International Journal of Food Microbiology 131 (2009) 178–182 Contents lists available at ScienceDirect International Journal of Food Microbiology j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j f o o d m i c r o In vitro antimycotic activity of a Williopsis saturnus killer protein against food spoilage yeasts Marta Goretti a, Benedetta Turchetti a, Morena Buratta b, Eva Branda a, Lanfranco Corazzi b, Ann Vaughan-Martini a, Pietro Buzzini a,⁎ a b Dipartimento di Biologia Applicata — Sez, Microbiologia and Industrial Yeasts Collection DBVPG, Università di Perugia, Perugia 06121, Italy Dipartimento di Medicina Interna, Sezione di Biochimica, Università di Perugia, Perugia 06122, Italy a r t i c l e i n f o Article history: Received 27 October 2008 Received in revised form 24 January 2009 Accepted 18 February 2009 Keywords: Yeast killer protein Williopsis saturnus Food spoilage yeasts Preservative agents MIC a b s t r a c t The in vitro antimycotic activity of a purified killer protein (KT4561) secreted by a strain of Williopsis saturnus was tested against 310 yeast strains belonging to 21 food spoilage species of 14 genera (Candida, Debaryomyces, Dekkera, Hanseniaspora, Issatchenkia, Kazachstania, Kluyveromyces, Pichia, Rhodotorula, Saccharomyces, Schizosaccharomyces, Torulaspora, Yarrowia and Zygosaccharomyces). Minimum inhibitory concentration (MIC) determinations showed that over 65% of the target strains were susceptible to concentrations ≤ 32 µg/ ml of KT4561. Three conventional food-grade antimicrobial agents were used as controls: 41, 33 and 40% of the target strains were sensitive to ≤ 512 mg/ml of ethyl 3-hydroxybenzoate (E214), potassium sorbate (E202) or potassium metabisulphite (E224), respectively. The susceptibility of food spoilage yeasts towards KT4561, E214, E202 and E224 was species- and strain-dependent. In most cases KT4561 exhibited MIC values several orders of magnitude lower (100 to 100,000 times) than those observed for E214, E202 and E224. With only a few exceptions, the activity of KT4561 was pH-, ethanol-, glucose- and NaCl-independent. The present study demonstrates the potential of this yeast killer protein as a novel and natural control agent against food spoilage yeasts. © 2009 Elsevier B.V. All rights reserved. 1. Introduction Since time immemorial yeasts have been important key microorganisms in many food industries. Nevertheless, they can also cause spoilage and undesirable changes in a wide range of foods, especially those that are processed, preserved and refrigerated (Loureiro and Querol, 1999; Loureiro and Malfeito-Ferreira, 2003). The predominance of certain species is normally related to the chemical–physical properties of the food. For example, the most prevalent species isolated from dairy products belong to the genera Candida, Debaryomyces, Kluyveromyces, Pichia, Rhodotorula, Saccharomyces and Yarrowia (Fleet, 1992; Jakobsen and Narvhus, 1996; Pitt and Hocking, 1997; Viljoen, 2001). The typical spoilage yeast species in different foods have been recently reviewed by Stratford (2006): Zygossacharomyces bailii and Issatchenkia orientalis in foods characterized by low pH; Dekkera bruxellensis, Pichia anomala and Pichia membranifaciens in wines, Dekkera anomala in beer; Yarrowia lipolytica in cheeses; Debaryomyces hansenii and Zygosacharomyces spp. in foods ⁎ Corresponding author. Dipartimento di Biologia Applicata, Sezione di Microbiologia, Università di Perugia, Borgo XX Giugno, I 06121 Perugia, Italy. Tel.: +39 075 5856455; fax: +39 075 5856470. E-mail address: pbuzzini@unipg.it (P. Buzzini). 0168-1605/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2009.02.013 with high NaCl and sugar concentrations. Some other species (of the Saccharomyces cerevisiae complex, for example) were considered to be opportunistic. Food spoilage is a serious problem for the food industry as it can make products unacceptable to the consumer, and can result in economic losses and potentially serious health hazards. Many spoilage yeasts develop when good manufacturing procedures are not employed (e.g. poor factory hygiene, missing or insufficient preservatives, inadequate pasteurizing temperatures, and/or use of poor quality raw materials). (Fleet, 1992; Viljoen et al., 2003; Stratford, 2006). Food-grade antimicrobial compounds, in particular sorbic and benzoic acid derivatives, are routinely used for prolonging shelf-life and the preservation of food quality by inhibiting spoilage microorganisms (Battey et al., 2002; Papadimitriou et al., 2007). Nevertheless, some yeasts have been shown to be resistant to many chemical preservatives (Fleet, 1992; Thomas, 1993; Tudor and Board, 1993; Lambert and Stratford, 1999; Battey et al., 2002; Hazan et al., 2004). In this background, an alternative approach could involve the use of killer proteins produced by selected yeasts that can neutralize the activities of these undesired microorganisms in foods. The role of killer yeasts as producers of proteins (toxins) exhibiting antimycotic activity is well documented (Golubev, 2006), and their use for the 179 M. Goretti et al. / International Journal of Food Microbiology 131 (2009) 178–182 Table 1 In vitro susceptibilities of 310 food spoilage yeast strains belonging to the genera Candida, Debaryomyces, Dekkera, Hanseniaspora, Issatchenkia, Kazachstania, Kluyveromyces, Pichia, Rhodotorula, Saccharomyces, Schizosaccharomyces, Torulaspora, Yarrowia and Zygosaccharomyces to KT4561, ethyl 3-hydroxybenzoate (E214), potassium sorbate (E202) and potassium metabisulphite (E224). Table 1 (continued) Species (number of strains) Yarrowia lipolytica (14) Zygosaccharomyces bisporus (8) Antimycotic agents KT4561 Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Debaryomyces KT4561 hansenii (20) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Dekkera KT4561 anomala (4) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Dekkera KT4561 bruxellensis (15) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Hanseniaspora KT4561 uvarum (15) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Issatchenkia KT4561 orientalis (24) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Kazachstania KT4561 exigua (11) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Kluyveromyces KT4561 lactis (19) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Kluyveromyces KT4561 marxianus (23) Ethyl 3-hydroxybenzoate Potassium sorbate potassium metabisulphite Pichia anomala (13) KT4561 Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Pichia KT4561 membranifaciens (16) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Rhodotorula KT4561 mucilaginosa (9) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Saccharomyces KT4561 bayanus (10) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Saccharomyces KT4561 cerevisiae (26) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Saccharomyces KT4561 pastorianus (9) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Schizosaccharomyces KT4561 pombe (12) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Torulaspora KT4561 delbrueckii (18) Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite Candida parapsilosis (13) MIC (µg/ml) MIC50 MIC90 % of susceptible strains 1 N 32 N512 N512 N512 16 N512 N512 512 2 N512 128 256 32 N512 N512 N512 N 32 N512 N512 512 0.031 N512 512 N512 2 N512 N512 N512 N 32 N512 512 N512 0.125 256 N512 512 0.5 N512 256 N512 16 N512 N512 512 N 32 512 128 N512 32 N512 N512 512 0.062 256 N512 512 1 512 N512 128 N 32 512 256 256 N 32 N512 N512 N512 N 32 N 512 N 512 N 512 N 32 N 512 N 512 N 512 N 32 N 512 N 512 N 512 32 N 512 N 512 N 512 N 32 N 512 N 512 N 512 0.25 N 512 512 N 512 4 N 512 N 512 N 512 N 32 N 512 N 512 N 512 0.25 512 N 512 N 512 0.5 N 512 512 N 512 N 32 N 512 N 512 N 512 N 32 N 512 512 N 512 N 32 N 512 N 512 N 512 0.125 512 N 512 N 512 1 512 N 512 N 512 N 32 N 512 512 512 N 32 N 512 N 512 N 512 0 7.6 30.8 46.1 85 20 15 60 50 0 25 25 93.3 26.7 20 33.3 0 13.3 40 0 100 0 88 53.33 100 9.1 27.3 9.1 0 5.3 68.4 5.3 100 100 4.2 79.2 93.3 0 93.3 0 87.5 0 12.5 18.7 0 77.7 100 44.4 60 20 30 50 100 100 3.8 50 100 100 44.4 88.9 0 50 100 100 0 16.7 5.5 0 Species (number of strains) Zygosaccharomyces bailii (26) Zygosaccharomyces microellipsoides (5) Antimycotic agents KT4561 Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite KT4561 Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite KT4561 Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite KT4561 Ethyl 3-hydroxybenzoate Potassium sorbate Potassium metabisulphite MIC50 MIC (µg/ml) MIC90 % of susceptible strains 1 16 N512 N512 N512 32 N512 N512 128 0.25 256 N512 512 0.25 128 256 N512 32 N 512 N 512 N 512 N32 N 512 N 512 512 0.5 512 N 512 N 512 16 N 512 N 512 N 512 100 14.3 7.1 0 62.5 0 0 100 100 100 3.8 61.5 100 80 80 0 MIC = minimum inhibitory concentration. KT4561 = purified killer toxin produced by Williopsis saturnus DBVPG 4561. 1 = percentage of strains susceptible to the following concentrations: KT4561 ≤ 32 µg/ml; ethyl 3-hydroxybenzoate ≤512 µg/ml; potassium sorbate ≤512 µg/ml; potassium metabisulphite ≤512 µg/ml. control of food spoilage yeasts has been proposed by many workers (Llorente et al., 1997; Kitamoto et al., 1999; Lowes et al., 2000; Fredlund et al., 2002; Fleet, 2003; Comitini et al., 2004a,b). The present study investigates the antimycotic activity of a killer protein secreted by Williopsis saturnus strain DBVPG 4561 (labelled KT4561) against a large panel of target yeasts belonging to food spoilage species. 2. Materials and methods 2.1. Chemicals and antimycotic compounds Diethylaminoethyl (DEAE)-biogel A and sodium dodecyl sulphate– polyacrilamide gel electrophoresis (SDS–PAGE) reagents were from Bio Rad (Life Science Research Group, Hercules, CA, USA). Food-grade antimicrobial compounds currently used as preservative agents [ethyl 3-hydroxybenzoate, potassium sorbate and potassium metabisulphite, (commercially labeled as E214, E202 and E224, respectively)] were obtained from Sigma (Sigma-Aldrich Corp. St. Louis, MO, USA). On the basis of the International Numbering System (INS), these are included in the list of accepted food additives in EU countries (http:// www.foodlaw.rdg.ac.uk). 2.2. Yeast strains Williopsis saturnus strain DBVPG 4561, previously studied for its ability to secrete a killer protein (KT4561) exhibiting a wide antimycotic activity against various pathogenic yeasts (Buzzini et al., 2004), was employed as the killer toxin producer. Three-hundred and ten yeast strains belonging to 21 food spoilage species (Stratford, 2006) of 14 genera (Candida, Debaryomyces, Dekkera, Hanseniaspora, Issatchenkia, Kazachstania, Kluyveromyces, Pichia, Rhodotorula, Saccharomyces, Schizosaccharomyces, Torulaspora, Yarrowia and Zygosaccharomyces) were used as target strains (each species included its type strain). All (killer and target) cultures were obtained from the Industrial Yeasts Collection DBVPG of Perugia (Italy) (http://www. agr.unipg.it/dbvpg). Strains were sub-cultured on YEPG (yeast extract 10 g/l, peptone 10 g/l, glucose 20 g/l) agar slants. 2.3. Preparation and purification of crude KT4561 KT4561 was produced according to Buzzini and Martini (2000). After production, the protein was purified by selective precipitation, elution in DEAE-Biogel A and concentration in a collodion-bag 180 M. Goretti et al. / International Journal of Food Microbiology 131 (2009) 178–182 0 to 400 mg/ml, and from 0 to 80 mg/ml, respectively. Fractional inhibitory concentrations (FICs) were calculated as follows (Pillai et al., 2005): apparatus (Sartorius AG, Goettingen, Germany) according to Buzzini et al. (2004). The presence of a single band in the sole active fraction was confirmed by using SDS–PAGE analysis (Laemmli, 1970) and stained with silver nitrate (Giulian et al., 1983). The presence of killer activity in different fractions after each purification and elution step was checked by using the agar diffusion well bioassay (ADWB) method on killer medium (KM: yeast extract 10 g/l, peptone 20 g/l, glucose 20 g/ l, buffered at pH 4.5 with citrate–phosphate buffer) agar + methylene blue 30 mg/l (Buzzini and Martini, 2000). Protein concentration of the active fraction was determined by the Coomassie–Blue method (Bradford, 1976). - FIC of KT4561 = MIC of KT4561 + ethanol (or glucose or NaCl, used in combination) / MIC of KT4561 alone; - FIC of ethanol (or glucose or NaCl) = MIC of ethanol (or glucose or NaCl) + KT4561 (used in combination) / MIC of ethanol (or glucose or NaCl) alone. The FIC index (Σ FIC) was obtained as follows: Σ FIC=FIC of KT4561+ FIC of ethanol (or glucose or NaCl). On the basis of the current literature (Odds, 2003), the presence of a synergistic interaction between two active compounds is characterized by a FIC index of 0.5. On the contrary, a FIC index from N0.5 to 4 indicates only additive interaction whereas a FIC index N4 indicates the presence of antagonism. 2.4. In vitro antimycotic susceptibility tests The minimum inhibitory concentrations (MIC) of KT4561, ethyl 3hydroxybenzoate, potassium sorbate and potassium metabisulphite were determined in the Roswell Park Memorial Institute (RPMI) 1640 medium, pH 4.5 (Sigma, USA), in 96-well microtitre plates (Corning Inc., USA) by using the broth microdilution method of Amsterdam (2005). Ranges of two-fold serial concentrations were as follows: KT4561 = from 0.031 to 32 µg/ml; ethyl 3-hydroxybenzoate, potassium sorbate and potassium metabisulphite = from 1 to 512 mg/ml. Plates were incubated at 20 °C for 48 h. MIC 50 and MIC 90 (concentrations where 50% and 90% of isolates tested were inhibited) were determined as reported by Amsterdam (2005). The effect of pH on the antimycotic activity of KT4561 towards the type strains of all susceptible species was assessed in RPMI 1640 medium (buffered at increasing pH with 10 mM citrate–phosphate buffer) according to Amsterdam (2005). The effects of increasing concentrations of ethanol (range 20–80 mg/ml), glucose (100– 400 mg/ml) or NaCl (range 20–80 mg/ml) were also evaluated in RPMI 1640 medium at 20 °C for 48 h. All susceptibility tests were carried out in duplicate. No discrepant results were obtained in over 96% of the trials. Any conflicting tests (lesser than 4%) were repeated in triplicate. Only data with at least a 66% agreement for each isolate were taken into consideration. In order to determine the presence of synergistic, additive or antagonistic interactions between KT4561 and ethanol, glucose or NaCl checkerboard combination tests were determined in RPMI 1640 in 96-well microtitre plates according to the method of Pillai et al. (2005). KT4561 was tested at concentrations from 0 to 32 µg/ml, whereas ethanol, glucose or NaCl were from 0 to 80 mg/ml, from 2.5. Assessment of fungicidal effect of KT4561 Cells of K. marxianus (type strain DBVPG 6165, corresponding to CBS 712), chosen as model, grown for 24 h at 20 °C on YEPG agar slants and suspended in sterile distilled water, were inoculated in RPMI 1640, pH 5.0 (final concentration about 1 × 106 cells/ml). Aliquots of purified KT4561 were added to obtain two-fold increasing concentrations (from 0.031 to 1 µg/ml). Over an incubation period of 36 h at 20 °C, samples were collected and plated on YEPG agar dishes. Colonies originating from viable cells were counted in YEPG agar, after incubation at 20 °C for 48–72 h. A control test (without KT4561) was also included. 3. Results In vitro susceptibilities to KT4561, ethyl 3-hydroxybenzoate, potassium sorbate and potassium metabisulphite are reported in Table 1. KT4561 was active (MIC50 and MIC90 ≤ 32 µg/ml) against strains of Debaryomyces hansenii (85% of strains were susceptible), D. anomala (50%), Dekkera bruxellensis (93.3%), Issatchenkia orientalis (100%), Kazachstania exigua [former Saccharomyces exiguus, (Kurtzman, 2003)] (100%), Kluyveromyces marxianus (100%), Pichia anomala (93.3%), P. membranifaciens (87.5%), Saccharomyces bayanus (60%), S. cerevisiae (100%), S. pastorianus (100%), Yarrowia lipolytica (100%), Zygosaccharomyces bisporus (62.5%), Z. bailii (100%) and Z. microellipsoides (100%). Table 2 Results of chequerboard titration studies for the susceptibility of type strains of Dekkera bruxellensis, Yarrowia lipolytica and Zygosaccharomyces bisporus to the combination KT4561+ ethanol, KT4561+ glucose and KT4561+ NaCl. Species and type strain Dekkera bruxellensis DBVPG 6706 Yarrowia lipolytica DBVPG 6053 MIC KT4561 alone (µg/ml) Ethanol alone (mg/ml) KT4561/ethanol used in combination (µg/ml–mg/ml) 32 8 80 60 8/80 1/60 KT4561 alone (µg/ml) Glucose alone (mg/ml) KT4561/glucose used in combination (µg/ml–mg/ml) 8 400 4/400 KT4561 alone (µg/ml) NaCl alone (mg/ml) KT4561/NaCl used in combination (µg/ml–mg/ml) 32 8 8 80 80 80 8/80 4/80 4/80 MIC Zygosaccharomyces bisporus DBVPG 6103 MIC Dekkera bruxellensis DBVPG 6706 Yarrowia lipolytica DBVPG 6053 Zygosaccharomyces bisporus DBVPG 6103 FIC of KT4561 FIC of ethanol Σ FIC 0.25 0.12 1 1 1.25 1.12 FIC of KT4561 FIC of glucose Σ FIC 0.5 1 1.5 FIC of KT4561 FIC of NaCl Σ FIC 0.25 0.5 0.5 1 1 1 1.25 1.5 1.5 KT4561 = purified killer toxin produced by Williopsis saturnus DBVPG 4561. MIC = minimum inhibitory concentration; FIC = fractional inhibitory concentration, Σ FIC = FIC index (calculated as reported in the text). M. Goretti et al. / International Journal of Food Microbiology 131 (2009) 178–182 On the contrary, no strains of Candida parapsilosis, Hanseniaspora uvarum, K. lactis, Rhodotorula mucilaginosa, Schizosaccharomyces pombe or Torulaspora delbrueckii were inhibited by the killer toxin (Table 1). By considering the results under cumulative form, over 65% of the spoilage yeast studied were inhibited by concentrations ≤32 µg/ml of KT4561. Significantly different susceptibilities were observed for different species. For example, I. orientalis, K. exigua, K. marxianus, P. anomala, S. cerevisiae, S. pastorianus, and Z. bailii exhibited MIC50 and MIC90 of several orders of magnitude lower (from 6.4 to 256 times) than those observed in other sensitive species (Table 1). About 40% of the strains studied were inhibited by ethyl 3hydroxybenzoate and potassium metabisulphite (MIC50 and MIC90 512 g/ml). On the contrary, only about one third was sensitive to potassium sorbate in the concentrations used in this study. In all cases the MICs observed were several orders of magnitude higher (from 64 to N512 mg/ml) than those found for KT4561. The activity of KT4561 was found to be pH-, ethanol-, glucose- and NaCl-independent. Only the type strains of D. bruxellensis, Y. lipolytica and Z. bisporus exhibited an increased susceptibility to KT4561 under high concentrations of ethanol, glucose or NaCl. In order to determine if this increased activity was the result of a synergism between the killer protein and ethanol (or glucose or NaCl), a chequerboard titration was carried out. As reported in Table 2, none of the combinations (KT4561 + ethanol, KT4561 + glucose or KT4561 + NaCl) exhibited synergy, but only additive interactions (Σ FIC N 0.5) in all of the combinations. The characterization of KT4561 activity against the type strain (DBVPG 6165) of the species K. marxianus indicates that fungistatic or fungicidal effects apparently depend on the protein concentration. After 36 h exposure at 0.25 µg/ml of killer protein, a reduction (about two logarithmic cycles) of viable cells of K. marxianus DBVPG 6165 was observed. By contrast, a KT4561 concentration of 0.125 µg/ml was only fungistatic. 4. Discussion Although the killer phenomenon has been known since the 1960's, the mechanisms of action of only a few killer proteins (e.g. K1 and K28, both produced by S. cerevisiae) have been extensively studied. K1 appears to cause cell membrane damage through the formation of an energy-independent link between the toxin and a cell wall receptor in the (1→6)–β–D–glucan complex, and the creation of an additional energy-dependent link between the toxin and a receptor in the cytoplasmic membrane (Breinig et al., 2002). This is followed by an increased permeability of the membrane, the loss of H+, K+ and ATP, and subsequent cell death (Marquina et al., 2002). On the contrary, the cell wall receptor for the K28 toxin is a mannoprotein with a yet to be identified membrane receptor (Breinig et al., 2002). Once the toxin has gained access into the cell, K28 acts directly in the nucleus causing G1/S cell-cycle arrest and cell apoptosis (Schmitt et al., 1996; Reiter et al., 2005). The exact mechanism by which KT4561 exerts its killing activity against spoilage yeasts remains unclear. However, previous studies (Hodgson et al., 1995) have reported that killer proteins secreted by Williopsis spp. strains act in a similar way to KI protein. We therefore postulate that a similar mechanism could explain the observed in vitro activity of KT4561 towards spoilage yeasts. The dose-response effect observed for the type strain of the species K. marxianus (chosen as a model strain for a preliminary study on the fungistatic/fungicidal activity of KT4561) is similar to that previously observed in a protein produced by Kluyveromyces phaffii (Ciani and Fatichenti, 2001). Hodgson et al. (1995) reported that, when a killer protein acts in a dose-response way, this could be in agreement with the hypothesis that a critical number of proteins molecules are associated with (or bound to) sensitive cells to elicit a lethal or inhibitory event. Although 181 we admit that the result observed for K. marxianus DBVPG 6165 relates to only the yeast tested, we could speculate that this evidence is apparently consistent with the hypothesis of the formation of an energy-independent link between the protein and a cell wall receptor (Golubev, 2006). On the contrary, the mechanism of action of the food-grade antimicrobial compounds ethyl 3-hydroxybenzoate, potassium sorbate and potassium metabisulphite is well known. Papadimitriou et al. (2007) report that after entrance into the cell in an undissociated form, the compounds dissociate. This then causes an accumulation of protons and a subsequent acidification of the cytoplasm, while a high anion accumulation probably generates an abnormally increased osmotic pressure (Piper et al., 2001). A possible mechanism of defense in resistant strains has been postulated to be an induced protonpumping system in the cell membrane which allows growth even at high concentrations of these preservatives (Holyoak et al., 1996; Piper et al., 1998). Another study hypothesized that some strains might be able to degrade compounds such as sorbic and benzoic acids (Mollapour and Piper, 2001). The consequence of these adaptive responses to antimicrobial compounds is that the inhibition of some spoilage yeasts often requires concentrations that are near (or above) legal limits (Lambert and Stratford, 1999; Hazan et al., 2004). The results of the present study regarding the susceptibility of several food spoilage yeasts species towards ethyl 3-hydroxybenzoate, potassium sorbate and potassium metabisulphite appear to corroborate this assumption. Taking into account that pH has a big effect on efficacy of ethyl 3hydroxybenzoate, potassium sorbate and potassium metabisulphite, their in vitro antimycotic activity has been checked by using a medium buffered at pH 4.5, because this value is comparable with that observed in a large amount of fermented foods (e.g. dairy products and cured meat). In this light, and considering the fact that consumers prefer foods with reduced amounts of chemical preservatives (Papadimitriou et al., 2007), the killer protein KT4561 could possibly be a good alternative. The results of this study show, in fact, that the toxin exhibited MIC values of several orders of magnitude lower (about from 100 to 100,000 times) (Table 1) than antimicrobials of current use in the food industry. The susceptibility of food spoilage yeasts towards KT4561, ethyl 3hydroxybenzoate, potassium sorbate and potassium metabisulphite was both species- and strain-dependent. In particular, large differences in KT4561 susceptibility were found between species (and even within strains belonging to a same species). Based on the above mentioned hypothesis suggesting that the mechanism of action for KT4561 could be similar to that of K1 protein, the different patterns of susceptibility could be due to the existence of two distinct mechanisms of insensitivity to killer proteins: resistance and immunity (Golubev, 2006). Resistant yeasts lack the specific wall receptors necessary for the absorption (and hence the action) of killer proteins: this phenomenon is taxa-related. On the contrary, immunity appears to be conferred at the cytoplasmic membrane level by a component acting as a competitive inhibitor of killer proteins by saturating membrane receptors. Thus, the lack of susceptibility caused by immunity is clone-related. The partial overlapping of these two mechanisms give sometimes a complicated picture of killer protein-susceptibility relationships (Golubev, 2006). The results of this study suggest that KT4561 could be a versatile agent for the control of some food spoilage yeasts species, in particular those belonging to the genera Debaryomyces, Dekkera, Issatchenkia, Kazachstania, Kluyveromyces, Pichia, Saccharomyces, Yarrowia and Zygosaccharomyces. This is demonstrated by a wide antimicrobial spectrum together with a persistence (or even increase) of killing ability at different pHs (3 – 6), or in varying concentrations of ethanol (20–80 mg/ml), glucose (100–400 mg/ml) or NaCl (20–80 mg/ml) as can be found in processed foods. In the case of a hypothetical use of KT4561 for the control of some food spoilage yeasts, due to the large differences in susceptibility 182 M. Goretti et al. / International Journal of Food Microbiology 131 (2009) 178–182 found within strains belonging to a same species, MIC90 could be seen as base for calculating the appropriate protein concentration. The application of KT4561 in foods can occur, however, only after thorough investigations have been conducted in order to resolve a series of still open questions: i) the stability of the purified protein in foods (e.g. resistance to food and/or microbial proteases), ii) the non-toxic nature of the toxin for human consumption in the concentrations effective for antimicrobial activities, and, consequently, iii) the authorization of the use of such yeast proteins in foods by regulatory authorities. Acknowledgements Authors are grateful to G. Lo Prete and E. Sansoni for their appreciable technical assistance in the course of this study. References Amsterdam, D., 2005. Susceptibility testing of antimicrobials in liquid media. In: Lorian, V. (Ed.), Antibiotics in Laboratory Medicine. Williams and Wilkins, Philadelphia, pp. 61–143. Battey, A.S., Duffy, S., Schaffner, D.W., 2002. Modeling yeast spoilage in cold-filled readyto-drink beverages with Saccharomyces cerevisiae, Zygosaccharomyces bailii, and Candida lipolytica. Applied and Environmental Microbiology 68, 1901–1906. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248–254. Breinig, F., Tipper, D.J., Schmitt, M.J., 2002. Kre1p, the plasma membrane receptor for the yeast K1 viral toxin. Cell 108, 395–405. Buzzini, P., Martini, A., 2000. Differential growth inhibition as a tool to increase the discriminating power of killer toxin sensitivity in fingerprinting of yeasts. FEMS Microbiology Letters 193, 31–36. Buzzini, P., Corazzi, L., Turchetti, B., Buratta, M., Martini, A., 2004. Characterization of the in vitro antimycotic activity of a novel killer protein from Williopsis saturnus DBVPG 4561 against emerging pathogenic yeasts. FEMS Microbiology Letters 238, 359–365. Ciani, M., Fatichenti, F., 2001. Killer toxin of Kluyveromyces phaffii DBVPG 6076 as a biopreservative agent to control apiculate wine yeasts. Applied and Environmental Microbiology 67, 3058–3063. Comitini, F., De Ingeniis, J., Pepe, L., Mannazzu, I., Ciani, M., 2004a. Pichia anomala and Kluyveromyces wickerhamii killer toxins as new tools against Dekkera/Brettanomyces spoilage yeasts. FEMS Microbiology Letters 238, 235–240. Comitini, F., Di Pietro, N., Zacchi, L., Mannazzu, I., Ciani, M., 2004b. Kluyveromyces phaffii killer toxin active against wine spoilage yeasts: purification and characterization. Microbiology 150, 2535–2541. Fleet, G.H., 1992. Spoilage yeasts. Critical Review of Biotechnology 12, 1–44. Fleet, G.H., 2003. Yeast interactions and wine flavour. International Journal of Food Microbiology 86, 11–22. Fredlund, E., Druvefors, U., Boysen, M.E., Lingsten, K.J., Schnürer, J., 2002. Physiological characteristics of the biocontrol yeast Pichia anomala J121. FEMS Yeast Research 2, 395–402. Giulian, G.G., Moss, R.L., Greaser, M., 1983. Improved methodology for analysis and quantitation of proteins in one-dimensional silver-stained slab gels. Analytical Biochemistry 129, 277–287. Golubev, W.I., 2006. Antagonistic interactions among yeasts. In: Rosa, C.A., Péter, G. (Eds.), Biodiversity and Ecophysiology of Yeasts. Springer, Berlin, pp. 197–219. Hazan, R., Levine, A., Abeliovich, H., 2004. Benzoic Acid, a weak organic acid food preservative, exerts specific effects on intracellular membrane trafficking pathways in Saccharomyces cerevisiae. Applied and Environmental Microbiology 70, 4449–4457. Hodgson, V.J., Button, D., Walker, G.M., 1995. Anti-Candida activity of a novel killer toxin from the yeast Williopsis mrakii. Microbiology 141, 2003–2012. Holyoak, C.D., Stratford, M., McMullin, Z., Cole, M.B., Crimmins, K., Brown, A.J.P., Coote, P.J., 1996. Activity of the plasma membrane H+-ATPase and optimal glycolytic flux are required for rapid adaptation and growth of Saccharomyces cerevisiae in the presence of the weak-acid preservative sorbic acid. Applied and Environmental Microbiology 62, 3158–3164. Jakobsen, M., Narvhus, J., 1996. Yeasts and their possible beneficial and negative effects on the quality of dairy products. International Dairy Journal 6, 755–768. Kitamoto, H.K., Hasebe, A., Ohmomo, S., Suto, E.G., Muraki, M., Iimura, Y., 1999. Prevention of aerobic spoilage of maize silage by a genetically modified killer yeast, Kluyveromyces lactis, defective in the ability to grow on lactic acid. Applied and Environmental Microbiology 65, 4697–4700. Kurtzman, C.P., 2003. Phylogenetic circumscription of Saccharomyces, Kluyveromyces and other members of the Saccharomycetaceae, and the proposal of the new genera Lachancea, Nakaseomyces, Naumovia, Vanderwaltozyma and Zygotorulaspora. FEMS Yeast Research 4, 233–245. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680–685. Lambert, R.J., Stratford, M., 1999. Weak-acid preservatives: modelling microbial inhibition and response. Journal of Applied Microbiology 86, 157–164. Llorente, P., Marquina, D., Santos, A., Peinado, J.M., Spencer-Martins, I., 1997. Effect of salt on the killer phenotype of yeasts from olive brines. Applied and Environmental Microbiology 63, 1165–1167. Loureiro, V., Querol, A., 1999. The prevalence and control of spoilage yeasts in foods and beverages. Trends in Food Science and Technology 10, 156–165. Loureiro, V., Malfeito-Ferreira, M., 2003. Spoilage yeasts in the wine industry. International Journal of Food Microbiology 86, 23–50. Lowes, K.F., Shearman, C.A., Payne, J., Mackenzie, D., Archer, D.B., Merry, R.J., Gasson, M.J., 2000. Prevention of yeast spoilage in feed and food by the yeast mycocin HMK. Applied and Environmental Microbiology 66, 1066–1076. Marquina, D., Santos, A., Peinado, J.M., 2002. Biology of killer yeasts. International Microbiology 5, 65–71. Mollapour, M., Piper, P.W., 2001. The ZbYME2 gene from the food spoilage yeast Zygosaccharomyces bailii confers not only YME2 functions in Saccharomyces cerevisiae, but also the capacity for catabolism of sorbate and benzoate, two major weak organic acid preservatives. Molecular Microbiology 42, 919–930. Odds, F.C., 2003. Synergy, antagonism, and what the chequerboard put between them. Journal of Antimicrobials and Chemotherapy 52, 1. Papadimitriou, M.N.B., Resende, C., Kuchler, K., Brul, S., 2007. High Pdr12 levels in spoilage yeast (Saccharomyces cerevisiae) correlate directly with sorbic acid levels in the culture medium but are not sufficient to provide cells with acquired resistance to the food preservative. International Journal of Food Microbiology 113, 173–179. Pillai, S.K., Moellering, R.C., Eliopoulos, G.M., 2005. Antimicrobial combination. In: Lorian, V. (Ed.), Antibiotics in Laboratory Medicine. Williams and Wilkins, Philadelphia, pp. 365–424. Piper, P., Mahe, Y., Thompson, S., Pandjaitan, R., Holyoak, C., Egner, R., Muhlbauer, M., Coote, P., Kuchler, K., 1998. The Pdr12 ABC transporter is required for the development of weak organic acid resistance in yeast. EMBO Journal 17, 4257–4265. Piper, P., Calderon, C.O., Hatzixanthis, K., Mollapour, M., 2001. Weak acid adaptation: the stress response that confers yeasts with resistance to organic acid food preservatives. Microbiology 147, 2635–2642. Pitt, J.I., Hocking, A.D., 1997. Fungi and Food Spoilage. Blackie Academic and Professional, London. Reiter, J., Herker, E., Madeo, F., Schmitt, M.J., 2005. Viral killer toxins induce caspasemediated apoptosis in yeast. Journal of Cell Biology 168, 353–358. Schmitt, M.J., Klavehn, P., Wang, J., Schonig, I., Tipper, D.J., 1996. Cell cycle studies on the mode of action of yeast K28 killer toxin. Microbiology 142, 2655–2662. Stratford, M., 2006. Food and beverage spoilage yeasts. In: Querol, A., Fleet, G.H. (Eds.), Yeasts in Food and Beverage. Springer, Berlin, pp. 335–379. Thomas, D.S., 1993. In: Rose, A.H., Harrison, J.S. (Eds.), Yeasts as spoilage organisms in beverages. . The Yeasts, vol. 5. Academic Press, London, pp. 517–561. Tudor, E.A., Board, R.G., 1993. In: Rose, A.H., Harrison, J.S. (Eds.), Food-spoilage yeasts. . The Yeasts, vol. 5. Academic Press, London, pp. 435–516. Viljoen, B.C., 2001. The interaction between yeasts and bacteria in dairy environments. International Journal of Food Microbiology 69, 37–44. Viljoen, B.C., Lourens-Hattingha, A., Ikalafenga, B., Peter, G., 2003. Temperature abuse initiating yeast growth in yoghurt. Food Research International 36, 193–197.