Phenotypic Traits of Wound Competence of Postharvest Biocontrol Yeasts and Potential of These Microorganisms for Prevention/ Detoxification of Mycotoxin Accumulation

Knowledge of the mechanisms of action of biocontrol agents (BCAs) is crucial to identify underlying pivotal traits, whose enhancement could boost antagonist activity and encourage the use of BCAs as alternative/integration of fungicides. We had previously shown that resistance of postharvest BCAs to oxidative stress is a possible key mechanism for their wound competence to successfully outcompete wound pathogens in stored apples. In this work we describe recent studies aimed at elucidating the phenotypic traits underlying wound competence of postharvest biocontrol yeasts. Further, we show that some BCAs display a potential activity in addressing other major issues of food safety: contamination of apple-based food products with Patulin and of wine with Ochratoxin A, mycotoxins produced by the fungal pathogens of apples and grape Penicillium expansum and Aspergillus carbonarius, respectively. INTRODUCTION Antagonist yeasts act as biological control agents (BCAs) of postharvest pathogens of horticultural and fruit products and are becoming promising altemativeslintegrations to the use of fungicides. Higher efficiency and reliability of BCAs' performances is needed for encouraging their utilisation and commercialisation. In this regard, it is crucial to understand their modes of action (Janisiewicz and Korsten, 2002). We have previously shown that wound competence of postharvest BCAs acting on apples relies on their resistance to reactive oxygen species (ROS) generated by wounded apple tissue, and that such resistance appears as a possible key mechanism to successfully outcompete wound pathogens such as Botrytis cinerea and Penicillium expansum (Castoria et a1., 2003). In order to elucidate the phenotypic traits underlying this mechanism(s), we compared 2 biocontrol yeasts. Cryptococcus laurentii strain LS28 and Rhodotorula glutinis strain LS 11, with higher and lower wound competence and antagonist activity, respectively. for a) their intracellular catalase (CAT) and superoxide dismutase (SOD) activities, and b) for the antioxidant potential of their cul ture filtrates. Besides fungicide and chemical contamination, mycotoxins represent another major issue of food safety. Patulin is synthesised by P. expansum that causes accumulation of this mycotoxin in infected pome fruits and, as a consequence, in juices and pome fruit-based baby foods (Drusch and Ragab, 2003). This toxin is known to display diverse toxic effects: immunosuppression, embryo and maternal toxicity in mice, genotoxicity (IARC, 1986; Bourdiol and Escoula, 1990). Ochratoxin A (OTA) is a mycotoxin produced by several Penicillium and Aspergillus spp., frequently detected as a contaminant of cereals. This toxin is carcinogenic to rodents and has teratogenic, immunotoxic and possible neurotoxic and genotoxic properties (Creppy, 1999). Recently, it has been reported as a frequent contaminant of red wines, especially in Southern Europe (Battilani el al., 2003). The fungal pathogen responsible for wine contamination is A. Proc. 5'1; Int. Postharvest Symp. Eds. F. Mencarelli and P. Tonutti Acta Hort. 682, ISHS 2005 2147 carbonarius section Sigl'i. that attacks berries of wine-grape both in the field and during storage (Battilani et a1.. 2003: Delage et a1.. 2003). The highest tolerable levels have been set both for Patulin and OTA (EC Regulation -472 2002 and 1425;2003). Here we report the ability of the BCA R. gill! inis LS 11 to decrease in vitro rccoverv of Patulin from its culture medium. and to lower toxin accumulation in stored apples. "as a part of its antagonistic activity. Further. we show the results of preliminary experiments focusing on the antagonist activity of another BC A. Aureobasidium putlulans strain LS30. against A. carbonarius on wine grapes. MATERIALS A:\D METHODS Antioxidant Enzyme Activities and Extracellular Antioxidant Potential 4 Overnight' cultures in LB (Lilly-Barnett) medium of the BCAs LS 11 and LS28 DC. were centrifuged at 5000 rpm for 15 min at Supernatants were used for determination of extracellular antioxidant potential (equivalents of Ascorbic acid) without previous dialysis. Cell pellets of each strain were resuspended in lysis buffer (50 mM K­ phosphate buffer pH 7.0 containing 1 mM phenylmethylsulfonyl fluoride, 1 ug mLI pcpstatin and 400 mg acid-washed glass beads). Cells were disrupted by sonication on ice and cell debris was removed by centrifugation at 4 ''C, Aliquots of intracellular extracts were filtered through 0.22 urn filters and then dialysed, Residual sample volumes were measured and freeze-dried. Samples were re-suspended in 500 Jll of sterilised bi-distilled water and their soluble protein content measured by the method of Bradford (1976). This extract was used for determination of intracellular CAT and SOD activities. Catalase activity was measured by the method of Beers and Sizer (1952). The rate of disappearance of Ho02 was measured spectrophotometrically at 240 nm after adding enzyme extract to SO mM K-Phosphate buffer pH 7.0 amended with 0.05% of H20 2. Superoxide disrnuiasc activity was determined by the method of Sutherland and Leannonth (1997), with slight modifications, by measuring the inhibition of xrr (2.3-bis (2-methoxy-4-nitro-5-sulphopheny1)-5-[(phenylamin0 )carbony1~ -2H -tetrazoli urn hydro­ xide, sodium salt) reduction at 470 run. Catalase and SOD activity assays were performed twice with three different enzyme extracts. Antioxidant potentials were measured in aliquots of cell free culture filtrates by using N,N-dimcthyl-p-phenylenediamine (DMPD) and expressed as umole equivalents of Ascorbic acid per g of freeze-dried filtrate (Fogliano et al., 1999). Mean values [rom data of two experiments were separated by the least significant difference (LSD) test (P = 0.05). Biological Control Trials Antagonistic activity of lS II against P. cxpansum was assessed at 23°C on apples Annurca (data not shown), as previously described (Castoria et al., 2001). Antagonistic activity of LS30 against A. carbonarius was assessed at the same temperature on wine grape berries cv, MontepuJciano as described by Zahavi et al. (2000) with some modifications, and compared to treatments of berries with 0.8 mg mL· 1 Switch! (Novartis). The experiments were performed twice. BCAs and Mvcotoxins BiocOtitrol strains LS 11 and LS30 were assessed for their in vitro capability of reducing Parulin and OTA levels, respectively. Recoveries of Patulin were determined in LS I 1 culture filtrates following ethyl acetate extraction, and subsequent HPLC analyses (AOAC, 2000). Patulin was measured after 10 days of in vitro incubation of LS 11 with 0, 250 and 500 ppm of toxin at 23°C. Ochratoxin A was measured by HPLC analysis (Visconti et al., 1999) of unextracted medium after 6 davs of incubation of LS30 with I flg mL" OIA at 23'C (data not shown). . Patulin accumulation in vivo, i.e. in fruits from biological control trials (see above), was determined after 4 and 6 days from inoculation.