Several techniques apart from PCR have been developed to detect specific DNA and rDNA targets. Fluorescence in situ hybridization (FISH) is a method that targets singular or groups of isolates in a community and operates on the basis of using different fluorescently labelled probes that hybridize to specific target regions (e.g. 16S rRNA) in a chemically fixed cell sample, followed by fluorescent microscopy (Bokulich et al., 2012a; Bottari et al., 2006). Based on the fluorescent signals observed, populations of different cells can be assessed. However, this method is limited as to how many unique cell populations it can identify and is thus more suited for targeted-analysis. FISH can also be coupled to Flow Cytometry (FCM), which performs automated cell sorting based on fluorescent signals, which allows for the quantitation of different cells within a population. Together, these methods provide a semi-automated means of acquiring quantitative data within a few days for isolates of interest without the need for excessive pre-processing (e.g. DNA extraction). However, expensive equipment is required and the probes needed are often not commercially available (Bokulich et al., 2012a). Therefore, FISH and FISH-FCM do not lend themselves well to in-house application for the brewery, yet provide interesting data when performed as an out-sourced procedure or in a research setting. Thus far, this methodology has only been well developed for characterization of yeast populations, with few reported studies of application to BSR LAB (Meng et al., 2012; Thelen et al., 2002, 2004). However, recent work involving cider fermentations showed that FCM could distinguish and separate mixed yeast and bacterial cultures based on membrane integrity and esterase activity, and could identify different physiological states resulting from differences in fermentation conditions, thus having interesting implications for beer fermentations (Herrero et al., 2006).
While FISH (and/or FISH-FCM) are targeted- analysis methods, given they are limited by the number of probes that can be used during experimentation, denaturing gradient gel electrophoresis (DGGE) allows for a more robust identification of microbial community members through the 16S rRNA gene, and has been applied to beer-related LAB (Bokulich et al., 2012a; Manzano et al., 2005; Tsuchiya et al., 1994). This method uses universal PCR primers to amplify specific DNA sequences in a community, then separates the amplicons in a polyacrylamide gels containing a gradient of urea and formamide on the basis of differences in GC content (melting temperature), thereby allowing detection of DNA sequence heterogeneity in microbial communities (Bokulich et al., 2012a; Muyzer et al., 1993). Again, this method has limited use within the brewery in that it is technically difficult and requires DNA extraction, and has a detection threshold that is often above the cell concentration found in beer samples (Cocolin et al., 2001). Further, it requires subsequent processing and sequencing steps following the gel separation to produce accurate identification of the bacteria yielding the resolved bands, making it a laborious process fraught with the inherent errors and biases related to PCR amplification and DNA extraction (Bokulich et al., 2012a; Cocolin et al., 2001; de Lipthay et al., 2004).
Another very useful method for assaying microbial community diversity is Terminal Restriction Fragment Length Polymorphism (TRFLP). Universal PCR primers targeting the 16S rRNA gene that have been fluorescently labelled are used to amplify this DNA region from a mixed culture. Amplicons are then purified and in separate reactions, digested by one or more restriction enzymes, followed by capillary electrophoresis. The separation of the fluorescently labelled DNA fragments allows for unique patterns to emerge for a given organism (Bokulich et al., 2012a; Liu et al., 1997). This method is flexible in terms of its ability to provide either high throughput data or more targeted analysis of mixed microbial communities, and is relative easy to use with low cost, making it a more attractive option for routine use in contaminant surveillance within breweries (Bokulich and Mills, 2012a; Bokulich et al., 2012a). Further, this method can be adapted to provide greater resolution for specific BSR LAB targets through modification of the target sequences and restriction enzymes used (Bokulich et al., 2015).