Fig. 8.31 Routine bioluminescence data from rinses post-CiP of keg lines (redrawn from Quain, 1999).

For the same reasons that the technology 'took off in breweries, ATP bioluminescence is increasingly being used in validating the hygienic status of dispense in trade outlets. The immediacy and practicality of such testing clearly lends itself to validating line cleaning and auditing the hygiene of dispense spouts and associated fittings. Given the challenge of improving and maintaining trade hygiene (see Section 8.2.2), it is likely that ATP bioluminescence will eventually have a comparable impact in this arena as it has in the brewery. Relatively early examples of the application of this technology in beer dispense can be found in Storgards (1996), Storgards and Haikara (1996) and Orive i Camprubi (1996). More recent data using the Biotrace Clean-Trace™ technology (Fig. 8.33), has assessed the microbial loading of the internal surface of 50 dispense lines pre- and post-cleaning. Comparison of this RLU data and that reported above for keg line and FV (Figs 8.31, 8.32), suggests that the standards for what is clean are similar.

Fig. 8.32 Routine bioluminescence data from rinses post CiP of FVs lines (redrawn from Quain, 1999).
Line nos

Fig. 8.33 Surface bioluminescence of dispense lines pre- and post-cleaning (unpublished results Gareth Lang, Biotrace Limited).

Product testing. The lower sensitivity and lack of selectivity have hampered the application of bioluminescence in final product testing. This is of little general significance as final product testing is of declining significance as prevention (QA) supersedes inspection (QC) in brewing microbiology. However, the more stringent shelf-life demands of export products, sterile filtered products and occasional need for 'positive release', have driven the development of specific bioluminescence protocols that accelerate conventional microbiological testing (for a review see Thompson & Jones, 1997). To improve sensitivity of detection, these hybrid protocols invariably require sample concentration (by filtration) and incubation, albeit of reduced duration.

Avis and Smith (1989) described an early application of ATP bioluminescence to the testing of unpasteurised products in PET (polyethylene terephthalate). Shortly thereafter Miller and Galston (1989), described a home made protocol that reduced detection times for aerobes from three days to one day and anaerobes from seven to three days. A commercial kit (Bev-Trace™) is now available and finds application in the early positive release of export products. Using this approach, testing for anaerobes is reduced from seven days to three.

Developments. Although a relatively young technology, bioluminescence is evolving rapidly (see Simpson (1999) for a review of recent innovations in ATP testing). There are encouraging signs that manufacturers are moving toward a universal description of bioluminescence in terms of ATP concentration. More rudimentary but important developments include the exploitation of 'caged' ATP (PhotoQuant™) in the calibration of bioluminometers and possible inclusion of ATP testing in the proficiency testing scheme, BAPS (see Section

The development of AutoTrack marked the advent of continuous-flow lumino-metry (CFL) which enables the continuous monitoring (equivalent to 60 measurements per minute) of ATP during a CiP cycle (Brady, 1999). AutoTrack will enable the optimisation and fine-tuning of CiP cycles so that they are effective - both in terms of cleaning and utilities usage. This approach heralded the commercial exploitation of adenylate kinase (AK) which promises to facilitate a step change in the sensitivity of detection. AK technology offers much and exploits the presence of adenylate kinase in microbial cells (both yeast and bacteria). This new approach exploits the presence of AK by measuring the formation of ATP from ADP.

A lot of development effort remains focused on real-time testing of product. One approach (Davies et al., 1995) has been to improve detection sensitivity by exploiting cross-flow filtration to concentrate large volumes (1000 ml) of beer. Although a promising route to real-time testing, the sensitivity (10 bacterial cells per ml of raw beer) requires further improvement. Perhaps AK technology will deliver this step change.

Another approach - the MicroStar-RMDS (rapid microbe detection system) - has been described (Takahashi et al., 1999). This innovative, semi-automated approach has been installed in Sapporo's breweries and offers product testing within 24 hours of sampling. Dowhanick et al. (1995) described a route to truly real-time product testing using a prototype Millipore RMD (rapid micro detection) system. This mix of technologies is reported to detect one to 200 yeast cells in less than 10 minutes although detection of bacteria may be less rapid. This innovative technology couples real-time imaging (CCD camera) of micro-organisms trapped in novel 'hydro-phobically gridded' membranes with fibre optics image intensification and (inevitably) computer data analysis. Although clearly highly promising, Dowhanick et al. (1995) hint that further development of the RMD approach may hinge around an 'economic cost-benefit evaluation'. Other rapid methods. The focus of the 'new microbiology' here and elsewhere, has been on bioluminescence. However, in the early 1980s bioluminescence was one of many putative 'real-time' or 'rapid' technologies that came to the attention of microbiologists. Although bioluminescence won this particular race it is interesting to speculate which of the new crop of methods described below (Section 8.3.4) will come out as the next 'bioluminescence' and change brewing microbiology once more.

With hindsight, the paper by Kilgour and Day at the 1983 EBC recognised the need for change in brewing microbiology. In addition to bioluminescence, they evaluated two other fledgling technologies, DEFT (direct epifluorescent filter technique) and an electrometric method based on growth-related changes in media conductance detected using a 'Malthus' instrument. In passing, it is interesting to note that unlike bioluminescence, these two approaches enabled some degree of identification of the offending micro-organisms. Together these three approaches and to a lesser extent methods such as ELISA (enzyme linked immunosorbent assays) provided the agenda for microbiological research for much of the 1980s and early 1990s.

The 'microcolony' and DEFT methods accelerate detection of micro-organisms by staining them with fluorescent dyes, which can be more readily visualised under a fluorescent microscope. Although frequently described separately, the two methods have much in common. Both involve sample concentration on a membrane, staining and counting the colonies. However DEFT is much quicker in requiring no incubation step prior to analysis whereas the microcolony approach require incubation overnight (aerobes) or for up to three days (anaerobes). In the case of DEFT the membrane is made of polycarbonate and the fluorescent dye is the DNA stain acridine orange. In passing, DEFT and its simpler relative DEM (direct epifluorescent microscopy) are routinely used in visualising and quantifying attached micro-organisms in biofilms (Holah et al., 1988). With the microcolony approach, black membranes are used to improve the visualisation of the optical brighteners (as used in domestic washing powders) which are used to stain the microcolonies.

The implementation of both methods has suffered from their dependency on microscopy. As noted by Storgards et al. (1997) 'microscopic counting is time consuming and tedious, leads to operator fatigue and limits the capacity' of both approaches. Accordingly, both approaches have been automated (see Parker (1989) for an early example) with varying degrees of success and cost! For details, the interested reader should consult Simpson (1991), Russell and Dowhanick (1996) and Storgards et al. (1997).

The other approaches that promised but never quite delivered in brewing microbiology were the impedimetric methods. This electrometric method exploited the changes in the electrical behaviour of the growth medium that occur in response to microbial metabolism. Without going into detail (see Russell & Dowhanick, 1996), microbial growth could be detected by an increase in impedance or decrease in conductivity and capacitance. Although the sensitivity was relatively poor (detection of 10 bacteria ml 1 required 18 hours' incubation), Evans (1982) described how this approach reduced the time required by a 'forcing test' from weeks to days! Kyriakides and Thurston (1989) enthusiastically 'threw the book' at this technology with particular emphasis on the opportunity to replace conventional plate counts. Unfortunately, the conclusions from this work were lukewarm, with the growth medium lacking the required selectivity to support the growth of the desired contaminants.

Although ELISA is described elsewhere (see Section, it is worth touching on the application of immunoassay described by Ziola and colleagues. Using monoclonal antibodies, membrane-filter-based immunofluorescent antibody tests have been worked up for Pediococcus species (Whiting et al., 1992), Pectinatus cerevisiiphilus (Gares et al., 1993) and Lactobacillus species (Whiting et al., 1999). Although the sensitivity is not quite good enough, this technology may yet be fine-tuned to offer a rapid microbiological solution.

8.3.4 Methods - current developments.

The thrust of much of today's microbiological innovation and development mirrors that described elsewhere (Section 4.2.6) for the differentiation and identification of brewing yeasts. Genetic approaches have been grabbed enthusiastically in the hunt for truly rapid detection and identification of beer spoilage organisms. Methods include PCR (see Section and, more recently, ribotyping, which is restricted to bacteria. Although less to the fore, there are signs that phenotypic methods are now entering the fray. Phenotypic methods. Pyrolysis mass spectroscopy (PyMS) and Fourier transform infrared spectroscopy (FT-IR) provide a different spin to rapid microbiology. Both exploit differences in cell composition and both require sophisticated data handling to draw out what are frequently revealing results. Both approaches are genuinely simple and rapid (two minutes for PyMS and 10 seconds for FT-IR) and potentially offer a definitive solution to identification of colonies on plates. Both approaches have found a niche in the differentiation of brewing yeast and accordingly are described elsewhere: Section (PyMS) and Section (FT-IR).

Of the two, PyMS has a higher profile finding particular application in the differentiation of clinically important bacteria (Taylor et al., 1998; Barshick et al., 1999). Preliminary work (Quain, 1999) has used both PyMS and FT-IR to good effect to track the location of environmental isolates of Lactobacillus. As discussed below this work has provided insight into the complexity of the different strains of bacterial species. It is anticipated that the use of these approaches in brewing microbiology will grow. Genotypic methods. The PCR (polymerase chain reaction) approach is now a staple method of molecular biology (for background see Section 4.2.6). For reviews of the application of PCR in brewing microbiology, the interested reader should consult Dowhanick (1995), Russell and Dowhanick (1996) and Storgards et al. (1997). Although PCR has been applied with a vengeance to needs of brewing microbiology (see Table 8.23 for examples), there is a suggestion that the 'bubble has burst' for this technology. Although now more kit-based (and simpler!), the use of PCR in brewing has been frustrated by contamination issues (resulting in the amplification of the wrong bit of DNA) and a variety of the inhibition effects. Most damning of all are the contradictory objectives of simplicity and sensitivity. As succinctly put by Storgards et al. (1997) 'if a simple protocol for sample treatment is used it is not possible, at present, to achieve the required sensitivity'. Although not down and out, PCR requires further effort and commitment to make the transition from the research bench to the brewery QA department.

Table 8.23 Some examples of the use of PCR in the detection and identification of beer spoilage organisms.





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