Grape-derived aroma and flavor precursors exist partially as non-volatile, sugar-bound glycosides. Hydrolysis of these compounds may modify sensory attributes and potentially enhance wine quality. In the first study, four levels of shoot thinning (control, 20, 25, and 30 shoots per meter) with and without basal leaf removal (2-4 leaves per shoot) were established on mid-wire (90 cm), bilateral cordon-trained, mature Cabernet Sauvignon (Vitis vinifera L.) grapevines in eastern Virginia in 1996 to determine the effects on grape chemistry, glycoconjugates, and conjugate fractions. Reduced shoot density generally resulted in higher berry weight and lower soluble solids (°Brix) at each sampling date. Titratable acidity and pH were generally unaffected by shoot thinning. The 25 shoots per meter treatment displayed the greatest rate of increase in total, red-free, and phenolic-free glycoconjugates, expressed as glycosides (μmol).. Leaf removal resulted in increased pH, total phenolics, and total anthocyanins at each sampling date and a higher concentration of total, red-free, and phenolic-free glycosides.

In a second study, three crop levels [high (6.4 and 5.3 kg/vine), medium (5.1 and 4.9 kg/vine), and low (3.2 and 2.6 kg/vine) ] were established on mature Cabernet Sauvignon grapevines during the 1995 and 1997 seasons, respectively. Cluster thinning of vines trained to a mid-wire (90cm), bilateral cordon-system was performed by hand three weeks post-bloom to determine the effects on grape glycoconjugates and conjugate fractions (expressed as glycosyl-glucose). In 1995, reduced crop level resulted in higher soluble solids concentration, pH, and total and red-free glycosides but did not affect berry weight or titratable acidity. In 1997, the reduced crop level treatment had higher berry weight and lower soluble solids, sugar per berry, and anthocyanins compared with the high treatment throughout the sampling period. The low treatment had the highest concentration of total, red-free, and phenolic-free glycosides per gram of fresh fruit weight on the last sampling date and the highest total, red-free, and phenolic-free glycosides per gram of fresh fruit weight when compared at similar soluble solids concentrations. Duo-trio significance testing resulted in no sensory differences among the treatments in 1997.

In a third study, Pinot noir (Vitis vinifera L.) wines were inoculated with one of six genetically different strains of Brettanomyces intermedius (Ave, M, 216, Vin 1, Vin 4, and Vin5). Wines stored sur lie and those racked immediately following the completion of secondary fermentation were analyzed to determine the influence of B. intermedius strains on total, red-free, and phenolic-free glycoside concentrations (estimated by the analysis of glycosyl-glucose), and on selected free volatiles. Sur lie wines inoculated with strain Vin 4 and racked wines inoculated with Vin 4 and Vin 5 had the lowest total glycoside concentration. Hydrolysis of red-free glycosides appeared greatest in sur lie wines inoculated with Vin 4 and racked wines inoculated with Vin 4 and Vin 5. Wines stored sur lie that were inoculated with M and Vin 1 and racked wines inoculated with Vin 1, Vin 4, and Vin 5 had the lowest concentration of phenolic-free glycosides. Wines were analyzed for volatile compounds known to be produced by Brettanomyces spp. Inoculated wines were found to have detectable concentrations of ethyl-2-methylbutyrate, isoamyl alcohol, ethyldecanoate, isovaleric acid, guaiacol, 2-pheylethanol, 4-ethylguaiacol and 4-ethylphenol. There were significant differences in the concentrations of these compounds among strains. Duo-trio testing demonstrated sensory differences between the control and all inoculated wines. Differences were also found between wines inoculated with strains Ave and Vin 5, strains M and 216, and strains M and Vin 4.