"The making of red wine, which involves the skins and pips as well as the juice of grapes,
Downloaded by [Monash University Library] at 06:42 02 June 2013
leaves extra substances dissolved, above all tannin. This gives the wine the special quality of hardness of drying up the mouth. These extras need time to resolve themselves to carry out slow and obscure chemical changes which make all the difference in the world to the eventual glass of red wine." So wrote Hugh Johnson83 in his famous book Wine. In fact, few subjects are more likely to cause a more heated discussion among the wine cogniscenti and the connoisseurs of claret than the subjective comparisons made between the great vintages. While a certain amount of tannin is both desirable and essential in any claret to give the wine body, longevity, and backbone, it is nevertheless possible, according to various authorities, to have too much. Where vintages have an apparent excess of tannin, the reason is invariably the weather, particularly when there are extreme variations. Thus, apart from fully ripening the grapes, a spell of intense summer heat thickens their skins — a major source of both tannin and pigments (anthocyanidins). If these factors are combined with, say, a relatively small concentrated yield, then this produces a claret that starts life with, among other attributes, a deep color and a high tannin content. The question as the wine matures over the next decade or so is whether the claret will retain sufficient "fruit and flavor" to balance the tannin content or whether it will retain a hard unyielding backbone of tannin throughout.
The processes involved in the aging of red wines have been of an unending fascination to man for centuries. There is no doubt that during this transformation the polyphenols (tannins) and pigments (anthocyanidins) which are present are of great importance.84-90
However, progress toward defining their precise roles in molecular terms has been slow. In an admirable review of the state of the art written in 1969, Singleton and Esau91 commented:
"Unfortunately, what we would really like to know—the specific amount of each individual substance and how it participates and changes in wine in each type of storage reaction — requires much more detailed knowledge than is now available." However, now, almost two decades later, it is possible to suggest, at least in broad outline, the part that oligomeric and polymeric procyanidins (condensed tannins) probably play in the aging process.
In the stalks, skins, seeds, and, to a lesser extent, the leaves of Vitis vinifera, in addition to the various associated flavonols and hydroxystilbenes, the balance of polyphenols is present as proanthocyanidins of largely oligomeric and polymeric forms, some of which are soluble and others {vide supra) which are not. ( + )-Catechin (1 la) is the principle flavan-3-ol present and (— )-epicatechin (lib) is present in lesser amount; procyanidins B-l (cat-4-|3-8-epicat, 35) and B-2 (epicat-4-|3-8-epicat, 36) are the principal diastereoisomeric procyanidin dimers present. Treatment of the polymeric procyanidins with acid gives both cyanidin (37) and delphinidin (38) in the approximate ratio 4:1. This evidence clearly implies that in the vinification of red wines, the expressed juices in the fermentation will contain the flavan-3-ols ( l l a , b), some procyanidin dimers, and, depending on their solubility, various oli-gomeric proanthocyanidins of differing molecular size and constitution.8889 On the basis of present evidence, these are likely to include species such as (7 ) composed entirely of flavan-3-ol structural units and those such as (16) in which the flavan-flavan-3-ol oligomers are bound to saccharide structures (vide supra). In order to fully comprehend the processes that occur during the aging of red wines, it is essential to mention briefly the two characteristic chemical reactions of pro(antho)cyanidins: (1) the acid-catalyzed rupture of the interflavan bonds of pro(antho)cyanidins92 and (2) the facile electrophilic substitution of the phloroglucinol " A "
ring of the flavan-3-ol units which comprise the oligomeric structures. The first of these reactions has been referred to previously (vide supra, J7 —»j8), and two examples of the second type of chemical reactivity are shown in Figure 18: (1) the reaction with aldehydes, under acid catalysis, which leads to condensation products [the reaction shown is that of vanillin with a flavan-3-ol and is the basis of the familiar color reaction for flavan-3-ol metabolites in which a red color, probably due to the quinonoid intermediate (39), is formed93], and (2) the reaction with the flavan carbocation which leads to procyanidin dimers and
Downloaded by [Monash University Library] at 06:42 02 June 2013
Volume 27, Issue 1 (1988) 27
A' ring
Procyanidins biomimelic synthesis FIGURE 18. Electrophilic substitution of flavan-3-ols.
higher oligomers — biomimetic synthesis.94 Under the mildly acidic (pH 3 to 4) conditions which appertain in wines, both of these familiar and characteristic types of carbon-carbon bond-breaking and bond-making processes are believed to occur.
'H
(Procyanidin B-1 , 35) epicatechin 4f}-»8 catechin
•OH
( Procyanidin B-2,36 ) epicatechin 4fl-*-8 epicatechin
OH
(37, R=H .cyanidin ) (SB.R^OH.delphinidin)
Downloaded by [Monash University Library] at 06:42 02 June 2013
( 0 )
OH (37)
HO ""OH
(11b,3R , epicatedYin)
FIGURE 19. Procyanidin B-2 — acid-catalyzed breakdown.
The hydrolytic decomposition of various procyanidins has been subject to a detailed kinetic analysis;95 the reaction is a specific acid-catalyzed one which is first order in hydrogen ion concentration. Mechanistically, the rate-determining step is the protonation of the phloro-glucinol type " A " ring in the flavan-3-ol oligomer (Figure 19). Extrapolation of the detailed kinetic data obtained for procyanidin B-2 (Figure 19) has permitted a rough estimate for the hydrolytic rate constant at pH 4.0 (such as appertain in wines and beers) and at 25°C to be made — Kobs = 6 X 10~6 hr"1. For example, in acetate buffer solutions at pH 4.1 at 25°C both procyanidin B-2 (36) and a soluble oligomeric procyanidin fraction from hawthorn (Crataegus monogyna) at concentrations at 50 to 100 ppm gave steadily over 7 weeks a fine brick-red precipitate (phlobaphen). This heterogeneous polymer results presumably from the interplay of the two characteristic carbon-carbon bond-breaking (7 —» S) and bond-making (Figure 18) processes referred to above and is depicted schematically in Figure 20. In the case of red wines, as high molecular weight oligomers are formed and precipitate from solution, then the equilibrium moves away from the various soluble oligomeric forms (7 and 16) and the associated astringency is lost from the beverage.
Aging of red wines first in oak barrels and then in corked glass bottles produces other desirable effects on quality, taste, and flavor which may also lead to the loss of the astringent polyphenols from the wine. The color, for example, shifts from a purplish-red in the young wine to the more fiery amber of the mature wine, and ultimately to the tawny hues of a long-aged wine. This is, at one and the same time, one of the most readily appreciated visual aspects of wine aging and yet one of the least well understood.86 Somers90 has studied this problem in great detail. During the aging of red wines, the grape anthocyanins responsible for the initial color of the wine are displaced progressively and irreversibly by more stable polymeric pigments which for a wine within the first year may account for up to 50% of the observed color density. These new pigment forms are, moreover, much less sensitive to changes in pH. Somers has suggested, on the basis of the model reactions of flavylium salts discovered by Jurd and Waiss, that the anthocyanins, such as occur in wines, are susceptible to attack by nucleophilic reagents including phenolic substrates such as ( + )-catechin. The sequence of reactions by which the color of young wines may be altered and stabilized during aging is thought therefore to be of the type shown in Figure 21. In this case, electrophilic substitution into the oligomeric flavan-3-ol structure is carried out by the anthocyanin molecule leading by involvement of a red-ox reaction to the condensed antho-cyanin species (40). (Somers suggested the more stable deprotonated quinonoid structure for this oligomeric pigment.) The enhanced stability of these pigments is believed to be due
Downloaded by [Monash University Library] at 06:42 02 June 2013
Volume 27, Issue 1 (1988) 29
Precipitation
FIGURE 20. Proanthocyanidin oligomers — acid-catalyzed polymerization.
to the aryl substitution of the anthocyanin nucleus at the 4-position. Clearly, further sub-stitution of the anthocyanin is also possible by reaction with an additional oligomeric pro-cyanidin (16). Alternatively, polymerization initiated by the acid-catalyzed bond-making or bond-breaking processes discussed above (Figure 20) would lead ultimately to precipitation of the polymeric pigmented species and at the same time to loss of astringency in the wine in an entirely analogous manner to that described previously.
Finally, another electrophilic condensation reaction that is probably of significance in the aging of red wines with the associated loss of astringency is that with acetaldehyde.96 This reaction is entirely analogous to that described above with vanillin (Figure 18), and the cross-linking of separate oligomeric flavan-3-ol molecules in solution would eventually lead to molecular species of sufficient size to precipitate from solution (Figure 22). Oxidation and its control have been primary considerations in both wine making and aging for more than a century. Riberau-Gayon97 showed that a wine under barrel storage would be expected
Downloaded by [Monash University Library] at 06:42 02 June 2013
anlhocyanin
Procyanidin oligomers(ejg.l6 )
5-sugar
(40)
FIGURE 21. Proanthocyanidins — electrophilic substitution and polymerization.
Reaction with anthocyanins in red wines.
to adsorb about 30 mg O2 per litre per year and even in the corked bottle there is evidence for oxygen penetration. Acetaldehyde formation in wine aging presumably derives from the oxidation of ethanol and in a very early series of papers it was clearly implied that it subsequently combined with and precipitated the tannins and pigments of red wine. The typical cross-linking reactions suggested in Figure 22 are almost certainly the manner in which this final mode for the loss of the astringent principles (oligomeric procyanidins) of red wine takes place. A complete understanding, particularly at the quantitative level, of the manner in which each particular substrate in red wine participates and changes during maturation is not yet possible. However, the evidence presently available strongly suggests that each of the processes outlined in Figures 20 to 22 is strongly implicated in the poly-merization of polyphenols which undoubtedly underlies the processes involved in the loss of astringency from red wines on aging.
Downloaded by [Monash University Library] at 06:42 02 June 2013
Volume 27, Issue 1 (1988) 31
(16)
Proanthocyanidin oligomers in fruil tissue-soluble'
Me
'cross-1 inking'
— reduces solubility,
& hence astringency.
FIGURE 22. Proanthocyanidins — electrophilic substitution and polymerization.
Reaction with acetaldehyde in red wines and fruit.