Many microbes—including bacteria and fungi—can affect the food and drink we consume, for better and for worse. Some spoil food, making it less tasty or even harmful to health. However, microbes can also be important ingredients: for example, yeast ferments malted barley sugars to make the alcohol and flavor of beer.

Nowadays, many beers are made under carefully controlled conditions, where the only microbes in the beer should be the strain of yeast added to the barley sugars. A more traditional ‘coolship’ method can be used to make sour beers; the barley sugars cool in an open-topped vessel and are fermented by the yeast and bacteria found naturally on the raw ingredients and in the surrounding environment.

Relatively little was known about how microbes spread around and adapt to living inside buildings. Now, Bokulich et al. have used a range of molecular and statistical techniques to examine how bacteria and fungi are dispersed throughout a North American brewery that produces beer using both conventional and coolship brewing techniques. Most of the microbes found in the building originated from the raw ingredients used to make the beer, with different parts of the brewery containing different species. Over the course of a year, some species spread to new parts of the building; a statistical method predicted the sources of these microbes, and revealed some key areas and features of the brewery that affect microbial transfer.

Bokulich et al. also looked at the spread of genes that enable their bacterial hosts to spoil beer, including those that protect bacteria from the antimicrobial action of the hops that flavor many beers. Lactic acid bacteria are the main cause of beer spoilage and so are usually to be avoided in breweries, but are also a normal ingredient in sour beer. In the brewery Bokulich et al. investigated, beer-spoilage and hop-resistance genes were found throughout the brewery, even in areas not used to produce sour beer. However, little beer spoilage occurred.

The techniques used by Bokulich et al. to track the spread of microbes and their detrimental genes could be used in the future to understand how microbes adapt to other indoor environments. Indeed, Bokulich et al. suggest that breweries could be used as models to safely understand the factors that influence microbial movement in any food-production facility as well as other building environments.

Microbial activity is an inherent feature of food-processing systems, influencing the quality and healthfulness of foods for human consumption. Like other indoor environments, the building materials, substrates, and physiochemical conditions encountered by microbes in food-processing facilities differ dramatically from the outdoor conditions to which microbial life evolved (Kelley and Gilbert, 2013). How this has impacted the adaptation and ecological assemblages of microbes in food systems is largely unexplored. While the microbial communities of other indoor environments, such as homes (Lax et al., 2014), office buildings (Kembel et al., 2014), and hospitals (Bokulich et al., 2013a), can influence the health of their inhabitants, food-production facilities represent a unique type of built environment wherein microbial activities are intimately tied to product-quality outcomes. Thus, the sensory and safety effects of microbial growth in food-production streams have a much broader impact, such that consumer enjoyment and health can be linked to the hygiene and processing decisions at the food facility. Furthermore, both beneficial and detrimental microbial activities are well defined in foods, making these systems a useful model for exploring microbial ecosystem dynamics beyond the scope of contaminant mitigation.

Fermented foods, including beer, have the additional distinction that microbial activity is central to their production, responsible for necessary transformations as well as product spoilage (Bokulich et al., 2012a; Bokulich and Bamforth, 2013). Most modern food-fermentation practices occur under relatively aseptic conditions, employing pure starter cultures in their production, and environmental contamination represents a prevalent threat to product integrity (Bokulich and Bamforth, 2013). Beer is typically produced through fermentation of malted barley sugars (wort) to alcohol by pure strains of Saccharomyces cerevisiae or Saccharomyces pastorianus, and any additional organisms—including cross-contamination from different Saccharomyces strains used in the same facility—are considered contaminants (Bokulich and Bamforth, 2013). In traditional fermentation production practices, conversely, microbial communities introduced from raw materials and processing environments are central to the fermentation process. For example, in the production of coolship (lambic-style) beers, a type of sour beer, no starter cultures are added to the wort; instead, boiled wort is allowed to cool overnight in a shallow, open-top vessel known as a coolship where indigenous microbiota are presumably introduced to the product, initiating fermentation (Van Oevelen et al., 1977; Bokulich et al., 2012b; Spitaels et al., 2014). The unique succession of indigenous microbiota in coolship ale fermentations sets them apart chemically and sensorially from other beers (Van Oevelen et al., 1976; Spaepen et al., 1978), making coolship breweries a particularly interesting system for tracking microbial populations in food-processing environments.

Microbial spoilage genes are also well defined in brewing environments and can be studied in situ without representing a direct threat to public health. Beer is protected from wholesale microbial contamination through its alcoholic, low-pH, and antimicrobial properties, and modern sanitary technologies and practices minimize the threat of spoilage organisms that have evolved specifically to grow in beer (Bokulich and Bamforth, 2013). Nevertheless, biofilms and other environmental reservoirs remain potential sources of microbial contamination in breweries (Timke et al., 2005; Storgards et al., 2006; Timke et al., 2008; Mamvura et al., 2011; Matoulkova et al., 2012). As wort is boiled prior to fermentation, the primary reservoir for spoilage microorganisms in beer production is the brewery environment. Lactic acid bacteria (LAB) are of particular concern, as some members of this clade resist hop antimicrobial compounds, enabling growth and spoilage of beer through acidification, hazes, and off-flavors (Suzuki et al., 2006; Bokulich and Bamforth, 2013). Hop iso-alpha-acids, the primary antimicrobial compounds in beer, kill most Gram-positive bacteria by acting as ionophores, dissipating proton-motive force across cell walls (Simpson and Fernandez, 1992, 1994; Simpson, 1993a), and via oxidative stress (Behr and Vogel, 2010). Very few resistance mechanisms have been proposed (Suzuki et al., 2006), primarily being the multi-drug transporters horA (Sami et al., 1997; Sakamoto et al., 2001) and horC (Suzuki et al., 2005; Iijima et al., 2006), the transcriptional regulator of horC, horB (Suzuki et al., 2005; Iijima et al., 2006), and hitA, a divalent cation transporter (Hayashi et al., 2001) that imports manganese into the cell. These genes are all located on plasmids and transmit via horizontal gene transfer (Suzuki et al., 2005, 2006). The frequency and transmission of these and other beer-spoilage genes within processing environments—or any other reservoir for spoilage microbes—have yet to be tested.

Here, we employ mixed molecular approaches (Bokulich and Mills, 2012b) and a Bayesian modeling method to interrogate the seasonal sources, reservoirs, and transmission of bacteria, fungi, and beer-spoilage genes within a North American brewery over the course of one year. Given the inherent role of environmental microbiota in conducting coolship ale fermentations, this serves as a model system for studying mechanisms for microbial transfer within food-processing systems. Information on population and gene flow extends to other food-processing systems where environmental microbiota are involved positively or negatively in the production, stability, and safety of human nutrition.

The microbial ecosystems of breweries are inherently linked to successful product outcomes, impacting the qualities and healthfulness of beer. In conventional brewing, any microbe not intentionally inoculated is a contaminant (Bokulich and Bamforth, 2013). Conversely, in coolship ale and other sour beer brewing, adventitious microbiota are integral to the process. Thus, we studied microbial ecosystem dynamics in a brewery that produces conventional, sour, and coolship beers, in order to observe food–ecosystem interactions from a dual perspective. First, how do microbial community assemblages and spoilage-gene frequencies change over time with respect to contamination issues in a conventional brewery? Second, how do these same elements interact with the production of sour beers?

Production environment microbiota are driven by substrate contact

Short-amplicon marker-gene sequencing was employed to survey the bacterial and fungal consortia inhabiting the entire brewery environment. A total of 501 samples were collected during three seasons, representing the main processing surfaces and equipment used throughout the brewing process (Figure 1). Beta-diversity (between sample) comparisons provide useful assessments of the taxonomic similarity between different sites. Bray–Curtis dissimilarity of complete microbial profiles reveals that many samples cluster by processing room and substrate type regardless of season (Figures 2–3). Thus, fermenter samples cluster, associated with Bacillaceae; cellar production areas, associated with Micrococcaceae (including the beer-spoiling genera Kocuria and Micrococcus); wort, malt, and hotside (wort-preparation) surfaces, associated with Enterobacteriaceae, Leuconostocaceae, Candida santamariae, Pichia, and Rhodotorula; barrel-room floor samples, associated with Staphylococcaceae and Carnobacteriaceae; and beer samples, associated with Lactobacillaceae and Enterobacteriaceae. Barrels cluster, associated with Aspergillus, Eurotium, and Penicillium; coolship and barrel-room samples with Cryptococcus and Cladosporium. These taxonomic trends each demonstrate significant site associations (Kruskal–Wallis Bonferroni-corrected p < 0.05). S. cerevisiae was common throughout the brewery, but especially in the fermentation cellar. LAB and acetic acid bacteria were found sporadically at different sites and times, including on and near packaging equipment and fermenters inoculated with LAB (Figure 3).

Figure 1

Download asset Open asset

Figure 2

Download asset Open asset

Figure 3

Download asset Open asset

To further examine this relationship, the Bayesian technique sourcetracker (Knights et al., 2010) was used to test whether raw substrates may act as sources for the microbial consortia of brewery surfaces. This tool predicts the relative proportion of contamination in sink samples (in this case brewery surfaces) from microbial sources (raw ingredients and extraneous sources). Raw substrates (grain, hops, yeast, beer) and extraneous sources (human skin, outdoor air, soil, saliva, feces, freshwater, ocean water) from previous studies (Caporaso et al., 2011; Bowers et al., 2012; Caporaso et al., 2012; Bowers et al., 2013) were tested as microbial sources. Results reveal distinct patterns of contamination across seasons (Figure 4). Grains were predicted as the largest microbial contributor to hotside areas, almost all surfaces in the coolship room, and areas of the cellar away from the main fermentation area. Hops were predicted as the major contributor to cellar fermentation areas and fermentation equipment. Yeast was predicted as the highest contributor to fermenters, conditioning tanks, and packaging equipment. Beer was predicted to be a common contaminant around fermenters and barrels within the cellar. Skin was a minor contributor to some surfaces in the cellar. Other surfaces, including most barrels and barrel-room surfaces, were most influenced by unknown sources. Outdoor air, soil, saliva, feces, freshwater, and ocean water were predicted to provide only a very low level of contamination (<0.001 relative abundance). These results suggest that raw substrates are the main contaminant sources within the brewery, compared to extraneous sources. However, it is important to note that these predictions do not indicate a causative role for contamination. These predicted source/sink relationships could alternatively suggest that both are actually contaminated by another, untested source (e.g, fruit flies or other vectors could transfer microbes between these and other surfaces within the brewery). Nevertheless, these predictions highlight potential sources of contamination, or at least shared microbial transmission patterns, between substrates and equipment/surfaces within the brewery.

Figure 4

Download asset Open asset

These findings suggest that substrate contact drives the microbial communities of different brewery surfaces. Consequently, raw materials (grain, hops, yeast) may be important vectors for any spoilage organisms encountered in the facility. This substrate-centric community structuring has also been observed in other food facilities (Bokulich and Mills, 2013a; Bokulich et al., 2013b, 2014). However, the brewery surface niches clustered throughout time with only subtle changes across seasons, unlike wineries, which were significantly impacted by seasonal factors (Bokulich et al., 2013b). This reflects the divergent production schemes of these two foods: whereas most beers are produced year-round, wine is an inherently seasonal product. The raw materials for beer are stored and processed throughout the year in a continual production schedule, creating a stable environment on equipment surfaces that interact with these same raw substrates on a daily basis. Instead, brewery environments may function more like cheese-making environments, where facility-specific ‘house’ microbial communities form on equipment surfaces in response to idiosyncrasies in the indoor environment (Bokulich and Mills, 2013a). Such a possibility would have interesting implications for sour beer breweries, and comparative studies of lambic and coolship breweries could offer insight into the brewery-specific flavor profiles displayed in these beers.

Physical mapping illustrates microbial dispersal in processing facility

By mapping surface samples to their physical location within the brewery, a spatial model emerges of microbial dispersion across brewery surfaces over time (Figures 5–7). Several populations visibly spread with time, most prominently S. cerevisiae, which progressed from dominating fermentation and packaging areas only in Autumn to being the most abundant fungus detected across the brewery in Spring and Summer. Likewise, C. santamariae displayed high abundance on hotside surfaces in Autumn but became more abundant and spread to nearby sections of the cellar in Spring and Summer (Figure 5). As building measurements were not taken during these times, the factors driving these changes cannot be assessed with these data, but warming temperatures and increasing humidity in the Spring and Summer months could be associated phenomena. Other taxa demonstrated more localized patterns of dispersion, such as Micrococcus and Kocuria, which appeared to spread the most around floors and other surfaces in cellar, barrel room, and packaging areas (Figure 6). Here, regular contact with beer runoff diluted with rinse water may support growth of these populations, which are associated with spoilage in low-alcohol beers (Back, 1981). High populations of Acetobacter and Lactobacillus were found more disparately, and specifically in areas where high volumes of wort and beer may be encountered: on conveyor belts and floors below packaging areas, hotside and cellar area sinks, and on sample ports for isolated fermenters and kegs (Figure 6). On these sites microbial communities can contact undiluted beer, for example, drips in the basin below the packaging-line belt, wort and beer collected for specific gravity (sugar concentration) measurements then dumped in sinks, and sampling ports on fermenters. This follows the known behavior of these bacteria, which can spoil undiluted, higher-alcohol beers under appropriate conditions (Bokulich and Bamforth, 2013).

Figure 5

Download asset Open asset

Figure 6

Download asset Open asset

Figure 7

Download asset Open asset

These results illustrate the progressive dissemination of microbes in space and time within a functioning food-processing environment. Microbes not only associate with specific substrates, they exhibit patterns of dispersion within confined regions of the brewery. This yields useful insight into the transmission behavior of these organisms, and especially taxa associated with beer spoilage, through physical space. Microbial confinement within discrete zones suggests that physical barriers (e.g., walls) and physiochemical conditions (e.g., humidity and temperature control) can staunch the spread of some microbes. Non-production surfaces that encounter beer and/or waste streams, such as floors, sinks, and grain-handling equipment, are typically only cleaned by hose and may accumulate substrates for supporting microbial growth and survival. Aerosols and splashing occur regularly in a production brewery (indeed, hosing down floors is one obvious cause), increasing the likelihood for microbial spread and colonization from these surfaces onto other surfaces, including equipment. Human and insect traffic can also increase the rate of dispersal between these surfaces (e.g., from one room to another). Physical partitions appear to inhibit this passage as observed in this brewery and, where possible, may reduce the probability for contamination. This is particularly important in packaging areas, where microbial dispersal can directly contaminate sanitary surfaces and open packages, increasing the likelihood of spoilage in finished product. Separating packaging rooms from cellar and hotside operations and using local partitions, such as blast shields, may help protect packaging surfaces and beer from such routes of contamination. Further investigations into the relationship between building materials, spatial area, indoor conditions (e.g., airflow, humidity, and personnel traffic), and the rate and extent of microbial dispersion could yield important findings for optimal brewery design from the perspective of microbial control. These findings are instructive but indeed should not be interpreted with alarm: they illustrate microbial dispersal in a functioning brewery that rarely suffers from any cases of product contamination.

Lactic acid bacteria profiles are also driven by substrate exposure

Different species of LAB are the principal spoilage bacteria in beer fermentations as well as important members of sour beer fermentations. However, short 16S rRNA gene amplicons are frequently inadequate to resolve reliable species-level identifications (Bokulich and Mills, 2012a). Therefore, we used LAB-TRFLP (Bokulich and Mills, 2012a) to characterize genus- and species-level LAB community compositions in a select subset of samples. This included raw material and beer samples, as well as surfaces on which detection of Lactobacillales by marker-gene sequencing was particularly high.

Results indicate that different surface and sample types exhibit distinct lactic acid bacterial patterns, corresponding to the substrates encountered at that site or contained in that sample (Figure 8). Wort samples contained a mixture of Lactobacillus delbrueckii, Lactobacillus sakei, Lactobacillus hilgardii, Leuconostoc mesenteroides, Lactococcus lactis, Streptococcus sp., and Bacillus sp. A, most of which were only rarely detected in other fermenting and bottled beer samples. Many of these species are also rarely found in beers (Bokulich and Bamforth, 2013), but instead appear associated with grain, hence their detection in wort. Coolship and fermenting sour beers (in this case coolship beers produced from different wort types) were dominated by Pediococcus and/or L. lindneri, corroborating previous studies of coolship beers in this brewery (Bokulich et al., 2012b). Fermenters and barrel surfaces that contacted these fermentations near the time of sampling exhibited similar communities, though Lactobacillus brevis and Lactobacillus sp. A were more common on these surfaces than in the beers or on other surfaces. Other sour and barrel-aged beers contained unique profiles, with involvement of other Lactobacillus species only weakly detected in coolship beers. Floor and packaging area surfaces contained a more diverse mixture of LAB, but primarily the L. lindneri, L. brevis, and L. delbrueckii detected in the wort and beer samples. Interestingly, only Pediococcus was detected on grain samples, though only weak amplification could be had from these samples, suggesting low LAB populations or inhibition of PCR by grain polyphenols, possibly suppressing the detection of less abundant populations. Hop pellet samples also contained a mixture of different LAB populations, including Pediococcus, L. lindneri, and L. brevis.

Figure 8

Download asset Open asset

These results illustrate that substrate drives the composition of LAB communities as well as whole microbial communities and highlight the risk of cross-contamination between different equipment surfaces. The detection of Lactobacillus spp. on both filler heads (only one of which is used for sour beer packaging) makes this observation all the more pertinent. This observation is not likely to be an exceptional case; cross-contamination between processing areas is very likely the prevailing cause of spoilage in any brewery, where microbial biofilms have been previously reported even on packaging equipment (Timke et al., 2005; Storgards et al., 2006; Timke et al., 2008; Mamvura et al., 2011; Matoulkova et al., 2012). This observation underlines the need for constant hygiene surveillance in breweries but is hardly cause for alarm, if the exceedingly low incidence of microbial spoilage in modern brewing practice is any indication.

Beer contact predicts spoilage-gene distribution on brewery surfaces

Beer-spoiling LAB possess several mechanisms that support their growth and survival in beer (Suzuki et al., 2006). Several hop-resistance genes are principal among these, counteracting the antimicrobial effects of iso-a-acids (hop bittering resins) (Simpson, 1993b; Simpson and Fernandez, 1994). Hop-resistant LAB typically contain several of these genes, including horA (Sami et al., 1997; Sakamoto et al., 2001), horB (Suzuki et al., 2005; Iijima et al., 2006), horC (Suzuki et al., 2005; Iijima et al., 2006), and hitA (Hayashi et al., 2001), which are upregulated during growth in beer (Bergsveinson et al., 2012; Pittet et al., 2013). However, the presence of these genes within brewery environments and on brewing equipment has not been described previously. Thus, we sought to quantify the abundance of these beer-spoilage genes on brewing equipment and in beer during different times in conjunction with microbial community profiles.

Results demonstrate high gene frequencies on several surfaces within the brewery (Figure 9). Sour beer samples contained the highest counts, between 2.0 × 10⁴–4.8 × 10⁴ copies/µl of horC, but fermenter and packaging area surfaces (filler heads, below bottling line belt, and packaging sink) also registered between 2.8–7.8 × 10² copies/cm². None of these alleles were detected on hop samples, keg samples, or one barrel bung (stopper) sample, though 1.1 × 10³ total copies/cm² were detected on a keg faucet used for attaching kegs to beer lines at the brewery. Among the genes analyzed, horC was the most abundant (Figure 9), which is interesting when considered in the context of previous work showing that presence of this gene correlates with increased hop-tolerance and beer-spoilage ability (Fujii et al., 2005; Iijima et al., 2006; Bergsveinson et al., 2012) and the plasmid carrying horC is the most important for supporting growth of Lactobacillus brevis in beer (Bergsveinson et al., 2015). The preferential expression of this gene observed in these previous studies and the relative increased abundance with which it was found in this study suggests horC is an important gene for facilitating beer-spoilage and is consequently selected for in the brewery environment, particularly in areas where sour beers are produced. The purported transcriptional regulator of horC, horB, was detected at stable ratios relative to horC in all samples, supporting this putative function (Iijima et al., 2006). The least frequently observed hop-resistance gene, hitA, is involved in manganese transport (Hayashi et al., 2001), supporting resistance against manganese depletion by iso-a-acids (Behr and Vogel, 2010). Other studies have observed similarly low frequencies of hitA presence and expression in LAB relative to the other hop-resistance genes (Haakensen and Ziola, 2008; Bergsveinson et al., 2012).

Figure 9

Download asset Open asset

The spoilage genes horA, horB, and horC all display high degrees of intercorrelation (Pearson's r = 0.83–1.0, p < 0.01) and significant but lesser correlation to hitA (r = 0.48–0.64, p ≤ 0.04). All spoilage genes except for hitA demonstrate significant correlation with bulk detection of Lactobacillales via 16S rRNA gene sequencing (r = 0.53–0.74, p ≤ 0.03), while bulk Lactobacillales and all spoilage genes but horA correlate significantly with L. lindneri detection via LAB-TRFLP (r = 0.48–0.77, p ≤ 0.04). The only gene correlated with Pediococcus abundance via LAB-TRFLP was horA (r = 0.57, p = 0.01), consistent with previous observations that horA is the primary known hop-resistance gene observed in Pediococcus spp. (Haakensen and Ziola, 2008). Interestingly, no resistance genes correlated with L. brevis, strains of which are among the most common brewery contaminants and most commonly positive for hop-resistance genes (Haakensen and Ziola, 2008). This likely reflects the strains detected in this brewery only, and L. brevis was only a minor constituent of sour beers and processing surfaces (Figure 7).

These results are the first indication of hop-resistance-gene abundance within a brewery environment. As the primary reservoir for spoilage microbes in beer production, tracking spoilage genes on brewing surfaces and materials is important for understanding contamination risks arising from the environment. Sour beer contained the highest gene abundance, which is predictable given that the LAB growing in these (coolship) beers must be adapted to resist hop antimicrobials. Likewise, detection on barrel surfaces is also expected, given their regular and direct contact with these sour beers. Fermenters and packaging-line surfaces also exhibited fairly high levels of hop-resistance genes, with the highest detection on surfaces that contact sour beers and unsanitary surfaces (below packaging belt, packaging-line sink). The only sanitary, conventional beer-making surface among these is the second filler head, which only exhibited 4 cumulative copies/cm² of the resistance genes, compared to 2.1 × 10² in the first filler head, which is used for packaging sour and barrel-aged beers. Nevertheless, this second filler head is also a sanitary surface ready for packaging, and should be free of contaminants. DNA was used for ddPCR and thus these counts may represent dead cells present on the equipment surfaces; the relative proportion of viable cells is unknown. If the resistance genes detected by ddPCR do in fact represent any viable cells, this demonstrates the resilience of these bacterial populations on equipment surfaces that regularly contact ‘contaminated’ (in this case, intentionally contaminated) beers. These findings argue for careful separation of equipment used in conventional beer-making from that used for sour beers: equipment that contacts fermenting and finished sour beer or barrels, including pumps, hoses, and especially packaging equipment, is best dedicated to sour beer production and should not be used to handle conventional beers.

No hop-resistance genes were detected on hop pellets, though LAB were present, indicating that hops are probably not a significant source of beer-spoilage bacteria. Iso-alpha-acids, the antimicrobial compounds released from hops and against which hop-resistance genes confer protection, are generated by the breakdown of humulones during boiling, and are not present in raw hops (Steenackers et al., 2015). Undissociated humulones are significantly less inhibitory to LAB, and the antimicrobial effects of hops are dependent upon acidic conditions (Simpson and Smith, 1992). Thus, the selective pressure to acquire and maintain hop-resistance genes only exists in the presence of hopped beer, not in raw hops. Only two hop pellets were tested for hop-resistance genes, but this finding suggests that the brewery environment itself is the site of hop-resistance-gene propagation. Larger studies of hop-resistance genes on brewery surfaces and brewing materials will illuminate the role of environmental vs raw material contamination in hop-resistance-gene transfer, and sites, conditions, and mechanisms of transfer within the brewery environment.

Tracking spoilage-gene flow across brewery surfaces presents a unique opportunity for understanding spoilage dynamics within food-processing systems and other built environments in general. This approach will facilitate understanding how spoilage resistance propagates within production environments and the reservoirs and vectors encouraging its spread, yielding novel insight for controlling spoilage—focused on gene transmission rather than taxonomic populations. Moving forward, this facility-surveillance model opens many questions and possibilities for mapping microbial spoilage dynamics within food-production systems. What relationship do indoor environmental factors, building design, and surface materials have with microbial transmission? What roles do cleaning and other processes play in controlling contamination on a facility-wide scale? What other biomarkers are associated with contamination and how does the spoilage-allele landscape behave over time and in response to these conditions? The potential advantages of studying microbiology of breweries are not confined to food systems alone, and lessons learned here may aid our understanding of microbial communities in other built environments. For example, the hop-resistance genes include ABC multi-drug transporters similar to other antimicrobial-resistance genes (Sakamoto et al., 2001). Studying and manipulating their transmission within breweries may aid understanding of similar gene-transfer events involved in pathogenesis in hospitals, homes, water systems, or public environments where in situ modeling of such genes is unfeasible or a potential health hazard. Breweries are a useful model for testing general theories of microbial transmission and spoilage-gene dispersal in situ for functional indoor environments and food systems, as beer spoilage is not actually detrimental to human health—merely human pleasure.

1. 1. Bokulich NA

   (2015) Bokulich_brewery_surfaces

   ID 10105. Available on login at Qiita (http://qiita.colorado.edu/).

   http://qiita.colorado.edu/study/description/10105

1. 1. Bokulich NA

   (2013) bokulich_quality_dataset9

   ID 1689. Available on login at Qiita (http://qiita.colorado.edu/).

   http://qiita.colorado.edu/study/description/1689

1. 1. Abarenkov K
   2. Henrik Nilsson R
   3. Larsson KH
   4. Alexander IJ
   5. Eberhardt U
   6. Erland S
   7. Høiland K
   8. Kjøller R
   9. Larsson E
   10. Pennanen T
   11. Sen R
   12. Taylor AF
   13. Tedersoo L
   14. Ursing BM
   15. Vrålstad T
   16. Liimatainen K
   17. Peintner U
   18. Kõljalg U

   (2010) The UNITE database for molecular identification of fungi - recent updates and future perspectives

   New Phytologist 186:281–285.

   https://doi.org/10.1111/j.1469-8137.2009.03160.x

   • Google Scholar
2. 1. Abdo Z
   2. Schüette UM
   3. Bent SJ
   4. Williams CJ
   5. Forney LJ
   6. Joyce P

   (2006) Statistical methods for characterizing diversity of microbial communities by analysis of terminal restriction fragment length polymorphisms of 16S rRNA genes

   Environmental Microbiology 8:929–938.

   https://doi.org/10.1111/j.1462-2920.2005.00959.x

   • Google Scholar
3. 1. Altschul SF
   2. Gish W
   3. Miller W
   4. Myers EW
   5. Lipman DJ

   (1990) Basic local alignment search tool

   Journal of Molecular Biology 215:403–410.

   https://doi.org/10.1016/S0022-2836(05)80360-2

   • Google Scholar
4. 1. Anderson MJ

   (2001)

   A new method for non-parametric multivariate analysis of variance

   Austral Ecology 26:32–46.

   • Google Scholar
5. 1. Back W

   (1981)

   Beer spoilage bacteria. Taxonomy of beer spoilage bacteria. Gram positive species

   Monatsschr Brauwiss 34:267–276.

   • Google Scholar
6. 1. Behr J
   2. Vogel RF

   (2010) Mechanisms of hop inhibition include the transmembrane redox reaction

   Applied and Environmental Microbiology 76:142–149.

   https://doi.org/10.1128/AEM.01693-09

   • Google Scholar
7. 1. Bergsveinson J
   2. Pittet V
   3. Ziola B

   (2012) RT-qPCR analysis of putative beer-spoilage gene expression during growth of Lactobacillus brevis BSO 464 and Pediococcus claussenii ATCC BAA-344T in beer

   Applied Microbiology and Biotechnology 96:461–470.

   https://doi.org/10.1007/s00253-012-4334-3

   • Google Scholar
8. 1. Bergsveinson J
   2. Baecker N
   3. Pittet V
   4. Ziola B

   (2015) Role of plasmids in Lactobacillus brevis BSO 464 hop tolerance and beer spoilage

   Applied and Environmental Microbiology 81:1234–1241.

   https://doi.org/10.1128/AEM.02870-14

   • Google Scholar
9. 1. Bokulich NA
   2. Bamforth CW

   (2013) The microbiology of malting and brewing

   Microbiology and Molecular Biology Reviews 77:157–172.

   https://doi.org/10.1128/MMBR.00060-12

   • Google Scholar
10. 1. Bokulich NA
    2. Mills DA

    (2012a) Differentiation of mixed lactic acid bacteria communities in beverage fermentations using targeted terminal restriction fragment length polymorphism

    Food Microbiology 31:126–132.

    https://doi.org/10.1016/j.fm.2012.1002.1007

    • Google Scholar
11. 1. Bokulich NA
    2. Mills DA

    (2012b) Next-generation approaches to the microbial ecology of food fermentations

    BMB Reports 45:377–389.

    https://doi.org/10.5483/BMBRep.2012.45.7.148

    • Google Scholar
12. 1. Bokulich NA
    2. Mills DA

    (2013a) Facility-specific ‘house’ microbiome drives microbial landscapes of artisan cheesemaking plants

    Applied and Environmental Microbiology 79:5214–5123.

    https://doi.org/10.1128/AEM.00934-13

    • Google Scholar
13. 1. Bokulich NA
    2. Mills DA

    (2013b) Improved selection of internal transcribed spacer-specific primers enables quantitative, ultra-high-throughput profiling of fungal communities

    Applied and Environmental Microbiology 79:2519–2526.

    https://doi.org/10.1128/AEM.03870-12

    • Google Scholar
14. 1. Bokulich NA
    2. Bamforth CW
    3. Mills DA

    (2012a)

    A review of molecular methods for microbial community profiling of beer and wine

    Journal of the American Society of Brewing Chemists 70:150–162.

    • Google Scholar
15. 1. Bokulich NA
    2. Bamforth CW
    3. Mills DA

    (2012b) Brewhouse-resident microbiota are responsible for multi-stage fermentation of American coolship ale

    PLOS ONE 7:e35507.

    https://doi.org/10.1371/journal.pone.0035507

    • Google Scholar
16. 1. Bokulich NA
    2. Mills DA
    3. Underwood MA

    (2013a) Surface microbes in the neonatal intensive care unit: changes with routine cleaning and over time

    Journal of Clinical Microbiology 51:2617–2624.

    https://doi.org/10.1128/JCM.00898-13

    • Google Scholar
17. 1. Bokulich NA
    2. Ohta M
    3. Richardson P
    4. Mills DA

    (2013b) Monitoring seasonal changes in winery-resident microbiota

    PLOS ONE 8:e66437.

    https://doi.org/10.1371/journal.pone.0066437

    • Google Scholar
18. 1. Bokulich NA
    2. Subramanian S
    3. Faith JJ
    4. Gevers D
    5. Gordon JI
    6. Knight R
    7. Mills DA
    8. Caporaso JG

    (2013c) Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing

    Nature Methods 10:57–59.

    https://doi.org/10.1038/nmeth.2276

    • Google Scholar
19. 1. Bokulich NA
    2. Ohta M
    3. Lee M
    4. Mills DA

    (2014) Indigenous bacteria and fungi drive traditional kimoto sake fermentations

    Applied and Environmental Microbiology 80:5522–5529.

    https://doi.org/10.1128/AEM.00663-14

    • Google Scholar
20. 1. Bowers RM
    2. Clements N
    3. Emerson JB
    4. Wiedinmyer C
    5. Hannigan MP
    6. Fierer N

    (2013) Seasonal variability in bacterial and fungal diversity of the near-surface atmosphere

    Environmental Science & Technology 47:12097–12106.

    https://doi.org/10.1021/es402970s

    • Google Scholar
21. 1. Bowers RM
    2. McCubbin IB
    3. Hallar AG
    4. Fierer N

    (2012) Seasonal variability in airborne bacterial communities at a high-elevation site

    Atmospheric Environment 50:41–49.

    https://doi.org/10.1016/j.atmosenv.2012.01.005

    • Google Scholar
22. 1. Caporaso JG
    2. Bittinger K
    3. Bushman FD
    4. DeSantis T
    5. Andersen GL
    6. Knight R

    (2010a) PyNAST: a flexible tool for aligning sequences to a template alignment

    Bioinformatics 26:266–267.

    https://doi.org/10.1093/bioinformatics/btp636

    • Google Scholar
23. 1. Caporaso JG
    2. Kuczynski J
    3. Stombaugh J
    4. Bittinger K
    5. Bushman FD
    6. Costello EK
    7. Fierer N
    8. Peña AG
    9. Goodrich JK
    10. Gordon JI
    11. Huttley GA
    12. Kelley ST
    13. Knights D
    14. Koenig JE
    15. Ley RE
    16. Lozupone CA
    17. McDonald D
    18. Muegge BD
    19. Pirrung M
    20. Reeder J
    21. Sevinsky JR
    22. Turnbaugh PJ
    23. Walters WA
    24. Widmann J
    25. Yatsunenko T
    26. Zaneveld J
    27. Knight R

    (2010b) QIIME allows analysis of high-throughput community sequence data

    Nature Methods 7:335–336.

    https://doi.org/10.1038/nmeth.f.303

    • Google Scholar
24. 1. Caporaso JG
    2. Lauber CL
    3. Walters WA
    4. Berg-Lyons D
    5. Huntley J
    6. Fierer N
    7. Owens SM
    8. Betley J
    9. Fraser L
    10. Bauer M
    11. Gormley N
    12. Gilbert JA
    13. Smith G
    14. Knight R

    (2012) Ultra-high-throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms

    The ISME Journal 6:1621–1624.

    https://doi.org/10.1038/ismej.2012.8

    • Google Scholar
25. 1. Caporaso JG
    2. Lauber CL
    3. Walters WA
    4. Berg-Lyons D
    5. Lozupone CA
    6. Turnbaugh PJ
    7. Fierer N
    8. Knight R

    (2011) Global patterns of 16S rRNA diversity at a depth of millions of sequences per sample

    Proceedings of the National Academy of Sciences of USA 108:4516–4522.

    https://doi.org/10.1073/pnas.1000080107

    • Google Scholar
26. 1. Cole JR
    2. Wang Q
    3. Fish JA
    4. Chai B
    5. McGarrel DM
    6. Sun Y
    7. Brown CT
    8. Porras-Alfaro A
    9. Kuske CR
    10. Tiedje JM

    (2014) Ribosomal Database Project: data and tools for high throughput rRNA analysis

    Nucleic Acids Research 42:D633–D642.

    https://doi.org/10.1093/nar/gkt1244

    • Google Scholar
27. 1. Edgar RC

    (2010) Search and clustering orders of magnitude faster than BLAST

    Bioinformatics 26:2460–2461.

    https://doi.org/10.1093/bioinformatics/btq461

    • Google Scholar
28. 1. Fujii T
    2. Nakashima K
    3. Hayashi N

    (2005) Random amplified polymorphic DNA-PCR based cloning of markers to identify the beer-spoilage strains of Lactobacillus brevis, Pediococcus damnosus, Lactobacillus collinoides and Lactobacillus coryniformis

    Journal of Applied Microbiology 98:1209–1220.

    https://doi.org/10.1111/j.1365-2672.2005.02558.x

    • Google Scholar
29. 1. Gonzalez A
    2. Stombaugh J
    3. Lauber CL
    4. Fierer N
    5. Knight R

    (2012) SitePainter: a tool for exploring biogeographical patterns

    Bioinformatics 28:436–438.

    https://doi.org/10.1093/bioinformatics/btr685

    • Google Scholar
30. 1. Haakensen M
    2. Ziola B

    (2008) Identification of novel horA-harbouring bacteria capable of spoiling beer

    Canadian Journal of Microbiology 54:321–325.

    https://doi.org/10.1139/W08-007

    • Google Scholar
31. 1. Haas BJ
    2. Gevers D
    3. Earl AM
    4. Feldgarden M
    5. Ward DV
    6. Giannoukos G
    7. Ciulla DM
    8. Tabbaa D
    9. Highlander SK
    10. Sodergren E
    11. Methe B
    12. DeSantis TZ
    13. Petrosino JF
    14. Knight R
    15. Birren BW

    (2011) Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons

    Genome Research 21:494–504.

    https://doi.org/10.1101/gr.112730.110

    • Google Scholar
32. 1. Hayashi N
    2. Ito M
    3. Horiike S
    4. Taguchi H

    (2001) Molecular cloning of a putative divalent-cation transporter gene as a new genetic marker for the identification of Lactobacillus brevis strains capable of growing in beer

    Applied Microbiology and Biotechnology 55:596–603.

    https://doi.org/10.1007/s002530100600

    • Google Scholar
33. 1. Iijima K
    2. Suzuki K
    3. Ozaki K
    4. Yamashita H

    (2006) horC confers beer-spoilage ability on hop-sensitive Lactobacillus brevis ABBC45cc

    Journal of Applied Microbiology 100:1282–1288.

    https://doi.org/10.1111/j.1365-2672.2006.02869.x

    • Google Scholar
34. 1. Kelley ST
    2. Gilbert JA

    (2013) Studying the microbiology of the indoor environment

    Genome Biology 14:202.

    https://doi.org/10.1186/gb-2013-14-2-202

    • Google Scholar
35. 1. Kembel SW
    2. Meadow JF
    3. O'Connor TK
    4. Mhuireach G
    5. Northcutt D
    6. Kline J
    7. Moriyama M
    8. Brown GZ
    9. Bohannan BJ
    10. Green JL

    (2014) Architectural design drives the biogeography of indoor bacterial communities

    PLOS ONE 9:e87093.

    https://doi.org/10.1371/journal.pone.0087093

    • Google Scholar
36. 1. Knights D
    2. Kuczynski J
    3. Charlson ES
    4. Zaneveld J
    5. Mozer MC
    6. Collman RG
    7. Bushman FD
    8. Knight R
    9. Kelley ST

    (2010) Bayesian community-wide culture-independent microbial source tracking

    Nature Methods 8:761–763.

    https://doi.org/10.1038/nmeth.1650

    • Google Scholar
37. 1. Lax S
    2. Smith DP
    3. Hampton-Marcell J
    4. Owens SM
    5. Handley KM
    6. Scott NM
    7. Gibbons SM
    8. Larsen P
    9. Shogan BD
    10. Weiss S
    11. Metcalf JL
    12. Ursell LK
    13. Vasquez-Baeza Y
    14. Van Treuren W
    15. Hasan NA
    16. Gibson MK
    17. Colwell R
    18. Dantas G
    19. Knight R
    20. Gilbert JA

    (2014) Longitudinal analysis of microbial interaction between humans and the indoor environment

    Science 345:1048–1052.

    https://doi.org/10.1126/science.1254529

    • Google Scholar
38. 1. Mamvura TA
    2. Iyuke SE
    3. Cluett JD
    4. Paterson AE

    (2011) Soil films in the beverage industry: a review

    Journal of the Institute of Brewing 117:608–616.

    https://doi.org/10.1002/j.2050-0416.2011.tb00511.x

    • Google Scholar
39. 1. Matoulkova D
    2. Kosar K
    3. Slaby M
    4. Sigler K

    (2012)

    Occurrence and species distribution of strictly anaerobic bacterium Pectinatus in brewery bottling halls

    American Society of Brewing Chemists 70:262–267.

    • Google Scholar
40. 1. McDonald D
    2. Price MN
    3. Goodrich J
    4. Nawrocki EP
    5. DeSantis TZ
    6. Probst A
    7. Andersen GL
    8. Knight R
    9. Hugenholtz P

    (2012) An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea

    The ISME Journal 6:610–618.

    https://doi.org/10.1038/ismej.2011.139

    • Google Scholar
41. 1. Pittet V
    2. Phister TG
    3. Ziola B

    (2013) Transcriptome sequence and plasmid copy number analysis of the brewery isolate Pediococcus claussenii ATCC BAA-344T during growth in beer

    PLOS ONE 8:e73627.

    https://doi.org/10.1371/journal.pone.0073627

    • Google Scholar
42. 1. Sakamoto K
    2. Margolles A
    3. van Veen HW
    4. Konings WN

    (2001) Hop resistance in the beer spoilage bacterium lactobacillus brevis is Mediated by the ATP-Binding Cassette multidrug transporter HorA

    Journal of Bacteriology 183:5371–5375.

    https://doi.org/10.1128/JB.183.18.5371-5375.2001

    • Google Scholar
43. 1. Sami M
    2. Yamashita H
    3. Hirono T
    4. Kadokura H
    5. Kitamoto K
    6. Yoda K

    (1997) Hop-resistant Lactobacillus brevis contains a novel plasmid harboring a multidrug resistance-like gene

    Journal of Fermentation and Bioengineering 84:1–6.

    https://doi.org/10.1016/S0922-338X(97)82778-X

    • Google Scholar
44. 1. Shyu C
    2. Soule T
    3. Bent SJ
    4. Forster JA
    5. Forney LJ

    (2007) MiCA: a web-based tool for the analysis of microbial communities based on terminal-restriction fragment length polymorphisms of 16S and 18S rRNA genes

    Journal of Microbial Ecology 53:562–570.

    https://doi.org/10.1007/s00248-006-9106-0

    • Google Scholar
45. 1. Simpson WJ

    (1993a) Ionophoric action of trans-isohumulone on Lactobacillus brevis

    The Journal of General and Applied Microbiology 139:1041–1045.

    https://doi.org/10.1099/00221287-139-5-1041

    • Google Scholar
46. 1. Simpson WJ

    (1993b) Studies on the sensitivity of lactic acid bacteria to hop bitter acids

    Journal of the Institute of Brewing 99:405–411.

    https://doi.org/10.1002/j.2050-0416.1993.tb01180.x

    • Google Scholar
47. 1. Simpson WJ
    2. Fernandez JL

    (1992) Selection of beer-spoilage lactic-acid bacteria and Induction of their ability to grow in beer

    Letters in Applied Microbiology 14:13–16.

    https://doi.org/10.1111/j.1472-765X.1992.tb00636.x

    • Google Scholar
48. 1. Simpson WJ
    2. Fernandez JL

    (1994)

    Mechanism of resistance of lactic acid bacteria to trans-isohumulone

    American Society of Brewing Chemists 52:9–11.

    • Google Scholar
49. 1. Simpson WJ
    2. Smith AR

    (1992) Factors affecting antibacterial activity of hop compounds and their derivatives

    The Journal of Applied Bacteriology 72:327–334.

    https://doi.org/10.1111/j.1365-2672.1992.tb01843.x

    • Google Scholar
50. 1. Spaepen M
    2. Vanoevelen D
    3. Verachtert H

    (1978) Fatty acids and esters produced during spontaneous fermentation of lambic and gueuze

    Journal of the Institute of Brewing 84:278–282.

    https://doi.org/10.1002/j.2050-0416.1978.tb03888.x

    • Google Scholar
51. 1. Spitaels F
    2. Wieme AD
    3. Janssens M
    4. Aerts M
    5. Daniel HM
    6. Van Landschoot A
    7. De Vuyst L
    8. Vandamme P

    (2014) The microbial diversity of traditional spontaneously fermented lambic beer

    PLOS ONE 9:e95384.

    https://doi.org/10.1371/journal.pone.0095384

    • Google Scholar
52. 1. Steenackers B
    2. De Cooman L
    3. De Vos D

    (2015) Chemical transformations of characteristic hop secondary metabolites in relation to beer properties and the brewing process: a review

    Food Chemistry 172:742–756.

    https://doi.org/10.1016/j.foodchem.2014.09.139

    • Google Scholar
53. 1. Storgards E
    2. Tapani K
    3. Hartwall P
    4. Saleva R
    5. Suihko ML

    (2006)

    Microbial attachment and biofilm formation in brewery bottling plants

    Journal of the American Society of Brewing Chemists 64:8–15.

    • Google Scholar
54. 1. Suzuki K
    2. Iijima K
    3. Sakamoto K
    4. Sami M
    5. Yamashita H

    (2006) A review of hop resistance in beer spoilage lactic acid bacteria

    Journal of the Institute of Brewing 112:173–191.

    https://doi.org/10.1002/j.2050-0416.2006.tb00247.x

    • Google Scholar
55. 1. Suzuki K
    2. Sami M
    3. Ozaki K
    4. Yamashita H

    (2005)

    Comparative study of two plasmids, pRH45 and pRH 20690 isolated from beer-spoilage Lactobacillus brevis ABBC45 and L. lindneri DSM 20690T

    American Society of Brewing Chemists 63:11–16.

    • Google Scholar
56. 1. Timke M
    2. Wang-Lieu NQ
    3. Altendorf K
    4. Lipski A

    (2005) Community structure and diversity of biofilms from a beer bottling plant as revealed using 16S rRNA gene clone libraries

    Applied and Environmental Microbiology 71:6446–6452.

    https://doi.org/10.1128/AEM.71.10.6446-6452.2005

    • Google Scholar
57. 1. Timke M
    2. Wang-Lieu NQ
    3. Altendorf K
    4. Lipski A

    (2008) Identity, beer spoiling and biofilm forming potential of yeasts from beer bottling plant associated biofilms

    Antonie Van Leeuwenhoek 93:151–161.

    https://doi.org/10.1007/s10482-007-9189-8

    • Google Scholar
58. 1. Van Oevelen D
    2. Delescaille F
    3. Verachtert H

    (1976) Synthesis of aroma compounds during spontantous fermentation of lambic and gueuze

    Journal of the Institute of Brewing 82:322–326.

    https://doi.org/10.1002/j.2050-0416.1975.tb06953.x

    • Google Scholar
59. 1. Van Oevelen D
    2. Spaepen M
    3. Timmermans P
    4. Verachtert H

    (1977) Microbial aspects of spontaneous fermentation in production of lambic and gueuze

    Journal of the Institute of Brewing 83:356–360.

    https://doi.org/10.1002/j.2050-0416.1977.tb03825.x

    • Google Scholar
60. 1. Wang Q
    2. Garrity GM
    3. Tiedje JM
    4. Cole JR

    (2007) Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy

    Applied and Environmental Microbiology 73:5261–5267.

    https://doi.org/10.1128/AEM.00062-07

    • Google Scholar

Author details

1. Nicholas A Bokulich

   1. Department of Food Science and Technology, University of California, Davis, Davis, United States
   2. Department of Viticulture and Enology, University of California, Davis, Davis, United States
   3. Foods for Health Institute, University of California, Davis, Davis, United States

   Present address

   Department of Medicine, New York University Langone Medical Center, New York, United States

   Contribution

   NAB, Conception and design, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

2. Jordyn Bergsveinson

   Department of Pathology and Laboratory Medicine, University of Saskatchewan, Saskatoon, Canada

   Contribution

   JB, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

3. Barry Ziola

   Department of Pathology and Laboratory Medicine, University of Saskatchewan, Saskatoon, Canada

   Contribution

   BZ, Acquisition of data, Analysis and interpretation of data, Drafting or revising the article

   Competing interests

   The authors declare that no competing interests exist.

   "This ORCID iD identifies the author of this article:" 0000-0003-4836-7576

4. David A Mills

   1. Department of Food Science and Technology, University of California, Davis, Davis, United States
   2. Department of Viticulture and Enology, University of California, Davis, Davis, United States
   3. Foods for Health Institute, University of California, Davis, Davis, United States

   Contribution

   DAM, Conception and design, Drafting or revising the article

   For correspondence

   damills@ucdavis.edu

   Competing interests

   The authors declare that no competing interests exist.

• 5,416

  views

• 870

  downloads

• 75

  citations

Views, downloads and citations are aggregated across all versions of this paper published by eLife.