Ethanol tolerance

The ability of yeast to withstand high concentrations of ethanol is obviously key to successful fermentation, particularly in the case of high gravity brewing using very concentrated worts (see Section 2.5). A number of factors may be considered as contributors to ethanol tolerance:

(1) Genetic components

(2) Toxicity of ethanol (or intermediates)

(3) Influence of physical environment

(4) Influence of physiological condition of yeast

(5) Influence of wort composition

It is generally accepted, although usually not proven, that there are genotypic differences in yeast strains, which predispose some but not all to withstand high ethanol concentrations. Two components may be recognised in this regard: the maximum concentration that may be generated during fermentation that does not compromise the viability of the crop and the maximum rate of ethanol production. These two elements are reflections of, first, the ability of the yeast to withstand the stress imposed by exogenous ethanol and, second, the susceptibility of the pathways leading to ethanol formation to feedback inhibition by the product. Brewing strains are considered to be moderately tolerant of ethanol, typically being used in fermentations in which up to 8% v/v ethanol is formed. Strains of Saccharomyces used for wine production are considered as being more tolerant of ethanol, performing fermentations which generate 10-15% v/v ethanol; in some extreme cases saké yeasts produce more than 20% v/v ethanol.

Such classifications assume that ethanol is actually inhibitory or toxic to yeast cells, or at least some aspects of their metabolism which relate to fermentation performance. Of course, if this is so then it is equally possible that intermediates of ethanol production or other products of fermentation are exerting deleterious effects in addition to, or instead of, ethanol. Conversely, ethanol tolerance may reflect differences in the ability of yeast strains to ferment very concentrated worts. In this regard the controlling factor may be ability to grow at reduced water activity and not in the presence of elevated ethanol concentrations, although these may be related. Assuming that ethanol exerts toxic effects on yeast cells it is probable that physical environmental conditions such as temperature, pressure, and pH will modulate the severity of the effects.

The purely genetic elements of ethanol tolerance, if they exist, will be modified by phenotypic factors. The condition of the pitching yeast is known to influence fermentation performance and therefore it is likely that ethanol tolerance will also be affected. In particular, the ability of yeast to withstand stresses is modulated by physiological condition. In this regard the conditions of storage of yeast in the interval between cropping and re-pitching will be influential. Further modifications to the yeast phenotype are possible due to the composition of the wort.

3.6.1 Ethanol formation during fermentation

During batch fermentation there are progressive changes in the rates of ethanol production. Thus, there is an initial lag phase, which corresponds with the passage of the yeast from lag to exponential growth. During the latter period the rate of ethanol formation reaches a maximum. Soon after yeast growth ceases the rate of ethanol production declines (see Fig. 3.1). It is assumed that the causes of the decline in rates of ethanol production are a consequence of a combination of nutrient depletion and toxic effects of ethanol.

Dombek and Ingram (1987) measured rates of ethanol production during batch fermentation of 20% glucose, which resulted in the formation of more than 10% v/v ethanol. The authors reported that specific activities of glycolytic and ethanologenic enzymes, measured in vivo, remained high throughout fermentation. However, rates of ethanol production, assessed off-line in terms of rates of C02 evolution, declined progressively as ethanol accumulated. Since removal of the ethanol did not restore rates of ethanologenesis it was concluded that the effects had other physiological causes, possibly related to irreversible ethanol-induced damage. In a later paper the same authors (Dombek & Ingram, 1988; Alterthum et al., 1989) concluded that there was a decline in glycolytic activity and this was due to changes in the pool sizes of adenine nucleotides. In particular, hexokinase was strongly inhibited by increased AMP concentration.

Others have reported control of glycolysis at other parts of the pathway. For example, the possibility of regulation of phosphofructokinase by mechanisms such as adenylate energy charge as well as other metabolites such as citrate and isocitrate has been reported (Sols, 1981). Others have disputed this thesis. The necessity for phosphofructokinase for ethanol formation has been questioned (BreitenbachSchmitt et al., 1984, Heinisch & Zimmermann, 1985). The same group demonstrated that mutant strains with increased expression of glycolytic genes did not permit rates of ethanol production greater than the wild type (Schaaf et al., 1989).

The control of carbon flux at the level of pyruvate has perhaps more plausibly been implicated as being influential in ethanol accumulation. Sharma and Tauro (1987) considered that strains of S. cerevisiae could be classified on the basis of their ability for rapid and slow ethanol production. These authors reported that the rapid ethanol producers had greater specific activities of pyruvate decarboxylase and lower aldehyde dehydrogenase compared to the slow producing class. The importance of control of carbon flow through pyruvate has been discussed with regard to the Crab tree effect (Section 3.4). However, it is interesting to note that during brewery fermentation pyruvate is excreted into the beer (Coote & Kirsop, 1974). The production of exogenous pyruvate is transient and the peak appears during the period of maximum ethanol production (Fig. 3.10). This implies that in brewing fermentations, at least for most of the time, dissimilation of sugars to the level of pyruvate is more rapid than carbon flow from pyruvate to ethanol and other products. Since ethanol accumulation occurs simultaneously with formation of extracellular pyruvate this also suggests that carbon flux via pyruvate decarboxylase is more rapid than that through other pyruvate-utilising pathways.

3.6.2 Ethanol toxicity

Several other authors have reported inhibitory effects due to ethanol. For example, Thatipamala et al. (1992) found that in batch culture an increase in ethanol con-

Fig. 3.10 Pattern of formation of extra-cellular pyruvate during the course of a high-gravity (1060) lager fermentation. The pitching rate was 12 x 106 cells ml-1, the initial dissolved oxygen concentration was 18 ppm and the temperature was maintained at a constant 15°C throughout (Boulton and Box, unpublished data).

Fig. 3.10 Pattern of formation of extra-cellular pyruvate during the course of a high-gravity (1060) lager fermentation. The pitching rate was 12 x 106 cells ml-1, the initial dissolved oxygen concentration was 18 ppm and the temperature was maintained at a constant 15°C throughout (Boulton and Box, unpublished data).

centration was responsible for an instantaneous decrease in biomass yield. Product inhibition was found to occur when the substrate concentration was higher than 150 g 1 1. In order that ethanol may exert its inhibitory effects, it has been suggested that intracellular ethanol accumulation takes place during fermentation. Nagoda-withana and Steinkraus (1976) reported that ethanol generated by yeast was more cytotoxic than exogenous ethanol added at the same concentration. These authors suggested that this was due to an increased intracellular ethanol concentration which accumulated during rapid fermentation. Several other authors confirmed accumulation of ethanol (Novak et al., 1981; Loureiro & Ferrera, 1983; Strehaiano & Goma, 1983; Legmann & Marglith, 1986).

Others have concluded that there is no accumulation of ethanol during fermentation (Dombek & Ingram, 1986; Jones, 1988). D'Amore et al. (1988) reported that up to 3 hours after the start of fermentation the intracellular concentration of ethanol was greater than that measured in the growth medium. However, after 12 hours both intracellular and extracellular concentrations were similar. It now appears that the plasma membrane is freely permeable to ethanol and intracellular accumulation only occurs when the rate of fermentation is very rapid and rates of production exceed those of diffusion out into the medium. This occurs rarely, usually only in early fermentation.

The explanation for the observation that ethanol generated intracellularly is more toxic than that added exogenously has been ascribed to the effects of other metabolites related to ethanol formation and which may exert toxic effects in addition to ethanol. Viegas et al. (1985) provided evidence that the powerful detergent properties of short chain fatty acids produced during fermentation exerted synergistic toxic effects with ethanol on yeast cells. Okolo et al. (1987) reported that higher alcohols increased the toxicity of ethanol. Others have described inhibitory effects due to acetate, which were not due simply to pH (Pampulha & Loureiro, 1989; Pampulha & Loureiro-Dias, 1990; Phowchinda et al., 1995). Perhaps most convincingly it has been argued that acetaldehyde, the immediate precursor of ethanol, is an order of magnitude more toxic than ethanol (Jones, 1987, 1989).

Several mechanisms for ethanol toxicity have been described, both non-specific and those in which specific cellular sites of action have been identified. D'Amore et al. (1988) concluded that ethanol toxicity was a non-specific effect in which increased intracellular ethanol produced increased osmotic pressure. However, these authors concluded that nutrient limitation was responsible for the observed reduction in growth and fermentation rates. Jones and Greenfield (1987) also considered that nonspecific osmotic effects were also contributory to ethanol toxicity. However, specific intracellular targets for ethanol have also been reported. In particular, ethanol is reported to exert toxic effects on cell membranes. Salgueiro et al. (1988) described the ethanol-induced leakage of amino acids and other UV-absorbing cellular components. It was concluded that ethanol tolerance and resistance to leakage could be correlated; furthermore, supplementation of growth media with the leaked material resulted in an improvement of alcoholic fermentation. Lloyd et al. (1993) described inhibition of transport systems in yeast by ethanol. Jimenez and Benitez (1987) reported that ethanol tolerant enological yeast strains were able to adapt membrane structure such that the toxic effects of ethanol could be withstood. Less tolerant strains were unable to undergo this adaptation and growth in the presence of ethanol resulted in increased but reversible sensitivity. The reversible effect required an energy source and was abolished by cycloheximide. This was taken to imply that protein components of the membrane might contribute to ethanol tolerance.

More recent work has shown that it is the lipid components of membranes that confer ethanol tolerance. Novotny et al. (1992) demonstrated that in chemostat cultures manipulation of carbon to nitrogen ratios which resulted in an increased membrane content of 5,7-unsaturated sterols was accompanied by increased resistance of the population to ethanol-induced death. Similarly, Alexandre et al. (1994) concluded that ethanol tolerance was associated with altered membrane fluidity. This was achieved by an increase in the proportion of ergosterol and unsaturated fatty acids, together with maintenance of phospholipid biosynthesis. Presumably, this would explain why elevated temperature increases the toxic effects of ethanol, since the former is associated with greater saturation of membrane lipids. A further response of yeast to ethanol may be accumulation of trehalose. The ability of this disaccharide to stabilise membranes has been described (Section 3.4.2.2).

Another target for ethanol toxicity is the mitochondrion (see Section 4.1.2.3). Aguilera & Benitez (1985) reported that yeast cells lacking mitochondria were more susceptible to ethanol inhibition. It is widely recognised that ethanol is a mutagen of the mitochondrial genome (see Section 4.3.2.7). Thus, Bandas and Zakharov (1980) noted that 24% v/v ethanol resulted in a five-fold increase in the generation of petite mutants in cultures of S. cerevisiae compared to the spontaneous mutation rate. This mutation is the most commonly occurring type in yeast during brewing fermentations, and, furthermore, such strains exhibit poor fermentation properties (Ernandes et al., 1993).

Costa et al. (1997) provided evidence that ethanol tolerance in S. cerevisiae required the presence of an active Mn-superoxide dismutase. This is the mitochondrial form of this enzyme and this was taken to indicate that the formation of oxygen radicals were implicated in the mechanism of ethanol toxicity. It is notable that the stress response of yeast (see Section 3.4.2.2), which may be triggered by heat shock, is associated with induction of enzyme systems which are involved in protection against oxidative stress (Stephen et al., 1995). Furthermore, there is overlap between cellular responses to heat and ethanol shock (Piper, 1995). Interestingly, the mitochondrial superoxide dismutase is not present in brewing yeasts under anaerobic conditions and requires several hours' exposure to oxygen before it is induced (Clarkson et al., 1991). The implication of the findings regarding oxygen radicals is that ethanol should be less toxic to yeast under anaerobic conditions. Furthermore, this aspect of ethanol tolerance should not be of significance in brewing fermentations since aerobiosis occurs only when ethanol concentrations are low. Certainly, the oxygenation process described in Section 6.4.2.2 subjected yeast, suspended in beer, to prolonged aerobiosis. There was no observed decrease in viability during the course of this treatment (Fig. 6.23). Furthermore, under these conditions the Mn superoxide dismutase would not be expressed (see Clarkson et al., 1991). Therefore if the toxic effects of ethanol are associated with the formation of oxygen radicals then the cytosolic and constitutive Cu-Zn superoxide dismutase, or other mechanisms, must have been active and effective.

Ethanol tolerance can be influenced by wort composition. The effect of wort oxygen concentration on the formation of sterols and unsaturated fatty acids and how these may then modulate membrane structure and ethanol tolerance are obvious from the previous discussion. Of course, the implication is that ethanol will be present in fermenter at high concentration when growth has ceased and unsaturated fatty acid and sterol concentrations will be minimal. Again the apparent toxicity of ethanol during the oxygenation process alluded to above, suggests that lack of sterol is not a critical factor with all yeast strains. Nevertheless, the synthesis of trehalose associated with the oxygenation process perhaps indicates both the level of stress associated with the process and the mechanism for dealing with it.

Several reports indicate that metal ions can moderate the inhibitory effects of ethanol. Dombek and Ingram (1986) observed that supplementation of a glucose medium with 0.5 mM magnesium resulted in prolonged exponential growth and a smaller reduction in fermentation rate compared to an unsupplemented control. Ciesarova et al. (1996) made similar observations but extended the investigations to include the effects of magnesium and calcium. They concluded that, under non-growing conditions in the presence of 10% v/v ethanol, both metal ions exerted protective effects. In fermentation, growth was stimulated by magnesium in the presence and absence of added ethanol, to a greater extent than calcium when added singly. Ethanol fermentation was more efficient when both metal ions were present.

It is suggested that some worts may contain less than optimal concentrations of these metals. Two groups of workers (Walker et al., 1996; Rees & Stewart, 1997) concluded that the ratio of magnesium to calcium was influential. With six yeast strains high magnesium to calcium ratios favoured rapid initial fermentation rates and improved ethanol yields. If the ratio was reversed ethanol production declined and uptake of maltotriose uptake was reduced. In the case of lager strains, maltose uptake was also affected. Conversely, Bromberg et al. (1997) noted stimulation of fermentation only by the addition of zinc; manganese, calcium and magnesium addition had no effect.

The mechanism by which metal ions may relieve ethanol toxicity is not known. However, magnesium in particular is a co-factor in numerous enzymes, including several involved in glycolysis. Furthermore, other divalent cations can compete for the magnesium binding sites. The metal ion contents of worts will depend upon the mineral composition of the brewing liquor, as well as those metals released from other raw materials. In addition, chelating agents in worts will serve to modulate the concentrations of metals available to yeasts. It may be readily appreciated, therefore, that metal ions may not always be limiting, depending on the particular wort under investigation.

Brew Your Own Beer

Brew Your Own Beer

Discover How To Become Your Own Brew Master, With Brew Your Own Beer. It takes more than a recipe to make a great beer. Just using the right ingredients doesn't mean your beer will taste like it was meant to. Most of the time it’s the way a beer is made and served that makes it either an exceptional beer or one that gets dumped into the nearest flower pot.

Get My Free Ebook


Post a comment