A REVIEW Malolactic fermentation in wine beyond deacidification

Transcription

1 Journal of Applied Microbiology 2002, 92, A REVIEW Malolactic fermentation in wine beyond deacidification S.-Q. Liu New Zealand Dairy Research Institute, Palmerston North, New Zealand 2001/186: received 10 July 2001, revised 30 October 2001 and accepted 13 November 2001 S. - Q. L I U Introduction, Citrate fermentation, Metabolism of carbohydrates, Metabolism of mono- and disaccharides, Metabolism of polysaccharides, Metabolism of polyols, Catabolism of aldehydes, Hydrolysis of glycosides, Degradation of phenolic acids, Synthesis and hydrolysis of esters, Lipolysis, Proteolysis and peptidolysis, Amino acid catabolism, Sensory impact, Health implications, Formation of amines, Formation of ethyl carbamate precursors, Formation of glyoxal and methylglyoxal, Conclusions, References, INTRODUCTION Malolactic fermentation (MLF) in wine is a secondary fermentation that usually occurs at the end of alcoholic fermentation by yeasts, although it sometimes occurs earlier. It is practically a biological process of wine deacidification in which the dicarboxylic L-malic acid (malate) is converted to the monocarboxylic L-lactic acid (lactate) and carbon dioxide (Davis et al. 1985). Deacidification is particularly desirable for high-acid wine produced in cool-climate regions, such as New Zealand and Canada. This process is normally carried out by lactic acid bacteria (LAB) isolated from wine, including Oenococcus oeni (formerly Leuconostoc oenos; Dicks et al. 1995), Lactobacillus spp. and Pediococcus spp. (Wibowo et al. 1985). Various technologies, such as bioreactors with high-density cells and immobilized cells or enzymes, have been developed to facilitate wine deacidification (Maicas 2001). Oenococcus oeni is the preferred species used to conduct MLF due to its acid tolerance and flavour profile produced. In addition to its occurrence in wine, MLF occurs in other fermented beverages, such as cider (Carr 1987; Jarvis et al. 1995). Numerous reviews on MLF have been published since the 1960s, covering the history of research on MLF, taxonomy, ecology, genetics and physiology of wine LAB and the oenological significance of MLF (Bartowsky and Henschke Correspondence to: S.-Q. Liu, New Zealand Dairy Research Institute, Palmerston North, New Zealand ( shao.liu@nzdri.org.nz). 1995; Kunkee 1967, 1974, 1991; Lafon-Lafourcade 1983; Davis et al. 1985; Wibowo et al. 1985; Henick-Kling 1988, 1993, 1995; Henschke 1993; van Vuuren and Dicks 1993; Lonvaud-Funel 1995, 1999; Versari et al. 1999). However, a review that is specifically focused on the metabolic capacity of wine LAB and its impact on wine quality is still lacking. Furthermore, new information on the metabolism of wine LAB has emerged since the publication of these reviews. It is now accepted that the role of MLF is more than just a deacidification process, although deacidification via MLF is still a primary objective in wine fermentation in cool-climate regions. The complexity and diversity of the metabolic activity of LAB suggest that MLF may affect wine quality both positively and negatively. Knowledge of the metabolism of wine LAB is necessary to assess the impact of MLF on wine quality. The primary focus of this review is to evaluate the current knowledge of the metabolism of wine LAB (predominantly oenococci) with a view to stimulating further research in this area. The author also aims to assess the potential impact of MLF on wine flavour, mouthfeel/body and human health in relation to the metabolism of wine LAB. References are made with respect to the metabolism of LAB from other fermented beverages and foods, such as cider and cheese. 2. CITRATE FERMENTATION Citrate is one of the major organic acids in grape juice and wine, besides malate and tartrate. As with citrate-fermenting ª 2002 The Society for Applied Microbiology

2 590 S.-Q. LIU dairy LAB, many wine LAB are known to metabolize citrate. The metabolism of citrate by dairy LAB has been well characterized (Hugenholtz 1993). In contrast, the fermentation of citrate by wine LAB has received relatively little attention. Wine LAB follow a similar metabolic pathway to that of dairy LAB in the metabolism of citrate. Using nuclear magnetic resonance (NMR) spectroscopy and a radiolabelling technique, Ramos et al. (1995) have demonstrated that the metabolic pathway of citrate metabolism in O. oeni is essentially the same as that in citrate-fermenting dairy LAB. Citrate is transformed to lactate, acetate, diacetyl, acetoin and 2,3-butanediol. A small amount of citrate is converted to aspartate via oxaloacetate and aspartate aminotransferase. The formation of aspartate from citrate has also been reported in dairy Leuc. mesenteroides (Marty-Teysset et al. 1996), although not in dairy Lactococcus lactis subsp. lactis biovar diacetylactis. The degradation of citrate by wine lactobacilli has been reported but has not been studied in detail, whereas wine pediococci do not degrade citrate (Davis et al. 1986a; Liu et al. 1995a). The co-fermentation of citrate and glucose in O. oeni is physiologically important for this bacterium. This co-metabolism of citrate glucose has been shown to enhance the growth rate and biomass yield of this bacterium, which result from increased ATP synthesis both by substrate-level phosphorylation via acetate kinase and by a chemiosmotic mechanism (proton motive force) (Salou et al. 1994; Ramos and Santos 1996). The growth stimulation of citrate sugar co-fermentation by the same mechanisms has also been reported in citrate-fermenting dairy LAB (Cogan 1987; Schmitt and Diviès 1991; Ramos et al. 1994; Bandell et al. 1998). From a winemaker s point of view, the co-metabolism of citrate sugar increases the formation of the volatile acid (acetate) in wine, which can affect the wine aroma detrimentally if present at excessive levels. The most important oenological significance associated with citrate fermentation is the production of diacetyl, an aroma compound with a buttery flavour note. In general, wines that have undergone MLF have higher concentrations of diacetyl (Martineau et al. 1995). The final level of diacetyl in wine is affected by a number of factors, such as bacterial strain, wine type, sulphur dioxide and oxygen (Nielsen and Richelieu 1999; Martineau and Henick-Kling 1995a). It should be pointed out that diacetyl is formed chemically from the oxidative decarboxylation of a-acetolactate, an unstable intermediary compound produced during citrate metabolism (Ramos et al. 1995). Martineau and Henick- Kling (1995b) observed the utilization of diacetyl by oenococci. This is not surprising, since many LAB, including oenococci, contain diacetyl reductase that converts the flavourful diacetyl to the much less flavourful acetoin and 2,3-butanediol (Seitz et al. 1963; Ramos et al. 1995; Rattray et al. 2000). The formation of diacetyl by LAB enhances the buttery aroma of wine, whereas the reduction of diacetyl decreases the buttery aroma. As mentioned above, several factors can influence diacetyl formation and reduction. Oxygenation, high concentrations of citrate and sugars, lower temperature (18 C), removal of yeast cells before MLF and low inoculation rate favour the production of diacetyl (Martineau et al. 1995). On the contrary, the presence of viable yeast cells during MLF, prolonged contact with LAB and addition of SO 2 cause diacetyl reduction. 3. METABOLISM OF CARBOHYDRATES 3.1 Metabolism of mono- and disaccharides The metabolism of carbohydrates (mono- and disaccharides) in LAB has been reviewed elsewhere (Kandler 1983; Axelsson 1993). For completeness, only a brief discussion of carbohydrate fermentation is given below in relation to wine LAB and MLF. Wine contains a range of monosaccharides (pentoses and hexoses) and disaccharides, with arabinose, glucose, fructose and trehalose being the major sugars (Liu and Davis 1994). The utilization of sugars by wine LAB as carbon and energy sources during MLF has been demonstrated in a number of studies and there exist species and strain differences in sugar utilization (Davis et al. 1986a, b; Salou et al. 1994; Liu et al. 1995a). Glucose and trehalose are generally preferred over other sugars (Liu 1990; Liu et al. 1995a). The metabolic pathways of sugars have not been fully elucidated in wine LAB, especially oenococci (Garvie 1986). Presumably, wine LAB follow similar pathways to those of other LAB, which are summarized briefly below. Pentoses are metabolized by heterofermentative and facultative homofermentative LAB via the pentose phosphate pathway with the formation of 1 mole each of lactate and acetate and 2 moles ATP per mole pentose used. Homofermentative LAB ferment hexoses through the Embden-Meyerhof-Parnas (EMP) pathway to 2 moles each of lactate and ATP per mole hexose utilized. However, hexoses are fermented by heterofermentative LAB via the phosphoketolase pathway to 1 mole each of lactate, ethanol, CO 2 and ATP per mole hexose consumed. Being a ketose, fructose can also act as an electron acceptor and is reduced to mannitol. Consequently, acetyl phosphate formed during hexose fermentation is converted to acetate instead of being reduced to ethanol, generating an additional ATP (Pilone et al. 1991; Salou et al. 1994). Besides fructose, heterofermentative LAB can use other substances as electron acceptors, such as oxygen and pyruvate, resulting in the production of acetate and additional ATP. The fermentation of disaccharides (trehalose and sucrose) by wine LAB has not been studied. It is not clear how the

3 MALOLACTIC FERMENTATION IN WINE 591 disaccharides are metabolized: hydrolysis or conversion to monosaccharides by a hydrolase(s) or phosphorylase(s). The resultant monosaccharides then enter the common pathways of sugar fermentation after being released from the disaccharides. 3.2 Metabolism of polysaccharides Little is known about the formation of exocellular polysaccharides by wine LAB. Some wine LAB are known to cause ropiness in wine. For example, Pediococcus damnosus, isolated from a ropy wine, produces a b-d-glucan composed of a trisaccharide repeating unit of D-glucose (Llaubères et al. 1990; Lonvaud-Funel et al. 1993). In contrast, a great deal of information is available on the exocellular polysaccharides produced by LAB of other origins, such as fermented dairy products (Cerning 1990; Malik et al. 1994; Sikkema and Oba 1998). Wine LAB may be able to catabolize polysaccharides besides biosynthesis. Oenococcus oeni has been shown to possess extracellular b(1- > 3) glucanase activity (Guilloux- Benatier et al. 2000). This report provides the first evidence that O. oeni has the potential to degrade polysaccharides, such as glucan. Increases in the concentrations of glucose and fructose during MLF in wine have been reported and various explanations are given, for example, acid hydrolysis of sucrose, trehalose and phenolic glucosides (Costello et al. 1985; Davis et al. 1986a, b). Based on the above evidence of glucanase activity exhibited by oenococci, the enzymatic hydrolysis of polysaccharides by wine LAB may, at least partially, account for the above-mentioned production of glucose and fructose during MLF. Clearly, further work is required to study the polysaccharide-hydrolysing enzymes in a wide range of wine LAB. Polysaccharides can affect wine processing, such as filtration, due to the increased viscosity. Polysaccharides may also affect sensory properties of wine, such as mouthfeel and body. Excessive levels of polysaccharides in wine are undesirable in terms of causing ropiness. However, moderate levels of polysaccharides may be beneficial for wine body and mouthfeel. This aspect remains to be exploited in wine fermentation. 3.3 Metabolism of polyols Oenococcus oeni is classified as being heterofermentative. However, the heterofermentative pathway of glucose has not been fully confirmed in this bacterium (Garvie 1986). Using the 13 C-NMR technique, Veiga-da-Cunha et al. (1992) confirmed the biosynthesis of glycerol and erythritol from glucose in O. oeni and that the ratio of erythritol to glycerol is dependent on the oxygen level. The production of glycerol, erythritol and other polyols by oenococci and other wine LAB has also been observed by other researchers (Firme et al. 1994; Liu et al. 1995a). According to Veigada-Cunha et al. (1993), the pathway of erythritol formation from glucose involves the isomerization of glucose 6-phosphate to fructose 6-phosphate, the cleavage of the latter to produce erythrose 4-phosphate and acetyl phosphate, the reduction of erythrose 4-phosphate to erythritol 4-phosphate and the hydrolysis of erythritol 4-phosphate to form erythritol. Therefore, the formation of polyols is essentially an alternative pathway to NAD(P)H reoxidation in addition to the conventional ethanol pathway. CoA-SH deficiency is apparently the reason for the shift to the formation of erythritol, acetate and glycerol from glucose when there is a lack of pantothenic acid, because phosphotransacetylase and acetaldehyde dehydrogenase activities become limiting under pantothenic acid deficiency (Richter et al. 2001). Mannitol, one of the major polyols in wine, is formed by the reduction of fructose, as discussed above (Section 3.1). Some wine lactobacilli are known to degrade glycerol and mannitol, two of the main polyols in wine. Lactobacillus brevis and Lact. buchneri, isolated from spoiled wine, can metabolize glycerol in the presence of glucose or fructose, resulting in the formation of 3-hydroxy propanal (also known as 3-hydroxy propionaldehyde, 3-HPA), which is subsequently reduced to 1,3-propanediol (Schutz and Radler 1984a, b). 3-Hydroxy propionaldehyde is a precursor to acrolein, a bitter compound found in alcoholic beverages, such as wine and cider. The conversion of glycerol to 3-HPA in co-metabolism with glucose or fructose is not restricted to wine lactobacilli. Lact. collinoides, isolated from spoiled cider and fermented apple juice, is also able to do so (Claisse and Lonvaud-Funel 2000; Sauvageot et al. 2000). It should be stressed that glycerol in the absence of glucose or fructose does not serve as a sole carbon and energy source to support growth of any of these lactobacilli. Physiologically, the co-metabolism of sugar and glycerol is important to these lactobacilli, as additional ATP is generated from acetyl phosphate (Veiga-da-Cunha and Foster 1992). Contrary to glycerol, mannitol can serve as a sole carbon and energy source to support growth of Lact. plantarum isolated from fermented vegetables (Chen et al. 1983) and wine (Davis et al. 1988; Liu et al. 1995a). However, the catabolism of mannitol by Lact. plantarum requires the presence of oxygen (aerobic metabolism) or compounds such as citrate and a-ketoacids that can directly or indirectly act as electron acceptors (anaerobic metabolism) (Chen and McFeeters 1986a,b; McFeeters and Chen 1986). Lact. plantarum, isolated from fermented green beans, ferments 1 mole mannitol via mannitol dehydrogenase to 2 moles lactate with the formation of other compounds, such as ethanol, and the type of such compounds formed is

4 592 S.-Q. LIU dependent on the presence of a particular electron acceptor (McFeeters and Chen 1986). The fermentative pathways of mannitol in Lact. plantarum isolated from wine have not been investigated. Presumably, similar pathways are also operative in this bacterium. The metabolism of polyols by wine LAB is expected to have oenological significance. The production of polyols may influence sensory properties of wine, such as body and mouthfeel, as well as wine filtration. The formation of acrolein from glycerol can impart a bitter taste to wine. From the standpoint of microbial stability, a successive growth of Lact. plantarum may occur after MLF by oenococci, because the wine ph is increased and mannitol is produced by oenococci. Mannitol may then be fermented by Lact. plantarum to produce excessive levels of lactate, presenting a potential spoilage problem. In the wine industry, mannitol is considered a bacterial spoilage product and the presence of sorbitol (glucitol) is an indication of addition of fruits, particularly apples (Amerine et al. 1980). Further, the presence of erythritol and arabitol in table wine is often associated with mouldinfected grapes (Ough and Amerine 1988). From the discussion above, it is clear that wine LAB also play a role in the biosynthesis of polyols. 4. CATABOLISM OF ALDEHYDES Wine contains various volatile aldehydes that contribute important sensory properties (de Revel and Bertrand 1993; Ebeler and Spaulding 1999). Among the aldehydes, acetaldehyde is quantitatively the most abundant aroma compound present in wine. The formation and significance of acetaldehyde in wine have been reviewed elsewhere (Liu and Pilone 2000). Briefly, acetaldehyde is produced mainly by yeasts during alcoholic fermentation and this compound can affect wine ageing and colour stability. In addition, acetaldehyde, hexanal, cis-hexen-3-al and trans-hexen-2-al cause the green, grassy and vegetative off-aroma in wine (Huis in t Veld 1996; Ebeler and Spaulding 1999; Kotseridis and Baumes 2000; Liu and Pilone 2000). In view of the importance of aldehydes to wine aroma, the removal of aldehydes is necessary when there is an excess. Traditionally in the wine industry, SO 2 is added to bind acetaldehyde when this compound is in excess (Somers 1998). It has recently been found that some LAB (especially oenococci) can catabolize acetaldehyde, converting it to ethanol and acetate (Osborne et al. 2000; Osborne 2000). This offers an alternative approach to SO 2 addition to remove acetaldehyde. The removal of other aldehydes that can cause off-aroma may also be important and, presumably, wine LAB (especially oenococci) can also metabolize these aldehydes. This is supported by the fact that dairy leuconostocs are able to reduce propanal to propanol (Keenan 1968). In addition to acetaldehyde utilization, LAB may produce acetaldehyde. For example, dairy lactococci and lactobacilli can produce acetaldehyde (Liu et al. 1997; Ott et al. 2000). It is not known whether wine LAB can produce acetaldehyde. Further research in this area will be fruitful. 5. HYDROLYSIS OF GLYCOSIDES Monoterpenes are important aromatic compounds present in grapes and wines (Ebeler 2001). Anthocyanins are the main pigments of red grapes and red wines. However, large amounts of monoterpenes and anthocyanins in grapes and wines are glycosidically bound to sugars, such as glucose (Stahl-Biskup et al. 1993; Mazza 1995; Baltenweck-Guyot et al. 2000; Ebeler 2001). The sugar-bound monoterpenes are non-volatile and flavourless. A glycosidase (e.g. b-glucosidase) hydrolyses the sugar-bound monoterpenes to release the volatile, aromatic monoterpenes as well as the sugar. In contrast, a glycosidase (also referred to as anthocyanase) hydrolyses the sugar-bound anthocyanins to liberate the sugar and the corresponding anthocyanidin, the latter spontaneously converts brown or colourless compounds (Blom 1983). In other words, an anthocyanase has a decolourizing effect. Much attention has been devoted to the b-glucosidase of wine yeasts, non-saccharomyces yeasts in particular (Rosi et al. 1994; Mateo and Di Stefano 1997; McMahon et al. 1999; Manzanares et al. 2000). By comparison, the glycosidase (b-glucosidase) in wine LAB has received little attention. McMahon et al. (1999) detected low b-glucosidase activity in O. oeni OSU and weak a-lrhamnopyranosidase activity in O. oeni Viniflora oenos. In contrast, Grimaldi et al. (2000) found readily detectable activities of b-glucosidase in 11 commercial preparations of oenococci. These findings suggest that wine LAB have the potential to hydrolyse glycoconjugates to affect wine aroma and colour. The presence of glycosidase activity in LAB from sources other than wine is known. De Cort et al. (1994) purified an a-glucosidase from Lact. brevis isolated from beer. Gueguen et al. (1997) purified and characterized a b-glucosidase from Leuc. mesenteroides isolated from cassava. More importantly, Lact. plantarum strains from cassava fermentation contain b-glucosidase that can degrade cyanogenic glycosides (Lei et al. 1999). These enzymes, or the LAB from which the enzymes were isolated, may be used to detoxify food plants, such as cassava, that contain cyanogenic glycosides by way of enzymatic hydrolysis or fermentation. The b-glucosidase activity of wine lactobacilli and pediococci has not been studied. This area merits further study.

5 MALOLACTIC FERMENTATION IN WINE DEGRADATION OF PHENOLIC ACIDS Phenolic acids (mainly ferulic and p-coumaric acids) are natural constituents of grape juice and wine. These phenolic compounds can be decarboxylated microbially during fermentation into volatile phenols such as 4-ethylguaiacol and 4-ethylphenol. The volatile phenols can contribute to wine aroma positively or negatively, dependent on the concentration, due to their low detection thresholds and their distinct flavour. Whereas wine yeasts are, to a large extent, responsible for the decarboxylation of phenolic acids (Shinohara et al. 2000), wine LAB may also be able to do this. It has been observed that the concentration of volatile phenols increased markedly during MLF (Etiévant et al. 1989), suggesting that wine LAB might be involved. This is verified by the catabolism of ferulic and p-coumaric acids by several wine lactobacilli and pediococci and the detection of corresponding volatile phenols (4-ethylguaiacol and 4-ethylphenol) (Cavin et al. 1993). The metabolism of phenolic acids, p-coumaric acids in particular, by Lact. plantarum has been well characterized (Barthelmebs et al. 2000, 2001). In contrast, the metabolism of phenolic acids by other wine LAB has not been characterized and warrants further study. Interestingly, lactobacilli isolated from malt whisky fermentation can also decarboxylate phenolic acids into volatile phenols (van Beek and Priest 2000). However, phenolic compounds can favourably and unfavourably affect the physiology and growth of wine LAB, dependent on the concentration and type of phenolic compounds (Vivas et al. 1997, 2000; Reguant et al. 2000, Alberto et al. 2001). 7. SYNTHESIS AND HYDROLYSIS OF ESTERS Esters, such as ethyl acetate and C 4 to C 10 fatty acid ethyl esters, are largely, if not exclusively, responsible for the fruity aroma of wine (Ebeler 2001). Yeasts are known to produce these fatty acid ethyl esters via esterases (esterification) and alcohol acetyltransferases during alcoholic fermentation (Suomalainen 1981; Yoshioka and Hashimoto 1981; Vianna and Ebeler 2001; Inoue et al. 1997; Bardi et al. 1998; Lilly et al. 2000). However, there is a lack of knowledge of ester formation by wine LAB during MLF. Tentative evidence is available that ethyl esters, such as ethyl acetate, ethyl lactate, ethyl hexanoate and ethyl octanoate, are formed during MLF (Dittrich 1987; de Revel et al. 1999; Delaquis et al. 2000). This suggests that wine LAB possess the ability to synthesize esters, which needs to be verified definitively by studying the enzyme system(s) of ester synthesis in these bacteria. In contrast, the ability of LAB and pseudomonads of dairy origin to esterify alcohols and fatty acids to form esters has been defined. For example, dairy isolates of lactococci, lactobacilli, streptococci and Pseudomonas fragi can form ethyl butanoate and ethyl hexanoate from ethanol and butanoic and hexanoic acids (Hosono and Elliott 1974; Hosono et al. 1974; Morgan 1976; Liu et al. 1998). Esterases are involved not only in ester synthesis but also ester hydrolysis in an aqueous environment. Dairy LAB contain esterases that can hydrolyse mono-, di- and triacylglycerols (Holland and Coolbear 1996; Gobbetti et al. 1997; Liu et al. 2001). Davis et al. (1988) showed that most strains of wine LAB oenococci, pediococci and lactobacilli examined had esterase activities. It is not clear whether these esterases can hydrolyse fatty acid ethyl esters. Knowledge of ester formation and hydrolysis by wine LAB would help in the understanding of the impact of MLF on the fruity aroma of wine (see below). It should be pointed out that the concentrations of some esters decrease while others increase during storage of wines, presumably due to acid hydrolysis and chemical esterification (Shinohara and Watanabe 1981; Ebeler 2001). 8. LIPOLYSIS Wine contains lipids that are comprised of tri-, di- and monoacylglycerols. Some of these lipids are released from yeast cells upon autolysis during alcoholic fermentation (Pueyo et al. 2000). The breakdown of lipids (lipolysis) may affect wine flavour by producing volatile fatty acids, which have low sensory thresholds (Brennand et al. 1989). Volatile fatty acids are natural components of alcoholic beverages, such as cider and wine (Blanco-Gomis et al. 2001; Shinohara 1985), and excessive levels of these compounds can have a negative influence on the sensory properties of cider and wine. There is a lack of information on the lipolytic system of wine LAB. A study by Davis et al. (1988) has shown that some wine LAB contain esterase and/or lipase activities. This suggests that wine LAB have the potential to hydrolyse lipids, but further work is needed in this area. By comparison, the lipolytic system in cheese LAB is well characterized and these LAB have esterases and lipases that can hydrolyse tri-, di- and monoacylglycerols (Holland and Coolbear 1996; Chich et al. 1997; Gobbetti et al. 1997; Liu et al. 2001). 9. PROTEOLYSIS AND PEPTIDOLYSIS Wine contains proteins that may be broken down by bacterial proteases and peptidases to generate peptides and amino acids to impact on wine flavour and stability. Although Davis et al. (1988) failed to detect protease activity in several wine LAB examined, including several strains of oenococci, pediococci and lactobacilli, the production of extracellular proteases by strains of O. oeni has

6 594 S.-Q. LIU been subsequently demonstrated (Rollan et al. 1993). These oenococcal proteases have also been partially characterized (Rollan et al. 1995a; Farias et al. 1996). Thus, wine LAB do appear to have the potential to break down wine proteins. Nonetheless, the proteolytic activity in oenococci is straindependent and not widespread among oenococci (Leitao et al. 2000). Manca de Nadra et al. (1997, 1999) have demonstrated the production of peptides and amino acids during MLF in both red and white wines by O. oeni X 2 L, the strain that produces extracellular proteases. It is not clear how the peptides may affect the wine taste, since some peptides can cause bitterness in cheeses (Visser 1993; Thomas and Pritchard 1987; Pritchard and Coolbear 1993). It is also not known how proteolysis during MLF may affect wine stability and haze formation. Clearly, these areas warrant further research. In spite of the above-mentioned studies, the proteolytic system of wine LAB remains poorly defined. In contrast, the proteolytic system of dairy LAB is well characterized. For detailed information, the reader is referred to review articles on this topic by Thomas and Pritchard (1987), Pritchard and Coolbear (1993), Visser (1993), Law and Mulholland (1995), Poolman et al. (1995) and Kunji et al. (1996). 10. AMINO ACID CATABOLISM The primary reactions of microbial catabolism of amino acids include decarboxylation, transamination, deamination and desulphuration (Hemme et al. 1982). Decarboxylation of amino acids leads to the formation of carbon dioxide and amines, the latter can have implications for human health (see below). Amino acids and a-keto acids are generated through the transamination of amino acids. Deamination of amino acids results in the production of ammonia and a-keto acids. Volatile sulphur compounds are produced via desulphuration of sulphur-containing amino acids, such as methionine and cysteine. The secondary reactions of amino acid catabolism involve the conversion of the above compounds (amines, a-keto acids and amino acids) to aldehydes. The reduction of the aldehydes to alcohols and/ or their oxidation to acids constitutes the final reactions of amino acid transformation. Few studies have been conducted on the catabolism of amino acids by wine LAB, with the exception of arginine (see below). The catabolism of amino acids by wine LAB is expected to have a significant impact on wine quality, given that a range of compounds can be produced, such as aldehydes, alcohols and acids, in addition to amines. For comparison, increasing research attention is being directed toward the metabolism of amino acids by cheese LAB because of its potential impact on cheese quality (flavour and safety). For details on the catabolism of amino acids by cheese LAB, the reader is referred to several recent review articles on this topic (Christensen et al. 1999; Weimer et al. 1999; McSweeney and Sousa 2000; Yvon and Rijnen 2001). A range of amino acids is found in wine (Sponholz 1991; Lehtonen 1996). The concentration of some amino acids increases, while that of others decreases during MLF (Davis et al. 1986a, 1986b). However, few studies have linked the utilization of specific amino acids with single strains of wine LAB. Arginine is a major amino acid present in wine. The catabolism of arginine by wine LAB and its practical significance has been studied in detail and is the subject of a recent review (Liu and Pilone 1998). Arginine-degrading wine LAB catabolize arginine via the arginine deiminase pathway (ADI) (Liu et al. 1996). Nonetheless, a strain of Lact. hilgardii can apparently degrade arginine by two pathways: arginine deiminase and arginine decarboxylase (Arena and Manca de Nadra 2001). This strain, however, may be atypical. Arginine degradation via the ADI pathway produces energy (ATP) which is beneficial for survival and growth of LAB, including wine LAB (Liu 1993; Champomier-Vergès et al. 1999; Stuart et al. 1999; Tonon and Lonvaud-Funel 2000; Tonon et al. 2001). Factors affecting arginine degradation include LAB strain, ph, arginine concentration and sugar type (Liu et al. 1995b; Granchi et al. 1998; Mira de Orduña et al. 2000a, b, 2001). Citrulline excreted during arginine degradation and ornithine can also be catabolized by some wine LAB (Liu et al. 1994; Arena et al. 1999; Mira de Orduña et al. 2000a). Some strains of O. oeni can catabolize serine with the formation of ammonia (Granchi et al. 1998). The catabolism of serine in LAB, including wine LAB, has not been studied. It is likely that serine is degraded by serine dehydratase (deaminase). This enzyme converts serine to ammonia and pyruvate, as reported in Brevibacterium linens of cheese origin (Hamouy and Desmazeaud 1985). Pyruvate can be further metabolized to formate, acetate, carbon dioxide, ethanol and/or diacetyl, being dependent on the presence of the particular enzyme system (Axelsson 1993). The utilization of tyrosine and phenylalanine by Lact. plantarum during MLF has been observed (Liu et al. 1995a). Whereas the metabolic pathway of the two amino acids in this species of Lactobacillus is not known, it is possible that they are decarboxylated to form the corresponding amines, tyramine and phenylethylamine, given the prevalence of the two amines associated with lactobacilli in wine (see below). The influence of aspartic acid on the growth and metabolism of malic acid and glucose by O. oeni was investigated by Vasserot et al. (2001). Although low concentrations (< 0Æ3 mmol l 1 ) of aspartic acid stimulated oenococcal growth, high concentrations (> 6 mmol l 1 )of

7 MALOLACTIC FERMENTATION IN WINE 595 this amino acid inhibited growth with corresponding reduced degradation of malic acid but increased utilization of glucose. Interestingly, an excessive amount of aspartic acid resulted in an overproduction of acetic acid and decreased ethanol formation. The reason for this is not known, because the metabolic pathway of aspartic acid in oenococci has not been elucidated. 11. SENSORY IMPACT The impact of MLF on the taste of wine as a result of deacidification is well recognized but the effect of MLF on wine aroma and mouthfeel/body is ill-defined. Davis et al. (1985) reviewed the contribution of MLF to wine aroma and found no consistent impact of MLF on wine aroma. Since 1985, further work has been carried out to investigate the effect of MLF on sensory properties of wine using a more stringent panel training method, gas chromatography (GC) olfactometry and GC mass spectrometry (McDaniel et al. 1987; Rodriguez et al. 1990; Henick-Kling 1993; Henick-Kling et al. 1994; Laurent et al. 1994; Sauvageot and Vivier 1997; Delaquis et al. 2000). Based on these studies, the current consensus is that MLF can indeed affect wine aroma and add complexity to wine flavour. These reports also indicate that the impact of MLF on wine flavour varies with wine LAB and wine type. Despite the significant influence of MLF on wine aroma, only certain wine attributes modified during MLF can be related to the production or utilization of specific chemical compounds by wine LAB. According to Henick-Kling (1993) and Henick-Kling et al. (1994), MLF enhances the fruity aroma and buttery note but reduces the vegetative, green/grassy aroma. The enhanced fruitiness may be the result of the formation of esters by wine LAB, as discussed in Section 7. The increased buttery note is known to arise from diacetyl produced from citrate fermentation by wine LAB (see Section 2). The reduction in vegetative, green/ grassy aroma may be due to the catabolism of aldehydes by wine LAB, as reviewed in Section 4. Besides aroma, MLF is believed to increase the body and mouthfeel of wine and give a longer after-taste (Henick-Kling et al. 1994). This may be ascribable to the production of polyols and polysaccharides by wine LAB (see Section 3). In addition to the fruity and buttery notes, other flavour characteristics associated with MLF are described as floral, nutty, yeasty, oaky, sweaty, spicy, roasted, toasty, vanilla, smoky, earthy, bitter, ropy and honey (Henick-Kling 1993; Henick-Kling et al. 1994; Laurent et al. 1994; Sauvageot and Vivier 1997). Further research is required to relate the wine attribute(s) altered during MLF to the production and/or degradation of a specific chemical compound(s) by wine LAB. This information will enable the winemaker to match the right strain of wine LAB to the right wine type so as to maximize or minimize a particular flavour attribute. 12. HEALTH IMPLICATIONS 12.1 Formation of amines As discussed above (Section 10), some LAB possess decarboxylases that decarboxylate amino acids to form corresponding amines and carbon dioxide. Amines are toxic substances that have deleterious effects on human health (Shalaby 1996). They are found in a range of fermented foods and beverages, such as fermented fish, cheeses, beer and meat products (Stratton et al. 1991; Shalaby 1996). Some lactobacilli, lactococci and leuconostocs isolated from beer, cheeses and meat starter cultures are known to produce histamine and tyramine via the decarboxylation of the corresponding amino acids, histidine and tyrosine (Zee et al. 1981; Maijala 1993; Gonzalez de Llano et al. 1998). As a fermented product, wine also contains amines. The major amines found in wine are histamine, tyramine, putrescine and phenylethylamine (Radler and Fath 1991; Lehtonen 1996; Lonvaud-Funel 2001). The role of wine LAB and MLF in the biogenesis of amines has now been defined, since the decarboxylation of histidine to histamine and tyrosine to tyramine has been demonstrated with single strains of lactobacilli and oenococci. Detailed information on the role of wine LAB in the formation of biogenic amines can be found in a recent review by Lonvaud-Funel (2001). The wine LAB vary in their ability to produce amines from amino acids. In wine, it appears that the lactobacilli and pediococci are the main producers of amines, although some oenococci can also produce amines (Delfini 1989; Farias et al. 1993; Leitao et al. 2000). Coton et al. (1998b) have found that histidine decarboxylase activity is also common in oenococci. Recent studies indicate that oenococci are primarily responsible for histamine formation and lactobacilli cause the formation of tyramine in wine (Farias et al. 1993; Coton et al. 1998b; Moreno-Arribas et al. 2000). Histidine decarboxylase from O. oenoi 9204 has been purified and characterized (Rollan et al. 1995b; Coton et al. 1998a). In addition, tyrosine decarboxylase from Lact. brevis IOEB 9809 has also been purified and characterized (Moreno-Arribas and Lonvaud-Funel 1999, 2001) Formation of ethyl carbamate precursors Ethyl carbamate (EC) is an animal carcinogen found in many fermented foods and beverages, including wine (Ough 1976). It is formed through the chemical reaction of ethanol and an EC precursor, such as citrulline, urea or carbamyl phosphate (Ough et al. 1988). Citrulline is an intermediate in the degradation of arginine by wine LAB. The excretion

8 596 S.-Q. LIU of citrulline during arginine degradation appears to be common among the arginine-degrading wine LAB (Granchi et al. 1998; Mira de Orduna et al. 2000a). Indeed, a good correlation has been demonstrated between the excretion of citrulline and the formation of EC during the degradation of arginine by the wine LAB O. oeni and Lact. buchneri (Liu et al. 1994). Therefore, arginine degradation is a potential source of the EC precursor citrulline. Further information on EC formation in wine is available elsewhere (Liu and Pilone 1998). In addition to citrulline, carbamyl phosphate is also an EC precursor (see above). To date, the excretion of carbamyl phosphate during arginine degradation has not been reported. Carbamyl phosphate is also a precursor to pyrimidine and some LAB can synthesize carbamyl phosphate from glutamine, bicarbonate and ATP (Nicoloff et al. 2001). Therefore, this is potentially another source of EC precursor Formation of glyoxal and methylglyoxal Glyoxal and methylglyoxal found in wine are toxic compounds and, therefore, have implications for human health. It appears that oenococci can produce glyoxal and methylglyoxal during MLF (de Revel and Bertrand 1993, 1994). However, definitive studies are required to link the production of glyoxal and methylglyoxal with specific strains of oenococci and other wine LAB. 13. CONCLUSIONS The role of MLF in wine is more than deacidification. The metabolic potential of wine LAB is diverse and complex. More fundamental studies are required to elucidate the metabolism of wine LAB, such as amino acid metabolism, proteolysis and peptidolysis, ester synthesis and hydrolysis, lipolysis, catabolism of aldehydes, polysaccharide formation, metabolism of polyols and hydrolysis of glycosides. Knowledge of the metabolism of wine LAB will help understand the impact of MLF on wine quality and allow better control of MLF during winemaking. 14. REFERENCES Alberto, M.R., Farias, M.E. and Manca de Nadra, M.C. (2001) Effect of gallic acid and catechin on Lactobacillus hilgardii 5W. Growth and metabolism of organic compounds. Journal of Agricultural and Food Chemistry 49, Amerine, M.A., Berg, H.W., Ough, C.S., Kunkee, R.E., Singleton, V.L. and Web, A.D. (1980) The Technology of Winemaking, pp and 217. Westport, CT: AVI Publishing. Arena, M.E. and Manca de Nadra, M.C. (2001) Biogenic amine production by Lactobacillus. Journal of Applied Microbiology 90, Arena, M.E., Saguir, F.M. and Manca de Nadra, M.C. (1999) Arginine, citrulline and ornithine metabolism by lactic acid bacteria from wine. International Journal of Food Microbiology 52, Axelsson, L. (1993) Lactic acid bacteria: classification and physiology. In Lactic Acid Bacteria: Microbiology and Functional Aspects, 2nd edn. ed. Salminen, S. and von Wright, A. pp New York: Marcel Dekker. Baltenweck-Guyot, R., Trendel, J.-M., Albrecht, P. and Schaeffer, A. (2000) Glycosides and phenylpropanoid glycerol in Vitis vinifera Cv. Gewurztraminer wine. Journal of Agricultural and Food Chemistry 48, Bandell, M., Lhotte, M.E., Marty-Teysset, C., Veyrat, A., Prévost, H., Dartois, V., Diviès, C., Konings, W.N. and Lolkema, J.S. (1998) Mechanism of the citrate transporters in carbohydrate and citrate cometabolism in Lactococcus and Leuconostoc species. Applied and Environmental Microbiology 64, Bardi, L., Crivelli, C. and Marzona, M. (1998) Esterase activity and release of ethyl esters of medium-chain fatty acids by Saccharomyces cerevisiae during anaerobic growth. Canadian Journal of Microbiology 44, Barthelmebs, L., Divies, C. and Cavin, J.-F. (2000) Knockout of the p-coumarate decarboxylase gene from Lactobacillus plantarum reveals the existence of two other inducible enzymatic activities involved in phenolic acid metabolism. Applied and Environmental Microbiology 66, Barthelmebs, L., Diviès, C. and Cavin, J.-F. (2001) Molecular characterization of the phenolic acid metabolism in the lactic acid bacteria Lactobacillus plantarum. Lait 81, Bartowsky, E.J. and Henschke, P.A. (1995) Malolactic fermentation and wine flavour. Australian Grapegrower and Winemaker Annual Technical Issue, van Beek, S. and Priest, F. (2000) Decarboxylation of substituted cinnamic acids by lactic acid bacteria isolated during malt whisky fermentation. Applied and Environmental Microbiology 66, Blanco-Gomis, D., Mangas Alonso, J.J., Cabrales, I.M. and Abrodo, P.A. (2001) Gas chromatographic analysis of total fatty acids in cider. Journal of Agricultural and Food Chemistry 49, Blom, H. (1983) Partial characterization of a thermostable anthocyanin b-glycosidase from Aspergillus niger. Food Chemistry 12, Brennand, C.P., Ha, J.K. and Lindsay, R.C. (1989) Aroma properties and other thresholds of some branched-chain and other minor volatile fatty acids occurring in milkfat and meat lipids. Journal of Sensory Studies 4, Carr, J.G. (1987) The microbiology of ciders and wines. In Essays in Agricultural and Food Microbiology ed. Norris, J.R. and Pettipher, G.L. pp New York: John Wiley. Cavin, J.-F., Andioc, V., Etievant, P.X. and Divies, C. (1993) Ability of wine lactic acid bacteria to metabolize phenol carboxylic acids. American Journal of Enology and Viticulture 44, Cerning, J. (1990) Exocellular polysaccharides produced by lactic acid bacteria. FEMS Microbiology Reviews 87, Champomier-Vergès, M.-C., Zuñiga, M., Morel-Deville, F., Pérez- Martinez, G., Zagorec, M. and Ehrlich, S.D. (1999) Relationships between arginine degradation, ph and survival in Lactobacillus sakei. FEMS Microbiology Letters 180,

9 MALOLACTIC FERMENTATION IN WINE 597 Chen, K.H. and McFeeters, R.F. (1986a) Utilization of electron acceptors for anaerobic metabolism by Lactobacillus plantarum. Enzymes and intermediates in the utilization of citrate. Food Microbiology 3, Chen, K.H. and McFeeters, R.F. (1986b) Utilization of electron acceptors for anaerobic mannitol metabolism by Lactobacillus plantarum. Reduction of alpha-keto acids. Food Microbiology 3, Chen, K.H., McFeeters, R.F. and Fleming, H.P. (1983) Fermentation characteristics of heterolactic acid bacteria in green bean juice. Journal of Food Science 48, Chich, J.-F., Marchesseau, K. and Gripon, J.-C. (1997) Intracellular esterase from Lactococcus lactis subsp. lactis NCDO 763: Purification and characterization. International Dairy Journal 7, Christensen, J.E., Dudley, E.G., Pederson, J.A. and Steele, J.L. (1999) Peptidases and amino acid catabolism in lactic acid bacteria. Antonie Van Leeuwenhoek 76, Claisse, O. and Lonvaud-Funel, A. (2000) Assimilation of glycerol by a strain of Lactobacillus collinoides isolated from cider. Food Microbiology 17, Cogan, T.M. (1987) Co-metabolism of citrate and glucose by Leuconostoc spp.: effects on growth, substrates and products. Journal of Applied Bacteriology 63, Costello, P.J., Monk, P.R. and Lee, T.H. (1985) An evaluation of two commercial Leuconostoc oenos for induction of malolactic fermentation under winery conditions. Food Technology in Australia 37, Coton, E., Rollan, G., Bertrand, A. and Lonvaud-Funel, A. (1998b) Histamine-producing lactic acid bacteria in wines: early detection, frequency, and distribution. American Journal of Enology and Viticulture 49, Coton, E., Rollan, G.C. and Lonvaud-Funel, A. (1998a) Histidine carboxylase of Leuconostoc oenos 9204: purification, kinetic properties, cloning and nucleotide sequence of the hdc gene. Journal of Applied Microbiology 84, Davis, C.R., Wibowo, D., Eschenbruch, R., Lee, T.H. and Fleet, G.H. (1985) Practical implications of malolactic fermentation: a review. American Journal of Enology and Viticulture 36, Davis, C.R., Wibowo, D., Fleet, G.H. and Lee, T.H. (1986a) Growth and metabolism of lactic acid bacteria during and after malolactic fermentation of wines at different ph. Applied and Environmental Microbiology 51, Davis, C.R., Wibowo, D., Fleet, G.H. and Lee, T.H. (1986b) Growth and metabolism of lactic acid bacteria during fermentation and conservation of some Australian wines. Food Technology in Australia 38, Davis, C.R., Wibowo, D., Fleet, G.H. and Lee, T.H. (1988) Properties of wine lactic acid bacteria: their potential oenological significance. American Journal of Enology and Viticulture 39, De Cort, S., Shantha Kumara, H.M.C. and Verachert, H. (1994) Localization and characterization of alpha-glucosidase activity in Lactobacillus brevis. Applied and Environmental Microbiology 60, Delaquis, P., Cliff, M., King, M., Girard, B., Hall, J. and Reynolds, A. (2000) Effect of two commercial malolactic cultures on the chemical and sensory properties of Chancellor wines vinified with different yeasts and fermentation temperatures. American Journal of Enology and Viticulture 51, Delfini, C. (1989) Ability of wine lactic acid bacteria to produce histamine. Sciences des Aliments 9, de Revel, G. and Bertrand, A. (1993) A method for the detection of carbonyl compounds in wine: glyoxal and methylglyoxal. Journal of the Science of Food and Agriculture 61, de Revel, G. and Bertrand, A. (1994) Dicarbonyl compounds and their reduction products in wine. Identification of wine aldehydes. In Trends in Flavour Research, Proceedings of the 7th Weurman Flavour Research Symposium, June, ed. Maarse, H. and van der Heiji, D.G. pp Zeist, The Netherlands: Elsevier. de Revel, G., Martin, N., Pripis-Nicolau, L., Lonvaud-Funel, A. and Bertrand, A. (1999) Contribution to the knowledge of malolactic fermentation influence on wine aroma. Journal of Agricultural and Food Chemistry 47, Dicks, L.M.T., Dellaglio, F. and Collins, M.D. (1995) Proposal to reclassify Leuconostoc oenos as Oenococcus oeni [corrig.] gen. nov., comb. nov. International Journal of Systematic Bacteriology 45, Dittrich, H.H. (1987) Mikrobiologie des Weines. Handbuch der Lebensmitteltechnologie, 2nd edn. Stuttgart: Ulmer. Cited in Henick-Kling, T. (1993) Malolactic fermentation. In Wine Microbiology and Biotechnology ed. Fleet, G.H. pp Amsterdam: Harwood Academic. Ebeler, S.E. (2001) Analytical chemistry: unlocking the secrets of wine flavor. Food Reviews International 17, Ebeler, S.E. and Spaulding, R.S. (1999) Characterization and measurement of aldehydes in wine. In Chemistry of Wine Flavor ed. Waterhouse, A.L. and Ebeler, S.E. ACS Sumposium Series 714, Washington, DC: American Chemical Society. Etiévant, P.X., Issanchou, S., Marie, S., Ducruet, V. and Flanzy, C. (1989) Sensory impact of volatile phenols on red wine aroma: influence of carbonic maceration and time of storage. Sciences des Aliments 9, Farias, M.E., Manca de Nadra, M.C., Rollan, G.C. and Strasser de Saad, A.M. (1993) Histidine decarboxylase activity in lactic acid bacteria from wine. Journal International des Sciences de la Vigne et du Vin 27, Farias, M.E., Rollan, G.C. and Manca de Nadra, M.C. (1996) Influence of nutritional factors on the protease production by Leuconostoc oenos from wine. Journal of Applied Bacteriology 81, Firme, M.P., Leitao, M.C. and San Romão, M.V. (1994) The metabolism of sugar and malic acid by Leuconostoc oenos: effect of malic acid, ph and aeration conditions. Journal of Applied Bacteriology 76, Garvie, E.I. (1986) Genus Leuconostoc. In Bergey s Manual of Systematic Bacteriology, Vol. 2. ed. Sneath, P.H.A. pp Baltimore, MA: Williams & Wilkins. Gobbetti, M., Fox, P.F. and Stepaniak, L. (1997) Isolation and characterization of a tributyrin esterase from Lactobacillus plantarum Journal of Dairy Science 80, Gonzalez de Llano, D., Cuesta, P. and Rodriguez, A. (1998) Biogenic amine production by wild lactococcal and leuconostoc strains. Letters in Applied Microbiology 26, Granchi, L., Paperi, R., Rosellini, D. and Vincenzini, M. (1998) Strain variation of arginine catabolism among malolactic Oenococcus oeni strains of wine origin. Italian Journal of Food Science 10,

10 598 S.-Q. LIU Grimaldi, A., McLean, H. and Jiranek, V. (2000) Identification and partial characterization of glycosidic activities of commercial strains of the lactic acid bacterium, Oenococcus oeni. American Journal of Enology and Viticulture 51, Gueguen, Y., Chemardin, P., Labrot, P., Arnaud, A. and Galzy, P. (1997) Purification and characterization of an intracellular b-glucosidase from a new strain of Leuconostoc mesenteroides isolated from cassava. Journal of Applied Microbiology 82, Guilloux-Benatier, M., Pageault, O., Man, A. and Feuillat, M. (2000) Lysis of yeast cells by Oenococcus oeni enzymes. Journal of Industrial Microbiology and Biotechnology 25, Hamouy, D. and Desmazeaud, M.J. (1985) Characterization of an L-serine dehydratase activity in permeabilized cells of Brevibacterium linens ATCC Le Lait 65, Hemme, D., Bouillanne, C., Métro, F. and Desmazeaud, M.-J. (1982) Microbial catabolism of amino acids during cheese ripening. Sciences des Aliments 2, Henick-Kling, T. (1988) Yeast and bacterial control in winemaking. In Modern Methods of Plant Analysis New Series, Vol. 6. Wine Analysis. ed. Linskens, H.F. and Jackson, J.F. pp Berlin: Springer- Verlag. Henick-Kling, T. (1993) Malolactic fermentation. In Wine Microbiology and Biotechnology ed. Fleet, G.H. pp Amsterdam: Harwood Academic. Henick-Kling, T. (1995) Control of malolactic fermentation in wine: energetics, flavour modification and methods of starter culture preparation. Journal of Applied Bacteriology Symposium Supplement 79, 29S 37S. Henick-Kling, T., Acree, T.E., Krieger, S.A., Laurent, M.-H. and Edinger, W.D. (1994) Modification of wine flavor by malolactic fermentation. Wine East 4, 8 15 and Henschke, P.A. (1993) An overview of malolactic fermentation research. Australian-New Zealand Wine Industry Journal 8, Holland, R. and Coolbear, T. (1996) Purification of tributyrin esterase from Lactococcus lactis subsp. cremoris E8. Journal of Dairy Research 63, Hosono, A. and Elliott, J.A. (1974) Properties of crude ethylesterforming enzyme preparations from some lactic acid bacteria and psychrotrophic bacteria. Journal of Dairy Science 57, Hosono, A., Elliott, J.A. and McGugan, W.A. (1974) Production of ethylesters by some lactic acid bacteria and psychrotrophic bacteria. Journal of Dairy Science 57, Hugenholtz, J. (1993) Citrate metabolism in lactic acid bacteria. FEMS Microbiology Reviews 12, Huis in t Veld, J.H.J. (1996) Microbial and biochemical spoilage of foods: an overview. International Journal of Food Microbiology 33, Inoue, Y., Trevanichi, S., Fukuda, K., Izawa, S., Wakai, Y. and Kimura, A. (1997) Roles of esterase and alcohol acetyltransferase on production of isoamyl acetate in Hansenula mrakii. Journal of Agricultural and Food Chemistry 45, Jarvis, B., Forster, M.J. and Kinsella, W.P. (1995) Factors affecting the development of cider flavour. Journal of Applied Bacteriology Supplement 79, 5S 18S. Kandler, O. (1983) Carbohydrate metabolism in lactic acid bacteria. Antonie Van Leeuwenhoek 49, Keenan, T.W. (1968) Production of acetic acid and other volatile compounds by Leuconostoc dextranicum. Applied Microbiology 16, Kotseridis, Y. and Baumes, R. (2000) Identification of impact odorants in Bordeaux red grape juices, in the commercial yeast used for its fermentation, and in the produced wine. Journal of Agricultural and Food Chemistry 48, Kunji, E.R.S., Mierau, I., Hagting, A., Poolman, B. and Konings, W.N. (1996) The proteolytic systems of lactic acid bacteria. Antonie Van Leeuwenhoek 70, Kunkee, R.E. (1967) Malolactic fermentation. Advances in Applied Microbiology 9, Kunkee, R.E. (1974) Malolactic fermentation in winemaking. In Chemistry of Winemaking ed. Webb, A.D. Advances in Chemistry Series 137, Washington, DC: American Chemical Society. Kunkee, R.E. (1991) Some roles of malic acid in the malolactic fermentation in winemaking. FEMS Microbiology Reviews 88, Lafon-Lafourcade, S. (1983) Wine and brandy. In Biotechnology, Vol. 5. ed. Rehm, H.J. and Reed, G. pp Weinheim: Verlag Chemie. Laurent, M.-H., Henick-Kling, T. and Acree, T. (1994) Changes in the aroma and odor of Chardonnay wine due to malolactic fermentation. Wein Wissenschaft 49, Law, B.A. and Mulholland, F. (1995) Enzymology of lactococci in relation to flavour development from milk proteins. International Dairy Journal 5, Lehtonen, P. (1996) Determination of amines and amino acids in wine a review. American Journal of Enology and Viticulture 47, Lei, V., Amoa-Awua, W.K.A. and Brimer, L. (1999) Degradation of cyanogenic glycosides by Lactobacillus plantarum strains from spontaneous cassava fermentation and other microorganisms. International Journal of Food Microbiology 53, Leitao, M.C., Teixeira, H.C., Barreto Crespo, M.T. and San Romão, M.V. (2000) Biogenic amines occurrence in wine. Amino acid decarboxylase and proteolytic activities expression by Oenococcus oeni. Journal of Agricultural and Food Chemistry 48, Lilly, M., Lambrechts, M.G. and Pretorius, I.S. (2000) Effect of increased yeast alcohol acetyltransferase activity on flavor profiles of wine and distillates. Applied and Environmental Microbiology 66, Liu, S.-Q. (1990) Growth and Metabolism of Lactic Acid Bacteria in a Model Wine System and a Red Wine with Emphasis on Carbohydrate Metabolism. Master s Thesis, Massey University, Palmerston North, New Zealand. Liu, S.-Q. (1993) Arginine Metabolism in Malolactic Wine Lactic Acid Bacteria and its Oenological Implications. PhD Thesis, Massey University, Palmerston North, New Zealand. Liu, S.-Q., Asmundson, R.V., Holland, R. and Crow, V.L. (1997) Acetaldehyde metabolism by Leuconostocs mesenteroides subsp. cremoris under stress conditions. International Dairy Journal 7, Liu, S.-Q. and Davis, C.R. (1994) Analysis of wine carbohydrates using capillary gas-liquid chromatography. American Journal of Enology and Viticulture 45, Liu, S.-Q., Davis, C.R. and Brooks, J.D. (1995a) Growth and metabolism of selected lactic acid bacteria in synthetic wine. American Journal of Enology and Viticulture 46,