Making invisible things visible: Do yeast cells stick to cork during bottle aging?

Yeast do not seem to form biofilms on the bottoms of corks when they’re used, rather than metal crown caps, to secure Champagne bottles during their in-bottle secondary fermentation. This, at least, is the conclusion of an article in the current issue of the American Journal of Enology and Viticulture (paywall), in which Burgundy-based authors investigated the question for the sake of understanding whether Champagne producers, some of whom are using cork for their longer-aging wines, risked upsetting in-bottle fermentation dynamics. After a year of bottle fermentation and aging, a few cells apparently got caught in porous crevices of the cork, but systematic growth presaging the tenacity of a biofilm wasn’t happening.

That finding will no doubt interest the odd sparkling wine producer. Much more interesting is the method they used to make “invisible” cells visible and so reach that conclusion.

How do you determine whether a microbial biofilm is growing somewhere?

Before it reaches scrape-it-off-with-a-fingernail thickness, the answer is usually microscopy. But what kind of microscopy? Let’s say you want to see the very earliest stages of what might become a biofilm. You want to visualize individual yeast cells sticking to a rough, craggy surface with lots of crevices for hiding—the cork. Pointing a light microscope at the cork won’t do. The kind that you probably used in school to see cells suspended in liquid (from your cheek, or bacteria from your teeth), and that many small wine and brewery labs use to check for live cells requires that whatever you’re looking at be thin enough for light to pass through. Slicing corks into sections thin enough for light microscopy might destroy or displace any would-be-biofilm-forming yeast cells. Moreover, can you imagine scanning the bottom of cork after cork, continually asking yourself: is that a yeast cell, or a bit of cork shaped like a yeast cell? A digital image analysis program trained to recognize yeast cells might address the latter problem (with some degree of error), but the whole task clearly calls for a more sophisticated technique.

When your sample isn’t transparent enough for light microscopy, fluorescence microscopy can be an alternative; instead of visualizing light passing through the sample, fluorescence microscopy relies on whateveritis you’re examining absorbing and then emitting light—fluorescing—which can then be picked up by the microscope’s detector.* Conveniently, cork and other plant matter comes with a built-in fluorescent molecule; lignin, a rigid polymer and major contributor to the stiffness of wood and bark, is easy to see under the microscope.

Yeast, on the other hand, aren’t automatically fluorescent; they need to be made that way. In this study, after pulling corks from Champagne bottle, the scientists rinsed the corks in water (to dislodge any yeast that weren’t committedly attached to the cork but just hanging nearby), then doused them in an intensely fluorescent molecule (DAPI) that penetrates cell membranes and attaches firmly to DNA in living or recently-living cells. Cork is so thoroughly dead that it no longer contains much intact DNA. Consequently, under the microscope, lignin in the cork fluoresces one color and yeast cells fluoresce in a different color, making them easily distinguishable even when tiny yeast cells tuck into the English muffin-like structure of the cork.

The verb “to look” gets casually tossed around a lot in experimental research. “We’re going to look to see whether yeast cells adhere to corks during bottle fermentations.” Or, “I’m going to look at what metaphors are used to make living cells into engineering materials in synthetic biology labs.”** But “look” is almost always a sloppy short-cut, a kludge that doesn’t describe what you actually need to do to “see” the phenomenon you’re looking at. Research is about making the invisible visible, applying specialized techniques to see more than what is typically visible to the casual observer employing the proverbial naked eye.

Unpacking “look” is essential. First, saying that you want to “look” at something in an experiment is only the beginning of a lot of work to determine what you actually need to do. Second, the way you “look” determines what you see. In this study, the yeast that counted, the yeast that could be seen, were cells with intact DNA, stuck firmly enough to a cork not to be easily rinsed off. We have no idea whether those cells were metabolically active, growing or capable of growing, surrounded by dead cell detritus. And since the fluorescent stain binds to any DNA, we can assume that most of what shows up will be the S. cerevisiae strain used to innoculate the wine, but any other cells that happen to be present will also show up.

All of those considerations would be more important here if the scientists had found lots of cells (which might then warrant being “looked” at in more detail) instead of finding almost none. The more important point in this case? “Looking” is neither as easy nor as “unbiased” as it might seem.


*For a fairly readable review of how fluorescence microscopy works in much more detail, see this open-access article from Cold Spring Harbor Protocols.

**On which I’ll have a paper or two coming out this year analyzing how metaphors shape the way synthetic yeast is being built.

Leave a Reply

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