Population heterogeneity and genetic instability
Whilst the direct relevance of the quiescence story is not clear, it is a further reminder of the heterogeneity of microbial populations. This has not always been recognized. On reflection, the vast majority of studies of yeast physiology and metabolism in the brewing world have been performed at a ‘bulk' level yielding an ‘average' measurement. Indeed, rather than being homogeneous, the dawning realization is that populations of single-strain, pure-culture yeast in fermenter, handling and storage are heterogeneous. This is a generic microbiological principle, such that microorganisms exhibit cellular individuality that ‘even when genetic and environmental differences between cells are reduced as much as possible, single cells differ from each other with respect to gene expression and other phenotypic traits' (Ackerman, 2015). The bottom line is that heterogeneity of genetically uniform populations is considered to confer a selective advantage during stress and in response to fluctuating environments.
As a subject, ‘heterogeneity' has been flagged most notably at the macro scale, where yeast distribution in fermenter is heterogeneous, the extent of which varies with time and yeast strain/flocculence (Boulton et al., 2007). Heterogeneity is increasingly recognized as the size of individual yeast cells varies with chronological age/bud scars (Barker and Smart, 1996), growth rate (Johnston et al., 1979) and ploidy (Galitski et al., 1999). Better recognition of phenotypic heterogeneity has come through cell sorting and flow cytometry (for a review see Davey and Kell, 1996), which has demonstrated heterogeneity in brewing yeast for internal pH (Imai and Ohno, 1995). Heterogeneity of cellular glycogen content during fermentation has been shown with ‘individual cell spectroscopy' (Cahill et al., 2000).
Another source of phenotypic heterogeneity is genetic instability. This reflects the pressures of adaptive evolution ‘where populations must secure a margin of genome variability that allows for the adjustment to new environmental conditions' (Skoneczna et al., 2015). Instability is ‘a given' occurring via a variety of routes and driven by a number of factors. Whatever their shape, mutations are spontaneous during DNA replication and ‘if a mutation is beneficial in the environment, allowing the cell and its descendants to proliferate more rapidly, that lineage will begin to increase in relative abundance in the population' (Gresham, 2015). As populations evolve, lineages with beneficial mutations increase at the expense of cells without such mutations. Less predictably, lineages with greater ‘fitness' then predominate, winning out and over time dominating the population. The hitherto difficulty in estimating the extent and complexity of spontaneous mutation has underplayed its importance, scale and significance (Zhu et al., 2014).
In brewing, genetic instability has had a somewhat chequered history, which has undermined widespread acceptance. However, without (typically) being mentioned explicitly, it is one of the drivers that has led to the management of the number of yeast cycles or generations that are replenished through the yeast supply process. Early sightings of instability stem from a series of publications from Guinness between 1963 and 1996 (for details see Boulton and Quain, 2001) that detailed ‘spontaneous' changes in maltotriose utilization and marked switches in flocculence. Whether this reflects a benefit in terms of fitness or simply process visibility and detection is not clear. A landmark paper (Casey, 1996), using the then new technique of karyotyping reported changes in the fingerprints of production lager yeast sampled and stored between 1958 and 1985. More specifically, two publications from different Japanese brewers reported the detection of flocculence changes during serial repitching in production (Jibiki et al., 2001; Sato et al., 2001). Another report (Quain, 2006a) detailed the tortuous journey from the observation of cropped yeast with enhanced floc- culence, demonstration of genetic change through DNA fingerprinting, the isolation of seemingly the same genetic variant some 7 years later and in a brewery more than 400 km away, and culminating in the demonstration that both variants (of the same strain) carried an additional copy of chromosome VII (Table 3.6). Conversely, Powell and Diacetis (2007) - in an extended production-scale study with an ale and lager yeast for, respectively, 98 and 135 generations - showed no genetic changes or changes in fermentation performance.
Chromosomal loss (or addition) is one of a number or routes that are associated with genetic instability in aneuploid and polyploid yeasts. There are different perspectives whether this results in reduced fitness (Storchova, 2014) or ‘large fitness gains' which ‘accelerate evolutionary adaptation' (Selmecki et al., 2015). Although beyond the scope of this chapter, it is tempting to build on the anecdotal preponderance of lager strains exhibiting genetic instability as being a consequence of the hybrid genome as well as aneuploidy. Of course this may reflect the relatively young evolutionary ‘age' of lager yeasts or the enhanced visibility through the dominance of lager-type beer production volumes over ale.
Less heralded but long recognized as part of the ‘genetic variants' story is the petite mutation. First identified in France by Ephrussi et al. (1949), and named to reflect the atypically phenotypic small size of colonies on agar plates. Petites are respiratory deficient as a consequence of alterations or part deletion of mitochondrial DNA. Such Rho- or p- mutants grow more slowly on fermentable sugars and are unable to grow aerobically on nonfermentable carbon sources (e.g. glycerol, ethanol etc.) Mutants without mitochondrial DNA (p0), although rare, have been found in cropped yeast (Lawrence et al., 2013) but are often lethal in cells growing aerobically. Deletion or rearrangement of mitochondrial DNA (p-) impacts on the genes for transfer and ribosomal RNAs, thus blocking the synthesis of proteins in mitochondria. Further, the mitochondria contribute to the biosynthesis of haem, amino acids, nucleotides and fatty acid metabolism.
Generically, the petite mutation is considered to be ‘spontaneous' and ‘natural' and is present in brewery fermentations at a frequency of ca. 0.1-1 to 6% or more (Silhankova et al., 1970; Smart, 2007; Lawrence et al., 2013) of the yeast cell population. Strain susceptibility is (as always) a variable that is associated with many and various causes, including ‘stress' (e.g. oxidative), cone storage and serial repitching (Lawrence et al., 2013). The frequency of petites is estimated ‘to be 106-108 times more frequent than other mutational types' (Silhankova et al., 1970).
In passing, another spontaneous (and stable) mutation of brewing yeasts results in glycogen deficiency at a frequency of < 0.8%. Growth of these
Table 3.6 Unravelling genetic change in a production yeast
mutants was linked with elevated levels of petites (2-9%) (Chester, 1967) and conversely analysis of ‘normal' petites were found to have ‘appreciably less glycogen than those of wild-type yeast (Chester, 1968).
Whatever the spontaneous mutation, it is salutary to reflect that a frequency of 1%, pitching yeast at 15 x 106/ml contains a substantial petite population of 1.5 x 105/ml. Whilst rare, there are reports that the frequency of petites can be much higher. For example, reported issues in multisite brand matching were associated with a brewery where ca 4% of yeast harvested from fermenters and up to 50% of stored yeast was respiratory deficient (Morrison and Suggett, 1983). Beers from the problem brewery were characterized by elevated diacetyl and reduced levels of esters together with reduced yeast viability. The issue was seemingly associated with protracted yeast storage (4-7 days at 4°C) that, when replaced with 24 hours at 4°C, reduced the occurrence of petites to 1% or less. Whilst an extreme and somewhat surprising demonstration from some 40 years ago - which sadly received no subsequent elaboration - petites remain a concern during brewery yeast processing.
Today the generic threat of petites are arguably a greater concern, as they are much less likely to be detected through testing or established through problem-solving or root-cause analysis. This sad state of affairs reflects reduced technical resources, a diluted knowledge base and a lack of appreciation of what is a relatively niche element of brewery yeast performance. Indeed, our understanding of petites in brewery fermentations and handling is something of a ‘grey' area as - under anaerobic, catabolite-repressed conditions - respiratory capability is not possible and the organelle is cytologically underdeveloped (and often termed ‘promitochondria'). However, despite this, petites with compromised mitochondrial function and performance are not fit for purpose. This is intriguing and is counter to the aforementioned selection of genetic variants in terms of enhanced fitness.
As with chromosomal instability, new insights from applied yeast genomics pose new questions as to the formation of petites in the two distinct lineages of S. pastorianus. These investigations build on the observation that the Group II ‘strain generated a lower amount of respiratory deficient ‘petite' cells' (Walther et al., 2014). Further, with the mitochondria of lager yeast being derived from the S. eubayanus parent (Dunn and Sherlock, 2008) - and not S. cerevisiae - there is a case for a fundamental difference in frequency of petite mutations in ale and lager yeasts.