Growing & Winemaking


The Role of Bacteria in Stuck Fermentations

March 2016
by Richard Carey
Switching Rate

Hundreds of articles about stuck fermentations cover how to avoid them as well as restart them. It is an art that relies on experience and intuition about what went wrong with a particular fermentation. The traditional methods that help to avoid stuck fermentations include assessment of spoilage organism potential, nutritional status of the initial must and viability of the yeast used relative to the style of wine desired.

Many laboratories servicing the wine industry offer assessment panels to test a winery’s grapes, but in most cases winemakers don’t use that service until it is too late. The sooner the winemaker understands there is a problem, the easier it will be to begin a triage of the situation and the better the resulting wine will be.

Even when a winemaker thinks all bases are covered, occasionally fermentation for a certain wine won’t perform the same way it did in previous seasons. As a wine consultant and winemaker, this has always been a bothersome situation to resolve: Why would this wine, this season, not complete fermentation? This was the case with wines from a recent client of mine. The winery had a series of wines with fermentations that had started normally. Then, sometime after the lag phase had passed and the yeast was on the concluding side of the fermentation curve, the rate slowed to a crawl and apparently stopped at about 2° Brix. After trying all the traditional paths to restart the fermentation, they asked me to figure out what might be causing this situation.

When I reviewed the wines organoleptically, I found no off odors. Analysis showed volatile acidity was below any threshold and, therefore, not a problem. VA is one of the more frequent causes of stuck fermentations at this level, and remedial efforts that would exacerbate that situation should be avoided.

However, one of the techniques the client had tried was to introduce lysozyme, which is frequently added to reduce the bacterial load in a fermentation. Lysozyme is most effective against gram-positive bacteria, which meant that after the lysozyme addition, the wine still could have had Acetobacter or other acetic acid-producing bacteria present.

Macro oxygenation is a technique frequently used to help fermentations. Yeast are facultative organisms, meaning they can survive both in aerobic and anaerobic environments. In a normal fermentation during lag phase, yeast grows rapidly and lives off the dissolved oxygen that is in the must. Eventually the organism population exceeds the capacity of the must to bring more oxygen in from pumping over, and so the yeast switch from aerobic metabolism to anaerobic metabolism. That is when augmentation of yeast growth is helped by injection of up to 60 mg/L per day of oxygen into a must, which can encourage the yeast to continue to grow at maximal rates. This technique is not without some peril, as Acetobacter are not facultative organisms. Withdrawing oxygen slows and then stops growth of the acetic acid bacteria during fermentation.

If you know that there are Acetobacter organisms present, you need to monitor any additions of oxygen to maintain the health of the wine. It was with much caution that I broached the possibility of adding oxygen to these fermentations. At this time I remembered a news brief published by Linda Bisson from the University of California, Davis, about a new concept of yeast metabolic regulation by bacteria that involved prions.

Use of the environment
The magazine Cell published an article by Daniel F. Jarosz et al. titled “An Evolutionarily Conserved Prion-like Element Converts Wild Fungi from Metabolic Specialists to Generalists.” In the article, the authors described a novel means of controlling yeast use of glucose through secretion of a prion that switches the yeast from exclusively metabolizing glucose as their food source to becoming a generalist in use of available carbon sources for metabolic activity (see “Switching Rate Tuned to Ecological Niche”).

As described in the article, use of the prion acts as if it is a genetic mutation but instead functions as an epigenetic mechanism, causing a fundamental change in the attacked organism’s use of its environment. The prion control is niche-based and reversible (see “Bet-Hedging Properties of the [GAR+] Prion”).

The action of the yeast prions in restricted nutritional environments stresses the cellular protein-folding network. So, evolutionarily, the yeast has retained the ability to hedge its survival by being able to switch carbon sources based on its local environment, except that in the wine-fermentation environment, bacteria have caused this to happen at precisely the wrong time. It happens when resources are at their most critical level for the yeast.

This ability to switch carbon sources is an evolutionarily conserved trait of Saccharomyces cerevisiae. Several types of typical wine bacteria such as strains of Lactobacillis and a couple of Pediococcus have the ability to induce this switch in carbon sources. Once this switch occurs, it can remain present in a given culture for many generations of yeast growth. Over time and under presentation of a new environmental set of conditions, S. cerevisiae will be able to become a glucophial again. The process takes the [gar -] state (glucose repressed) where the yeast can ferment easily on glucose to the virtual exclusion of any other carbon source. Upon induction to the [GAR+] state (induced generalist), they switch to using a much wider array of carbon sources. However, within any population of a specific strain of yeast, there is a certain percentage that naturally switch at any time.

In grapes there is a mixture of glucose, fructose and other simple sugars. Since yeast prefer feeding only on glucose, that is where the initial growth starts. In any natural population of S. cerevisiae, a small percentage of that population is capable of fermenting fructose. That population segment can multiply during any given fermentation, so that at the end of a fermentation all the fructose is consumed, leaving only glucose as a food source. If the proper bacteria were present, they could exacerbate the [GAR+] state, increasing that percentage of the yeast population. With the other simple sugars exhausted and leaving only glucose, the yeast metabolic systems are incapable of switching back soon enough to take advantage of this food. The result is a stuck fermentation. Importantly the yeast are not dead, just not able to grow vigorously.

Back to the beginning
So how does this relate to my client’s fermentation? The client knew his fermentation was stuck. It had not responded to the normal buildup of other yeast nutrient additions and cultures. The lots of wine must have had some population of Lactobacillis/Pediococcus organisms in the mixture that induced the [GAR+] state in the fermentation, and the fermentation slowed to a crawl. The client, seeing the fermentation sticking, added lysozyme to the fermentation, which wiped out the gram-positive bacteria so that the gram-positive bacteria and yeast could not take advantage of and consume any remaining sugar. The yeasts were weakened by [GAR+] and had not reverted to [gar -], so they could not effectively use the low amounts of glucose remaining to finish fermentation.

This is when I received the call to solve the issue of why the fermentations wouldn’t complete. The one nutrient that was not added was oxygen. As my first test, I used the fermentation that had the greatest amount of sugar remaining. This wine had started at 23° Brix at harvest, and now, using a hydrometer, it measured 1.8° Brix. In order to add oxygen to the wine, I used a Parsec Oxygenious.

After about three days, the sugar level fell by 0.1° Brix. It continued to fall at the same rate until it got to -0.3° Brix. It was at that point that I noticed the first hint of acetaldehyde, and I immediately stopped the flow of oxygen. In the next three days the Brix fell to -1.8° Brix. An appropriate amount of SO2 was added, and now the wine smells and looks just like a freshly fermented Pinot Noir. There are no oxidative notes and no off-aromas or flavors.

The other wines had similar results. My thoughts about what was happening now return to the discussion above. I had not contemplated the glucose-repression phenomenon until well after this process finished and I began evaluating how this all came to pass. The rather unique environment that this trial presented had me initially worried about Acetobacter taking off in the presence of the excess of oxygen going into the wine. While I did not plate the wines, organoleptic evaluation showed no noticeable increase in volatile acidity.

It makes sense that the oxygen provided an extra kick to metabolism that could reactivate the aerobic pathways; it then took a longer than average time for the yeast population to re-accommodate to the new environment and start the final push to the end. The final sugar level at the end of fermentation for the Pinot Noir discussed above was 0.181 g/L glucose.

This exercise suggests that winemakers should follow their fermentation curves critically and look at changes in the curve’s slope during the later parts of the cycle. Once a winemaker has good data about the average curves for his or her wines, when a fermentation begins to show a different curve, early introduction of oxygen should keep it as robust as possible.

The bottom line is, don’t give up if a fermentation doesn’t go to completion. Perseverance and the delicate use of “macro” micro oxygenation can coax the pesky yeast to finish their job.

Richard Carey, Ph.D., is owner and wine consultant at Tamanend Wine Consulting in Lancaster, Pa. He has written numerous articles about new technologies for the grape and wine industry for Wine East and Wines & Vines.

Currently no comments posted for this article.