PROCESSING & ENOLOGY

NEW YORK WINE ANALYTICAL LABORATORY AND WINE DATA BANK

Thomas Henick-Kling and Ben Gavitt

Dept. of Food Science and Technology

NYS Agricultural Experiment Station, Cornell University, Geneva, NY

Forty-five clients utilized the NYS Wine Laboratory submitting 360 samples for 3,923 individual analyses. The clients were comprised of 35 commercial wineries, 4 hard cider companies, 1 scientific manufacturer, and 5 home winemakers.

As in years past, the major focus (98%) of the wine analyses were done to identify and solve problems. The other 2% were for quality control to assure stability of the wine after bottling. In addition to this, we helped a scientific equipment manufacturer develop and check automated methods for alcohol, SO2, and sugar analysis.

Consultation by phone,fax, e-mail

417 (BG) + 250 (est. for THK) quiries were addressed. About one-third of these telephone consultations were about product improvements. The rest dealt with technical problems winemakers were facing. Sterile bottling remains a major concern and problem of the industry. This year, a marked increase in "acid" questions mirrored the high acid 1996 vintage. The increase in reduced sulfur smells (defects) might be due to less ripe fruit and higher residues of sulfur sprays. The higher number of reduced sulfur defects discussed might also be the result of a better awareness of this common defect in all wine fermentations. Winemakers are more careful before bottling a wine with a detracting odor of reduced sulfur.

Success stories

A Finger Lakes winemaker noticed some off-flavors in a Riesling wine (4,000 gallons). A sample was sent to the NYSWAL . Upon examination it was noted that the wine was oxidized and also had a slight sulfur smell. We recommended that the wine be yeast fined, and the SO2 adjusted. After the procedure was explained, the winery completed the method successfully with significant improvement in taste and aroma. The wine was now able to be sold as a top selling varietal instead of blended or perhaps even thrown away, saving the winery thousands of dollars ($211,200 est. retail value vs $96,000).

A Hudson Valley winery had a problem with a large wine lot that had not been filtered correctly. After examining and testing a sample, the recommendation was made to uncork, adjust pH, and SO2 then sterile filter again and bottle. The winery realized saving of tens of thousands of dollars for a now salable product.

A large manufacturer of hard cider contacted us in early July with a severe contamination problem in their product. No product could be sent for over five weeks. We consulted with them until November and spent many hours analyzing their product and helping them set up a quality control plan at their facility. Over the five months that we consulted with this company, hundreds of thousands of dollars were saved.

Top

PRODUCTION OF FOOD-GRADE GRAPESEED OIL FROM POMACE

Thomas Henick-Kling and William D. Edinger

Dept. of Food Science & Technology

NYS Agricultural Experiment Station, Cornell University, Geneva, NY

An initial literature survey of the methods of production and the composition of grape seed oil was done. During the 1997 vintage we investigated practical methods for the separation of grape seeds from pomace and collected several different batches of grape seeds. These samples were frozen for later processing trials. Dr Edinger left to take a job with a winery in California. A new investigator was found who will continue the project in February 1998. We plan to expand the literature search for processing methods and complete the processing trials.

Top

YEASTS IN NEW YORK WINES:

Brettanomyces/Dekkera Yeast Identification By Genetic Techniques

Christoph M. Egli, Craig Mitrakul, Jonathan Licker, and Thomas Henick-Kling

Dept. of Food Science & Technology

New York State Agricultural Experiment Station, Cornell University, Geneva, NY

Brettanomyces/Dekkera yeasts are primarily known both as wine and beer spoilage organisms. Yet, in some wines, such as Bordeaux and Italian red wines, it might also add a desirable complexity. Brettanomyces, the anamorph form of the genus Dekkera, includes the five species B. bruxellensis, B. anomalus, B. custersianus, B. naardenensis, and B. nanus. These yeasts typically grow in low cell numbers during wine aging in barrels and tanks. Their aroma characteristics have been described as mousy, smoky, burnt plastic, barnyard, horse sweat, leather, wet wool.

To find out what flavors Brettanomyces yeast can produce, we studied in detail three wines that had been well described by a group of winemakers as having three different characteristics of ‘brettiness’. The wines were made from same grape variety, from the same vineyard sources, and in the same winery in the years 1989, 1992, and 1994. A preliminary microbiological analysis had shown that the wines contained some Brettanomyces-like yeasts and they all contained the compound 4-ethyl phenol which is produced by Brettanomyces. The three wines were evaluated by a panel of 13 experienced tasters using descriptors developed through Descriptive Analysis. The wines were rated significantly different for the descriptor ‘plastic’. Comparing the odor spectrum gas chromatograms, a general effect was observed. ‘Floral’, ‘fruit’ compounds were the dominant odors in the 1994 wine while ‘rancid’, ‘plastic’ odors accounted for one-third or less of the odor activity; in 1992 wine, the ‘floral’, ‘fruity’ compounds decrease to one-half or less of the odor activity while the ‘rancid’, ‘plastic’ compounds increase to two-thirds; in the 1989 wine, the ‘rancid’, ‘plastic" compounds were the dominant odors. In addition to the already known compound 4-ethyl phenol we found three more compounds that can give the wine ‘burnt’, ‘smoky’, ‘barnyard’ aromas.

From each of the wines, we isolated yeasts which were subsequently identified as Brettanomyces. Physiological tests using solid and liquid media were found not reliable. Misidentification occurred in 14% of the type strains and in more than 30% of the wine isolates. This finding highly argued for genotypic rather than phenotypic identification. Karyotyping (analysis of chromosome number and size) allowed identification of all individual isolates. However, the lack of any obvious intra-species regularity (chromosomes of a unique size) and the labor-intensive character of this procedure forced us to develop two independent and rapid PCR-based methods. ITS-PCR was applied by amplifying the two internal transcribed spacer (ITS) regions including the 5.8S rDNA from the rDNA repeats. Primary sequences of these regions sufficiently diverged to act as identification target sites. Specific oligonucleotides were designed for each Brettanomyces species and evaluated for specificity and reliability using the reference strains and wine isolates. No cross reaction products were detected when the specific primers were assayed in a PCR reaction with Brettanomyces strains of different species or other wine related non-Brettanomyces yeasts (Figure 1). Thus, PCR reactions using a combination of all four specific primers represented a specific and highly reproducible detection assay for Brettanomyces yeasts in general whereas use of single specific primers allowed for species-specific discrimination. The Brettanomyces yeast isolates from the three different wines were tested and shown to uniquely belong to the species B. bruxellensis. Confirmation for this finding came from RAPD-PCR.

Furthermore, use of the selected primers OPA 02, OPA 03, and OPA 09 allowed for strain discrimination within the species B. bruxellensis. The wine isolates turned out to be unique per vintage and were identical in the two vintages 1992 and 1994, but different in 1989.

In summary, we have developed two independent genetic techniques to quickly and reproducibly identify Brettanomyces yeasts on the strain level. A unique population of B. bruxellensis was found in all three infected wines tested. To study in more detail which aromas are produced by Brettanomyces yeasts we will use single strains isolated and characterized from these and other wines.

Figure 1. Identification of Brettanomyces yeast strains by PCR amplification of partial ITS elements. (A) Discrimination of the four species B. bruxellensis (CBS 72; lanes 1), B. anomalus (CBS 76; lanes 2), B. custersianus (CBS 4805; lanes 3), and B. naardenensis (CBS 6042; lanes 4). PCR reactions in lanes A contained the primer pair pITS1/pITS4, reactions in lanes B included the primer mix (pB2, pA1, pC1, pN1)/pITS4 in equimolar amounts. (B) Specificity proof for the Brettanomyces-primers came from testing the most abundant non-Brettanomyces yeasts in wine used as templates in the same reactions as described in (A). The yeast strains were in lane 1: S. cerevisiae, in lane 2: Cryptococcus curvatus, in lane 3: Rhodotorula glutinis, in lane 4: Torulaspora delbrueckii, in lane 5: Hanseniaspora uvarum, in lane 6: Candida guillermondii, in lane 7: Pichia kluyveri, and in lane 8: Pichia anomala. No DNA template was added in the negative control reactions (K-). The 100bp ladder (Promega) was taken as molecular weight marker (lane M).

back | wine | viticulture | publications | nysaes