556 C. Santos-Buelga and V. de Freitas
and Noble 2005; Horne et al. 2002), with which it seems to be inversely related
(Lesschaeve and Noble 2005; Condelli et al. 2006). Low-flow subjects felt astrin-
gency later and more intensively for a longer time than did high-flow subjects (Less-
chaeve and Noble 2005). This effect could be explained by the increase of the saliva
pH and the protein concentration decrease for higher saliva flow.
9D.3.5 Changes in Tannin Composition and Astringency During
Wine Aging
It is generally considered that wine astringency intensity is higher in young wine and
decreases during aging “softening” the wine. After being extracted from grapes,
phenolic compounds – and namely flavanols – involved in astringency perception
and anthocyanins responsible for wine colour undergo chemicalreactions yield-
ing new simple phenolic compounds and more complex tannin-like structures. The
latter could include in their structures anthocyanins or other chromophore groups
(‘polymeric pigments’). Different pathways have been proposed to explain some of
those chemical transformations that occur during wine aging, some of them revised
in previous chapter of this book.
It has been known for a long time that the oxidative polymerization of flavanols
could directly involve oxygen (catalyzed by traces of metal ions) via quinones, or
be mediated by acetaldehyde mainly arising from ethanol oxidation (Somers 1971;
Timberlake and Bridle 1976; Wildenradt and Singleton 1974; Rivas-Gonzalo et al.
1995). The role of acetaldehyde in the polyphenol transformation has been exten-
sively studied by different research groups and is well documented. Acetaldehyde-
mediated condensation between flavanols themselves (Peleg et al. 1999) or with
anthocyanins leads to ethyl-bridged oligomers (Timberlake and Bridle 1976) and
also to flavan-pyranoanthocyanins (Francia-Aricha et al. 1997; Mateus et al. 2003).
Thus, acetaldehyde formed in situ from ethanol oxidation should play an important
role in the transformation and polymerization (insolubilization) of water-soluble
proanthocyanidins, causing the loss of astringency. In wines there are other alde-
hydes such as furfural and benzaldehydethat could be involved in such reac-
tions, as it was demonstrated in wine-like model solutions (Es-Safi et al. 2002;
Nonier et al. 2006; Pissarra et al. 2004). On the other hand, wine tannin-like
polyphenolic compounds could derive from direct reactions between flavanols
and anthocyanins yielding anthocyanin-flavanol and flavanol-anthocyanin adducts
(Remy et al 2000; Salas et al. 2005). Some of these tannin-like polyphenolic com-
pounds have been suggested to be less bitter andastringent than the original precur-
sors (Vidal et al. 2003b, 2004a). As a result of this whole field of polycondensations,
tannins become consecutively larger and the biggest structure are presumed to pre-
cipitate forming an insoluble deposit, as it was often observed specially in red wines
that age in barrel or bottle (Haslam 1980; Singletonand Noble 1976).
An alternative point of view is that some tannins (proanthocyanidins) might be
depolymerised by acid catalysis, thereby becoming smaller and reducing astringency