CHARACTERIZATION AND REMOVAL OF UNSTABLE PROTEINS FROM GRAPE JUICE AND WINE (ULTRAFILTRATION, BENTONITE, ELECTROPHORESIS, GLYCOPROTEIN, HEAT)

Improved sensitive techniques were developed for the determination of soluble proteins in grape, juice and wine. These techniques which permitted direct sample analysis without prior dialysis or concentration, were used for: (1) characterization of heat unstable proteins, (2) evaluating the effect of bentonite fining and ultrafiltration (UF) on the soluble proteins and heat unstable proteins in Gewurztraminer and Riesling juices and wines. Protein molecular weights (MW), isoelectric points (pI) and glycoproteins were determined by using LDS-polyacrylamide gel electrophoresis and 2-dimensional isoelectric focusing (IF)-LDS electrophoretic techniques with silver staining as well as protein blotting for glycoprotein detection. Relative concentrations of proteins in stained gels were determined by laser scanning densitometry. Bentonite fining tends to remove higher pI (5.8-8.0), intermediate MW (32,000-45,000) protein fractions first. In general, it is necessary to remove the lower pI (4.1-5.8), lower MW (12,600 and 20,000-30,000) fractions and glycoproteins to "protein stabilize" wines to heat testing. Unstable proteins precipitated by heat test were recovered and analyzed. These proteins were mainly of low MW (<30,000) and primarily glycoproteins. Protein fractions with MW of greater than 14,000 were more heat sensitive than lower MW fractions. Wines and juice were ultrafiltered with Romicon and Millipore systems operated with membranes of "nominal MW cut-off (MWCO)" of 10,000-100,000 daltons. A progressive increase in membrane retention of soluble protein was observed with decreasing MWCO, up to 99% of wine protein being retained with membranes of 10,000 dalton MWCO. However, certain high MW protein fractions also remained in permeates even with 10,000 MWCO membranes. In the order of 3-20 mg protein/L frequently remained in UF wine permeates, this correlated with the periodic detection of heat instability and of a low bentonite requirement for "protein stability" as determined by sensitive heat testing. "Protein stability" could be obtained with MWCO of 10,000 and 30,000; however, when not obtained, reductions in the order of 80-95% in "bentonite demand" were achieved. Protein stabilization of wines by UF is similar to that by bentonite fining in that it is necessary to remove the lower MW (12,600 and 20,000-30,000), lower pI (4.1-5.8) fractions, and glycoproteins to stabilize wines to heat testing. UF membranes were more effective at retaining wine proteins than juice proteins. UF juices were more susceptible to heat induced haze formation than UF wines.