Belgian Co-ordinated Collections of Micro-organisms (Agro)industrial fungi & yeasts collection

12301 Parklawn Drive, Rockville, MD 20852, USA Tel: (+ 1) 301 881 2600 Fax: (+ 1) 301 816 4365 www. atcc. org/general. html

Yeast Division, Julianalaan, 2628 BD Delft, Netherlands Tel: ( + 31) 15 278 7466 Fax: (+31) 15 278 2355 www.cbs.knaw.nl/www/cbshome.html

Institute of Food Research, Research Park, Colney, Norwich,

NR4 7UA, UK Tel: ( + 44) 1603 255 000 Fax: ( + 44) 1603 458 414 www.ifrn.bbsrc.ac.uk/NCYC/Brew.html

VTT Biotechnology and Food Research, PO Box 1500, FIN-02044

VTT, Finland Tel: ( + 358) 9 456 5133 Fax: (+358) 9 455 2103 www.vtt.fi/bel/mib/prosmik/ekokelm.htm

BCCM™/MUCL, Mycothèque de l'Université Catholique de Louvain, Place Croix du Sud 3, B-1348, Louvain-la-Neeuve, Belgium Tel: ( + 32) 1047 3742 Fax: (+32) 1045 1501 http.www.belspo.be/bccm/mucl.htm

4.2.4 Differentiation - an introduction

Most - if not all breweries - need to identify with some certainty their production yeasts. The need is usually proportional to the complexity of brewery operations. For example, breweries who use many brewing yeasts (or mixed strains), or who franchise brew, or regularly propagate new lines of yeast will periodically require the assurance of strain identity. Unfortunately providing a brewing strain with a unique 'fingerprint' via some laboratory test has for many years frustrated microbiologists the world over. Strains that are so easily characterised in fermenter or in final product are difficult - seemingly impossible - to identify unambiguously via laboratory methods. Consequently, a multitude of approaches has, over the years, been developed to differentiate brewing production yeast strains. Many of the older methods, whilst not providing definitive identity, have found routine application in differentiation between strains. For many this was and is sufficient! However, where a more precise identification is required, the new approaches described in recent years at last offer the potential of unambiguously identifying and differentiating brewing yeasts.

For convenience, the methods for the differentiation of brewing yeast strains can be divided into 'traditional' and 'modern' methods. Traditional methods are generally based on what might be described as conventional brewing microbiology. Conversely, the modern methods owe little to brewing microbiology but to developments elsewhere. For example, the dramatic developments in the molecular biology of yeast have allowed the genetic differences between yeasts to be probed and 'fingerprinted'. A second fledgling route to differentiate brewing yeasts involves the application of sophisticated instrumentation/data handling which can be used to probe differences in cell composition between strains.

Unfortunately, despite much fevered activity, the take-up of the modern approaches to differentiation has been relatively poor. Such approaches still remain in the domain of 'research' with only a few reported applications in routine brewing QA (Wright et al., 1994; Quain, 1995; Casey, 1996). There are at least three reasons why traditional methods of differentiation have not been superseded by modern methods: (i) the capital and revenue costs of implementing modern methods are high, (ii) the required 'skill sets' are not normally found in brewery QA laboratories and (iii) the overall cost/risk/benefit analysis is not attractive.

Developments in the differentiation of brewing yeast strains have been subject to review in recent years (Quain, 1986; Casey et al., 1990; Meaden, 1990; Gutteridge & Priest, 1996; Russell & Dowhanick, 1996).

4.2.5 'Traditional' methods

The 'traditional' methods are tried and trusted approaches that offer some degree of strain identification and differentiation. They are not definitive or unequivocal. However these methods are generally simple, affordable and are 'fit for the purpose'. Such methods focus on the morphological, physiological and biochemical differences between strains of S. cerevisiae. Indeed some of these approaches exploit taxonomic differences whereas others relate more directly to brewing performance. Plate tests. Isolation of micro-organisms by growth on solid agar plates remains of pivotal importance in microbiology (see Section Despite this, our understanding of how yeast cells grow on agar plates has been scanty. This is changing with a series of publications that have begun to shed light on the growth of yeast colonies on solid media. For example a gene (IRR1) has been identified that is required for the formation of yeast colonies on solid media but has no role on growth in liquid culture (Kurlanzka et al. (1999). Elegant work by Meunier and Choder (1999) has quantified colony size such that 'solitary' colonies (an average of five colonies per 9 cm plate) contain greater than 109 cells. Plates with 250 (more typical) and 5000 (heavily loaded) colonies per plate contain 5 x 107 and 106 cells respectively. Surprisingly, the initial logarithmic growth rate resembles that in liquid culture and in 'solitary' colonies growth is biphasic with growth being focused on the periphery of the colony. Although this resembles the classic diauxie of growth on glucose in liquid culture, the second growth phase on plates does not reflect a transition from fermentative to oxidative growth. Consequently, this biphasic growth pattern is observed with cells growing oxidatively on ethanol or with respiratory deficient petite mutants.

Colony morphology and shape is highly organised. Although typically circular, and, in older colonies, pyramidal, there are reports of quite atypical colony shape. For example, pseudohyphal outgrowth of colonies (also implicated in the penetration of human tissues by pathogenic isolates of S. cerevisiae (see Section has been reported under specific conditions of nitrogen starvation (Wright et al., 1993). Perhaps more intriguing is the report (Engelberg et al., 1998) of stalk-like structures from the few yeast cells that survive UV irradiation of a lawn of cells on plates. These stalks can be as long as 3 cm and as wide as 4 mm. On resuspension in liquid and replating, the cells from the stalks grew as conventional colonies. This suggests that the cells in these fascinating structures are not mutants caused by the UV irradiation.

Unlike shaken liquid media, cells on plates are subject to a complexity of nutrient and metabolite gradients. It is interesting to speculate how nutrients are transported to cells at the top of 'multicellular' colonies, which are far removed from the agar surface. One mechanism - passive capillary movement - has been proposed (Meunier & Choder, 1999) although 'more active feeding mechanisms cannot be ruled out'. Indeed, the issue of cell and colony 'communication' has been enlivened by the report that ammonia acts as a signal between individual colonies of Saccharomyces and other yeast genera (Palkova et al., 1997). Colonies pulse ammonia toward neighbouring colonies, which, in turn, reciprocate pulsing ammonia in the return direction. This results in growth inhibition of facing parts of both colonies, which thereby orientate their growth towards areas that minimise the competition for nutrients.

The giant (or 'solitary') colonies have a 'highly organised morphology specific for a particular yeast strain' (Palkova et al., 1997). Indeed in brewing microbiology, the giant colony method is the most traditional of the 'traditional' methods of strain characterisation! Its beginnings can be traced back to 1893 (Hall, 1954) when Lindner reported that up to ten single yeast colonies on thick wort-gelatin plates gave rise to characteristic 'giant colonies' (Fig. 4.5) when incubated for c. three weeks at 18°C. Subsequent embellishments (Hall, 1954; Richards, 1967) added consistency and robustness to the method. Indeed, colonies could be classified into six groups (Richards, 1967) by the detailed analysis of colony appearance. With hindsight, the giant colony method seems a little archaic, the method had many devotees and was used successfully in breweries worldwide for many years. However, despite the method's virtues, the lengthy turnaround time for results is now clearly unacceptable in today's 'real time' world.

Chromogenic media is increasingly finding application in the differentiation and identification of micro-organisms, most notably for the 'rapid' differentiation of coliforms in water. In brewing, the use of colony colour to exploit differences in media acidification for the differentiation of yeasts has been somewhat limited. The most common application is in the use of WLN (see Section agar - a general purpose 'green' media used in the detection of aerobic yeasts and bacteria. Colonies on WLN typically range from light lime green through to dark green. In experienced hands this can offer some differentiation between brewing strains. For example, Lawrence (1983) noted that increasing the concentration in WLN of Bromocresol green exaggerates any differences in colony colour such that closely related brewing yeasts can be differentiated on colony colour and size.

Lager and ale strains of S. cerevisiae can be most effectively differentiated by colony colour after growth in media containing a chromogenic substrate ('X-a-gaF, 5-bromo-4-chloro-3-indoyl-a-D-galactoside) (Tubb & Liljestrom, 1986). Lager yeast expresses a a-galactosidase, which cleaves X-a-gal to form an insoluble blue-green dye. Consequently lager yeast colonies are blue whereas ale colonies remain cream. Unfortunately, for this is a useful and effective method, subsequent health and safety concerns about the solvent used to prepare X-a-gal (N,N'-dimethy-formamide) have relegated the method from routine usage. However, the demise of approach is premature and the identification of an alternative, food-grade solvent, propanediol, has given the method a new lease of life. This modified method can also be used for the 'real time' detection (through the blue coloration) of very low levels of lager yeast contamination in ale yeast slurries (but unfortunately not vice versa) (Wendy Box, unpublished observations).

The antibiotic cycloheximide (or 'actidione') inhibits cytoplasmic protein synthesis in yeasts and other eukaryotes. Resistance in yeasts is associated with a single amino acid (proline to glutamine) modification of the ribosomal protein, L41 (Kawai et al., 1992). In brewing microbiology (see Section, the inclusion of cycloheximide makes growth media selective for bacteria and 'wild' yeasts, as brewing strains are sensitive to low levels of the antibiotic (1 mg/1). There have been reports (Harris & Watson, 1968; Gilliland, 1971a; Lawrence, 1983) that resistance to cycloheximide can be used to differentiate between brewing strains. However, as noted by Gilliland (1971a), differentiation of strains at such low concentrations of cycloheximide (0.060.4 mg/1) requires care (cycloheximide is heat sensitive). Gilliland (1971a) also expressed further reservations about this approach as strains can vary in their resistance to cycloheximide.

A more comprehensive approach has been described (Simpson et al., 1992) which quantifies the inhibition of growth by six dyes and antibiotics (including cycloheximide). Discs impregnated with growth inhibitors are positioned around a ring, which is then placed on microbiological media seeded with a lawn of yeast. The degree of growth inhibition is quantified by measuring the zone of clearing around each inhibitor. Although many brewing strains exhibit the same inhibition 'fingerprint', some can be clearly differentiated. Like many methods, this approach is perhaps more useful for characterising 'wild' Saccharomyces and non-Saccharomyces yeasts. Flocculation tests. Flocculation tests remain one of the most popular approaches to brewery yeast differentiation (see Section 4.4.7 for a detailed review of 'flocculation'). This is hardly surprising as - even in closed vessels - flocculence remains one of the handful of characters used by brewers to describe the process performance of a yeast. However, despite very real differences in flocculence in fermenter, laboratory methods to differentiate strains on the basis of differing flocculence remain based on empirical approaches devised many years ago (Burns, 1941; Gilliland, 1951; Hough, 1957).

The various flocculation tests are described in Table 4.12. Of them, perhaps only the 'Gilliland' method remains in widespread use. This approach was developed some

Fig. 4.5 Examples of giant colonies: (a) ale. (b) lager and (c) wild yeast (all x 10 magnification).

Fig. 4.5 Examples of giant colonies: (a) ale. (b) lager and (c) wild yeast (all x 10 magnification).

years earlier (Donnelly & Hurley, 1996) by J.W. Tullo to monitor the proportion of the two strains used in Guinness fermentations. Indeed, in 1932 these methods revealed that the yeast consisted of four distinct varieties of yeast ranging from non-flocculent or 'dispersed' (Class I) to the heavily flocculent or 'filamentous' chain forming strains (Class IV). The typical fermentation performance of the four classes is shown in Fig. 4.6.

Although these methods (Gilliland, 1951) are inevitably retrospective, they have found routine application for monitoring pitching yeast purity (Gilliland, 1971a). Indeed the Gilliland method is still is use today at Guinness (Donnelly & Hurley, 1996) and in Bass Brewers. In the later case, the Gilliland method is used to track the proportion of two strains used in ale fermentations where one strain is Class I and the other is Class II. Remarkably, the Gilliland classification has even penetrated the very different world of advertising. A poster for 'Draught Bass' in the late 1980s noted that '... our two yeasts - named Gilliland I and II, after the chemist who classified them -have been together now for over 200 years'! Fermentation performance. Although fraught with difficulty, using 'fermentation performance' to characterise brewing yeasts is potentially a most rewarding approach. Not only is a yeast strain presumably identified but a whole host of useful data can be gleaned in terms of process performance and product quality. It is perhaps no surprise that the methods described to assess 'fermentation performance' are fundamentally the same! This reflects the difficulties encountered in attempting to mimic production fermentations in the laboratory. Although stirred

Table 4.12

brewing yeast and fermentation Flocculation tests used to differentiate brewing yeasts.



Burns (1941) • Laboratory wort fermentations.

• Flocculation one of a number of fermentation tests.

• Flocculation assessed in 'own beer' or distilled water.

• Flocculation in sodium acetate-acetic acid buffer (pH 4.6) is of 'cardinal significance'. Clumped yeast is quantified.

Gilliland (1951) • 50 colonies are inoculated into 50 x 5 ml wort cultures.

• After three days the cultures are examined and the appearance of the yeast sediment noted after (i) gently swirling the culture and, where the yeast remains compact, (ii) decanting all but 0.5 ml of the supernatant and then shaking the sediment.

• Four classes can be identified.

• Class I - compact sediment which on resuspending in 0.5 ml is completely homogeneous.

• Class II - compact sediment which on resuspending is distinctly granular.

• Class III - sediment peels away in large flakes/dense round clumps which will not disperse.

• Class IV - sediment consists of loose flakes/clumps of yeast.

Hough (1957) • Yeasts are characterised according to whether they flocculate (or not) under the below defined conditions.

• Washed yeast (0.5 g) is suspended in 10ml calcium chloride (0.1%, w/v).

• One aliquot (0.5ml) is adjusted to pH 3.5 and another to pH 5. Flocculence is assessed.

• Where flocculation is observed at pH 3.5, maltose is added (final concentration 10%, w/v). In some strains maltose disperses flocculant cells.

• Where flocculation is not found at pH 3.5, ethanol is added (final concentration 2%, v/v). Some strains then flocculate.

• Where flocculation is not observed at pH 5, a second strain of S. cerevisiae is added and co-flocculation assessed. Those coflocculating with NCYC 1108 are described as 'type I', those with NCYC 1109 as 'type II' and those failing to flocculate as 'type III'.

Fig. 4.6 The fermentation performance of yeasts from the four Gilliland flocculation classes (redrawn from Gilliland, 1951).


Class 1

Class II

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