GMO yeast in wine and how to find them

The vast majority of wine does not involve genetically modified organisms (GMOs). Let me repeat, the vast majority of wine does not involve GMOs. On to the rest of the story:

Whether wine contains genetically modified organisms (GMOs) is a question I’m asked often. In general, the answer is no. Genetically modified grapevines aren’t being used for commercial winemaking (though not for want of trying). Two genetically modified wine yeasts have crossed the commercial production threshold, but not worldwide. One, the un-charismatically named ECMo01, available only in the United States and Canada, has been engineered to produce an enzyme that degrades urea. That’s a useful property because urea in wine can become ethyl carbamate, which the World Health Organization thinks is probably carcinogenic enough to be worried about it.

The other, ML01 (which rolls off the tongue much more easily), is legal in the US and Canada as well as Moldova, and seems to have won more traction (though not, I dare say, because it’s available in Moldova). ML01 includes genes for two non-Saccharomyces cerevisiae proteins: a malate permease from fellow yeast Schizosaccharomyces pombe, and a malolactic gene from the lactic acid bacteria Oenococcus oeni. Together, those molecules allow ML01 to import malic acid into the cell and convert it into lactic acid, granting ML01 the rather magical ability to perform both alcoholic fermentation and malolactic fermentation simultaneously, all by itself. In addition to speed and convenience, this one-stop fermentation is advertised as a route to fewer wine headaches. Lactic acid bacteria can produce biogenic amines, which can produce headaches and other unpleasantries in sensitive people (I’m one of them); eliminating the need for those bacteria should eliminate the biogenic amines and those symptoms.

For reasons which are probably obvious, North American wineries using these GM yeasts don’t exactly go shouting that news from the rooftops, fewer headaches or not.

What if you wanted to identify whether or not any given wine was made with a genetically modified yeast? You’d go looking for the modified gene, right? This isn’t as simple as it sounds, and not just because genes are very small. The genes that distinguish ML01 and ECMo01 are also found in other common wine microorganisms; detecting ML01, for example, means ensuring that you’re not just detecting the presence of perfectly normal malolactic bacteria.

The authors of a recent paper in the International Journal of Food Microbiology handled these problems with a conceptually simple solution to identify ML01 in mixed microbial company. They used PCR – the polymerase chain reaction, or the standard means of “looking” for genes that constitutes the bread and butter of virtually every molecular biology lab these days. PCR amplifies a very specific DNA sequence, determined by “primers” that line up with the genetic sequence you’re looking for, so that even a tiny amount of that genetic sequence in a sample can be detected. By choosing those primers to line up with the joints at either end of the signature ML01 genes – the scars left over from its engineering procedure, essentially – they could target the engineered yeast to the exclusion of both O. oeni and unmodified S. cerevisiae. By using quantitative PCR, which adds some fancy fluorescent chemistry to the basic PCR process to provide a rough idea of how much of that specific genetic sequence is in a sample, they could distinguish between large quantities of ML01 indicating that it was used for primary fermentation versus small quantities suggesting accidental contamination

The goal, in this paper, was to offer a means of establishing whether GM yeasts are being used illicitly in countries where they’re illegal as well as a test against ML01 contamination in factories where it might be produced near non-GM strains.

That’s how to find GM yeast in wine if you’re a biologist. If you’re not, you’re left with less precise methods. One: exclude wine from outside North America and Moldova. Two: exclude organic wine and wine from companies which expressly declare themselves non-GM-users. Three: recognize that the general ethos of a winery probably gives you a good idea of how likely they are – or aren’t – likely to use ECMo01 or ML01. Four: invest in a PCR machine.

*See, for example, this Australian report from 2003, or this Cornell proposal from 1996, along with numerous research projects investigating the concept.


Have we domesticated yeast? Yes.

Common sense says that winemakers – and beer brewers, and bread bakers – were developing specialized Saccharomyces cerevisiae yeasts a good long while before Red Star marketed its first dried and packaged commercial product to the industry in 1965. Winemakers weren’t inoculating ferments with an aluminum foil packet they bought at the store, but that doesn’t mean they weren’t inoculating, maybe with a little bit of an already-active ferment, maybe just by having a conducive winery environment where the right kinds of yeast were happy to make a home. Either way, the yeast you’d find in any given winery or brewery weren’t the same as the yeast you’d pick up off the street, or the same as what you’d find in the next alcoholic beverage factory down the road.

Plenty of evidence, old and new, supports that story. But did those yeast become different simply because they were isolated from each other, like Darwin’s famous Galapagos finches? Or did they change because they became domesticated, because brewers and winemakers cultivated and selected them? In other words, what kind of difference did the humans make to the yeasts’ evolution?

The theory basically goes like this. If yeast populations developed in different ways just because they were physically separated, then their genomes should look like what you expect from “wild” yeast. If humans domesticated them, they should be less genetically fit, because they’ve grown accustomed to being specially cared for and protected by humans and have lost some of their capacity to live on their own.

Continue reading

Yeast odd-balls: Trying to understand flor

Flor yeast are something like that one strange guy with the office down the hall, the one who’s always pleasant, who has a kind word when the rest of the world seems out to either hate or bother you, who sometimes holds the door open when you’re carrying coffee, and who does goodness-only-knows-what when he closes the door to that office of his. You’re really glad he’s there, but he’s a mystery.

I’m stretching, but you get the point: flor yeast are useful, we’re glad they’re there for us, and we don’t understand them. Flor are the film-forming, surface-dwelling yeast best known for their role in making sherry, though they’re also key actors in Hungarian dry Szamorodni**, Xeres from Jurez, and Jura’s Vin Jaune. “Biological aging” — the yeast metabolizing and spitting out acetaldehyde, glycerol, a bit of ethanol, volatile acids, and some other stuff — is responsible for much of the unique and complex nutty-freshness these wines share. Though it’s entirely possible to purchase and inoculate flor yeast, the film often comes and goes on its own in traditional production schemes. All of this happens after ordinary alcoholic fermentation finishes.

Flor are Saccharomyces cerevisiae, but they’re evidently not the same yeast responsible for alcoholic fermentation in these wines. Previous flor research also tells us that being flor is more about a lifestyle choice than about the yeast’s innate characteristics. Being flor is about surviving under stress: the yeast develop as a “velum” or skin on a high-alcohol (fortified) wine with little remaining nutrients. Yeast strains that survive here are generally those that have been able to make a few rapid, key genetic changes that researchers see as a common pattern across flor. By definition, one of those changes is the ability to form a film or skin on the wine’s surface, which signifies that the yeast have become more hydrophobic than usual: they don’t mix well with water, so they stick to each other and find the least-liquidy environment they can…given that they’re living off of a liquid.

Flor aren’t exactly like that guy down the hall, unless he’s been the subject of 34 studies published in the American Journal of Enology and Viticulture. Even still, how and why flor come and go isn’t something we understand well, and neither is what exactly makes flor yeast flor in the first place.

Wine microbiology research is often very local: researchers study the vineyards and wines near them (they’re convenient, but they’re also, often, the primary interest of whichever regional body is helping to fund the research). A study of flor yeast recently published in PLoS One (which means that it’s open-access; no paywall!) is notable for including samples from Hungary (Tokaj), Spain (Jerez), France (Jura), and Italy (Sardinia). This group of French enologists set out to build a better genetic tree for these yeasts, both to see how they’re related and to establish more data about what makes them flor-ish. My genetics knowledge is too poor to comment on how their data look, but it’s satisfying to note that they used more than one method.

How they’re related — The strains from different locations formed genetically similar groups: all of the strains from Jura were more similar to each other than to the strains from Jerez, and Jura strains were similar to other strains from the same cellar, too. My genetics knowledge is too poor to comment on how their data look, but it’s satisfying to note that they used more than one method to judge relatedness and kept coming up with the same relationships.

What makes a flor yeast a flor yeast — The best that can be said here is that this study makes one more incremental step forward in solving a big, complex problems. The researchers found two genetic changes shared across all of their flor yeasts, involving the duplication of two known genes. We know they’re genes; we don’t know much about what they do. So goes the story of modern genetics, caught between the past of not understanding how DNA mapped to functions at all and the future in which (we hope and presume) we’ll know what all (heck; I think we’d settle for most) genes actually do. This study also reinforces some previous findings about genetic characteristics shared by flor which, again, all point toward them having some kind of shared ability to deal with their high-alcohol, low-nutrient, clingy lifestyle. Gosh, now I feel as though I’ve really insulted that guy down the hall…


** An earlier version of this piece wrongly named the Hungarian wine as Tokaj Hegyalja, which is how the authors of the paper name that wine. A reader working in the Tokaj Hegyalja wine industry helpfully pointed out that Tokaj Hegyalja refers to the region and that the name of the wine I’m calling out is actually dry Szamorodni. She suggests that good descriptions of this wine can be found on the producer Samuel Tinon’s website. Thanks, Katherine.