Info

Production 'generations' Fig. 4.20 Pumped solids for two production yeasts exhibiting a change in flocculence.

pieces of evidence suggest that the atypically flocculent population is genuinely different to the 'parental' population and that such observations are not some artefact of the process. Firstly, the populations differ in their response to calcium, which is required for flocculation (see Section 4.4.6.1). Washed cells of the atypical yeast are more susceptible to calcium-promoted flocculation than the normal yeast (see Fig. 4.21). Furthermore, there are small but definite genetic differences between the two yeasts. RFLP fingerprinting (Section 4.2.6.1) of the two strains (Wightman et al., 1996) using two different restriction enzymes has shown there to be three bands which differ in intensity between the two yeasts (Fig. 4.22). Although by no means definitive, these observations strongly suggest a genetic explanation for changes in flocculation in a production environment. What is not clear is how such events seemingly result in a step change in phenotype rather than a progressive but slow shift in behaviour.

Calcium (mg/l)

Fig. 4.21 Role of calcium in the flocculation of the yeasts with normal (A) and atypical (■) flocculence.

Calcium (mg/l)

Fig. 4.21 Role of calcium in the flocculation of the yeasts with normal (A) and atypical (■) flocculence.

Casey's (1996) observations on long-term culture stability are particularly interesting. Retrospective chromosomal analysis of Stroh Brewery yeast samples deposited on a yearly basis between 1958 and 1985 revealed seven different karyotypes. However, the changes were restricted to four chromosomes, I (230 kb), VI (270), X (745) and XI (667). Casey argues that positive selective pressures may drive these changes as these chromosomes carry genes that are important in yeast performance (I/flocculence, Vl/glycolyis, Xl/maltose utilisation and X/diacetyl production).

4.3.2.7 Mitochondrial instability. There is general acceptance about one genetic change in brewing yeast, the 'petite' (p ) mutation first identified in S. cerevisiae in 1949 by Ephrussi (Cox, 1995) by their atypically small colony size on agar plates. These mutants are respiratory deficient and, consequently, are unable to grow on oxidative substrates such as ethanol and glycerol. Correspondingly, unaffected cells

Fig. 4.22 RFLP DNA fingerprinting of the yeasts in Fig. 4.20. Normal yeast (lanes 7 and 9) and yeast with atypical flocculence (lanes 8 and 10) digested with different restriction enzymes (Pi/I lanes 7. 8 and Sail lanes 9. 10). Differences in the RFLP fingerprint identified by arrows. From Wightman et ai. (1996). with permission from the Society for Applied Bacteriology.

Fig. 4.22 RFLP DNA fingerprinting of the yeasts in Fig. 4.20. Normal yeast (lanes 7 and 9) and yeast with atypical flocculence (lanes 8 and 10) digested with different restriction enzymes (Pi/I lanes 7. 8 and Sail lanes 9. 10). Differences in the RFLP fingerprint identified by arrows. From Wightman et ai. (1996). with permission from the Society for Applied Bacteriology.

(p + ) which form normal colonies are described as being 'grande'. This genetic change does not effect the chromosomal DNA of the nucleus but the specialised DNA of the mitochondria.

Mitochondrial DNA is the poor relation of the yeast genome inasmuch that it was not subject to the collaborative and systematic sequencing of the genome project. However Foury el al. (1998) have subsequently published the sequence of mitochondrial DNA from the same haploid strain of S. cerevisiae as used in the genome project. At only 86 kb, mDNA is significantly smaller than the smallest chromosome (chromosome I at 230 kb). Although representing only 0.5% of the yeast genome (as measured by length or 'hypothetical' proteins), the mitochondrial genome is enriched (c. ten-fold) by genes that code for transfer RNA. Indeed, the mDNA codes for only about 25 identified proteins and seven hypothetical potential proteins (ORFs - see Section 4.3.2.1) as the vast majority (95%) of mitochondrial proteins are encoded by the nuclear genome (Dujon, 1981) and imported into the mitochondria.

The petite mutation is typically caused by loss (p°), gross alterations or, in the case of lager yeast petites, extensive deletion of mitochondrial (mt) DNA (Good el al., 1993). In the later situation, the remaining mtDNA is amplified so that the petite contains the same amount of mtDNA as the wild type yeast (Good el al., 1993; Piskur el al., 1998). The mutation is frequently described as being 'spontaneous' and 'natural' occurring in S. cerevisiae at a frequency of c. 1% (Silhankova el al., 1970a). However, in the laboratory, mutagens such as ethidium bromide (Meyer & Whit-taker, 1977) can be used to generate petites (Piskur el al., 1998).

Classically, petites are identified by the 'tetrazolium overlay' plate test of Ogur et al. (1957) as white colonies whereas grandes are red. Alternatively, and preferably, petites can be estimated by more functional approaches. For example, replica plating from a fermentative (glucose) to an oxidative (glycerol) carbon source or directly by classification as large (p+) or small (p ) colonies on media containing a limiting amount of fermentable carbohydrate (Piskur et al., 1998).

As ever, it is unwise to make sweeping generalisations about the genetics and physiology of relatively closely related species of yeast. For example, the three groupings that make up the genus Saccharomyces (see Section 4.2.1) differ in the size and stability of their mitochondrial genomes (Piskur et al., 1998). Both the Saccharomyces sensu stricto and Saccharomyces sensu lato form petites spontaneously but differ in susceptibility to ethidium bromide. The Saccharomyces sensu stricto group that contains S. cerevisiae has the biggest mtDNA (64-85 kb), and petite formation is induced by treatment with ethidium bromide. The mitochondrial genome is smaller (23-48 kb) in the Saccharomyces sensu lato group and ethidium bromide does not induce significant formation of petites. The third group, consisting of S. kluyveri, is quite distinct in not giving rise to petites, neither naturally nor after treatment with the mutagen, ethidium bromide.

To add further complexity, within S. cerevisiae, the frequency of the petite mutation can vary. For example, Donnelly and Hurley (1996) have reported that a population with 2-4% petites is 'acceptable for brewing purposes'. Conversely Morrison and Suggett (1983) reported that in one production lager strain, petites accounted for almost 50% of the population. In this case, petites were seemingly triggered by protracted yeast storage. Others have reported the induction of petites by the use (and presumably carryover in fermenter) of formaldehyde in the brewhouse to improve colloidal stability (Cowan et al., 1975). Finally, there may be more subtle reasons for any variation in susceptibility to petite formation. As discussed at length earlier in this chapter (Section 4.2.2), ale yeasts are exclusively S. cerevisiae whereas lager yeasts are a genetic hybrid between S. monacensis and S. cerevisiae. Interestingly, as shown by Piskur et al. (1998), the mtDNA of lager strains is exclusively derived from the S. bayanus parent which, although similar, differs from the mtDNA of S. cerevisiae. Consequently, differences in the nature and frequency of the petite mutation might be anticipated in lager and ale isolates of S. cerevisiae.

Arguably, the petite mutation should be of little consequence in brewery fermentations. Afterall, respiration - as noted elsewhere (see Section 3.4.2) - plays no direct role in brewery fermentation. Despite this, there is substantial evidence that petite mutants are undesirable. It is increasingly clear that the large primitive mitochondria ('promitochondria') found in anaerobic cells (Visser et al., 1995) are vital to cell metabolism during anaerobiosis (for a review see O'Connor-Cox et al., 1996). For example numerous enzymes are located in mitochondria, notably some of those involved in the citric acid cycle, sterol biosynthesis and amino acid synthesis (Visser et al., 1994). Further compelling evidence for the contribution of promitochondria to anaerobic metabolism of yeast comes from studies with bongkrekic acid, which inhibits ATP transport into mitochondria. Under anaerobic conditions, this is necessary as ATP cannot be supplied via respiration but is required to fuel transport of metabolites in and out of the mitochondria. On addition of bongkrekic acid, growth rate is reduced in anaerobic cultures, both continuous (Visser et al., 1994) and batch cultures (O'Connor-Cox et al., 1993). Other work has shown that a functional mitochondrial genome is important in tolerance to ethanol, with petites being more sensitive to the growth inhibitory effects of ethanol (Brown et al., 1984; Hutter & Oliver, 1998). Furthermore, exposure to high concentrations of ethanol (e.g. 18%,v/ v) can induce the formation of petites. Chi and Arneborg (1999) have extended this theme to show a relationship between petite formation and ethanol tolerance and membrane lipid composition. More ethanol-tolerant strains are less prone to the formation of petites. Compared to ethanol-sensitive strains, ethanol-tolerant yeasts have comparatively higher levels of phosphatidlycholine, ergosterol and long-chain unsaturated fatty acids. These differences, Chi and Arneborg (1999) argue, would reduce membrane fluidity, which in turn would protect against the fluidising effects of ethanol on membranes.

The applied significance of mitochondria is borne out by more applied work in fermentations with petites where beer quality and process performance are markedly affected (Silhankova et al., 1970b; Morrison & Suggett, 1983; Ernandes et al., 1993). Elevated levels of petites result in sluggish fermentation rates, reduced yeast growth and changes in flocculation. Corresponding beers have a skewed flavour with elevated levels of diacetyl and higher alcohols together with reduced concentrations of ethyl acetate.

4.3.3 Cell cycle

Like the rest of this chapter, the focus here is on polyploid brewing yeast strains and vegetative asexual reproduction. Sexual reproduction (mating or conjugation) and sporulation, which apply respectively to haploid and diploid cells, are not considered here. The interested reader is directed to various reviews for a description of these processes (Byers, 1981; Pringle & Hartwell, 1981; Wheals, 1987; Cox, 1995; Sprague, 1995).

Whether laboratory, baking or brewing strains, S. cerevisiae undergoes asexual reproduction via an asymmetric form of cell division (see Section 4.3.3.1) called 'budding'. One cell gives rise to two 'daughter' cells that are genetically identical to the original 'mother' cell. This simple relationship is central to the success of the yeast cell, be it industrially in a brewery fermenter, in a laboratory culture or environmentally on the surface of a grape. Of course, cell division is an ongoing process in that the progeny of cell division - the 'virgin' daughter cells - themselves divide becoming mother cells and so on (Fig. 4.23 ). However, cell division is not a linear, never-ending process. Cell division slows or stops ('arrests') when growth nutrients become limiting (see Section 4.3.3.3 Stationary phase), when cells age and become senescent (see Section 4.3.3.4 Ageing) or when cells die (see Section 4.3.3.5 Death).

4.3.3.1 Cell division. Other than a rather academic review by Duffus (1971), the yeast cell cycle has attracted little interest in the brewing literature. However, recent work on cell ageing (Section 4.3.3.4) in brewing yeast may well renew interest in this fundamentally important area.

The biology of the cell cycle of the budding yeast S. cerevisiae and, particularly, the

for General Microbiology).

fission yeast Schizosaccharomyces pombe, continue to provide rich pickings for cytologists, morphologists and geneticists. Consequently, an astonishing amount of information has been - and continues to be - published about the yeast cell cycle. As ever, the driver for this research is that S. cerevisiae is a model system for understanding eukaryote cell division. As noted by Jazwinski (1990), 'it has become abundantly clear that virtually any phenomenon involving individual eukaryotic cells can be profitably studied in yeast'. In this case, the key elements of the cell cycle appear to be highly conserved throughout evolution, so what happens in yeast is likely to find application in higher eukaryotes from plants to man. Here, coverage is limited to an overview of the fundamental steps of this process. A comprehensive review can be found by Wheals (1987) in The Yeasts.

The cell cycle is divided into a series of events or 'landmarks' that describe this remarkable process that creates a new genetically identical cell. Although inevitably an overview of this sort cannot do justice to the wonders of the cell cycle, the diagram of Harold (1995) (Fig. 4.23) succinctly captures the cycle landmarks together with cell morphogenesis. Normally, unless synchronised in some way, an actively growing population will contain cells at all stages of this cycle. The unbudded gap period or 'GI' phase represents the time prior to starting a new round of cell division. This commences at a point toward the end of GI termed 'START', a decision point that represents a collection of cellular events necessary for cell division. Although complex (see review by Sherlock and Rosamond, 1993), one of the questions posed before committing to the cell cycle is whether the environment contains adequate levels of nutrients. If there is, the next step is that of DNA synthesis ('S' phase), during which time the daughter cell 'bud' starts to develop. The post-synthetic G2 phase coincides

Chitin ring, neck filaments

Localized exocytosis and wall expansion

Chitin ring, neck filaments

Vesicle translocation and targeting

Fig. 4.24 Yeast cell morphogenesis (from Harold, 1995) (reproduced with permission from the Society for General Microbiology).

Vesicle translocation and targeting

Fig. 4.24 Yeast cell morphogenesis (from Harold, 1995) (reproduced with permission from the Society for General Microbiology).

with significant bud growth, which then culminates in mitosis ('M' phase), and nuclear division. The final steps of the cell cycle are (i) cytokinesis where daughter and mother cells are separated by septa and (ii) cell separation. The sites of attachment on mother and daughter cells are seen (under the microscope) as, respectively, bud and birth scars.

The physiology of bud growth continues to excite much interest. Not surprisingly, this complex sequence of events (see Fig. 4.24) requires numerous checkpoint controls. Even the site of a new bud is controlled, being either axial (haploid) or bipolar (diploid) (Chant, 1994). As noted by Harold (1995) 'expansion of the wall is patterned in space and time: to begin with, growth takes place chiefly at the tip of the forming bud, then it becomes uniformly distributed, and eventually expansion halts'. These events are driven through secretory vesicles that deliver the appropriate enzymes and precursors for wall growth. Intriguingly, there is a view that 'cables' of actin microfilaments track through the cytoplasm into the bud, both to facilitate and to target the vesicles (Harold, 1995). The chitin (Section 4.4.3.3) ring elaborated early bud development remains after separation as a bud scar. These are usefully visualised using fluorochromes such as primulin or Calcofluor (Streiblova, 1988).

Not surprisingly given its complexity, cell division is 'resource hungry'! According to Wheals (1987) there are about 70 cdc (cell-division cycle) mutations. The yeast genome project extends this commitment and allocates about 14% of the 3167 functional ORFs (see Fig. 4.17 ) to 'cell growth, cell division and DNA synthesis' (Mewes et al., 1997). Much of what is known about the cell cycle, although undeniably fascinating and worthy of discussion, is beyond the scope of this book. However, the role of yeast cell size in the cell cycle warrants comment.

At the end of the cell cycle, on separation, daughter cells are usually smaller than their mother cells (Fig. 4.25). Work in the late 1970s (Johnston et al., 1977) showed that the daughter cell had to increase its cell volume before it was capable of budding.

Fig. 4.25 The role of yeast cell size in the cell division cycle.

Fig. 4.25 The role of yeast cell size in the cell division cycle.

Accordingly, the G1 period of the daughter cell was longer than the mother cell whose size was more appropriate for entering the cell cycle. This asymmetric division has been crystallised in terms of the need to achieve a 'critical cell size' before a new round of division can start. Typically, there is no definitive 'critical cell size'! The work of Tyson and colleagues (Tyson et al., 1979) showed the mean cell size in aerobic cultures to increase 2.5 fold with increasing growth rate (doubling time from 450 to 75 minutes). Correspondingly, the 'critical size' varied with growth rate, being (at slow growth rates) almost half the volume of cells dividing quickly.

As something of an aside, mutants have been isolated which have been used to probe the relationship between cell size and division. These mutants - wee (S. pombe) (Nurse, 1975) and whi (S. cerevisiae) (Sudbery et al., 1980) - start budding at only half the cell size of wild type cells. Consequently, because of their small size, the mutants were named wee and whi. However, this is only part of the story. The wee designation also reflects on the Scottish heritage of this mutant (the work was done in Edinburgh) whereas whi stems from the whisky that was consumed in celebration of obtaining the mutant (Andrew Goodey, personal communication)!

4.3.3.2 Cell division and brewery fermentation. Cell cycle events in brewery fermentations have received little attention. This is a little surprising given the widespread and high-profile interest in the cell cycle over the last 25 years. Perhaps, in comparison to the genetic improvement of brewing yeast strains, it is seen to offer little obvious added value. Another likely explanation is the comparative technical complexity of brewery fermentations. This is a two-edged sword! On the one hand, there is the need to replicate as precisely as possible a brewery fermentation in a laboratory culture without, on the other hand, so many compromises that the conclusions cannot be extrapolated to the real world. This, together with the ever-changing cultural environment (notably oxygen) in brewery fermentation, is perhaps sufficient to dissuade substantial work in this area.

This is a pity as even laboratory 'brewery' fermentations can provide fascinating insights into cell cycle events. By definition having been recovered from a previous fermentation, pitching yeast is in stationary phase (see Section 4.3.3.3). On pitching into a single batch of air-saturated wort, the critical cell cycle events on which successful fermentation depends can be seen to unravel step by step (Quain, 1988). The shift from the effectively 'starved' conditions of yeast storage to fresh wort results in early events being seemingly synchronised (Fig. 4.26). The initially unbudded

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