Inquiring Winemaker


Ongoing Search for Low-Yielding Yeast

June 2011
by Tim Patterson
Three or four years ago, the high-octane section of the wine world buzzed with speculation that newer commercial yeast strains somehow yielded more alcohol than old standbys. By extension, some theorized, the yeast was responsible for rampaging alcohol levels in wines from California and elsewhere.

The short scientific response to this line of argument was no, the yeasts aren’t raising your alcohol levels, it’s all that sugar in the grapes you insist on hyper-ripening.

That debate has mostly disappeared from the public stage, replaced with a concern that’s almost a mirror image. If the yeasts aren’t making more alcohol, are there some strains that can yield less? Might a bit of inefficiency be a virtue?

And indeed the quest for low-yielding yeasts has been a major research preoccupation in both academic enology departments and commercial yeast research labs. The results so far: a lot of fascinating science, but no silver bullets.

Yeast 101
In fermentation mode, yeast converts one sugar molecule to two molecules of ethanol and two molecules of carbon dioxide. This activity is wired pretty deep into the genome; it’s what yeasts do. And herein is the problem with the theories about super-yeast that could procure more ethanol: They would somehow have to create 2.5 molecules of ethanol (or some such number) from the same sugar molecule, thus violating the laws of chemistry and of second-grade math.

Modern commercial yeast strains do have the ability to ferment high-Brix juices dry. Most have been selected for their ability to reliably complete a fermentation, on riper and riper grapes, but that increased ethanol tolerance does not change the conversion ratio: The molecule math stays the same.

In 2007, wine yeast producer Lallemand  conducted a study in which the ethanol yields of a total of 113 commercial yeast strains were measured. And among those who finished the job to dryness, the widest difference in final alcohol was 0.51%.

Additional studies, by the Australian Wine Research Institute and others, also have found the widest gap in commercial yeast strains to be in the 0.3-0.5% range.

Linda Bisson and her lab crew at UC Davis have taken the comparison contest a bit further. By including non-commercial Saccharomyces cerevisiae strains—vineyard isolates, strains that failed to succeed commercially, etc.—they have seen a range as wide as 1.5% among strains able to complete a fermentation. Since that disparity could mean the difference between an old-fashioned 13.5% Cabernet and a newfangled 15 percenter, it suggests that figuring out the metabolic differences could pay off.

Updating the 0.55 rule of thumb

All this focus on ethanol yields has led to a re-calibration of the rule of thumb on sugar-to-ethanol math. The old industry standard, multiplying the starting Brix by 0.55 to get the final percentage of alcohol, may have been accurate in the bad old days of uncontrolled, open-vat fermentations without any temperature control, when a good deal of alcohol went volatile and disappeared.

These days, with refrigeration, closed-tank fermentation for both whites and many reds, and gentler pump overs and punch downs, not as much ethanol is vaporizing. Something more like 0.60 or 0.62 is a better multiplier, and occasionally a fermentation gets close to the theoretical maximum of 0.65. Cool-fermenting whites in closed tanks may retain slightly more alcohol than warm, open-top Pinot fermentations, but there are enough other factors in the mix to make that less than a reliable generalization. Remember: This rule of thumb has climbed up not because the yeasts are making more alcohol, but because the winemakers are holding onto more of it after the yeast do their work.
Making less with more
Yeast are determined creatures, and if something in their genetic code or environment makes them produce less ethanol, they will produce more of something else instead. Those sugar molecules won’t just be frittered away.

One of the things yeast can produce more of is yeast. The carbon energy source in the glucose and fructose can be put into ethanol, or into more biomass. Bisson’s crew has observed that some of the high-ethanol-producing strains rapidly build up biomass during the early part of a fermentation and then switch off reproduction and concentrate on pumping out ethanol until they run out of fuel. Some low-ethanol producers just keep budding and birthing, finding a kind of safety in numbers and domination of the fermentation in relation to other microbes.

Somewhat ironically, the strains that go for biomass make themselves more vulnerable to ethanol toxicity, while the strains focused on ethanol enjoy greater resistance, even with their smaller numbers.

Besides making ethanol and carbon dioxide, yeast also produces small amounts of many other compounds. Strains can be isolated, created through conventional breeding or concocted by GMO techniques that yield less ethanol and more of something else, especially glycerol.

Some of the experimental research has involved crossing commercial and “wild” cerevisiae strains, some breeding with non-cerevisiae strains, some with non-Saccharomyces yeast. A number of these trials have been nominally successful, reducing alcohol yields by as much as 1.5%.

Trouble is, in order to get that much less alcohol, you’re likely to get way more glycerol than you bargained for. A little extra glycerol is great for red wine mouthfeel; a lot of it is not so nice. Here’s a portion of an email from Lallemand researcher Anne Julien-Ortiz:

“We know that through GMO strategy it is possible to push the yeast to overproduce glycerol by the overexpression of the genes GPD1 and GPD2; in this way, the final ethanol level will decrease a little bit. For example, to have about 1.5%-2% less alcohol, the yeast needs to produce at least 25g/L of glycerol—which is a lot! At the same time, the yeast will ove rproduce acetate (volatile acidity) as well. Once the genes ald4 and ald6 (responsible of this production) are deleted, there is another problem: accumulation of acetoin, which is bad for wine quality (very strong lactic notes, off flavor.)

“When you force the yeast to do something else, she will find another way, but one that won’t be especially positive for the wine sensory profile. So it’s really challenging.”

In brief, be careful when you mess with Mother Nature. Some examples of the low-ethanol, high-glycerol strategy yeasts (all of them created through conventional breeding) are commercially available, but none of them are taking the high-alcohol sector by storm. The consensus among the researchers I contacted—Bisson, Julien-Ortiz, Charlotte Gourraud of Laffort and Karien O’Kennedy of Anchor in South Africa—was that a great deal has been learned about yeast metabolism, but that no major breakthrough was on the horizon.

For the moment, the two best places to hold alcohol down are still in the vineyard or the reverse osmosis machine.

Tim Patterson is the author of the newly released “Home Winemaking for Dummies.” He writes about wine and makes his own in Berkeley, Calif. Years of experience as a journalist, combined with a contrarian streak, make him interested in getting to the bottom of wine stories, casting a critical eye on conventional wisdom in the process.

Currently no comments posted for this article.