Yeasts of the Brettanomyces/Dekkera genus can act as spoilage yeasts to create off-odors in wine, commonly described as wet animal, horse sweat, barnyard, burnt plastic, medicinal, Band-aid, mousy, metallic, etc.

Brett management is a source of ongoing debate among winemakers who must balance the risks of potential contamination producing negative sensory impacts versus the potential merits of Brett at controlled levels that may add complexity and stylistic character to certain wines. Inventories of Brett strains (or isolates) are being expanded and maintained, and studies have begun to analyze different Brett strains and correlate their potential positive and negative sensory effects in wines. Other recent developments in Brett research include new test methods for earlier Brett detection, and the ability to reduce or remove off-odors/off-flavors by filtering odor-active compounds from contaminated wines.

Background

Dekkera and Brettanomyces are genetically identical organisms, but Dekkera is the sporulating form of this yeast. Although the most common species found in wine is now identified as Dekkera bruxellensis, the organism and its associated problems in wine are still popularly called "Brett." Brett has been identified in all the world's wine producing regions. It is more commonly associated with red wines, but it can be found in white table wines, and it has been found in some sparkling wines. Brett is usually viewed as an aging problem, but studies by Professor Ken Fugelsang of California State University Fresno, show it can also be a fermentative yeast capable of producing 10 percent to 11 percent alcohol. Brett yeasts are able to develop anaerobically, but growth and fermentation can be enhanced when oxygen is present. Brett grows slowly and can ferment in wines with very low levels of fermentable sugars, including low levels of residual sugar in wines that would be considered "dry." Since fermentation activity can occur without forming a film or producing noticeable carbon dioxide, Brett often goes unnoticed until sensory impacts occur.

Studies done throughout the world show that Brett can be found at many locations within a winery, including at the crush pad and on grape receiving equipment, within fermenting must, within cellar buildings and within aging cooperage (where it is most commonly noticed, often too late). It appears that Brett can be spread from winery to winery through contaminated wine, equipment and barrels. Once established, insects, notably fruit flies (Drosophila) can also be vectors. A study performed at a Washington state winery and published in 2002 in the American Journal of Enology & Viticulture (AJEV) found Brettanomyces in air samples taken from the crush, tank, barrel and bottling rooms, indicating that the yeast can be spread through a winery by air currents. Estimated populations based on samples taken in this winery ranged from 3 colonies/250 L of air near the bottling room, to 100 colonies/250 L air in the crush room.

Formation of Odor-Active Compounds

The formation of volatile ethyl phenols from Brett in wine is based on the enzymatic conversion of vinyl phenols derived from cinnamic acids. Two volatile phenols formed by Brett in wine, 4-ethyl phenol (4-EP) and 4-ethyl guaiacol (4-EG), are associated with off-odors. 4-EP is associated with "medicinal" or "plastic" odors, and 4-EG individually has been described as "spicy," "smoky" or "bacon." A study published in 1999 in the American Chemical Society Symposium Series, performed at Cornell University with Cabernet Sauvignon wines with different levels of Brett using gas chromatography-olfactometry analysis found over 15 odor-active compounds. Not all of these compounds and odors were necessarily the direct result of Brett yeast activity, and the odors associated with some of these compounds could be considered desirable in some wines. This study indicated that what is considered "Brett" aroma in wine "is a complex mixture of odor-active compounds, including acids, alcohols, aldehydes, ketones, esters and phenolics."

Compounds associated with Brett activity and character, in addition to the volatile phenols 4-EP and 4-EG, are medium-chain fatty acids that include isovaleric, octanoic, dodecanoic, isobutyric and 2-methylbutyric acids; and other odor-active compounds that include guaiacol, 2-phenylethanol, isoamyl alcohol, cis-2-nonenal, trans-2-nonenal, B-damascenone and ethyl decanoate.

Fugelsang and Bruce Zoecklein performed a study to evaluate six strains of B. bruxellensis in Pinot Noir wines from Sonoma County (published in AJEV in 2003) to evaluate differences in positive and negative sensory effects of Brett growth based on strain. The study suggests that "differences may result from strain variation, length of time for Brettanomyces-wine contact, and the ratio of current to cumulative (total) cell exposure."

Lucy Joseph, curator of the University of California Davis wine yeast and bacteria collection, said the collection now includes over 50 different Brett isolates and more will continue to be added to the collection. Joseph said the collection serves as a tool available to researchers to understand Brett and how it functions, and to look at the diversity of Brett in relation to the odors it causes. "Some odors may be isolate dependent," Joseph said, "and sensory data is being collected for different Brett isolates that could be of interest to some people in the future." Eventually, Brett isolates that produce odors and flavors desired by some winemakers may one day be available and used to enhance characteristics in certain wines, just as fermentation yeasts and lactic acid bacteria are used today.

Improving Brett Testing and Detection

Due to the slow growth of Brettanomyces yeasts, the detection and isolation of Brettanomyces from other microorganisms in test samples can be a problem. Traditionally, plating of Brett cultures followed by microscopic evaluation has been the test method for determining and identifying Brett presence in wine. This method has several limitations: it can take up to 14 days to incubate the culture and confirm its presence, Brett yeast is not uniformly distributed in tanks or barrels from which samples are taken, and low cell viability due to SO₂ additions can result in false negatives. In addition, Brett can also be present as viable non-culturable cells.

Since 4-EP is unique in wine as a byproduct of Brett, it is a good indicator of Brett presence and activity and can be used to test for Brett in wine. Since 1993, ETS Laboratories in St. Helena has offered test analysis using gas chromatography/mass spectroscopy for 4-EP to identify Brett populations at low levels with a turnaround time as short as two days. Since the concentration of 4-EP is related to the concentration of Brett and its activity, this enables monitoring over a time period to evaluate if levels are increasing or decreasing, and if treatments to combat Brett, such as SO₂ additions, are effective.

ETS can also monitor for 4-EG, which is also an indicator of Brett, but is present in red wine in much lower quantities than 4-EP. Brett character perception is influenced by the concentrations of both 4-EP and 4-EG, and variation in 4-EG can explain why the perception of Brett character in a wine can be very different even when 4-EP levels remain the same. Analysis for both compounds offers a more complete analytical tool than testing for just one compound.

ETS Labs to Introduce Real-time PCR for Brett Detection and Monitoring

In recent years, knowledge of Brett genetics, and the development of real-time Polymerase Chain Reaction (PCR) analysis have been applied and tested to successfully detect and ennumerate low levels of Brettanomyces in wine in a much shorter time. ETS Laboratories will soon commercially offer an analytical service using real-time PCR with specific DNA assaid to identify Brettanomyces, as well as two other common spoilage organisms--Pediococcus and Lactobacillus. This method is based on amplification of genetic markers specific to each organism. With this advanced genetic assay and real-time analysis, ETS will be able to detect small populations within one day, including quantification of total Brett populations without culturing.

Torey Arvik, a molecular biologist with ETS that helped develop the analysis for commercial use said, "This is a very powerful tool for establishing the presence of Brett, which allows us to extract DNA from all populations of Brett in the sample, both culturable and viable non-culturable, and we can get the results one hour after the DNA is prepared."

Arvik further explained: "With this, winemakers don't have to wait for Brett to grow, as with standard culture plating that can take up to two weeks. This gives them a heads-up for potential problems in the future, so they can take earlier action to control or manage Brett before it becomes a problem." This analysis enables detection of Brett cells before 4-EP can be detected, and before the Brett populations grow. So it can be used in conjunction with the other analyses to monitor Brett growth and determine if control treatments are working. "This method is very complementary to plating and to our 4-EP and 4-EG analyses, it establishes a great baseline marker for monitoring," Arvik said. With this new tool, winemakers can determine the total number of Brettanomyces before, during or after the 4-EP/4-EG have reached sensory threshold levels, to make treatment decisions and to adjust treatment levels accordingly. In combination with plating, these tools cover all potential eventualities.

Brett Management

Earlier Brett detection provides a major benefit in managing Brett before it becomes a problem. However, Brett can still be difficult to control, and Brett management and treatment methods can have their own drawbacks. Depending on the wine processing stage in which treatments are used to inhibit Brett, these treatments, such as SO₂ additions, can also inhibit desired activity of other organisms such as Saccharomyces for fermentation, or lactic acid bacteria for malolactic fermentation. SO₂ may also cause sensory impacts.

Proper sanitation practices throughout the winery can assist with Brett control. Some general cellar practices used to inhibit and control unwanted Brett are reducing cellar temperatures, limiting oxygen by keeping barrels topped and making sulfite additions. Maintaining adequate levels of free SO₂ in wine, particularly during barrel aging, can inhibit Brett, but it can become established in the barrel in areas of limited SO₂ contact, such as around the bung hole, within the oak of the barrel and in the lees. Used barrels are one of the most common sources of Brett contamination in a winery, and are a common method of Brett transfer between wineries, via used barrel purchases. However, even new barrels can stimulate Brett growth, as some species can metabolize sugars such as cellibiose from new toasted barrels. Wineries that purchase wine from outside sources are advised to quarantine and test lots to confirm there is no Brett contamination, or sterile filter these wines before they enter the winery.

Sterile filtration is generally viewed as the most effective method to remove Brett cells and other microorganisms using a filtration membrane of 0.45microns. (One reference suggests that dormant, elongated forms of Brettanomyces cells may be able to pass through a 0.45 micron filtration.) If Brett it is not adequately removed, there is the risk that activity can occur in the bottle to yield off-odors once the bottle is opened. Filtration for Brett is seen as unfavorable for some red wines that could be stripped of positive flavor and aroma characteristics. Wineries that wish to avoid sterile filtration may instead try a clean racking and fining before bottling. A proven alternative to sterile filtration is the use of dimethyldicarbonate (DMDC), marketed by the trade name Velcorin. It tends to be used for high-end wines, or at larger-budget wineries, and has been effective against Brett without affecting wine taste. It requires special dosing equipment and special safety training for workers that use it. Scott Laboratories of Petaluma, California is the exclusive distributor of Velcorin for the wine industry.

Winetech's Brett
Off-Flavor Treatment

Winetech LLC of Napa, which provides mobile filtration services for wineries, now offers a filtration service to remove Brett-caused odor-active compounds 4-ethyl phenol and 4-ethyl guaiacol from wine before bottling. The process came about through Brett research and testing done by Enologica Vason in Italy. Winetech began using the process at California wineries in the fall of 2003. Winetech owner and winemaker Stefano Migotto said the process uses the same equipment that Winetech uses for VA removal, but the membrane and ion exchange media are different. The process has been used on red wines between barrel and bottle. Migotto said Brett yeast cells should first be removed, which the winery can do with sterile filtration, or Winetech can remove the yeast with its filtration equipment. If the yeast cells are not removed, there is the risk that they will continue to produce odor-active chemicals after bottling.

Winetech's 4-EP and 4-EG treatment is a two-step process that uses reverse osmosis and a treatment column. The wine is first filtered by reverse osmosis through a membrane and is separated into two parts--a permeate and a concentrate. The permeate, which contains the 4-EP and 4-EG, is then passed through an ion exchange treatment column to filter the two odor-active compounds. The permeate, with reduced levels of 4-EP and 4-EG is then recombined with the concentrate. Migotto explained, "When we do treatment we never remove all of the compounds present, it is usually around 50 percent." The odor/flavor compounds are removed to below their detectable thresholds, but caution is used beyond that due to the risk of stripping the wine of desireable flavor/aroma components. Migotto said a sample batch of up to 500 gal. is usually treated first, improving it from a level of noticeable off-odors and flavors to a point where they are not detectable. He said, "We may try to push the removal a bit further and see if the wine improves more or if it loses complexity." Migotto said a sample batch is done because every wine reacts differently. Based on the results of the sample batch treatment, the remaining wine is then treated for removal of the optimum proportion as determined by the test batch. Migotto said, "We recommend the process be done at least one month before bottling to give the wine enough time to come back together before going into the bottle." The filtration is done at working temperatures of 55 to 65 degrees F and can process from 400 to 550 gal./hour. Winetech requires a minimum quantity of 500 gallons.

Contact info

Winetech LLC
phone: 707-552-2080, 1-877- 4 WINETECH
fax: 707-552-2117
email: filtration@winetechllc.com

PO Box 6888 Napa, CA 94581
office and warehouse:
125 Klamath Court American Canyon, CA 94503

Vinovation, Inc.
phone: (707) 824-7900 • fax:(707) 824-7905
email: info@vinovation.com

office: 652 Petaluma Ave, Sebastopol, CA 95472
facility: 1365 Gravenstein Hwy. South, Sebastopol, CA 95472

ETS Laboratories
phone: 707-963-4806, fax: 707-963-1054
email: info@etslabs.com

899 Adams Street - Suite A
St. Helena CA 94574

Connell, Laurie, Henrik Stender, and Charles G. Edwards. 2002. Rapid Detection and Identification of Brettanomyces from Winery Air Samples Based on Peptide Nucleic Acid Analysis. Am. J. Enol. Vitic. 53:4, 322-324.

Fugelsang, Kenneth C. 1997. Wine Microbiology, Chapman & Hall, New York.

Fugelsang, Kenneth C., Bruce Zoecklein. 2003. Population Dynamics and Effects of Brettanomyces bruxellensis Strains on Pinot noir (Vitis vinifera L.) Wines. Am. J. Enol. Vitic. 54:4, 294-300.

Licker, J.L., T.E. Acree, and T. Henick-Kling. 1998. What is "Brett" (Brettanomyces) flavor? A preliminary investigation. Chemistry of Wine Flavor. ACS Symposium Series Vol. 714. A.L. Waterhouse and S.E. Ebeler (Eds.) pp 96-115. American Chemical Society, Washington, DC

Rossini, Ricardo. 2003. Contamination of Brettanomyces: the state of the art. Enologica Vason srl. Verona, Italy.

Zoecklein, Bruce W., Kenneth C. Fugelsang, Barry H.Gump, Fred S. Nury. 1999. Wine Analysis and Production. Chapman & Hall, New York.