428 M.A. Pozo-Bay ́ ́ on and G. Reineccius
8F.2.2.4 Macromolecules Derived from Wine Micro-Organisms
A major group of yeast macromolecules are the mannoproteins which represent ca.
35% of total wine polysaccharides. Thesemannoproteins can be classified into two
main groups, those secreted into the wine by yeast during alcoholic fermentation
and those released into wine due to yeast autolysis during aging on lees (Chalier
et al. 2007). These macromolecules play an important role in winemaking; for exam-
ple, they are involved in the improvement of tartaric acid stability, reduction of pro-
tein haze and the stabilization of wine colour and phenolic compounds among others
(Feuillat 2003). However, since the addition of mannoproteins under EU law is only
currently permitted for experimental trials (EU Regulation 1622/2000, art.41), other
alternatives, i.e. ingredients that provide mannoproteins but are approved for use
in wines (such as yeast derivatives, extracts and autolysates), are being used com-
mercially to improve wine organoleptic characteristics. It must be considered that
these alternatives bring in many other components (not just mannoproteins) which
may have a negative effect on wine flavor. They may contribute undesirable aroma
compounds, offer interactions that imbalance the wine, or result in new aroma com-
pounds being formed during wine making. There is a risk in using these materials
without adequate research on their overall effect.
Elaborating more on mannoproteins and their effects on wine aroma, Voilley
et al. (1990) studied the binding of volatile compounds on a yeast-wall:bentonite
mixture in finning experiments. The binding ofβ-ionone was higher (about 30%)
than the three other volatile compounds studied (n-hexanol, ethyl hexanoate, isoamyl
acetate). Lubbers et al. (1994a,b) found thatthese interactions were mainly hydropho-
bic in nature, although binding was dependent on the type of aroma compound and
nature of the substrate.
Interactions between mannoproteins from yeast cell walls and aroma compounds
have been studied by Langourieux and Crouzet (1997). They performed the exper-
iments with crude mannoproteins extracts and observed no effect on the activity
coefficient of isoamyl acetate, and a slight decrease on the activity coefficients of
ethyl hexanoate and limonene. However, when they purified the mannoproteins or
when they used a model glycopeptide, they did not observe any effect on limonene
volatility. If the synthetic peptide was heat treated (50◦C), they observed a slight
reduction on the activity coefficient of limonene. This was explained by an increase
in the hydrophobicity of the glycopeptide after the thermal treatment.
Chalier et al. (2007), using mannoprotein at levels usually found in wines (150
mg/L), compared the effect of a whole mannoprotein extract (isolated from a syn-
thetic medium subjected to alcoholic fermentation) to that of well characterized
different mannoproteins fractions. From the four wine aroma compounds studied
(isoamyl acetate, hexanol, ethyl hexanoate andβ-ionone), all except isoamyl acetate
showed a decrease in volatility (up to 80%) when mannoproteins were present
(Fig. 8F.3). They suggested that both the glycosidic and the peptidic parts of these
macromolecules may be responsible for the interaction. They also found that the
interactions of the whole mannoprotein extract Vs. mannoprotein fractions were
different, suggesting that the conformational and compositional structure of these