• Oenodev
• Useful Links
• Our offer
• Contact us

Brettanomyces

Study of red wine contamination by Brettanomyces yeast

• Introduction
• Equipment and Methods
• Results
• Conclusion
• Bibliography

Results

Brettanomyces can develop in anaerobiosis, but its growth and fermentation are facilitated in the p resence of oxygen. This leads to the idea that micro-oxygenation during elevage encourages the establishment of this contaminant. In order to verify this hypothesis, we attempted to evaluate the effect that micro-oxygenation, using various different doses of oxygen, can have on Brettanomyces growth in wine.

Breakdown of Brettanomyces populations

FIGURE 3 : Breakdown of Brettanomyces populations

Figure 3 clearly shows that Brettanomyces populations are spread out evenly in the vat, whatever the oxygen dose used. This yeast settles in the whole volume of the vat during growth. We can, however, see some shift in population, even in the unoxygenated vat. These results show that wine samples drawn from the surface in the vat are sufficiently representative for Brettanomyces detection. Taking samples from the surface of the wine during elevage is the simplest way, and can be done in sterile conditions using a syringe. It also means less risk of mistakes in result readings, should the drawing-off taps be contaminated.

Effect of micro-oxygenation on Brettanomyces growth in wine.

In all cases, there is a 10 day latency period, than a 10 day intense growth period, followed by a rapid decline phase. Counts taken after 60 days culture show that a residual population (about 50 CFU/mL) remains in suspension in all the inoculated vats.

After 15 days growth, in the case of oxygenated lots, we can clearly see that the higher the dose of oxygen used, the more intense the Brettanomyces growth. However, lack of oxygen does not prevent Brettanomyces growth. On the contrary, we can see the development is equivalent to, perhaps even higher than that obtained in the wine oxygenated at 10mL/L/month. This is important insofar as it shows that wine probably contains molecules that cause this yeast to develop, as it would in the presence of small quantities of oxygen. The Custer effect could play a role in Brettanomyces contamination of wine.

In Vat n° 1, the curve is the result of an "accentuated running" on the 17th day. We can see a sudden increase in Brettanomyces growth just after oxygen injection.

According to these results, it seems that in our case, a supply of 10mL/L/month is a "limit" value, below which Brettanomyces growth is equivalent to that obtained in anaerobiosis. Above this level, Brettanomyces growth is encouraged, and the cellular concentration obtained is 30% higher. The same is true of oxygen injected in one go (vat n° 1), despite a weaker overall quantity of injected oxygen (table 2). Thus, racking during Brettanomyces growth encourages it as much as an excessive continuous oxygen supply would. 

In all cases, the concentrations of dissolved oxygen measured are very weak, (0.16 mg/L). It is thus difficult to determine whether the yeast or the wine consumes the oxygen. However, tasting and free SO2 measurements can yield further information.

Table 2 shows that ethanal is only found in clean, oxygenated vats. There is thus a chemical interaction between the oxygen and wine compounds to produce ethanal.

In the contaminated vats, all or most of the oxygen was probably consumed by the yeast ; this is why there was no chemical reaction. Moreover, we can see that the level of free SO2 diminished more in clean, oxygenated vats, probably due to a chemical combination with the ethanal.

TABLE 2 : results of oxygen analysis, cellular concentration, free SO2 consumption and presence of ethanal through tasting of the wine from experimental vats.

It appears that the Brettanomyces yeast can consume oxygen at a quicker rate than wine can. Thus micro-oxygenation of a Brettanomyces-contaminated wine is completely counter productive.

Tasting

The "blind" tasting sessions 1 and 3 were carried out by a team of 6-8 inexperienced people. These people were basically asked to describe the predominant smell and taste sensation and to determine which lot they preferred and which lot they hated. The results show the frequency of each smell and taste sensation.

Tasting session n° 2 (also "blind") was carried out by two experienced tasters from Oenodev. In this case, marks were given according to each smell and taste sensation.

The first tasting session was held 15 days after inoculation, i.e. during the exponential phase of Brettanomyces growth. Generally speaking, the wines are evaluated on fruit and animal aroma (nose) and fruit savour (mouth) (4 represents 50% of the tasters). Vats n°s 2 (non inoculated, oxygen 5mL/L/month) and 8 (inoculated without oxygen) appear to be the strongest in "animal" (musk, venison, etc.) aroma ; there seems to be a certain confusion among the animal notes, probably because the tasters were not familiar with this type of aroma.

The animal note is not particularly present in the mouth. It is interesting to note that vat n° 7 (the most oxygenated and containing the most contaminant) was the vat most people preferred. It would seem that at this stage a certain aromatic balance exists ; the animal "flaw" doesn't seem to bother the taster.

The second tasting session was held 22 days after inoculation, i.e. during the stationary phase of Brettanomyces growth. Once again, the wines were evaluated on fruit and animal aroma (nose) and fruit savour (mouth) (top score = 5). Vats n°s 7 (inoculated, oxygenated at 20mL/L/month) and 8 (inoculated and unoxygenated) presented the most animal aroma to the nose.

We can see here the appearance of an animal taste for vat n° 8. Although the ratings are different, the tendency is still the same as before. There is, however, a general decline in aromatic complexity, and only the fruit and animal notes are perceived in the contaminated wines.

The third tasting session was carried out 29 days after inoculation, i.e. when Brettanomyces growth was on the downturn. This time, tasting profiles were completely different and the animal notes much more present. The wines were judged on the animal aroma (nose) and fruit and animal taste (mouth). The strongest animal aroma is to be found in the inoculated wines; the fruity, flowery and woody sensations disappear. Although the gustatory sensation is still essentially fruity, we can still see an increase in animal character, particularly in vats n°s 7 and 8. The vats preferred by the tasters were vats n°s 2 and 6 (uninoculated); the one they hated was vat n° 1 (racking). In the case of vat n° 1, the woody note, among others (humid, mouldy....) were considered unpleasant.

At this stage of the experiment, we find the characteristics usually attributed to Brettanomyces-contaminated wine : sweat, humidity, mould, solvent, leather...

These wines are not at all supple, and are particularly astringent. It would appear that no softening of the tannins occurred, quite the contrary, in fact. Only wine from the clean vats presented traces of ethanal.

<< Equipment and Methods | Conclusion >>

TOP

Research

• Brettanomyces
• Screening Brett
• Acetaldehyde

  Conferences & publications

• Full list

 

Customer section