18th August 2019
Translated from the original article in Catalan
A few months ago -April 2019- my friend Jordi Diloli, Professor and Archaeologist, shared a very surprising article (Aouizerat et al 2019) with me. It was echoed on the internet (Borschel-Dan 2019), and I will comment here.
“Resurrected” yeasts from 3,000 years ago
The group of researchers led by Ronen Hazan of the Hebrew University of Jerusalem took samples of 21 clay containers from various sites in present-day Israel from 2500 to 5000 years ago, from the Persian, Philistine and Egyptian (this is the oldest) periods. Archaeologists believed that these vessels contained fermented beverages such as beer or mead (Figure 1). The authors submerged the containers in a rich YPD medium, specific for growing yeasts and other fungi, and incubated them at room temperature for 7 days. Then, samples of this medium were spread on agar plates with the specific medium, and the resulting colonies were isolated for subsequent analyses (Aouizerat et al 2019).
Figure 1. Clay vessels from where the yeasts were isolated (Image of Judah Ari Gross, Times of Israel).
The isolates that were found were 6 strains of different yeast species, and one of which was Saccharomyces cerevisiae, specifically from a Philistine site dated 3000 years ago. Obviously, it is very surprising that living yeasts of such ancient remains have been isolated. For this reason, the authors of the work carried out a series of experiments that could confirm this unique fact and that the isolates were not a product of contamination.
Firstly Aouizerat et al (2019) showed that it is possible to isolate yeasts from clay vessels that have contained beer or wine after a certain time. They did so with containers with unfiltered beer buried for 3 weeks, and also with another vessel that had repeatedly contained wine but not used last 2 years. With these samples they developed the isolation methodology and in both cases they were able to isolate yeasts. No isolates were obtained from a control sample with filtered beer, therefore without yeasts.
To demonstrate that the isolates were originals of the old vessels because these had contained the fermented beverage, authors applied the same protocol with samples of other ceramics that were surely not for this purpose, and also with sediments near the containers. The result was clearly negative for these samples: only 2 isolated yeasts from 110 samples, while the mentioned 6 yeast strains were isolated from the 21 initial samples. That is, yeasts would be significantly more abundant in containers of alcoholic fermented beverages than in other related archaeological vessels or sediments around them.
Another argument that supports the hypothesis of this work was the identification of these 6 yeasts. Total DNA was obtained and processed to sequence the genomes and compare them with the databases. Two of them, from the Egyptian period, were identified as Saccharomyces delphensis, a species that has been isolated from African dried figs and is not at all common on soil. Therefore, this suggests the use of figs in the alcoholic beverages of these containers. Another isolate was identified as Rhodotorula, common pollutant yeast in African beers. Another was identified as Debaryomyces, a frequent yeast in traditional African sorghum beers. As said before, another isolate was identified as Saccharomyces cerevisiae, the yeast most used to make wine, beer or bread (Figure 2). In spite of this, the genetic sequence of this S. cerevisiae was clearly different from the strains most commonly used today, as commercial or laboratory strains, and therefore the possibility of contamination is excluded. And finally, the other isolate was identified as Hypopichia burtonii, previously isolated yeast from Ethiopian mead.
These genetic data, together with the phenotypic characterization -fermentative kinetics and other biochemical characteristics carried out with the isolates by Aouizerat et al (2019)- suggest that these yeasts actually come from an environment related to alcoholic beverages. These authors even elaborated beer with these isolates and some of them, especially the Saccharomyces, gave a very good analytical and sensory result.
Figure 2. Saccharomyces cerevisiae at the scanning electronic microscope (MD Murtey & P Ramasamy)
Aouizerat et al (2019) conclude that the isolates are descendants of yeasts that were originally used 3000 years ago, in large quantities and in repeated fermentations. This would have facilitated their survival in pore microenvironments of the ceramic matrix of these containers, and the microcolonies would have continued to grow minimally for millennia thanks to the humidity and residual nutrients. The authors make the analogy with some handmade beers where it is usual that the containers waste serve as starter for new productions.
Finally, the authors of this work speculate that it is possible to isolate microorganisms from archaeological remains, not only yeasts, and that in the case of bacteria it could even be easier, given the resistance characteristics of some of them, such as the sporulated ones.
Is there no previous similar work to that of Aouizerat et al (2019) ?
As we have seen, this is certainly a very surprising finding. Scientifically, the work is quite accurate and has been “approved” by the international community: the article is published in an open-access journal with prestige (mBio, high impact factor: 6.7), of the American Society for Microbiology, where all the articles are reviewed by a minimum of two experts, besides the editors. The results presented by the article seem very well worked, and the conclusions are well reasoned.
However, in my opinion it is still almost incredible, and it is strange that nothing like this has been found before. Maybe if someone else had previously tried to isolate such old microorganisms without getting them, perhaps it would not have been published ? Maybe nobody has previously tried to do something similar ? A “malicious” explanation might be that archaeologists have their own interests and microbiologists or molecular biologists have others, and that for this type of work the collaboration of both is needed. Well, it seems not being so, since there are a lot of studies on microorganisms from ancient remains, but they have been almost always focused on the detection and analysis of ancient DNA. These studies demonstrated the presence of certain microorganisms although they did not proceed to isolate them.
DNA gives evidence of microorganisms in ancient remains
In relation to yeasts, the oldest evidence is that ribosomal DNA of Saccharomyces cerevisiae has been obtained from residues found in Egyptian wine jars 5000 years old (Cavalieri et al 2003). It must be remembered that the oldest archaeological evidence of large-scale wine production has 7400 years, in north of the Zagros Mountains, in present-day Iran (McGovern et al 1986). As it is known, S. cerevisiae is also the bread and beer yeast, derived from cereals, but since neither S. cerevisiae nor its spores are aerial, surely the use of this yeast in fermented grape juice, as well as dates, figs or honey, historically preceded its use for brewing and bread (Cavalieri et al 2003). It is probable that the wine yeasts naturally occurring in damaged grapes (Mortimer & Polsinelli 1999) were used to ferment other cereal products such as cereals, and after centuries of selection for humans, they evolved into specific strains to ferment food and beverages from cereals.
The genomes of pathogenic microorganisms have also been studied in archaeological remains by means of new massive DNA sequencing techniques, in order to know to epidemic diseases of historical importance, such as black plague, tuberculosis, cholera or leprosy (Andam et al 2016). Logically, in these cases the archaeological remains are human ones, such as bones, teeth, coprolites or mummified tissues. In this way, for example, the phylogeny and evolution of Yersinia pestis strains causing the black plague have been recognized by remains of the Bronze Age (5000 years ago) and until the well-known epidemics of the 6th and 14th centuries (Bos et al 2011). Another well-known case is the Helicobacter pylori genome identified in the intestine of the Ötzi mummy, the iceman in the eastern Alps, 5300 years old (Maixner et al 2016).
DNA has also been isolated from specific bacteria of the human gut, such as Bifidobacterium and Bacteroides, to demonstrate the human presence in archaeological sediments 5000-12000 years old, in north east of Poland (Madeja et al 2009).
It should be remembered that DNA is degraded over time, and in fact it is more unstable than other cellular components. This macromolecule spontaneously suffers damage by oxidation, hydrolysis, and fragmentation in pieces that may be less than 100 bp. Most fossils or other biological remains of more than 100,000 years old no longer contain PCR-amplifiable DNA (Hofreiter et al 2001), although it seems that if the samples are extracted from frozen sediments, with constant temperatures below zero, DNA could be recovered from up to 400,000 years or a little longer (Willerslev et al 2003). In addition the tissues are colonized over time by fungi and bacteria that greatly reduce the relative amount of endogenous molecules and can contribute to giving false positives. The risk of contamination is very high and often this is not taken in account. Generally the DNA of the host that is analysed can be less than 1% of the total DNA found. All these factors complicate the DNA extraction, the construction of sequence libraries, the alignment of DNAs and the analysis of genomes (Andam et al 2016).
Surprisingly, there are a few published works where it is found old DNA of plants, animals and various microorganisms, some million years (My) old, even hundreds of My. The most remarkable are those obtained from amber samples of 20-40 My, and those obtained from a halite 250 My old. This would be comparable to the Jurassic Park fiction where almost non-degraded DNA from the dinosaurs of 100 My old “was recovered”.
Hebsgaard et al (2005) thoroughly reviewed all these more spectacular cases, with the conclusion that these works suffered from inadequate experimental approaches and inadequate authentication of the results. Therefore, there are great doubts as to whether DNA sequences and in some cases viable bacteria could survive such large geological times.
In addition, it is worrying that these works with so old DNA have not been replicated independently in order to confirm their authenticity, and that they did not show a relationship between the age of the sample and the persistence of DNA depending on the different types of bacteria (Willerslev et al 2004). In contrast, Willerslev et al studied the persistence of DNA in permafrost and they found a clear relationship of DNA degradation with time (Figure 3). As seen, DNA amount is very small beyond 100,000 years and it is hardly found beyond 1 My.
Figure 3. Persistence of not degraded bacterial DNA over time (kyr, thousands of years) maintained in permafrost, measured by fluorescence (Willerslev et al 2004).
When analysing the bacterial phyla of these DNA, Willerslev et al (2004) observed (Figure 4) that the most persistent are those of Arthrobacter, the main representative of Actinobacteria (high G+C gram-positive), followed by sporulated (Bacillaceae and Clostridiaceae), and finally the Gram-negative Proteobacteria.
Figure 4. Proportions of the main bacterial phyla (Actinobacteria in brown, sporulated in orange and Proteobacteria in blue) based on DNA obtained from permafrost samples, along time (kyr, thousands of years) (Willerslev et al 2004).
This increased persistence of non-sporulated Actinobacteria is surprising because sporulated bacteria have always been considered the most resistant of all types of cells. Although endospores have special adaptations such as proteins binding DNA to reduce the rate of genetic modifications, they do not have active metabolism or repair and their DNA will degrade over time. The mechanism of greater resistance of Actinobacteria is unknown, but there may be some activity and repair of DNA at temperatures below zero, and/or adaptations related to the dormant cells state (Willerslev et al 2004).
Anyway, the limit for PCR-amplifying the DNA would be between 400,000 years and 1.5 My for samples kept below zero, but this is much more unlikely in non-frozen materials, such as the amber of halite samples of million years, and much less likely to find viable cells from these samples so old (Willerslev et al 2004).
The same commented works where DNA of some millions of years (My) was found, are the most surprising cases of having “resurrected” microorganisms, basically bacteria: viable cells of the sporulated Bacillus from amber samples of 30 My (Cano & Borucki 1995), Staphylococcus also from amber of about 30 My (Lambert et al 1998), and the most spectacular case of Bacillus from an halite of 250 My (Vreeland et al 2000 ). This sporulated bacterium would have been in a hyper-saline environment of the last Permian and trapped in a salt crystal, surviving until now. In the case of Staphylococcus isolated from amber, in spite of not being sporulated, they are bacteria very resistant to extreme conditions, and which have been isolated also from ancient permafrost and very dry environments (Lambert et al 1998).
In spite of this, the revision of these cases by Hebsgaard et al (2005) concludes that none of them fulfilled the relative rate of molecular distance test, which is the probable rate of mutations calculated in comparison to related lineages. Therefore, these isolations are arguable and not reproduced. In addition, in the case of the 250 My Bacillus, it has been argued that the inclusion of bacteria in the halite could be the consequence of a subsequent recrystallization (Lowenstein et al 2011).
Another review on microorganism preservation records (Kennedy et al 1994) comments published cases up to 600 My, indicating that it is curious that there are several cases with more than 1 My, and also cases with less than 10,000 years ago, but there are very few cases of intermediate periods. These authors also point out the doubts raised by works with surviving bacteria so old, which would surely be artefacts or contaminations.
On the other hand, the most credible works are those of Abyzov et al (2006) and Soina et al (2004), which demonstrated the presence of several living microorganisms, both prokaryotes and eukaryotes (especially yeasts, but also some microalgae), in Antarctic ice samples that have some thousands of years. These authors combined classical microbiological methods, such as enrichment and isolation of colonies, together with epifluorescence microscopy, electronic microscopy, and molecular techniques. The bacteria found were Gram-positive (Micrococcus) and gram-negative (Arthrobacter), which are not sporulated, but they have cist-shaped dormant cells, which can survive while maintaining viability at temperatures below 0ºC for some thousands of years.
When geologically ancient DNA findings are published as well as viable cultures of ancient samples, the independent reproduction of the results by another laboratory is fundamental, to exclude any contamination from the same laboratory. In the case of having recovered living cells, it is necessary to demonstrate the reproducibility of the isolation, sequencing the genomes of the cultures obtained in independent laboratories from the same sample, and checking that in both cases the genomes coincide (Hebsgaard et al 2005).
From the remains of the Roman fort of Vindolanda, in the north of England, viable endospores of Thermoactinomyces, member of Bacillales (Unsworth et al 1977) have been recovered. They are about 1900 years old and the remains were a mixture of clay with straw and other vegetable materials. The authors propose to use these sporulated bacteria as indicators in archaeological studies.
Besides sporulated bacteria, there are several groups of non-sporulated ones for which anabiosis resistance abilities have been demonstrated. In particular, they have been isolated from permafrost and the tundra soil of Siberia of about 1 My (Suzina et al., 2006), in the limit of what we mentioned earlier (Willerslev et al 2004), which is quite difficult to believe. In order to study experimentally the formation of these anabiosis forms, Suzina et al incubated several gram-positive and gram-negative bacteria, and some archaea, in poor media with limiting nitrogen, and after a few months they obtained their dormant cells. They had cist structures, with capsule and a thickened cell wall, intramembranous particles and a condensed nucleoid (Figure 5). They also observed that these cysts did not have metabolic activity and supported stress factors such as lack of nutrients or heating.
Studying the permafrost isolates, they confirmed that there are cist structures very similar to those obtained in the laboratory, with multi-layer wall structures of up to 0.4 μm. In fact, these authors believe that most of the bacteria present in the permafrost and the tundra are in the form of a cyst (Suzina et al 2006).
Figure 5. Sections of a vegetative cell (a) of Micrococcus luteus and of a cyst cell (b) of the same bacterium, obtained after 9 months of culture in a medium limiting in nitrogen. C, microcapsule; CW, cell wall; OL1, 2, 3, outer layers of the cell wall; IL, inner layer of the wall; CM, cytoplasmic membrane; N, nucleoid. The bar measures 0.3 μm (Suzina et al., 2006).
Other “resurrected” yeasts and fungi
Besides the surprising mentioned article by Aouizerat et al. (2019), there are other few published cases of yeasts and other “resurrected” fungi such as the following.
Chicha is a beer-like beverage from corn, yellowish and slightly effervescent, elaborated and consumed by Andean populations for some thousands of years, whose traditional process has the peculiarity of using amylase of saliva for convert the starch into fermentable sugars. Fermentation traditionally took place in clay containers called “pondos”. From the remains of the chicha pondos from the Hipia culture in Quito (2100-2800 years old), various yeast were isolated, especially Candida, Pichia and Cryptococcus (Gomes et al 2009). Interestingly, some of these yeasts have been confirmed molecularly as Candida theae, similar to those isolated from contaminated Asian tea (Chang et al 2012). It is worth mentioning the absence of Saccharomyces in these ancient chicha, although today it is the main yeast, coming probably from beer and wine fermentation that led the Spaniards (Gomes et al 2009).
From Greenland ice samples of about 100,000 years (Ma et al 1999), several microorganisms were revived, such as bacteria (Micrococcus, Rhodotorula, Sarcina) and yeasts (Candida, Cryptococcus) and other fungi (Penicillium, Aspergillus). The authors also isolated the DNAs and demonstrated the phylogenetic relationship of the isolates. Once again, we see how ice provides a stable environment that facilitates the conservation of microorganisms and their DNA.
Raghukumar et al (2004) have recovered living Aspergillus (sporulated Ascomycota fungus) and other fungi from sediment samples of the deep-sea, about 5900 m deep in the Chagos trench, south of the Maldives, in the Indian Ocean. Based on the depth in the sediment and the present Radiolaria, authors estimated that they correspond to a minimum of 180,000 years, and up to 430,000 years in some samples. From the isolates identified as A. sydowii they obtained spores that germinated and grew in hydrostatic pressure equivalent to the depth of 5000 m, and at a temperature of 5ºC. With microscopy of epifluorescence and bright field, the fungal hypha and their relation to the particles of the sediment are clearly observed (Figure 6). It seems that this Aspergillus found in the deep-sea is the oldest fungus recovered alive so far. The authors suggest that preservation would have been possible thanks to high hydrostatic pressure, along with low temperature.
Figure 6. Photomicrographies of deep-sea sediment (5900 m) of the Indian Ocean with hyphae of Aspergillus sydowii and sediment particles. (a) epifluorescence microscopy combined with that of bright field; (b) epifluorescence (Raghukumar et al 2004).
One of the most surprising works, and hard to believe, is that of Kochkina et al (2001), where a lot of fungi of all kinds and bacteria, especially actinobacteria, were isolated from samples of permafrost from Russia, Canada and Antarctica reaching 3 My old. The authors even suggested that there is no limit of years to recover viable microorganisms. This article has had very little echo, and it is not even mentioned by later articles as Raghukumar et al. (2004).
As we have seen, evidence of DNA from no-living yeasts in ancient remains related to winemaking dates back to around 5000 years in ancient Egypt (Cavalieri et al 2003). Regarding other microorganisms, taking into account the natural degradation of DNA over time, it seems that the oldest samples would be about 400,000 years at most, in particular actinobacteria in frozen sediments such as permafrost (Willerslev et al 2003 ). Publications of bacterial DNA recovered from several millions years (up to 600 My) have many scientific concerns about their credibility and reliability (Kennedy et al 1994).
With regard to living yeast as those of 3000 years apparently isolated by Aouizerat et al (2019), it seems that Candida and others were isolated from containers to elaborate chicha about 2800 years old (Gomes et al 2009), although this reference is a review and the original work does not appear to have been published. Other authors (Abyzov et al 2006; Soina et al 2004) also find alive yeasts, without specifying which ones, in Antarctic ice samples of some thousands of years. More surprising are the isolated isolations of yeast and other fungi and bacteria from Greenland ice samples 100,000 years old (Ma et al 1999), as well as those of Aspergillus from the Indian Ocean seabed of about 180,000 years (Raghukumar et to 2004).
Regarding other “resurrected” microorganisms, some of the most reliable are the several Antarctic ice bacteria of some thousands of years (Abyzov et al 2006) and Thermoactinomyces spores of Roman remains 1900 years old (Unsworth et to 1977). Of the oldest, perhaps the anabiosis forms of bacteria conserved in permafrost a million years old (Suzina et al., 2006) would have certain likelihood. Curiously, these bacteria would be non-sporulated but they would have a cyst structure, with multi-layer walls and other intracellular modifications. The other findings of “resurrected” bacteria from more millions of years of amber or halite, just like their DNA and also because of this, are very hard to believe (Hebsgaard et al 2005).
Thinking in the cellular forms of resistance and anabiosis, as the bacterial endospores and the mentioned cysts, it must be remembered that yeasts, like many other fungi, have the ability to produce spores, in particular ascospores as they are Ascomycetes. Although these ascospores have a greater capacity for resistance than vegetative cells in dry conditions or other inhospitable environments and have a persistence in time, apparently there is no work (or I have not been found) related to the recovery of yeasts ascospores from ancient remains.
The work of Aouizerat et al (2019) makes no mention of the yeast spores, neither as a possible explanation of the yeast survival in these ancient remains. In fact, they propose that the microcolonies of yeasts on ceramics pores would have continued to grow minimally during these 3000 years thanks to the humidity and residual nutrients. Well, we do not know, and neither if the yeast ascospores have had any role.
Finally, we can believe the finding of Aouizerat et al (2019) is truth, but obviously further investigation in other similar archaeological samples must be done. This research should be done not only for yeasts, but also for bacteria of other fermented products. Besides considering the sporulated ones, other bacteria should be considered, that could survive thanks to the cell cysts or other forms of anabiosis.
Abyzov SS et al (2006) Super-long anabiosis of ancient microorganisms in ice and terrestrial models for development of methods to search for life on Mars, Europa and other planetary bodies. Adv Space Res 38, 1191-1197
Andam CP et al (2016) Microbial genomics of ancient plagues and outbreaks. Trends Microbiol 24, 978 –990
Aouizerat T et al (2019) Isolation and characterization of live yeast cells from ancient vessels as a tool on bio-archaeology. mBio 10, 2, 1-21
Borschel-Dan A (2019) Israeli scientists brew groundbreaking “ancient beer” from 5,000-year-old yeast. The Times of Israel, 22nd may 2019.
Bos KI et al (2011) A draft genome of Yersinia pestis from victims of the Black Death. Nature 478, 506–510
Cano, R.J. and Borucki, M.K. (1995) Revival and identification of bacterial spores in 25- to 40-million year-old Dominican amber. Science 268, 1060–1064
Cavalieri D et al (2003) Evidence for S. cerevisiae fermentation in ancient wine. J Mol Evol 57:S226-232
Chang CF et al (2012) Candida theae sp. nov., a new anamorphic beverage-associated member of the Lodderomyces clade. Int J Food Microbiol 153, 10-14.
Gomes FCO et al (2009) Traditional foods and beverages from South America: microbial communities and production strategies. Chapter 3 in Industrial Fermentation, ed. J Krause & O Fleischer, Nova Science Publishers.
Hofreiter M et al (2001) Ancient DNA. Nature Rev Genet 2, 353–359.
Kennedy MJ et al (1994) Preservation records of micro-organisms: evidence of the tenacity of life. Microbiology 140, 2513-2529.
Kochkina GA et al (2001) Survival of micromycetes and actinobacteria under conditions of long-term natural cryopreservation. Microbiology 70, 356-364
Lambert LH et al (1998) Staphylococcus succinus sp. nov., isolated from Dominican amber. Int J Syst Bacteriol 48, 511-518
Lowenstein TK et al (2011) Microbial communities in fluid inclusions and long-term survival in halite. GSA Today 21, 4-9
Ma L et al (1999) Revival and characterization of fungi from ancient polar ice. Mycologist 13, 70-73.
Madeja J et al (2009) Bacterial ancient DNA as an indicator of human presence in the past: its correlation with palynological and archaeological data. J Quaternary Sci 24, 317-321.
Maixner F et al. (2016) The 5300-year-old Helicobacter pylori genome of the Iceman. Science 351, 162–165
McGovern PE et al (1986) Neolithic resinated wine. Nature 381:480–481
Mortimer R & M Polsinelli (1999) On the origins of wine yeast. Res Microbiol 150, 199-204
Raghukumar C et al (2004) Buried in time: culturable fungi in a deep-sea sediment core from the Chagos Trench, Indian Ocean. Deep Sea Res Part I: Oceanog Res Papers 51, 1759-1768
Soina VS et al (2004) The structure of resting microbial populations in soil and subsoil permafrost. Astrobiology 4 (3), 345–358.
Suzina et al (2006) The structural bases of long-term anabiosis in non-spore-forming bacteria. Adv Space Res 38, 1209-1219.
Unsworth BA et al (1977) The Longevity of Thermoactinomycete Endospores in Natural Substrates. J Appl Microbiol 42, 45-52
Vreeland RH et al (2000) Isolation of a 250 milion-year-old halotolerant bacterium from a primary salt cristal. Nature 407, 897-900.
Willerslev E et al (2003) Diverse plant and animal DNA from Holocene and Pleistocene sedimentary records. Science 300, 791-795
28th October 2018
It is not easy to “live” in the beer
In principle, lactic acid bacteria (LAB) and many other bacteria and generally most microorganisms, do not have it easy to survive in beer or other alcoholic beverages such as wine. This is one of the main reasons why wines and beers have been from ancient times the safest ways to drink hygienically something similar to water and that it was not contaminated, apart from boiled waters, such as tea and other herbal infusions.
The reasons for the difficult survival of microorganisms in beer are ethanol, the pH quite acidic (around 4), the lack of nutrients due to the fact that the yeasts have assimilated them, the little dissolved oxygen, the high concentration of carbon dioxide (0.5% by weight / volume) and the presence of humulone derived compounds (Figure 1) of hops: iso-alpha-acids, up to 50 ppm, which are microbiocides. All these obstacles make it very difficult for any microorganism to thrive. The most susceptible beers of unwanted microbial growth are those where some of the mentioned obstacles are dampened: beers with a higher pH of 4.5, or with little ethanol or little CO2, or with added sugars – which are nutrients -, or with little amount of compounds derived from hops (Vriesekoop et al 2012).
Figure 1. Humulone (left) of the hop is degraded during beer elaboration to isohumulone (right) and other iso-alpha-acids, which are compounds bitter and microbiocides (Wikipedia; Sakamoto & Konings 2003)
The acid pH of the beer (slightly higher than the wine) inhibits many of the best-known pathogens (Figure 2). And the cases we see that could grow at this pH near 4 are inhibited by other factors such as ethanol.
Figure 2. Range of acid pH for the growth of various bacteria, compared to the typical beer pH (Menz et al 2009).
The “bad” lactic acid bacteria of beer
Despite what we have just seen, some bacteria, particularly some LAB, have been able to adapt evolutionarily to the strict beer conditions, and they can survive and spoil them. In particular, the most frequent harmful species against the quality of beers are Lactobacillus brevis and Pediococcus damnosus (Figure 3). The first is the most frequent, and it can give tastes and undesired aromas, as well as turbidity to the final product. P. damnosus has the advantage of growing at low temperatures, and it can also produce undesired aromas, such as diacetyl (Vriesekoop et al 2012). Some Pediococcus and Lactobacillus may adhere to yeast, inducing them to sediment, which delays fermentation (Suzuki 2011).
Figure 3. Lactobacillus brevis (left) and Pediococcus damnosus (right) at the electronic scanning microscope.
Some Pediococcus may also be responsible for the appearance of biological amines in some beers, at risk for the consumer. Amines in a certain concentration are toxic, they may be present in some fermented foods such as cheese, cold meat and alcoholic beverages such as wines and beers, and are produced by decarboxylation of amino acids by LAB. The level of tyramine and other amines has been used as a measure of quality in some Belgian beers made with LAB (Loret et al 2005).
Apart from these LAB, other bacteria related to problems of beer contamination are acetic acid bacteria such as Acetobacter, typically associated with oxygen intake in packaging or distribution. Other harmful bacteria are some enterobacteria, such as Shimwellia pseudoproteus or Citrobacter freundii, which proliferate in the early stages of fermentation, and produce butanediol, acetaldehyde and other unwanted aromatic compounds (Vriesekoop et al 2012). Other harmful bacteria for beer, especially when bottled, are Pectinatus and Megasphaera, which are strict anaerobes, of the clostridial family, and can produce hydrogen sulphide and short chain fatty acids, all of them unpleasant (Suzuki 2011 ).
The “good” lactic acid bacteria of beer
LAB are well known for being some of the microbes that most benefits contribute to the food production, on the one hand as an economic means of preserving food, and on the other hand to improve their quality and organoleptic characteristics. That’s why they are the main agents of fermented foods, along with yeasts. We have seen some of the LAB’s food benefits in other posts in this blog: prehistoric cheeses, or breast milk microbiota, and even wine bacteria.
Therefore, LAB also have a good role in the production of beers: in particular, as we will see below, in the production of acidified malt, and in some peculiar styles of beer such as the Belgian Lambic and the Berliner Weissbier.
As you know, malt is the raw material for making beer. The cereal is subjected to the malting process, where cereal grains, mainly barley, are germinated, the enzymes hydrolyse the starch into sugars, and all of this is then heated obtaining the must, the substrate solution which will be fermented by the yeasts ferment, producing ethanol and carbon dioxide.
The acidification of the malt, that is, with a lower pH, has the advantages of activating many important enzymes in malting, giving a lower viscosity to the malt and therefore to the final beer. Although adding mineral acids or commercial lactic acid can achieve acidification, it is often recommended or legislated a biological acidification, which is achieved by adding LAB. The use of LAB starter cultures is a relatively new process and in addition to the commented benefits on the quality of the malt, it has been shown to also inhibit unwanted molds that are a real problem in malting and that can give mycotoxins. The compounds produced by LAB that can inhibit the fungi are the same lactic acid and the consequent pH drop, bacteriocins, hydrogen peroxide, and other compounds not well known as perhaps some peptides (Lowe & Arendt 2004).
The most commonly LAB strains used to acidify malt are Lactobacillus amylolyticus previously isolated from the same malt. These strains are moderately thermophile, resistant to compounds derived from humulone, and they have the advantage of being amylolytic in addition to producing lactic acid, which lowers the pH (Vriersekoop et al 2012).
Beers with LAB participating in the fermentation, such as Lambic and Berliner Weissbier styles, belong to the type of spontaneous fermentation beers. The other types of controlled fermentation beers are the best-known Ale and Lager, both inoculated with specific yeasts. Ale beers are those of high fermentation, where Saccharomyces cerevisiae yeast used tends to remain on the surface and the fermentation temperature is above 15-20ºC. Lager ones are those of low fermentation, originally from Bavaria, where yeast S. pastorianus (S. carlsbergensis) tends to settle at the bottom of the fermenter and the temperature is between 7 and 13ºC.
Belgian Lambic beer
Traditional Belgian beers (in Dutch lambiek or lambik) are known for their sensorial characteristics due to LAB activity. They are traditional in Brussels itself and in the neighbouring region of Pajottenland, in the Zenne river valley, in the Flemish Brabant on the SW of the Belgian capital. One of the villages in this valley is Lembeek, which could be the origin of the name of this beer.
These beers of spontaneous fermentation represent the oldest style of making beer in the developed world, for some centuries. For a few years now (since around 2008), similar beers are made in the USA, called “American coolship ales” (Ray 2014).
Lambic beer is made with barley malt and a minimum of 30% of non-malted wheat. The cones of a special hops, completely dried and aged for 3 years, are added to the must. They are added not for their aroma or bitterness, but rather as antimicrobial, to prevent above all, the growth of gram-positive pathogenic bacteria in the fermentation broth.
Also to avoid these contaminants and to promote the microbiota typical of the Lambic fermentation, these beers are brewed only between October and May, since in summer there are too many harmful microorganisms in the air that could spoil the beer, and it is necessary to lower the temperature after boiling. Boiling of the must is done intensively, with an evaporation of 30%.
After boiling, the broth is left in open deposits, and in this way the microorganisms of the air present in the fermentation rooms of the brewery (usually at the top of the building) are acquired, and of the outside air, since the tradition says that the windows must be left open. It is assumed that the captured microbes are specific to the Zenne Valley. These open deposits are the koelschip in Dutch (coolship in English), like swimming pools (Figure 4). Being well open, with a lot of surface (about 6 x 6 m) and shallow depth (about 50 cm), they favour the collection of microbes from the room and from the outside. Another purpose of this form is the fastest cooling of boiled broth to start fermentation. They can be made of wood, copper, or stainless steel more recently.
Figure 4. Koelschip (in Dutch) or coolship in English, the open deposits, as swimming-pools, where the Lambic beer process begins (Brasserie Cantillon, Brussels).
The “inoculated” broth in this spontaneous way is left only one night in the coolship, and on the following day this must is pumped into fermentation tanks where there will stay a year, during which the sugar content will go down, up to about 30 g/L. Then it is transferred to oak barrels, previously used for sherry or port, and there it can be left for another two years, at temperatures of 15-25ºC. Some barrels are the same used since 100 years ago. The final product is a cloudy beer, with a pale yellow, very little CO2, dry, acidic, with about 6-8º of ethanol. It reminds a bit like the sherry and especially the cider, and with a slightly bitter taste (Jackson 1999).
In this long process of fermentation, up to 3 years, of course there is a diversity in the composition of the microbial population. In a first phase there is a certain predominance of Kloeckera yeasts and especially enterobacteria during the first month. After 2 months, Pediococcus damnosus and Saccharomyces spp. predominate, and alcoholic fermentation begins. After 6 months of fermentation the predominant yeast is Dekkera bruxellensis (Spitaels et al 2014), or what is the same, Brettanomyces (Kumara & Verachtert 1991), of which Dekkera is the sexual form.
Figure 5. Species of isolates in MRS and VRBG agar media, for lactic acid bacteria and enterobacteria respectively, during the process of making a Lambic beer. The number of isolates is given between brackets (Spitaels et al 2014).
We see (Figure 5) as in particular after 2 months the predominant bacterium is the LAB P. damnosus. It was appointed in the first studies as “P. cerevisiae“, but this name was finally not admitted because it included other species. The count of these in MRS is 104UFC per mL until the end of fermentation. Acidification seems to be rapidly taking place in the transition from the first stage to that of maturation, coinciding with the growth of P. damnosus, which produces lactic acid, although Dekkera/Brettanomyces and acetic acid bacteria also contribute to the acidification (Spitaels et to 2014).
In other trials with the American coolship ales (ACA) of Lambic style, Lactobacillus spp. have also been found, and in a metagenomic study (Bukolich et al 2012) of these ACA, DNA of several Lactobacillales has been detected. At the end of the process, a predominance of Pediococcus (Figure 6, panel C) was also observed. In the same figure in panel A we observe how the predominant unicellular fungus is also Dekkera/Brettanomyces.
Figure 6. TRFLP analysis (polymorphisms of lengths of PCR-amplified terminal restriction fragments) of total DNA extracted from the fermentation samples of ACA beers (similar to Lambic) during 3 years, using primers for: ITS1/ITS4 of 26S rDNA for yeasts (panel A), 16S rDNA for bacteria (panel B), and specific ones for LAB (panel C). Samples marked with * did not give amplification (Bukolich et al 2012).
Lambic derived beers: Gueuze, Faro, fruity and others
The basic Lambic, which is difficult to purchase, is only found in a few Brussels cafes and the production area. In fact, Lambic is the basis for elaborating the others, much more common to consume:
The Faro is a Lambic sweetened with brown sugar and sometimes with spices.
The fruity Lambic are those that have been added whole fruits or fruit syrup. They can be with bitter cherry (kriek), which are the most traditional, or with raspberry, peach, grapes, strawberry, and sometimes also apple or pineapple or apricot or other.
And finally, the Gueuze, which are sparkling and easy to find. They are made by mixing young Lambics (from 6 months to 1 year) with other more mature ones (2-3 years) in thick glass bottles similar to those of champagne or cava and left for a second fermentation with the remaining sugars from the young Lambic. This would have been begun by a mayor of Lembeek in 1870 that owned a brewery and applied the fermentation techniques in the bottle that had been successful in the Champagne some years before (Cervesa en català 2012). The word Gueuze can have the same etymological origin as gist(yeast in Flemish) and it could also refer to the fact that it produces bubbles of CO2, that is, gas (Jackson 1999). However, another historical version would be that this beer was called “Lambic de chez le gueux” (Welsh from poor people) because the mentioned mayor of Lembeek had similar socialist ideas to those of the “Parti des Gueus” founded by the Calvinists from Flanders in the 16th century to fight against the Spanish empire. And since beer is feminine in French, the gueuxfeminine is gueuze, here it is.
In this refermentation in the bottle the populations of Dekkera/Brettanomycesand LAB are maintained, although other unicellular fungi such as Candida, Hansenula, Pichia or Cryptococcus (Verachtert & Debourg 1999) appear in limited numbers.
Figure 7. Several beer Gueuze and fruity Lambic, mostly Belgian (from www.swanbournecellars.com.au/).
The Berliner Weissbier (Figure 8) is another beer relatively similar to Lambic ones. It is also brewed with an important part of wheat must, it is cloudy, acidic and with 3% ethanol. It is traditional in Berlin and the north of Germany, made from the s. XVI and the most popular alcoholic beverage in Berlin until the end of the s. XIX. It was called the “northern champagne” by the Napoleon’s soldiers. Spontaneous fermentation of must involves a mixture of Dekkera/Brettanomyces, Saccharomycesand hetero-fermentative Lactobacillus.
Figure 8. Berliner Weisse beer (from G-LO, @boozedancing wordpress).
Beers similar to Lambic brewed in Spain
In the same way that the commented American Coolship Ales, Lambic style beers are also made in many other countries and, in the case of Spain, coinciding with the boom of artisanal beers, they are also elaborated, especially the fruity Lambic ones. According to the Birrapedia website, 6 of these are currently being processed, all of which are cherries. Two of them are made in Lleida, one in Barcelona, one in Alicante, one in the Jerte valley, and another in Asturias.
Resistance of lactic acid bacteria from beer to hop compounds
Lactobacillus and Pediococcus, both bad and good we have seen, and other contaminating bacteria of beers, have the ability to withstand hop compounds, which, as we have seen, are natural microbiocides. This resistance can be due to various defence systems, both active and passive (Sakamoto & Konings 2003). The active systems include efflux pumps, such as HorA and HorC, which carry the iso-alpha-acids (Figure 1) out the cell. HorA does it with ATP consumption, and HorC using the proton driving force (Figure 9). The corresponding genes horA and horC were originally found in L. brevis, but later they were also found in L. lindneri, L. paracollinoides and in the best known P. damnosus(Suzuki et al., 2006).
Curiously, HorA shows a resemblance of 54% to OmrA, a membrane transporter of Oenococcus oeni, related to the tolerance of this bacterium from wine to ethanol and other stressors (Bourdineaud et al 2004) (See some more about O. oeni in my post on the bacteria of the vine and the wine). Therefore, it is probable that HorA also has functions of exclusion of other compounds aside from those of the hops. It has been seen that these horAand horC resistance genes and their flanking regions are well preserved and have sequences almost identical to the different species that have them. Therefore, it is very likely that some have been acquired from others by means of horizontal gene transfer, by plasmids or transposons, as is usual in many other bacteria (Suzuki 2011).
Figure 9. Mechanisms of resistance to hop compounds in Lactobacillus brevis (Suzuki 2011).
As we see in Figure 9, protons are pumped out by an ATPase, and the consumption of ATPs is compensated by forming it thanks to the consumption of substrates such as citrate, malate, pyruvate or arginine. Another mechanism of resistance, passive in this case, is the modification of the composition of membrane fatty acids, with the addition of more saturated ones, such as C16:0, which reduces the membrane fluidity and makes it difficult the entrance of the hop compounds. This also reminds us of the changes in membrane of O. oeni related to the resistance to ethanol (Margalef-Català et al 2016). The cell wall also changes its composition in the presence of the hop alpha-iso-acids, increasing the amount of high molecular weight lipoteichoic acid, which would also be a barrier. We also see (Figure 9) how hop compounds can lower the intracellular levels of Mn2+, and then a greater synthesis of Mn-dependent proteins is observed, and a greater capture of Mn2+ from outside. Finally, cells of L. brevis reduce their size when they are in beer (Figure 10), probably in order to decrease the extracellular surface, thus minimizing the effect of external toxic compounds (Suzuki 2011).
Figure 10. Effects of beer adaptation (left) in the size of Lactobacillus brevis cells compared to well grown cells in rich media MRS (right). The bars are 5 mm (Suzuki 2011).
All these mechanisms have been studied in L. brevis strains harmful to beer, but it is assumed that the resistance of beneficial bacteria from Lambic and others would be due to the same mechanisms, since they are of the same bacterial species.
As a conclusion to all said, we see that LAB have outstanding roles as beneficial in various aspects of brewery and malting, despite their most known role of harmful in the processing of the most common beers.
Birrapedia (seen 18 august 2018) Cervezas de tipo Fruit Lambic elaboradas en España. https://birrapedia.com/cervezas/del-tipo-fruit-lambic-elaboradas-en-espana
Bokulich NA et al (2012) Brewhouse resident microbiota are responsible for multi-stage fermentation of American Coolship Ale. PLoS One, 7, e35507
Bourdineaud J et al (2004) A bacterial gene homologous to ABC transporters protect Oenococcus oeni from ethanol and other stress factors in wine. Int J Food Microbiol 92, 1-14.
Cervesa en català (2012) Fitxes de degustació – Timmermans Gueuze Tradition http://cervesaencatala.blogspot.com.es/2012/06/fitxes-de-degustacio-timmermans-gueuze.html
Jackson, Michael (1999) Belgium’s great beers. Beer Hunter Online, July 30, 1999
Kumara HMCS & Verachtert H (1991) Identification of Lambic super attenuating micro-organisms by the use of selective antibiotics. J Inst Brew 97, 181-185
Loret S et al (2005) Levels of biogenic amines as a measure of the quality of the beer fermentation process: data from Belgian samples. Food Chem 89, 519-525
Lowe DP & Arendt EK (2004) The use and effects of lactic acid bacteria in malting and brewing with their relationships to antifungal activity, mycotoxins and gushing: a review. J Inst Brew 110, 163-180
Margalef-Català et al (2016) Protective role of glutathione addition against wine-related stress in Oenococcus oeni. Food Res Int 90, 8-15
Menz G et al (2009) Pathogens in beer, in Beer in Health and Disease Prevention, (Preedy, V. R. Ed.), 403–413, Academic Press, Amsterdam
Ray AL (2014) Coolships rising: the next frontier of sour beers in the U.S. First we feast 27 feb 2014
Sakamoto K & Konings WN (2003) Beer spoilage bacteria and hop resistance. Int J Food Microbiol 89, 105-124
Spitaels F et al (2014) The microbial diversity of traditional spontaneously fermented lambic beer. PLOS One 9, 4, e95384
Suzuki K et al (2006) A review of hop resistance in beer spoilage lactic acid bacteria. J Inst Brew 112, 173-191
Suzuki K (2011) 125th Anniversary Review: microbiological instability of beer caused by spoilage bacteria. J Inst Brew 117, 131-155
The Beer Wench (2008) My obsession with wild beers. Nov. 20, 2008 https://thecolumbuswench.wordpress.com/tag/lambic/
Verachtert H & Debourg A (1999) The production of gueuze and related refreshing acid beers. Cerevisia, 20, 37–41
Vriesekoop F et al (2012) 125th Anniversary review: Bacteria in brewing: the good, the bad and the ugly. J Inst Brew 118, 335-345