|
A Technical Guide for Wine Producers
|
|
RECENT ARTICLES | WYNBOER HOME
The Occurrence Of Bitterness In Wine: An Overview
Astringency, as opposed to bitterness, is a tactile sensation that may also be ascribed to the occurrence of phenols. Astringency is also described as a puckering, drying out sensation in the mouth. The sensation of astringency is possibly caused by the interaction of the phenolic components or tannins with glycoprotein in the saliva, causing a decrease in the "lubricating activity", hence the dry sensation (Sarni-Manchado & Cheynier, 1999). According to Gawel (1998) tannins are chemically defined as water soluble phenolic components with molecular masses ranging from 500 to 3000 causing the precipitation of alcaloids, gelatine and other proteins. The monomers of phenols (molecular mass below 500) are also astringent, although chemically they do not qualify as astringent substances.
The aim of this article is to give the reader an overview of the current information available in literature about the occurrence and causes of bitterness in white and red wines.
The "phenol factor"
a) General
The bitter taste of wine is ascribed mainly to the specific phenolic composition of wine which is directly dependent on the phenolic composition of grapes, as well as on winemaking techniques. Flavonoid phenols are the main cause of bitterness in wine. Anthocyanins only contribute to astringency in wine and not to bitterness. In some red wines the cause of bitterness is ascribed not only to the phenolic composition of the wine, but also to microbiological activity. The microbiological breakdown of glycerol forms acrolein as a product which causes bitterness in wine by binding with phenolic components (Singleton, 1995).
b) Flavonoid phenols
The phenol fraction of grapes and wine consists mainly of flavan-3-oles of which (-)-epicatechin and (+)-catechin (Fig. 1), as well as their dimers, oligomes and polymers, constitute the largest percentage. Flavan-3-oles occur mainly in the skin, seeds and stems of grapes. Alternative names encountered in the literature for polymeric flavan-3-oles, are procyanidins, proanthocyanidins, condensed tannins, flavans and flavolans. According to Gawel (1998) both monomeric and polymeric flavan-3-oles are bitter, as well as astringent. The monomer flavan-3-ole molecule contains a C6-C3-C6 carbon skeleton of which, as mentioned, (-)-epicatechin and (+)-catechin are the most important in grapes (Fig. 1). Other flavan-3-oles occur in smaller concentrations, e.g. (-)-epigallocatechin, (+)-gallocatechin and (-)-epicatechin-3-O-gallate. The monomeric forms of (-)-epicatechin and (+)-catechin occur mainly in the seeds of grapes, while the skins contain larger amounts of polymerised flavan-3-oles. Procyanidin B1, an epicatechin-catechin-dimer (Fig. 2), is the most important dimer found in red and white skins (Gawel, 1998).
All phenol fractions (fractioned according to size) are described as bitter, as well as astringent (Noble, 1990; 1994). Smaller phenolic compounds are described as more bitter and less astringent, while larger phenolic compounds are more astringent and less bitter (Noble, 1999; Peleg et al., 1999; Robichaud & Noble, 1990; Arnold et al., 1980). The ratio of bitterness to astringency is the biggest in monomers. As the monomers polymerise to tetramers (four linked monomers) (Fig. 3), causing the molecular mass to increase, the astringency, relative to the bitterness, increases more (Arnold et al., 1980). This notion is confirmed by Peleg et al. (1999), their results showing that the monomers are more bitter than trimers and that bitterness decreased and astringency increased with an increase in molecular size. Noble (1994) also found that a phenol fraction containing polymers of 6 and more flavonoids, was 25 to 30 times more bitter and astringent than the same concentration of monomers, but that the ratio of bitterness to astringency was highest in the monomers. Eventually further polymerisation of the phenols results in a decrease in astringency (Noble, 1990). The taste of young red wines, which contain a bigger amount of small oligomers (flavonoids with fewer than 4 units, mainly di- and trimers), is described as "hard" (bitter and astringent). Older red wines contain more polymerised phenols (polimers of 8 to 10 or more units) and their taste is described as "soft" (less bitter and mainly astringent) (Noble, 1994).
(-)-Epicatechin is significantly more bitter and astringent than its chiral isomer, (+)-catechin (Noble, 1999; Kallithraka et al., 1997; Thorngate & Noble, 1995). Grape dimers and trimers are polymers of (-)-epicatechin and (+)-catechin and the binding site of the dimers also has an influence on the bitterness of the dimer. So for example the catechin-catechin dimer with a C4-6 binding site, was more bitter than the catechin-catechin dimer with a C4-8 binding. The former is also more bitter than the catechin-epicatechin dimer bound with a C4-8 binding (Noble, 1999; Peleg et al., 1999).
c) Non-flavonoid phenols
In both red and white cultivars cinnamic acid derivatives (non-flavonod phenols with a C6-C3-carbon skeleton, Fig. 4) are encountered in the juice part of the vacuoles of the grape. Cinnamic acid derivatives occur in concentrations as high as 200 mg/l in grapes and are released in the juice during the pressing process (Noble, 1990).
Cinnamic acid derivatives are the most important phenolic components occurring in white wines that received minimal skin contact. The two most important cinnamic acid derivatives occurring in grapes are caffeoyl tartaric acid, comprising the highest concentration, and coumaroyl tartaric acid which occurs in low concentrations (Vrette et al., 1988). These two cinnamic acid derivatives are bitter in a watery solution, but in the same concentrations in white wine they make no significant contribution to bitterness. In red wines the concentration cinnamic acid derivatives is considerably lower than that of the flavonoid phenols and they probably only contribute to bitter nuances in the wine (Gawel, 1998).
With exposure of the juice to air, as well as with polyphenol oxidase activity, caffeoyl tartaric acid oxidates to produce trans-2-S-glutathionyl caffeoyl tartaric acid as product. The concentration of trans-2-S-glutathionyl caffeoyl tartaric acid occurring in the wine depends on the enzyme activity during juice processing. This component does not contribute to bitterness in wine (Vrette et al., 1988). According to Nagel & Graber (1988), wine made from the oxidised juice of Chardonnay and Weisser Riesling did indeed produce more bitter wine than cases where the juice was not oxidised. However, the reason is unknown.
According to Vrette et al. (1988), a bitter taste cannot be ascribed to cinnamic acid derivatives, but to flavonoid phenols instead; even so, the possibility cannot be ruled out that hydroxy cinnamic acids may contribute to bitterness in wine through a synergistic procedure.
d) The effect of vinification practices on phenolic composition of wine
The concentration of phenols in a wine has a direct influence on the bitterness, astringency and ageing potential of that wine. Factors that influence the phenol concentration of grapes (and therefore eventually the wine), are i.a. grape cultivar, region (cooler regions have higher phenol concentrations) and degree of ripeness of the grapes. If the stems are removed and the seeds are not damaged during the crushing process, phenols are extracted mainly from the skins. The vinification technique, e.g. the temperature of the grapes during crushing, the time of skin contact, fining and barrel maturation, have a direct influence on the concentration and kind of phenols that eventually occur in the wine (Noble, 1990).
In the course of red wine vinification, when fermentation occurs on the skins, the phenols are rapidly extracted during the first two to four days. The total phenol concentration of red wines usually ranges between 1000 and 3500 mg/l. Maximum phenol extraction is reached within 7 to 21 days of skin contact, whereafter the rate of extraction decreases (Auw et al., 1996; Noble, 1990). The maximum recovery of phenols from grapes is between 22% and 33% of the total phenols of the grape, which includes the stems and seeds. If additional skin contact is applied after maximum extraction, there is no further significant increase in bitterness and astringency (Noble, 1990). According to Castellari et al. (1998), wines that underwent micro-oxygenation during fermentation contained lower concentrations of total phenols and anthocyanins at the end of fermentation, compared to the control. The incorporation of oxygen during fermentation and maturation accelerates the polymerisation of phenols. During maturation (a slow process of oxidation) polymerisation of phenols occurs, eventually resulting in precipitation of the polymers (Noble, 1990). Young red wines described as "hard" (bitter and astringent), benefit from the polymerisation process in that the bitter, lower molecular mass polymerises phenols during maturation to phenols with a higher molecular mass. Such wines are "softer" (less bitter and mainly astringent) during maturation (Noble, 1994).
The total phenol concentration of white wines, vinified with minimal skin contact, is between 100 and 250 mg/l. Approximately 30 mg/l of this consists of flavonoid phenols (Fischer & Noble, 1994; Noble, 1990). With skin contact not only aroma compounds are extracted from the skin, the total flavonoid phenol concentration of the wine also increases drastically (for Chardonnay, up to 110 mg/l). After 16 to 24 hours skin contact, this increase did not in all instances result in a significant increase in bitterness in the wines. Even after applying skin contact in various cultivars for five days, the wines of two cultivars still did not display bitterness that differed significantly from the control (Noble, 1990).
Flavonoid phenols occur in low concentrations in oak. The most important phenolic compounds in oak are ellagitannins (polymers of gallic acid) (Fig 5). These tannins are extracted from the wood and may contribute to bitterness and hardness in the wine. It is especially prominent in the case of maturation of white wines in new oak barrels, particularly barrels made from European oak (Singleton, 1995). During wood maturation of wine slow oxidation occurs which results in polymerisation of the phenols. Eventually the phenols will precipitate as polymers, causing a decrease in bitterness and astringency in wine (Noble, 1990).
Tannins can be removed from wine by the application of phenol removing fining agents. Protein fining agents, e.g. egg white, gelatin, casein and isinglass may be used selectively to reduce bitterness and astringency and/or harshness in wines. Polyvinyl polypyrrolidone (PVPP) adsorbs smaller phenolic compounds selectively and is therefore applied to remove the smaller, more bitter phenols from wine (Noble, 1990).
Unknown causes of bitterness
Often wines made from muscat cultivars, such as Gewrztraminer and Weisser Riesling, are bitter and the wines require residual sugar to mask the bitterness. The bitterness is not correlated to the total phenol content of the wine. According to Noble (1994), Crespo and Singleton (1986) found a correlation between bitterness and the concentration of terpene glycocides, but this finding was recently contradicted in that the addition of double the amount of extracted glycocides from a Muscat d'Alexandrie wine to a model wine solution and to wine, did not have a significant influence on the bitterness (Noble, 1990). It is therefore not known yet what causes bitterness in muscat type wines.
Bitter substances that do not contribute to bitterness in wine
Tyramine, tyrosol and chlorogenic acid are also bitter phenols encountered in wine, but their concentrations are too low to make a substantial contribution to the bitterness of wine (Noble, 1990; 1994).
Spores of the bitter glycocide, esculin, which is extracted during barrel maturation, are encountered in such wine. However, the component occurs below its threshold value (3200 mg/l) in wine and does not therefore contribute to the bitterness of wine (Noble, 1994).
Factors influencing the observation of bitterness
An increase in the viscosity of wine does not have a significant influence on the observation of bitterness, while an increase in the degree of sweetness does indeed mask bitterness (Noble, 1999). Ethanol increases the intensity of the bitter taste, as well as the duration of the bitter sensation (Noble, 1999; Fischer & Noble, 1994; Noble, 1990). An increase in alcohol concentration resulted in an increase in the bitter sensation (Fischer & Noble, 1994). An increase in pH has no effect on the intensity of bitterness, but extends the duration thereof (Noble, 1990). A decrease in the pH has only a small and inconsequent influence on the observation of bitterness (Noble, 1999).
Tasters also differ in their ability to observe bitterness. Persons who experience bitterness intensely have a larger number of bitterness observation seats on the tongue than those who do not experience bitterness as intensely. Tasters who have a higher saliva flow rate, experience the bitter sensation later and not as intensely than the tasters who have a lower saliva flow rate (Noble, 1999). According to Noble (1995), it takes approximately 15 seconds for a taster to experience maximum intensity bitterness.
Phenols cause both an astringent and a bitter sensation in the taster. The influence of astringency therefore perplexes the observation of bitterness in the judges and there is also confusion about the two tastes. These factors confound research on bitterness.
Microbiological cause of bitterness
This phenomenon is far from new and already in 1873 Louis Pasteur ascribed bitterness in red wines to the presence of bar-shaped bacteria and the loss of glycerol. Later Rentschler and Tanner (1951) suggested that acrolein was formed by the bacterial breakdown of glycerol in wine and that acrolein reacted with phenols to produce a bitter product. The problem therefore occurs more readily in red wines containing higher total phenol concentrations than white wines (Sponholz, 1993).
Various bacterial species are bound by the breakdown of glycerol, although not all strains of a particular species are able to degrade glycerol. Only 1% of Oenococcus oeni, 12% of Pediococcus parvulus and 31% of Lactobacillus species could be linked to the breakdown of glycerol to form acrolein (Sponholz, 1993). Other species of bacteria that have been associated with acrolein formation, include Bacillus amacrylus (Rentschler & Tanner, 1951), Lactobacillus cellobiosus and Leuconostoc mesenteroides (Bartowsky & Henschke, 1995) .
Apart from ethanol and CO2, glycerol is formed by the yeast Saccharomyces cerevisiae as one of the by-products of alcoholic fermentation. The amount of glycerol formed by the yeast cell is dependent on the redox balance of the must, as well as the osmotic stress experienced by the yeast cell. An increase in the fermentation temperature causes the yeast cell to produce larger quantities of glycerol. The optimal fermentation temperature for glycerol production, however, is between 22řC and 32řC. A higher fermentation temperature is the most obvious reason why higher concentrations of glycerol are found in red wines. The pH of grape juice does not play a significant role in glycerol production. In high sugar concentrations, such as those in grapes at optimal ripeness, the yeast cell produces larger amounts of glycerol to compensate for the osmotic stress it experiences in the must of such grapes (Scanes et al., 1998).
The biochemical pathway for the formation of acrolein is initiated by a bacterial dehydratase-enzyme which converts glycerol to 3-hydroxipropionaldehyde. The aldehyde undergoes slow, spontaneous dehydration to acrolein in an acid medium (Fig. 6). The process occurs more rapidly at higher temperatures (Sponholz, 1993). Acrolein is an unsaturated aldehyde with the smell of horse-radish or mustard. This aroma is not usually noticeable in wine, but may be encountered in brandy since it is concentrated during distillation. Acrolein in itself is not bitter. The process by which wine becomes bitter involves the formation of acrolein from glycerol and the binding thereof with phenolic components in the wine to produce a bitter product (Rentschler & Tanner, 1951).
Sporadic cases of wine becoming bitter, presumably as a result of microbiological spoilage, have occurred recently. Possible causes might be that grapes are being pressed fairly ripe and that high sugar concentrations put high osmotic pressure on yeast cells in the fermentation medium. As a stress response, yeast cells form glycerol which serves as a building block for acrolein. When grapes are pressed very ripe, the pH of the grapes is usually also higher. A higher pH reduces the efficiency of the antibacterial action of added SO2. There is also a trend to make wines with a lower total SO2 than in the past. Higher fermentation temperatures are also common, especially with Pinotage. All these factors contribute to higher populations of undesirable bacteria in red wines which in time form acrolein and causes bitterness.
Summary
The phenol composition of wine is directly dependent on the phenol composition of grapes, as well as winemaking techniques. Non-flavonoid phenols (cinnamic acid derivatives) represent the main group of phenols in white wines that received minimal skin contact, but are also encountered in red wines. Cinnamic acid derivatives as well as their oxidation products do not make a significant contribution to bitterness in wine. The most important group of phenolic components causing bitterness in red and white wine, are flavonoid phenols (flavan-3-oles). These phenols are extracted from the skins, stems and seeds of grapes in the course of vinification. The vinification method may make a substantial contribution to bitterness in wine, since it has a direct influence on the phenolic composition of wine.
Monomers, dimers and trimers of the flavan-3-oles are more bitter and less astringent than the bigger polymers. During maturation the smaller phenolic components polymerise, with the result that the matured wines contain larger concentrations of the bigger polymers. The taste of matured wines is described as "soft" (astringent and less bitter), while the taste of young red wines, which contain larger concentrations of smaller phenolic combinations, is described as "hard" (bitter and astringent). Excessive maturation will result in a decrease in astringency as a result of the polymerisation and precipitation of tannins.
All phenol fractions are bitter as well as astringent. If the phenol fractions do not contribute independently to bitterness in wine, they might give rise to bitterness by synergistic action. The interaction between bitterness and astringency complicates the sensorial evaluation of bitterness, as well as the study of the effect of vinification techniques on phenols and the occurrence of bitterness. Not only do tasters differ physiologically in their ability to experience bitterness, bitterness is also influenced by the degree of sweetness and the alcohol concentration of the product.
Other well-known bitter substances such as esculin, chlorogenic acid, tyramine and tyrosol which occur in wine, are present in concentrations below their threshold values and do not therefore make a substantial contribution to bitterness. However, the cause of bitterness in muscat type wines is still not known.
It is also possible that a wine may become bitter as a result of the microbiological formation of acrolein. This problem is only encountered in red wines, however. Acrolein in itself is not bitter, but a bitter component is formed after binding to phenol fractions in the wine. Modern winemaking techniques in which grapes are pressed at a very ripe stage, less SO2 is added and warm fermentation temperatures are used, combined with high pH's, could possibly be conducive to the occurrence of this kind of bitterness. Further studies in this regard must still be conducted, since little is known about the specific bacteria and phenolic groups involved.
Queries may be addressed to the author at adele@infruit.agric.za, Tel. (021) 809 3091 or Fax (021) 809 3002.
Literature
Arnold, R.A., Noble, A.C. & Singleton, 1980. Bitterness and astringency of phenolic fractions in wine. J. Agric. Food Chem. 28, 675-678.
|
|
Wynboer is incorporated in WineLand, magazine of the SA wine producers.
Visit our sister sites:
|
| UP |
COPYRIGHT (C) 2000 WineLand |