The Family Tree of Yeast
Researchers Draw Genomic Tree of Saccharomyces cerevisiae
Per Kølster, who grows his own raw materials to brew beer in the countryside outside Copenhagen, Denmark, headed east several years ago to learn more about traditional farmhouse brewing. In Lithuania he made beer with a local farmer, and when it came time to pitch yeast, they walked to a neighboring farm to collect what they needed. On the way the farmer told Kølster not to say “thank you” for the yeast. He explained that because no one owns yeast, it must be available to anyone and saying “thank you” would disrupt this system.
Commercial breweries still share yeast, but not as many and not as often. Until 1980 a guard at the Carlsberg brewery gate in Copenhagen handed out small quantities from a “yeast tower” to locals who asked. Kølster points to this tower as an example of what the word “culture” can mean. It is quite opposite “the pasteurized approach to culture,” he wrote via email. “Diversity and resilience versus controlled purity and managed intensity.”
It would seem there’s more than one fermentation narrative in brewing these days. Some brewers are intent on exploring diverse flavors often described as sour or funky, the result of spontaneous fermentation, mixed fermentation or fermentation with yeast isolated from the wild. Others expect new pure strains from commercial laboratories to provide diversity. And there are those who are content with the flavor profiles available but are seeking varieties that can broadly be categorized as more efficient.
Serving all of these groups is a genetic family tree for Saccharomyces cerevisiae (one of 1,500 known species of yeast) drawn by research teams from White Labs in California and a Belgian genetics laboratory. Reconstructing what occurred in an older world, the genomic tree—which includes two distinct beer lineages—gives brewers the tools to create a new world.
“To me, this is just the beginning,” says White Labs founder Chris White. “Understanding the DNA is going to be the first part of the story. There is so much information we’ll be sifting through. There is much more to come out of this.”
The first results were published in the scientific journal Cell last September. A headline in The Economist summarized them with typical British restraint: “Domesticated Tipple.” The Washington Post was more expansive: “Beer yeast is tame. Wine yeast is wild. Draw your own conclusions.” The headlines suggest shrinking diversity, but that is not necessarily one of the conclusions of the research. Instead, the authors write that one of the results from differences between beer brewing and winemaking has been a “large genetic diversity within beer yeasts, while wine yeasts are genetically more homogeneous.”
Researchers in Australia came to a similar conclusion after sequencing the genomes of hundreds of strains of the wine yeast. They discovered yeast strains sold by different companies were almost genetically identical. “Our results show that only a limited branch of the yeast evolutionary tree is currently used in winemaking,” says Anthony Borneman of the Australian Wine Research Institute.
Scientists first sequenced the species of yeast used by brewers and bakers in 1996, determining the order of all 12,057,500 chemical subunits contained in the yeast’s nuclear DNA. It was a step toward sequencing the human genome, a project that took almost a decade and cost about $3 billion by the time it was completed in 2000. Advances in technology since made the process quicker and less expensive, cheap enough that in 2012 Illumina, a San Diego biotechnology company located not far from White Labs, sequenced 96 strains free of charge in order to test new machinery.
“Sequencing used to be the hard part. Not anymore,” White says. “Now it’s the time to assemble all the information, to link DNA to phenotype.”
Strains that are closely related genetically may behave quite differently in beer. “Things were more different than I expected,” says Troels Prahl, head of research and development at White Labs. “Part of me was hoping we find only a few (unique varieties), but we did not find many duplicates. The diversity is real. We’re not doing 100 strains every week just because we want to.”
The collaborative research began not long after White Labs started working on this project in 2012. Prahl was speaking at a conference where he learned a Belgian group was also exploring the phenotypic landscape of yeast. Together the two teams sequenced the genomes of 157 industrial S. cerevisiae isolates. This collection includes 102 commercial beer strains, 19 wine strains, 11 spirit strains, seven sake strains, seven strains isolated from spontaneous fermentations, five bioethanol strains, four bread strains and two laboratory strains. (When used to preface beer the word industrial often has negative connotations, but White explains scientists use it to describe yeast selected for distribution for commercial purposes.)
The results illustrate that modern S. cerevisiae yeasts (because they were the focus of the research, from here on they will simply be referred to as yeasts) are distinctly different from wild yeasts because of human selection and, sometimes, geography. Domesticated beer yeasts differ from wild strains, as well as those used for making wine or bread in several significant ways:
- They are more efficient at converting maltose (the malt sugar brewers create) into alcohol.
- They flocculate, or settle, better, resulting in beer that is more naturally bright.
- They no longer mate, so are less likely to change when brewers reuse them.
- Most do not produce phenols that may be described as smelling of anything between spicy and electrical tape.
Two Distinct Lineages
Although it wasn’t until the 19th century that scientists widely recognized that yeasts were microbes and responsible for fermentation, bakers, brewers and winemakers have been producing bread, beer and wine for thousands of years. They learned that backslopping, an appetizing term that describes a process in which they inoculated unfermented foods or drinks with a portion of fermented product, resulting in predictable fermentations. The report in Cell suggests that growing continuously in a man-made environment led to domestication, and that there are two distinct beer lineages. The first dates to between 1573 and 1604. This coincides historically with a shift from brewing primarily in the home to commercial brewing on a larger, in fact industrial, scale. The second lineage likely originated between 1645 and 1671.
The first, labeled Beer 1, includes three separate, geographically distinct, groups. It originated in the region around what is now Germany and Belgium, spread first to the United Kingdom and then to the United States. The data show that U.S. beer yeasts were imported from Europe during colonization, rather than stemming from indigenous wild U.S. yeast. They are most closely related to British beer yeasts, suggesting that the origin of U.S. brewing strains can be traced to the introduction of beer culture in the United States by early 17th-century British settlers.
“We deliberately obfuscated which yeast came from which specific brewery, to protect the brewers,” says Kevin Verstrepen, a yeast geneticist at the University of Leuven and the Flanders Institute for Biotechnology, because many brewers “consider their yeasts something of a top-secret ingredient.” Thus there is no branch of the tree labeled “Duvel” or “Sierra Nevada.” And even though the researchers have placed White Labs California Ale Yeast (001) in the Beer 1 group, that doesn’t mean that the other quintessential American ale yeast, Wyeast American Ale (1056), belongs there as well.
However, when Illumina first sequenced yeasts from White Labs, White had them compare a few strains from other labs with those from his own. “California ale yeast is so important to us we did it for fun,” he says, discussing 001 and 1056. “It turns out that they are different,” he says. “Which I’ve been saying all along.” However, when Illumina compared other strains that were said to come from the same industrial sources as White’s, they did turn out to be the same.
Beer 2 yeasts are more stress-tolerant and therefore more successful fermenting higher alcohol beers. They are more closely related to wine yeasts than Beer 1 strains; Beer 2 is placed next to wine in the circular polar format the researchers chose for their phylogenetic tree. Beer 2 includes one-fifth of brewing strains, but unlike Beer 1 they lack geographic structure and contain yeasts originating from Belgium, the United States, the United Kingdom, Germany and Eastern Europe. The research hints that two independent European “domestic events,” one of which is the origin of both Wine and Beer 2, resulted in the two beer groups.
A certain amount of mystery remains, particularly when it comes to one shared trait. They both contain Phenolic Off Flavor Positive (POF+) yeasts and Phenolic Off Flavor Negative (POF-) yeasts. POF+ yeasts interact with ferulic acids from malted grains to produce 4-vinyl guaiacol, which imparts a clovelike character to beer that is considered positive in German wheat beers (hefeweizens) and some beers of Belgian origin, but otherwise is generally an off flavor.
Wild yeasts are POF+, but the report indicates that medieval brewers must have selected POF- yeast early on, Verstrepen explains, because they were most likely the starting point for domestication. “The most intriguing questions we cannot answer yet is how they did this,” he says. Every fermentation will contain a few cells that turn POF- because of spontaneous mutation, but those would “only be one or a few cells in a gazillion POF+ cells.” Yet it happened for both the Beer 1 and Beer 2 families.
“The POF+ yeasts used for some beers seem to have regained their POF characteristics, rather than never losing them, likely because they crossed with a POF+ yeast, or at least received some part of DNA of one,” he says.
‘A New Way’
Scientists now have what amounts to cut-and-paste tools for altering DNA, so it would be easy enough to flip POF-related genes in any yeast on or off. Two years ago, a microbiologist in the same lab as Verstrepen told The New York Times, “Right now we have a few hundred genetically modified yeast strains patiently waiting in our laboratory’s freezer.” However, the brewing community is firmly united against genetic modification (GM). White Labs, for instance, has banned GM yeasts from its premises, for fear one might accidentally mix with one of its house strains. Verstrepen’s lab has taken a different approach to creating new strains. Even though its research found 40 percent of commercial beer strains incapable of reproducing sexually, and the others showed dramatically reduced sexual fertility, robots are at work in the lab to breed such strains. “We have really optimized the conditions so that strains that have very poor sexual cycles can still be persuaded to breed; it is all about tweaking the environment,” Verstrepen says. “I want to stress that this is a very natural process, completely the same as what farmers have been doing and are still doing. But the robots allow us to progress much faster, so that we can hopefully catch up to and even surpass the breeding of crops and livestock.”
The lab has already sold new strains to several breweries, some of them tiny and others multinationals. “What customers want changes. This is an example of that. We’re going to get yeast that will create styles we haven’t thought of yet,” White says. However, White Labs did not begin the project with a singular goal to produce new strains. “For me it was always about the science and information. We wanted to understand these strains better,” he says.
Speaking at Homebrew Con, the annual gathering of American homebrewers, last June in Baltimore, White talked about how quickly understanding was changing. Ten of the S. cerevisiae beer strains in the joint project are used for commercial production of lager beers rather than “traditional” lager strains, or Saccharomyces pastorianus, which is itself a hybrid of S. cerevisiae and S. eubayanus. Obviously, much more research is needed there as well. Lagers have long constituted the majority of beers brewed, but the origins of their yeast strains have been a mystery. S. eubayanus was only discovered in 2011, first in the wild in Patagonia, but elsewhere since. Jürgen Wendland, a former yeast biologist at the Carlsberg brewery, recently wrote that “genome sequencing of lager yeast is only at its early beginnings.”
White put it more dramatically in Baltimore. “Without unlocking the genetic information, we are still thinking like the 1860s,” he said. He showed a slide with S. cerevisiae-ale yeast-“top fermenting” on one side and S. pastorianus-lager yeast-“bottom fermenting” on the other. “I’m glad you’re coming to this talk because we are kind of on the brink,” he said. “This is the old way of talking about this. There is going to be a new way in the next few years.”
When some brewers talk about “wild yeast,” they include Brettanomyces, although they most often buy it from a laboratory, as well as Lactobacillus and Pediococcus, which are not yeast but bacteria. Jasper Akerboom, who oversees quality control at Lost Rhino Brewing in Virginia and also operated Bright Yeast Labs, is more specific.
He’s isolated more than 100 single strains of Saccharomyces cerevisiae, collected in the wild, that will ferment beer. All share what he calls outspoken character. “I have not found anything that is really mild,” he says. Instead they tend to be “wild,” estery and often needing to ferment at higher temperatures. Most of them do not flocculate very well. They are all POF+, so better suited for brewing saisons than pale ales.
They are much like ones brewers would have chosen 500 years ago—except those existed in a “mixed culture” that included more than one strain of S. cerevisiae as well as Brettanomyces and bacteria. This new interest in wild yeast provides scientists with an opportunity to trace how a strain develops as it becomes domesticated.
In Copenhagen, where Troels Prahl works, White Labs released a strain isolated from apples spontaneously fermented on a remote island off the coast of Denmark. “The goal is to end with [pure] local yeasts that are suited for brewing and safe for consumption for each of the seasons at a certain habitat or region,” says Prahl. The yeast ferments at a cool, lager-friendly temperature, creating a beer that is “nice and dry, but aromatic like no lager you’ve ever tasted.”
Stan Hieronymus gives full credit to Mike Karnowski of Zebulon Artisan Ales for wryly stating last year that, “Twenty years ago we had everything figured out.” Because now we know we don’t, which is what has made writing for All About Beer Magazine since 1993 so much fun.