In vino veritas: An elucidation of yeast-bacterial interactions in wine

By Abby Beltrame

            Any wine drinker will attest that there’s nothing better than a well-balanced, flavorful glass of (my favorite, red) wine.  Luscious flavors, slight acidity and tannins, a delicately smooth mouth feel, and bountiful enticing aromas all contribute to a successful pour.  Most wine drinkers also know, unfortunately, the smell and taste of a spoiled vintage.  How tragic!  What you may not know is that to achieve that perfect balance, or to prevent a spoiled disaster, a subtle and complicated relationship must be perfectly managed. 

The skins of these grapes are host to a myriad of bacteria
and yeast species. [Author’s photo]

            So what is it that makes the composition of a great wine so complicated and the task of creating such a wine so challenging? This is in large part due to that the flavors and aromas of wine are the result of microbial metabolic products.  The substrates for these metabolic processes are the compounds found in grape must, a liquid concoction derived from the pressing of grapes.  Must consists not only of grape juice but also the seeds, skins, and pulp.  Found naturally within this mixture are various microbes, a natural assortment of bacterial and yeast species¹.  The sensory-related components of the wine are highly dependent upon the types of microorganisms present and both their cooperative and antagonistic interactions.  These interactions depend on a number of microbial chemicals and byproducts, including ethanol, lactic acid, glycerol, and inhibitory toxins. 

            It is important to start at the most basic level- how is wine made?  In essence, there are two main steps required to make a palatable wine: a primary alcoholic fermentation and a secondary malolactic fermentation.  The primary fermentation is most commonly carried out by the yeast Saccharomyces cerevisiae, during which the yeast converts sugars into alcohol.  Bacteria are the key players in the secondary fermentation. Though not actually a fermentation, but called as such, malolactic fermentation (MLF) is the enzymatic conversion of malic acid to lactic acid¹.  MLF is accomplished by lactic acid bacteria (LAB), primarily by the bacterium Oenococcus oeni (formerly known as Leuconostoc oenos)¹.

            Essential to composing a tasty wine is balancing these yeast-bacteria interactions and the metabolic compounds produced by each organism.  Ethanol produced by yeast contributes to the pleasing, psychotropic effect found in any alcoholic beverage².  Glycerol, a simple alcohol, is a secondary product of yeast fermentation and is known to contribute to the texture of wine, lending a full-bodied smoothness, while also augmenting the wine’s flavor profile³.  Lactic acid production by LAB is equally important not only because it deacidifies and stabilizes the wine, but also because lactic acid imparts desired flavors and aromas that enhance the wine quality⁴. Even so, the acidic pH, high ethanol content, and low availability of nutrients that are byproducts of the primary fermentation by S. cerevisiae result in conditions that do not readily support LAB growth¹.  On the other hand, LAB can carry out additional non-MLF metabolic reactions that can result in off-putting aromas, undesirable flavors, and overall spoilage of the wine¹.  It is not surprising then, that the complex interactions between S. cerevisiae and LAB involve both inhibitory and stimulatory reactions by both the yeast and bacteria.

            The most commonly reported relationship between the two organisms is yeast inhibition of bacteria. Early scientific studies found that in grape juice with mixed cultures of S. cerevisiae and O. oeni, yeast has a shorter lag phase and grows at faster rate than bacteria⁵.  While ethanol production by yeast has long been known to inhibit the growth of LAB, high ethanol concentration does not seem to have a substantial effect on MLF activity alone⁴. Sulfur dioxide (SO₂) produced by yeast can also slow bacterial growth, reducing the risk of spoilage, yet also inhibits MLF⁴.  Other yeast metabolites, such as medium-chain fatty acids, have been shown to reduce bacterial growth and MLF ability⁴.  As a compounding effect, this inhibition is greatly increased at acidic pH.

            To augment this inhibition, certain strains of S. cerevisiae have been shown to encode genes for killer toxins that act specifically on competitive microorganisms, including LAB⁶.  These toxins, known as killer factors, are cell-surface proteins that act at receptors on the bacterial cell wall.  Strains of S. cerevisiae have been shown to produce three of these toxins while under winemaking conditions, and while they can efficiently inhibit bacteria, the toxins also restrain other yeast species⁶.  Traditionally, winemakers add additional SO₂ to wine to halt undesired microbial growth and prevent spoilage.  The use of killer toxin-producing yeast strains is promising because they can act on their own to control microbial overgrowth and replace or reduce the amount of SO₂ used in winemaking⁶.

            While it is commonly thought that yeast overpower bacteria in the winemaking process, LAB can in turn inhibit yeast.  In lower alcohol conditions, rapidly proliferating bacteria cause a faster death rate of the yeast⁵.  Interestingly, at a lower Brix (a winemaker’s measure of sugar content), although LAB declines more in lag phase when co-inoculated with yeast in grape juice, the bacteria has a faster logarithmic growth rate.  Thus at lower sugar concentrations, the bacteria ultimately reaches a higher clonal population, resulting in more rapid malolactic acid conversion⁵.

            Another bacterial effect is that many types of LAB metabolically degrade glycerol, resulting in disagreeable organoleptic properties³.  Some LAB, such as O. oeni, can produce glycerol under specific conditions, particularly when nutrients are limited³.  This can be used to the benefit of the winemaker, by allowing conditions that support glycerol production while minimizing bacterial overproliferation and spoilage of the wine.

            In addition to the inhibitory effects of wine yeast towards bacterial co-inhabitants, some evidence suggests that stimulatory effects also persist between the two groups⁴.  Yeast autolysis results in the release of many nitrogenous compounds, including whole proteins, small peptides and individual amino acids.  Some of these nitrogen isolates derived from wine have been shown to support LAB proliferation⁴.  Taken together with the strong inhibitory interactions, this supportive effect on the bacteria suggests a mechanism of high bacterial inhibition while the yeast is active during the primary fermentation phase, but then a transition to conditions that support bacterial MLF once the yeast population reaches stationary phase and subsequently degrades. 

            As a novice winemaker, or just as a wine-lover, it is beneficial to understand how to take advantage of these interactions in order to produce a superior wine. A recent study by Ale et al. focused on the interactions between S. cerevisiae, O. oeni, and another common wine yeast Kloeckera apiculata³.  After much experimentation using different concentrations of each organism and playing with inoculation times, the authors of the study propose a model inoculation scheme.  Their proposal promotes ideal alcoholic fermentations by yeast, while encouraging glycerol production and MLF by the bacteria.  Ale et al. recommend a simultaneous inoculation of 10⁶ CFUs/mL of S. cerevisiae with the same amount of K. apiculata, followed by a sequential inoculation of 10⁶ CFUs/mL O. oeni.  By heeding these recommendations, any winemaker can hopefully increase the chances of concocting a brilliantly balanced wine.

Pressing the grapes extracts the sugars for yeast to feed upon,
while also brings the many types of yeast and bacteria in
close contact, encouraging a suite of interactions. [Author’s photo]

            It is important to note that a whole chorus of microbes may be present on the skins of grapes and, as such, there are likely many more interactions that exist.  Also, these interactions are not isolated and often simultaneous, presenting further challenges in their study.  While lab investigations into these relationships often contain only S. cerevisiae and LAB in grape juice-containing media, the many additional species of microbes residing in the vineyard and winery contribute in their own ways to the success or spoilage of a wine.  More research to elucidate these relationships will allow winemakers across the globe to better harness the microbial processes that contribute to the delicious beverage that many of us love. Even so, the key relationships discussed here contribute greatly to the outcome of a wine, and I look forward to implementing them in my own winemaking practices, as should you! And with that, I will end this with a toast: Here’s to never sipping a spoiled wine again. Cheers!

References:

1. Moreno-Arribas, M.V. Wine Chemistry and Biochemistry (Springer Science+Business Media 2009) Chapter 2:27-57.

2. Pretorius, I.S., et al. The winemakers bug: From ancient wisdom to opening new vistas with frontier yeast science. Bioengineered Bugs 3, 147-156 (2012).

3. Ale, C. E., et al. Glycerol production by Oenococcus oeni during sequential and simultaneous cultures with wine yeast strains. Journal of Basic Microbiology 54, S200-S209 (2014).

4. Alexandre, H., et al. Saccharomyces cerevisiae- Oenococcus oeni interactions in wine: current knowledge and perspectives. International Journal of Food Microbiology 93, 141-154 (2004).

5. King, S.W. and Beelman, R.B. Metabolic interactions between Saccharomyces cerevisiae and Leuconostoc oenos in a model grape juice/wine system. American Society for Enology and Viticulture 37, 53-60 (1986).

6. Fernandez de Ullivarri, M., et al. Killer activity of Saccharomyces cerevisiae strains: partial characterization and strategies to improve the biocontrol efficacy in winemaking. Antonie van Leeuwenhoek 106, 865-878 (2014).