A fraction containing the mannoproteins released during fermentation from the winemaking strain of Saccharomyces cerevisiae, Maurivin PDM, was able to reduce the visible protein haze in white wine. This fraction of haze protective mannoprotein material (HPM) could be recovered by either ultrafiltration or ethanol precipitation. The kinetics of the release of both mannose- and glucose-containing polymers during the growth cycle of PDM were determined as a guide to the release of HPM. Active HPM was first detected in the culture supernatant when the cells were exponentially growing. HPM was also released into the medium under an environment simulating winemaking conditions by PDM cells during fermentation as well as during storage on yeast lees. Since the amounts of HPM released during fermentation are greater than those subsequently extracted from the cell wall, fermentation would be a more viable procedure than extraction from yeast cells for the commercial production of HPM. Yeast invertase, a mannoprotein with haze protective activity, was used as a model substrate to investigate the mechanism of haze protection. Invertase was found to reduce visible turbidity but not prevent protein precipitation. Invertase itself did not precipitate but remained soluble in the wine. On the basis of these observations, we propose that the mechanism of haze protection may be one of competition between HPM and wine proteins for unknown wine component(s), the latter being required for the formation of large insoluble aggregates of denatured protein. As the available concentration of these components decreases, due to the presence of HPM, the particle size of the haze decreases and thus visible turbidity declines.

Yeast-derived haze-protective mannoprotein material (HPM) offers protection to white wines from commercially unacceptable turbidities. HPM extraction methods have been evaluated using three winemaking strains of Saccharomyces cerevisiae. Digestion with Zymolyase of cells pretreated with DTE and EDTA gave the greatest yields of active material. Heat treatment of cells with SDS also released active material but the quantities were low. Treatment of the cells in an autoclave or with a French pressure device was less effective. A detailed study was conducted on the strain Maurivin PDM. SDS was not necessary to extract HPM from PDM; boiling the cells for 5 min in Tris buffer was sufficient. HPM could also be extracted with EDTA during the pretreatment of the cells prior to Zymolyase digestion. The data suggest that HPM was noncovalently linked to other cell wall components and loosely associated with the cell wall. An immunological investigation showed that a specific mannoprotein with haze-protective activity, HPF1, was located primarily on the outermost and innermost layers of the cell wall.

Various aspects of fungus-originated 0-apiosidase were studied. The production of the enzyme on various carbon sources by fungi appeared to be inducible as it was only produced when apiin, an apiosylglucoside, was present in the culture medium. The influence of apiin concentration, nitrogen source, and surfactants on enzyme production was studied. The enzyme was partially purified by filtration chromatography on Ultrogel AcA44 and ion-exchange chromatography on DEAE-Sepharose CLGB. The molecular weight of this enzyme was 38 OOO; K , and V,, values for p-nitrophenyl O-Dapiofuranoside were, respectively, 16 mM and 0.192 nkat/mg of protein. With regard to activity, the optimum p H and temperature were, respectively, 5.6 and 50 "C. The optimum p H of stability was 7. Na+, Mgz+, MnZ+, Cu+, Cu2+ have an inhibitor effect on 0-apiosidase activity. Conversely, the enzyme was inhibited neither by glucose nor by ethanol. The effect of 0-apiosidase toward apiosylglucosides of terpenols, grape flavor precursors, was tested.