Innovation in sparkling wine production: Trust the yeast

Numerous recent studies have been playing with how yeast can work above and beyond the usual call of duty in sparkling wine production. The Australian Wine Research Institute’s (AWRI) superb yeast biologist Jenny Bellon continues to convince yeast to reshape itself to our needs by breeding across the usual species lines.* Hybrid yeast (open-access article) with a Saccharomyces cerevisiae strain as one parent and a Saccharomyces mikatae or other close cousin Saccharomyces species as the other, generate different secondary metabolites compared with conventional straight-up S. cerevisiae strains, and we somehow end up interpreting that difference as “complexity,” and liking it.

The goal in those cases is to produce new and different (and better) flavors by using these more metabolically complex yeasts for tirage or in-bottle fermentation.

The interesting thing about tirage yeast, though, is that they do a good portion of their winemaking work after they die. While alive, yeast are useful for their insides: the enzymes they house convert sugar to alcohol and numerous other valuable metabolites. In dying, yeast are useful for their outsides: they release mannoproteins from their cell walls that improve wine quality in numerous ways, by enriching mouthfeel, by stabilizing mousse, and by adding lovely bready or toasted aromas. (Find more detail on those effects in this embarrassingly badly written article from 2012).

When yeast cells die, they don’t just turn off; enzymes split open the cell from the inside (autolysis), releasing good-for-winemaking compounds. However, autolysis happens inefficiently under standard winemaking conditions: yeast are most inclined to self-sacrifice around pH 5 and at warmer temperatures; sparkling wine generally sits below pH 3 and is fermented cool. Here’s where two recent scientific studies about innovating in sparkling wine production meet.

Temperature: A group of scientists in Franciacorta (paywall) worked with a local winery (Villa Crespia) to ask: if chilling (heat removal, more accurately, as they point out) consumes something like 90% of the energy the winery uses, and the reason for keeping the wine cool is to improve sensory qualities, then can we find a yeast that yields an equally good product at higher temperatures? Yeast adapt** to environmental changes quickly and well; ask in the right way and you’re likely to find one that suits your needs. They did. The article describes a S. cerevisiae strain that ferments at 19C just as well as a more conventional strain does at 15C: just as fast, producing only low levels of SO2, and with comparable sensory qualities. Their panel of experienced judges wasn’t able to reliably distinguish the cooler-fermented wine from the warmer-fermented one, and the warmer fermentation used 65% less energy.

This study reminds me of the numerous small wineries I’ve visited where winemakers don’t chill their white fermentations at all, because they don’t have the money to spend on that kind of equipment, and/or because it’s not a priority. Even if this study is only directly relevant for a small group of people, it’s a good reminder of two things. One: good developments can come from questioning your basic assumptions. Two: yeast don’t necessarily share your assumptions.

Autolysis: Before I go anthropomorphizing yeast too much, let me summarize a second article addressing the “we want more autolysis” problem more directly. If sparkling wine benefits from having more dead yeast cells in the bottle during secondary fermentation, why don’t we take advantage of yeast’s natural inclination to kill each other? Some yeast secrete attack molecules that kill other, susceptible yeast (any one strain’s toxins don’t work on that same strain, obviously, and so can only work on a subset of all strains). This “killer” phenomenon causes real problems when it happens unexpectedly and causes a stuck fermentation. But if we want more dead cells around, could a mixed-inoculation strategy combining killer strains with good fermentation capacities and susceptible strains (with good dying kinetics, you might say) produce a completed fermentation with desirable sensory characteristics? The authors of a paper (paywall) using Macabeo wines and yeast isolated from Spanish wineries say yes: combinations of killer + non-killer yeasts can complete sparkling wine secondary fermentation fast enough to be practical and generate higher numbers of autolysed cells at the same time. They’re a bit fuzzy on whether this strategy actually improved the resulting wine’s sensory characteristics, and I’m a bit fuzzy on the details: you’d think that the killer yeast would have to release its killing toxins fairly late in the game, after the susceptible yeast has had time to multiply; this seems to be the case, but the authors don’t go into the details.

Regardless, it’s a sensible idea: work with the characteristics of the yeast to accomplish some mutual human-yeast goal. Maybe we accomplish better things in winemaking when we trust the yeast?


*Yeast can have sex, but it’s not a necessity. Most yeast cells are born when a mother cell duplicates her DNA (that’s mitosis) and buds off that DNA with a bit of her cytoplasm (general cellular contents) to form a new daughter cell. However, some (not all) yeast can mate by sending

As for why yeast cells are always female? I imagine that it’s because mothers and not fathers give birth, so the parent cell must be female, and since the baby cell will eventually become a parent cell it must also be female unless we want to deal with some uncomfortable sex switching in the middle. But it’s also worth noting that, according to a native Mandarin speaker I met last week, “yeast” in Mandarin is a homonym (same sound, different spelling) for “grandmother.” And sometimes we speak of sourdough starters as sourdough “mothers.” So perhaps there’s something more to be said here about our feelings that yeast is somehow a nourishing, loving, care-taking organism?

**Or, rather, a very, very few yeast cells are subject to accidental genetic rearrangements that happen to suit new conditions well, and the rest die or are out-competed by the lucky ones.

2 thoughts on “Innovation in sparkling wine production: Trust the yeast

  1. Thank you for a very entertaining and logical article on yeast. And I assume you are aware that there is always a ‘mother’ used when making batches of vinegar and yoghurt.
    I look forward to reading more on your site.
    Many thanks, Karen

    • Yes, indeed, but vinegar and yogurt are the work of acetic acid and lactic acid bacteria, respectively, not yeast, so I didn’t include them here. We’ve developed intimate relationships with all manner of microbes.

Leave a Reply

Your email address will not be published. Required fields are marked *