Insights into the evolutionary history of brewing from 157 yeast genomes

The Homebrew Forum

Help Support The Homebrew Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

IainM

Landlord.
Joined
Oct 19, 2015
Messages
2,653
Reaction score
1,455
Location
Maidenhead
OK, so when I got into brewing I thought I'd go and find some papers on yeast genetics, expecting that hundreds of strains used in commercial brewing would have had their genomes sequenced. I was surprised to find that they there really wasn't much in the literature about this. Well, that has now changed with a paper that was published in Cell last week by teams from the US and Belgium, who sequenced and analysed the genomes of 157 species of yeast, mostly commercial brewing strains. It throws up a few interesting facts which, while it might not improve your beer, you might find interesting. For instance, it was previously thought that brewers mainly used two different species of yeast, top-fermenting ale yeasts and bottom-fermenting lager yeasts. However, the study revealed that 10 commercial lager yeasts were in fact S. cerevisiae, and not S. pastorianus. Anyway, I thought I'd summarise some of the take-home messages.

They found a lot of evidence of yeast domestication, which stretched back hundreds of years, but certainly not back to the dawn of brewing. This is the point at which people switched from fermenting with wild yeast to re-using yeasts from previous batches, and preferentially used yeasts that produced good beer. From this point on, the yeasts changed dramatically to fill their new niche away from the wild and in the nutrient-rich environment of worts and musts. They started jettisoning genes, even whole chromosomes, so that they had less DNA to replicate and thus replicate faster and, thankfully for us, outpace their wild brethren. They lost genes associated with living in the tougher conditions in the wild, and so brewers yeasts became unable to prosper outside of the sweet alcoholic liquids they had become accustomed to, and lost the ability to increase their genetic diversity by reproducing sexually. This effect was more pronounced in beer yeasts than wine yeasts, as the wine yeasts still had to survive in sub-optimal conditions between grape harvests, and ferment in more alcoholic and nutrient deficient conditions, while the beer yeasts had the pleasure of a less alcoholic and more nutrient rich medium, and due to the longevity of grain and year-long production of wort, didn't have to spend long times in hibernation.

Beer and sake yeasts saw an increase in the efficiency and number of copies of the MAL genes, which encode enzymes which break down maltose in wort and rice into simple sugars. They also saw mutations in genes in metabolic pathways which result in 'off' compounds, showing that our forefathers re-used the yeasts that made beers that tasted better, just as we re-use our favourite strains. More recently, wine yeasts have evolved a resistance to suphite, giving them a head start over wild yeasts in the must, the latter of which are more heavily stunned by the addition of campden tablets. In general, domestication also had profound impacts in the way the yeasts metabolise nitrogen and carbon, how they use ions, and, of course, their ability to flocctuate.

The authors build a evolutionary tree from the genomes, and showed that there were five main branches. The most ancient divergence is the Asian branch, containing species used to make sake, and which in fact are more closely related to strains developed from wild Asian strains for bio-ethanol production. The fact that sake and beer strains separately developed the same mechanisms to adapt to a high-maltose environment is a sign of convergent evolution; different lineages coming up with the same solution to a problem. Another set of lineages is the wine branch, mostly species used to make wine, but also a few strains used to make spirits and beer strains mixed in there. A third branch is the mixed branch, which contains all the bread yeasts, some spirit yeasts and 'wild' yeasts, and some beer yeasts. It probably isn't surprising that the beer yeasts in this group, which clearly aren't as well adapted to specific tasks as the others, are Belgian. The other two branches are the beer yeasts. Surprisingly, they are not at all closely related to each other, which indicates that there have been two separate domestication events for beer yeasts. This split, however, doesn't seem to be along geographic lines, as both groups had members from all main beer-producing regions: the decedents of both had spread across the world from Europe, their place of origin. While one of these branches of beer strains are more closely related to the wine strains, and tend to have a higher alcohol tolerance, the other lineage is the one that has been used the longest for making beer. Once established in Britain and Germany/Belgium, the strains continued to be used in these areas without much exchange between them, giving two main sub-branches, one German/Belgian and the other British. The British branch has another more recent split, in which the descendants all constitute American strains. This shows that American beer yeasts are not native to the US, but came with British settlers in the 17th Century.

The authors calculate that the diversity they observed in domesticated beer yeast spans around 75,000 generations of usage in making beer since the late 1500s, which corresponds to the time that beer making changed from small-scale home-brewing to large scale production, first in monasteries and then in breweries. They estimate about 150 generations per year, with three doublings per batch of beer. Most of this, of course, was before White Labs, Wyeast or any other company were storing strains, and so they were in constant use batch after batch. So next time you crack open a home-brew and hold it up to the light, you know that the yeast in the bottom is just one tip on a vast evolutionary tree, and has travelled the world and been shaped through the generations in the hands of monks and brewers. It is as much a product of your choice of grain and hops and it is the product of the choices these brewers made in the 25,000 or so batches of beer that came before it.
 
Actually, I say that it probably won't help with your brewing, but there is some interesting stuff in the phenotype data. For instance, although most the beers don't have a good salt tolerance, some of the strains, particularly ones descended from the more recent beer domestication event, thrive in a salty environment. They may be good candidates for making gose. It's a shame that the strains aren't reported in a way that would make it easy to find their equivalents available online.
 
Thanks Ian, I'll grab the article soon and give it a proper read but this is an excellent summary.

I believe this work has been done by White Labs. I heard an interview on the Beersmith podcast (I believe) with Chris White but it was still prepublication.

He did say that their intention is to be able to let people chose yeast strain based on phenotypes. So you could ask for a yeast with high alcohol tolerance and high ester production but low phenolic production they will be able to point you in the right direction.
 
Yeah, a couple of the authors are from White Labs, including Chris. The work seems to have been driven by the guys in Belgium though. Its strange that the product codes were scrubbed so we don't know which strain corresponds to whicin their catalog.
 
I have made it a sticky for the effort that went into posting it and the likes it got. :thumb:
 
Awesome post Ian! I'm super guilty when it comes to yeast of just sticking to us-05 and nottingham and only looking at grain/hop bill for flavor mixing. Really need to get into yeasts more. Anyone recommend a good book for the homebrewer?
 
Back
Top