'Killer' character of yeasts isolated from ethanolic fermentations

O cárater 'killer' de leveduras na fermentação etanólica

'Killer' character of yeasts isolated from ethanolic fermentations

Sandra Regina Ceccato-Antonini^1*; Luiz Carlos Moreira Cremonini^1,2; Christine Regenfuss^1,3

¹Depto. de Tecnologia Agroindustrial e Sócio-Economia Rural, Centro de Ciências Agrárias - UFSCar/Campus de Araras C.P. 153 - CEP: 13600-970 - Araras, SP.

²Bolsista do CNPq/PIBIC.

³Bolsista - Intercâmbio IAESTE.

*e-mail: [e-mail]

ABSTRACT: The number of killer, neutral and sensitive yeasts was determined from strains isolated from substrates related to alcoholic fermentations. From 113 isolates, 24 showed killer activity against NCYC 1006 (standard sensitive strain), while 30 were sensitive to NCYC 738 (standard killer strain), and 59 had no reaction in assays at 25-27°C. Two wild yeast strains of Saccharomyces cerevisiae and one of Candida colliculosa were tested against 10 standard killer strains and one standard sensitive strain in a cell x cell and well-test assays at four different pHs. None of the isolates displayed strong killer activity or were sensitive to the standard strains. All belonged to the neutral type. It was concluded that although the number of killer strains was high, this character cannot be used to protect ethanol fermentation processes against yeast contaminants like those which form cell clusters.

Key words: killer yeasts, ethanolic fermentation, wild yeast

O cárater 'killer' de leveduras na fermentação etanólica

RESUMO: Avaliou-se o número de linhagens 'killer', sensíveis e neutras em leveduras isoladas de substratos relacionados à fermentação etanólica. Das 113 linhagens, 24 mostraram atividade 'Killer' contra o isolado NCYC 1006 (padrão de sensibilidade), 30 foram sensíveis ao isolado NCYC 738 (padrão 'Killer') e 59 apresentaram reação neutra para ambos os isolados. Três cepas de leveduras selvagens do processo (duas de Sacch. cerevisiae e uma de Candida colliculosa, que formam cachos de células não separáveis por tratamentos químicos ou físicos), foram testadas contra 10 isolados padrões do tipo 'Killer' e um isolado padrão de sensibilidade. Os ensaios células x células e células X toxinas foram realizados em diferentes pH e a 30ºC. As três cepas contaminantes mostraram reação neutra a todos os isolados padrões testados. Apesar do alto número de linhagens 'Killer' entre aquelas testadas, concluiu-se que este caráter não pode ser utilizado na proteção do processo de fermentação etanólica contra essas leveduras selvagens contaminantes.

Palavras-chave: leveduras 'Killer', fermentação etanólica, levedura selvagem

INTRODUCTION

Killer yeasts produce extracellular toxins known to inhibit sensitive yeast strains. This character was first identified in Saccharomyces cerevisiae several years ago and was found in wine yeasts (Naumov et al., 1973; Heard and Fleet, 1987), in 'saké' yeasts (Imamura et al., 1974), in sugarcane molasses yeasts (Bonilla-Salinas et al., 1995) and in brewing yeasts (Maule & Thomas, 1973). Other yeast genera such as Candida, Debaryomyces, Hansenula, Kluyveromyces, Pichia, Torulopsis and Cryptococcus (Philliskirk & Young, 1975; Stumm et al., 1977; Rosini, 1985; Rosini & Cantini, 1987) also present the killer property.

In ethanol fermentation for fuel, the decrease in fermentative yield has been related to contamination by wild yeasts, identified as Saccharomyces and non-Saccharomyces strains, which can outnumber the inoculated yeasts. Since killer toxin- producing strains are being used commercially for the fermentation of beverages (Tredoux et al., 1986), this property has been suggested as viable to control such contaminants.

The objectives of this study were to check for killer yeasts from a collection of strains from ethanolic fermentations and to evaluate the possibility of using the killer yeast system in industrial fermentations.

MATERIAL AND METHODS

One hundred and thirteen strains of fermentative yeasts isolated from musts, mills, sugar cane and molasses, (maintained at 4 º C on filter paper or on GYMP agar slants) were screened for their killer, sensitive, or neutral character. To screen killer strains, it was utilized the conventional method of observing a clear killing zone around the test yeast streaked on YEPD-MB (0.003%) (pH 4.5) seeded with a sensitive strain (S. cerevisiae NCYC 1006) at 25-27º C for two days. The sensitive character was detected by the formation of a blue streak (dead cells stained by methylene blue) when a killer strain (S. cerevisiae NCYC 738 - K₂ type) was seeded on the background. The strains that did not respond to both standard strains were considered as neutral.

Two strains of S. cerevisiae and one of Candida colliculosa (TABLE 1), isolated as contaminants from fermentation musts of sugar cane juice, were tested against ten standard killer strains and one standard sensitive strain (TABLE 2), in cell x cell and well-test assays.

The seeded-agar-plate technique (cell x cell assay) was used with YEPD-MB agar buffered at different pHs (3.5; 4.0; 4.5; and 5.0) with 0.1 M phosphate-citrate buffer. Yeast suspensions were prepared in saline solution and spreaded, in aliquots of 0.1 ml, with Drigalsky loop onto the agar surface. Yeast strains to be assayed for their killer activity (TABLES 1 and 2) were streaked onto the agar surface and the plates incubated at 30ºC for 2 days and then, the presence/absence of inhibition and blue zones, due to the presence of dead cells, were evaluated.

To carry out the well-test assays, the strains listed in TABLES 1 and 2 were grown in 10 ml of YEPD medium, buffered to pH 4.6 with phosphate-citrate buffer, at 23-25ºC for 3 days, in erlenmeyer flasks, without shaking. The cell suspensions were filtered through 0.22 mm pore size membrane filters (Millipore). Eighty-ml samples of each filtrate were inoculated into wells (7-mm diameter) cut in the YEPD-MB agar medium buffered at different pHs (3.5; 4.0; 4.5; and 5.0) with 0.1 M phosphate-citrate buffer. The presence/absence of inhibition and blue zone of dead cells were evaluated after incubation for 2 days at 30ºC.

RESULTS AND DISCUSSION

The results showed that 24 cultures, out of 113 yeast strains, killed the sensitive standard strain, sensitive strains numbered 30, while 59 were neutral. The percentage of killer strains is high when compared with the results of other authors, around 1-2%, using isolates from fruits, sugar cane juice and musts from alcohol producing plants (Chung et al., 1989; Nascimento, 1994).

The evaluation of these killer yeast strains in ethanol production is an interesting approach. However, the presence of the killer property may be an interesting but not essential character for an ethanol-making starter, since the wild (contaminant) yeasts may be indifferent (neutral) to the killer factor.

Strains of yeasts are scored as killers when the inoculum/filtrate is surrounded by a clear zone, in which no growth of the sensitive strains occurs, bounded by a zone of dead cells which, in presence of methylene bue, stain dark blue (Young, 1987). Therefore, using this assumption, none of the yeast strains tested showed killer activity or were sensitive to the killer strains (TABLES 3 and 4). A clear zone of inhibition could be observed using yeast filtrates of CCA 361, 362 and 363 against some killer yeasts and the sensitive strain (NCYC 1006), as seen in TABLE 4, however, since no ring of dead cells is present, cell death is absent, and the growth arrest (inhibition) may be produced by antifungal metabolites other than yeast killer toxins.

The variation of pH value from 3.5 to 5.0 did not bring about significant results.

The most commonly temperature range used for incubation is 18-20^oC because, in general, killer toxins are rapidly inactivated at temperatures in excess of 20^oC. However, toxin material tends to be more thermostable in agar and may be incubated at 28-30^oC, for a short incubation time (one to two days) (Young and Yagiu, 1978). Ethanolic fermentation is mostly carried out at 30-32^oC and so, 30^oC was the temperature chosen for the test, in order to evaluate the action of killer toxins in the conditions of pH and temperature occurring in the fermentation tanks.

The majority of results obtained showed only a ring of dead cells (blue zone) around the strain or the well, probably due to a very weak killer activity at 30^oC (TABLES 3 and 4).

It can be concluded that the yeasts tested are neutral, that is, they do not produce toxins and are resistant to them. So, protection against spoilage by these wild yeasts using killer yeasts/toxins cannot be guaranteed. On the other hand, the predominance of these wild yeasts in the fermentation process cannot be atributed to their killer activity, although this character may be one of the factor influencing it.

In all distilleries from which these wild yeast strains were isolated, the concentration of total yeast cells during the fermentation was always higher than normal. Probably their higher and faster growth rate allows them to outnumber the starter yeasts, sometimes causing a decrease in fermentative yield (alcohol production) and operational troubles like the formation of a kind of 'foam'. Their cells are grouped in clusters like 'flocks', unseparated by chemical or physical treatments, remaining at the top of the fermentation musts (Ceccato-Antonini & Parazzi, 1996).

CONCLUSIONS

• the percentage (and number) of 'killer', sensitive and neutral character among strains of yeasts isolated from substrates related to ethanolic fermentation, was 21% (24), 27% (30) and 52% (59), respectively, in assays at 25^o-27^oC;

• three undesirable wild strains (two Sacch. cerevisiae and one Candida colliculosa, which present cell clusters unseparated by chemical or physical treatments) to fuel ethanol fermentation, showed neutral reaction to all `killer' standards tested, in cell X cell and well-test assays, at different pH;

• although the number of killer strains is considerably high, this character cannot be used to protect ethanolic fermentation against those wild yeasts.

ACKNOWLEDGEMENTS

The authors wish to thank CNPq and IAESTE for the oportunity of training for the second and third authors, respectively.

Received June 18, 1998

Accepted January 27, 1999

• BONILLA-SALINAS, M.; LAPPE, P.; ULLOA, M.; GARCIA-GARIBAY, M. GŐMEZ-RUIZ, L. Isolation and identification of killer yeasts from sugar cane molasses. Letters in Applied Microbiology, v.21, p.115-116, 1995.
• CECCATO-ANTONINI, S.R.; PARAZZI, C. Isolamento de levedura selvagem floculante e efeitos da contaminaçăo em processo de fermentaçăo etanólica contínua. In: CONGRESSO NACIONAL DA STAB, 6., Maceió, 1996. Anais Maceió: STAB, 1996. p.23-29.
• CHUNG, K.T.; BANG, K.W.; CHUNG, S.W.; SONG, H.I.; KIM, J.K. Isolation of the killer yeast and its characteristics. Korean Journal of Microbiology, v.27, n.4, p.415-421, 1989.
• GREEN, S.R.; GRAY, P.P. A differential procedure applicable to bacteriological investigation in brewing. Wallerstein Laboratories Communication, v.13, p.357-368, 1950.
• HEARD, G.M.; FLEET, G.H. Occurrence and growth of killer yeasts during wine fermentation. Applied and Environmental Microbiology, v.53, p.2171-2174, 1987.
• IMAMURA, T.; KAWAMOTO, M.; TAKAOKA, Y. Characteristics of main mash infected by killer yeast in sake brewing and the nature of its killer factor. Journal of Fermentation Technology, v.52, p.293-299, 1974.
• MAULE, A.P.; THOMAS, P.D. Strains of yeast lethal to brewery yeasts. Journal of the Institute of Brewing, v.79, p.137-141, 1973.
• NASCIMENTO, A M. Isolamento e seleçăo de leveduras produtoras de fator 'Killer' para aplicaçăo na produçăo de bebidas alcoólicas. Campinas, 1994. 74p. Dissertaçăo (Mestrado) - Faculdade de Engenharia de Alimentos, Universidade Estadual de Campinas.
• NAUMOV, G.I.; TYURINA, L.V.; BUR'YAN, N.I.; NAUMOVA, T.I. Wine making, an ecological niche of type K2 killer Saccharomycetes. Biologicheskie Nauki, v.16, p.103-107, 1973.
• PHILLISKIRK, G.; YOUNG, T.W. The occurrence of killer character in yeasts of various genera. Antonie van Leeuwenhoek, v.41, p.147-151, 1975.
• ROSINI, G. Interaction between killer strains of Hansenula anomala var. anomala and Saccharomyces cerevisiae yeast species. Canadian Journal of Microbiology, v.31, p.300-302, 1985.
• ROSINI, G.; CANTINI, M. Killer character in Kluyveromyces yeasts: activity on Kloeckera apiculata. FEMS Microbiology Letters, v.44, p.81-84, 1987.
• STUMM, C.; HERMANS, J.M.; MIDDELBEEK, E.J.; CROES, A.F.; VAN DE VRIES, G.J.M.L. Killer-sensitive relationships in yeasts from natural habitats. Antonie van Leeuwenhoek, v.43, p.125-128, 1977.
• TREDOUX, H.G.; TRACEY, R.P.; TROMP, A. Killer factor in wine yeasts and its effect on fermentation. South Africa Journal of Enology and Viticulture, v. 7, p.105-112, 1986.
• YOUNG, T.W. Killer yeasts. In: ROSE, A H.; HARRISON, J.J. (Ed.) The yeasts. London: Academic Press, 1987. p.131-164.
• YOUNG, T.W.; YAGIU, M. A comparison of the killer character in different yeasts and its classification. Antonie van Leeuwenhoek, v.44, p.59-77, 1978.

Publication Dates

• Publication in this collection
  17 Sept 1999
• Date of issue
  July 1999

History

• Received
  18 June 1998
• Accepted
  27 Jan 1999