Wisniewski, M.E., Wilson, C., Droby, S., Chalutz, E., El-Ghaouth, A. and Stevens, C. 2007. Postharvest Biocontrol: New Concepts and Applications. p.264-273 In: C. Vincent, M.S. Goettel and G. Lazarovits (eds.), Biological Control: A Global Perspective. CABI, Oxfordshire, UK.

767

Tables Table 1. Yeast species identified by PCR-RFLP and D1-D2 sequencing, and number of isolates obtained from A and B strategies.

Yeast species AMP1Restriction patterns2

(pb) Number of isolates3 (%)

Cfo I Hae III Hinf I A B Aureobasidium pullulans 600 170+150+90 450+150 280+180+130 11 (46) 4 (22) Cryptococcus albidus 630 380+290 500+70+60 350+160+120 1(4) 2 (11) Cryptococcus difluens 530 290+250 400+120 250+250 5 (21) - Cryptococcus victoriae 510 280+260 380+130 270+270 - 6 (33,5) Cryptoccocus weringae 620 350+280 480+90 260+240+150 - 3 (17) Cystofilobasidium infirmominiatum 620 280+90 500+70 270+180+160 - 1 (5,5) Pichia membranifaciens 480 170+100+80+50 330+90 280+200 1 (4) - Pichia philogaea 600 290+230+100 400+130+70 350+220+70 1 (4) - Rhodothorula laryngis 600 500+90 580 330+220+50 - 2 (11) Rhodotorula mucilaginosa 600 310+230+100 400+220 350+220 5 (21) - Total/chamber 24 (100) 18 (100) 1 Values refer to number of base pairs of the amplification product. 2 Values refer to number of base pairs of restriction fragments. 3 Yeasts obtained from both A and B strategies of isolation and selection. (A: Pear surfaces, B: Pear wound washings).

767

768

Table 2. Effect of individual yeasts on incidence, lesion diameter and % of inhibition of B. cinerea and P. expansum growth in dual

cultures at -1/0°C.

Antagonist source Yeast isolate

P. expansum B. cinerea Incidence

(%) Lesion

diameter1,2% inhibition dual culture2

Incidence(%)

Lesion diameter1,2

% inhibition dual culture2

60 days 14 days 30 days 7 days

Pear surfaces (A)

A. pullulans 33 3±5 a 44 bc 100 21±4 ab 52 ef C. albidus 100 22±2 de 51 e 100 28±1 bcd 27 ab C. difluens 100 9±3 ab 53 e 100 27±2 bcd 23 a

P. membranifaciens 100 19±1 d 40 bc 100 22±0,5 ab 36 cd P. philogaea 100 15±4 cd 33 a 100 24±5 bc 35 bcd

R. mucilaginosa 33 3±5 a 47 cd 100 19±4 a 46 de Control 100 24±6 de 100 27±2 bcd

100 days 14 days 100 days 7 days

Pear wounds (B°

A. pullulans 0 0 a 36 c 100 71±3de 25 b C. albidus 40 20±2 b 17,5 abc 100 65±5d 1,1 a

C. victoriae 0 0 a 15 ab 80 42±6b 0 a C. weringae 0 0 a 26,4 cd 80 32±1a 0 a

C. infirmominiatum 0 0 a 22,3 bc 100 55±5c 0 a R. laryngis 0 0 a 23,2 bc 100 71±5de 2,3 a

Control 100 28±1 c 100 88±5 f 1 Means of lesion diameters ± standard deviation (mm) on pear wounds inoculated with yeasts and pathogen or pathogen alone. 2 Values within a column followed by the same letter are not significantly different according to Fisher’s (P0,05).

768