- Email: [email protected]
The enumeration and identification of acetic acid bacteria from South African red wine fermentations
International Journal of Food Microbiology 74 (2002) 57 – 64 www.elsevier.com/locate/ijfoodmicro
The enumeration and identification of acetic acid bacteria from South African red wine fermentations W.J. Du Toit a, M.G. Lambrechts a,b,* a
Department of Viticulture and Oenology, University of Stellenbosch, Z.A.-7600, Stellenbosch, South Africa b Institute for Wine Biotechnology, University of Stellenbosch, Z.A.-7600, Stellenbosch, South Africa Received 29 May 2001; received in revised form 15 August 2001; accepted 6 October 2001
Abstract Acetic acid bacteria are microorganisms that can profoundly influence the quality of wine. Surprisingly, little research has been done on these microorganisms in the winemaking field. The object of this study was to investigate the occurrence of acetic acid bacteria in South African red wine fermentations and to identify the dominant species occurring. Acetic acid bacteria were isolated and enumerated from small-scale and commercial red must fermentations in 1998 and 1999, respectively. The initial occurrence of acetic acid bacteria in the must was shown to vary with cell numbers ranging from 106 – 107 to 104 – 105 cfu/ml for the 1998 and 1999 musts, respectively. The acetic acid bacteria decreased to 102 – 103 cfu/ml in musts having a low pH ( V 3.6), whereas in some musts having a high pH ( 3.7), the cell numbers increased during fermentation. During the process of cold soaking, the cell numbers of acetic acid bacteria also increased until inoculation with commercial wine yeast. Gluconobacter oxydans dominated in the fresh must and Acetobacter pasteurianus and A. liquefaciens during fermentation. This study showed that A. liquefaciens and A. hansenii were present in significant numbers, which has not been reported before. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Acetic acid bacteria; Red wine fermentations; pH
1. Introduction Acetic acid bacteria are divided into the genera Gluconobacter, Acetobacter and Frateuria (Holt et al., 1994). Of these, Gluconobacter oxydans, Acetobacter aceti, A. pasteurianus, A. liquefaciens and A. hansenii are normally associated with grapes and wine. The main species observed on unspoiled grapes and must is G. oxydans (Blackwood et al., 1969; Joyeux et * Corresponding author. Distell, Adam Tas Road, P.O. Box 46, Stellenbosch 7600, South Africa. Tel.: +27-21-809-7322; fax: +2721-886-5414. E-mail address: [email protected] ( M.G. Lambrechts).
al., 1984a). This species prefers a sugary rich environment and usually dies off during alcoholic fermentations. Acetobacter species prefer ethanol as carbon source (De Ley et al., 1984) and usually dominate during the later stages of fermentation and in wine (Drysdale and Fleet, 1985; Joyeux et al., 1984a). Acetic acid bacteria, although able to produce high amounts of acetic acid from ethanol, have not been of a major winemaking concern because of their requirement of oxygen for growth. The anaerobic conditions that generally prevail in wine and the presence of free sulfur dioxide are usually considered to be an adequate means of avoiding acetic acid bacterial spoilage. Therefore, surprisingly little research has been done
0168-1605/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 0 5 ( 0 1 ) 0 0 7 1 5 - 2
58
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
on the role of acetic acid bacteria in the winemaking process. However, in recent years, the ability of acetic acid bacteria to affect wine quality has been the subject of renewed interest. Joyeux et al. (1984a) have shown that acetic acid bacteria are able to survive and even grow during the winemaking process. Furthermore, acetic acid bacteria can produce other compounds, apart from acetic acid, that can influence wine quality (Drysdale and Fleet, 1989a). The objectives of this study were to investigate the occurrence of acetic acid bacteria in different South African red wine fermentations and to identify the dominant acetic acid bacteria strains. This study forms part of a research program to determine the origin of volatile acidity in South African red wines.
2. Materials and methods 2.1. Fermentations monitored Acetic acid bacteria were isolated and enumerated from red wine fermentations during the 1998 and 1999 harvesting seasons. During the 1998 season, six small-scale fermentations of Cabernet Sauvignon grapes were monitored. These fermentations consisted of 110-kg grapes each and were conducted in duplicate at 15, 22 and 30 C. At each fermentation temperature, the grapes were divided into two batches prior to yeast inoculation and only one half of the grapes were sulfured (50 mg/kg). The different musts were fermented on the skins until 0 B. The occurrence of acetic acid bacteria was determined in the must at 22, 18, 12, 6 and 0 B during the fermentation. During the 1999 harvesting season, bacteria were sampled from fermenting red musts at six commercial wine cellars in the Stellenbosch, Paarl, Franschhoek and Durbanville areas. Sampling was done prior to yeast inoculation, during the fermentation (between 10 and 14 B) and at the end of fermentation (between 0 and 5 B). Two of the cellars applied the technique of cold soaking prior to yeast inoculation. During cold soaking, which is a type of skin maceration, the must and skins were kept together between 15 and 18 C for 3 days. Five samples were taken from these musts.
2.2. Isolation and identification of acetic acid bacteria Acetic acid bacteria were isolated by plating 100 ml of a dilution series of juice or fermenting juice onto GYC medium [glucose (5% m/v), yeast extract (1% m/v), CaCO3 (3% m/v) and agar (2% m/v)] and Mannitol medium [mannitol (2.5% m/v), yeast extract (1% m/v) and agar (1.5% m/v)]. Growth of yeast and lactic acid bacteria (which normally exhibit weak growth on these types of media) were eliminated by adding 50 mg/l pimaricin (Delvocid, Gist-Brocades) and 50 mg/l nisin, respectively, to the media. Each dilution was plated out in triplicate and the plates incubated at 30 C for 6 days. Representative colonies were isolated and Gram staining and catalase tests were performed on each colony. Representative Gram ( ) rods and catalase (+) colonies were stored on GYC slants at 4 C and transferred monthly or kept at 80 C in 40% glycerol until identification. One hundred and fifteen acetic acid bacterial isolates were identified up to species level with the following biochemical and physiological tests: the ability to overoxidize ethanol, growth on sodium acetate, ketogenesis of glycerol, growth on dulcitol and production of a brown pigment on GYC medium (Drysdale and Fleet, 1988). Thirty-three of these isolates, chosen randomly, were also identified using numerical analysis of whole cell proteins by SDSPAGE (Pot et al., 1994). These 33 strains were grown on GYC agar for 70 h at 30 C. Cell-free extracts were prepared by washing the bacteria from the GYC agar with NaPBS buffer [consisting of 0.115% (m/v) Na2HPO4.12H2O, 0.023% (m/v) NaH2PO4.2H2O and 0.8% (m/v) NaCl, pH 7.3]. The bacteria were then centrifuged for 10 min at 10 000 rpm, the supernatant was discarded, the pellet was washed twice with 30 ml NaPBS buffer, and finally, 70 mg of the bacteria was placed in an eppendorf tube to which 1.26 ml of Sample Treatment Buffer [0.75% (m/v) Tris, C2H6OS (5% (v/v), pH 6.8], without sodium dodecyl sulphate (SDS), was added. The samples were vortexed and 0.14 ml of SDS [20% (m/v)] was added and vortexed. The mixture was then heated to 95 C for 10 min, cooled and centrifuged at 10 000 rpm for 5 min. The supernatant, which contained the extracted proteins, was removed and stored at 80 C. The denatured fraction was used for SDS-PAGE analysis. Registration of the protein electropherograms, normalization of
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
59
Table 1 Composition of musts used to monitor the evolution of acetic acid bacteria during red wine fermentations Year and area
Cultivar
Sugar concentration (B)a
Titratable acidity (g/l)
pH
SO2 dosage (mg/kg)b
1998 Stellenbosch (US) 1999 Stellenbosch (SI) Stellenbosch (SII) Stellenbosch (SIII) Paarl (PI) Franschhoek (FI) Durbanville (DI)
Cabernet Sauvignon Pinotage Merlot Merlot Cabernet franc Merlot Merlot
22.0 24.3 24.6 25.2 24.0 25.2 23.4
4.9 5.7 6.8 5.4 4.0 5.8 6.2
3.76 3.50 3.41 3.75 3.75 3.71 3.40
50 50 50 40 30 40 50
a b
B: Degree Balling where 1 B = 10 g/l sugar. Amount added at crushing.
at the higher fermentation temperatures. The number of acetic acid bacteria decreased from 106 –107 cfu/ml prior to yeast inoculation to 103 – 104 and 102 – 103 cfu/ml in the middle and at the end of the fermentation, respectively. During the 1999 harvesting season, fermenting musts from six commercial cellars were monitored for the occurrence and growth of acetic acid bacteria. The area, cultivar and general composition of the musts are shown in Table 1. In Fig. 1, the occurrence of these bacteria during fermentation is shown. It is clear that the acetic acid bacteria numbers decreased drastically from the beginning to the middle of fermentation in the low pH musts (pH < 3.6), but this did not happen in the higher pH musts (pH < 3.7). In the two high pH musts, the acetic acid bacteria numbers were slightly lower or even higher at the end of fermentation. In the low pH musts, these numbers were,
the densitometric traces, grouping of strains by the Pearson product moment correlation coefficient (r) and UPGMA cluster analysis were performed (Pot et al., 1994) using the Gelcompar version 4 software.
3. Results 3.1. Occurrence of acetic acid bacteria in red wine fermentations During the 1998 season, the growth of acetic acid bacteria was followed in Cabernet Sauvignon must. The composition of the must is shown in Table 1. The numbers of acetic acid bacteria as well as the dominant species during fermentation at the different temperatures are given in Table 2. The addition of sulfur dioxide influenced the acetic acid bacteria, especially
Table 2 Levels (cfu/ml) and species of acetic acid bacteria in Cabernet Sauvignon must fermented at 15, 22 and 30 C during 1998 Fermentation stage (B) 22 18 12
6 0
30 (C)
22 (C)
SO2
+ SO2 6
3.4 10 , G. oxydans 1.2 106, G. oxydans 6.0 103, A. aceti; 6.0 103, A. liquefaciens 1.0 103, A. pasteurianus 1.9 102, A. pasteurianus
15 (C)
SO2 6
1.2 10 , G. oxydans 3.2 103, G. oxydans 4.4 103, A. pasteurianus
1.4 103, A. pasteurianus 3.6 103, A. pasteurianus
+ SO2 6
8.5 10 , G. oxydans 3.2 106, G. oxydans 2.0 104, A. pasteurianus; 1.1 104, A. aceti 3.6 103, A. pasteurianus 3.2 102, A. pasteurianus
SO2 6
+ SO2 7
1.6 10 , G. oxydans 1.6 106, G. oxydans 7 103, A. pasteurianus
1.6 10 , G. oxydans 1.4 106, G. oxydans 3.1 104, A. aceti
1.4 106, G. oxydans 1.1 106, G. oxydans 3.0 104, A. pasteurianus
9.3 103, A. pasteurianus 1.0 103, A. pasteurianus
7.0 103, A. pasteurianus 1.3 103, A. pasteurianus
3.3 104, A. pasteurianus 4.2 103, A. pasteurianus
60
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
The growth of acetic acid bacteria in musts that underwent cold soaking prior to yeast inoculation is shown in Fig. 2. Both these winemakers incubated the must between 15 and 18 C during cold soaking and both added 40 to 50 mg/kg SO2 just after the grapes were crushed. During cold soaking, the population of acetic acid bacteria increased in the SIII must from 103 to almost 105 cfu/ml. After yeast inoculation, the bacteria quickly died off during fermentation, with a slight increase in cell numbers occurring at the end of fermentation. In the SI must, the acetic acid bacteria numbers remained almost the same during skin maceration and followed the same pattern as that of the SIII must during fermentation. 3.2. Identification of the isolated acetic acid bacteria
Fig. 1. Growth of acetic acid bacteria in different commercial red wine fermentations during the 1999 harvesting season. The pH values of the different musts are included.
however, between 102 and 103 cfu/ml lower than in the beginning of fermentation. G. oxydans dominated again in the freshly pressed must, except in the FI fermentation where only A. pasteurianus was detected (Table 3). During the course of fermentation, A. pasteurianus and A. liquefaciens dominated, with A. aceti and A. hansenii occurring irregularly and at lower cell numbers.
All Gram-negative, catalase-positive rods were preliminarily identified as acetic acid bacteria according to the biochemical and physiological tests described in the Materials and methods. The percentage of the 115 acetic acid bacterial strains belonging to G. oxydans, A. pasteurianus, A. hansenii, A. aceti and A. liquefaciens is indicated in Table 4. The identification of 33 of the above acetic acid bacteria was confirmed by the whole cell protein gel electrophoresis. A dendrogram of the clustering of the identified species is shown in Fig. 3.
Table 3 Occurrence of acetic acid bacterial species in commercial red wine fermentations during the 1999 harvesting season Area
Fermentation stage Beginning
Middle
End
SI SII
G. oxydans G. oxydans, A. liquefaciens
A. pasteurianus A. liquefaciens, A. hansenii
SIII
G. oxydans
PI FI
G. oxydans, A. pasteurianus A. pasteurianus
A. pasteurianus G. oxydans, A. liquefaciens, A. aceti A. pasteurianus, A. liquefaciens A. pasteurianus A. pasteurianus, A. liquefaciens
DI
G. oxydans
A. pasteurianus, A. liquefaciens, A. hansenii A. liquefaciens
A. liquefaciens
A. pasteurianus A. pasteurianus
Strains indicated in bold dominated at that stage of the fermentation.
Fig. 2. Effect of cold soaking and yeast inoculation on acetic acid bacterial growth. The grape must were incubated at between 15 and 18 C during the time of maceration.
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
61
Table 4 Biochemical and physiological tests used for the identification of acetic acid bacterial isolates Biochemical/ physiological test
EtOH oxidation Growth on sodium acetate Ketogenises from glycerol Growth on dulcitol Production of brown pigment
Number of isolates Protein cluster giving a (+) result for this test
I. II. III. IV. V. A. hansenii A. liquefaciens A. pasteurianus A. aceti G. oxydans
Percent of isolates 4 identified as strain
12
57
6
21
+ + + +
+ + +
+ +
+ + +
+
84 80 48 5 14
Five clusters with the SDS-PAGE numerical analysis of whole cell proteins were obtained. Cluster I, belonging to A. hansenii, also grew on dulcitol as carbon source. This is one of the tests described in Drysdale and Fleet (1988) to distinguish this Acetobacter strain from others. Cluster II, belonging to A. liquefaciens,
+
produced a brown pigment on GYC medium and Cluster III (A. pasteurianus) was ketogenesis-negative. These are tests used to identify these two species, respectively. These two clusters contained by far the most numerous species. Cluster IV belonged to A. aceti and Cluster V to G. oxydans, the latter being unable to
Fig. 3. Dendrogram showing the clustering analysis of acetic acid bacteria based on protein patterns of the red wine isolates and reference strains of Gluconobacter and Acetobacter spp. Grouping was performed by the unweighted average pair-group method. Reference strains were obtained from the Deutsche Sammlung von Mikroorganismen, Braunschweig, Germany.
62
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
overoxidize ethanol and grow on sodium acetate. These acetic acid bacteria are grouped into five, i.e. G. oxydans, A. pasteurianus, A. liquefaciens, A. hansenii and A. aceti well-defined clusters at r 0.68, which were related to each other at r 0.58. A large variation in strains is observed within each cluster if one considers the low r value of 0.68.
4. Discussion Acetic acid bacteria had long been believed to play little, if any, role during winemaking operations due to their aerobic nature (Drysdale and Fleet, 1988). Winemaking in general is an anaerobic process and the growth and survival of these bacteria under this and other unfavourable conditions like high ethanol concentrations, low pH and high SO2 concentrations seems unlikely. Some studies have, however, shown that acetic acid bacteria can survive during fermentation and the following operations in the winemaking process, such as malolactic fermentation and during maturation of the wine (Joyeux et al., 1984a; Drysdale and Fleet, 1985). Acetic acid bacteria had been shown to contribute significantly to volatile acidity in must and wine, and the production of acetic acid by these bacteria may thus also contribute to sluggish or stuck fermentations (Joyeux et al., 1984b; Drysdale and Fleet, 1989b). 4.1. Occurrence of acetic acid bacteria Acetic acid bacteria occurred at high numbers on grapes and the freshly pressed must in 1998. These numbers are higher than those expected from healthy grapes, but spoiled grapes can easily harbour these counts. The only species isolated at this stage was G. oxydans, which normally occurs on unspoiled grapes (Splittstoesser and Churney, 1992; Sponholz, 1993). The survival of this strain during the early stages of fermentation can only be explained by its low ethanol tolerance (De Ley et al., 1984). The higher toxicity of SO2 at 30 C could be explained by the bacteria being metabolically more active at this temperature, thus accumulating and metabolizing the SO2 faster. In the middle of the fermentation, A. aceti and A. liquefaciens occurred, especially in the fermentations containing no SO2. In the fermentations where SO2 had
been added, A. pasteurianus dominated and this strain also dominated towards the end of the fermentation. This could be due to the fact that the A. aceti and A. liquefaciens strains that occurred in these fermentations were more SO2-sensitive than the A. pasteurianus strain. Although the SO2 should not have had any effect on the bacteria at this stage of fermentation due to acetaldehyde binding it, the initial population of A. aceti and A. liquefaciens before fermentation commenced might have been affected more by the SO2 than was the case with A. pasteurianus. The dominance of A. pasteurianus towards the end of fermentation might also be due to the higher ethanol tolerance of this strain (De Ley et al., 1984). During the 1999 season, the bacterial counts were lower than in the 1998 fermentations. This difference in acetic acid bacteria counts was also observed in base wines for brandy distillation in a study undertaken at the Institute for Wine Biotechnology during these two seasons (unpublished data). G. oxydans dominated in most of the fresh musts, but an A. pasteurianus strain was found to dominate the FI must. It has not been reported before that an Acetobacter strain can dominate in a fresh, unspoilt must. The more ethanoltolerant Acetobacter strains dominated again the G. oxydans strains during fermentation. As in the 1998 season, A. pasteurianus dominated in some of the fermentations, with A. liquefaciens dominating others. This is a different scenario from that observed in French and Australian wineries where A. aceti and A. pasteurianus seemed to dominate (Joyeux et al., 1984a; Drysdale and Fleet, 1985). The growth of acetic acid bacteria during fermentation also correlated with the initial pH of the must. In the low pH musts, the decrease in acetic acid bacterial counts was more pronounced than in the high pH must (refer to Table 1). At a lower pH value, more of the free SO2 exists in the molecular form, which is the active toxic form towards bacteria. Furthermore, ethanol toxicity is known to increase at a lower pH (Dupuy and Maugenet, 1963). In one of the high pH musts, FI, the acetic acid bacterial counts were even higher at the end of the fermentation than at the beginning, indicating that acetic acid bacteria can grow during the alcoholic fermentation. The production of ethanol by the yeast has been shown to stimulate the growth of some Acetobacter strains that prefer ethanol over glucose as a carbon source (Swings and De Ley,
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
1981). This also raises the question if these bacteria are as aerobic as thought before, seeing that they are able to survive until the end of alcoholic fermentation. Acetic acid bacteria have been shown to be able to use quinones instead of oxygen as electron acceptors (Adlercreutz and Mattiasson, 1984). The growth of acetic acid bacteria in the different commercial fermentations also confirms the advantages of having a must with a low pH and the addition of SO2 to the must prior to alcoholic fermentation. However, in the SIII fermentation, which had a high pH, the acetic acid bacterial numbers also decreased drastically after yeast inoculation (Fig. 2). The sensitivity of acetic acid bacteria to pH, SO2 and ethanol is therefore also strain-dependent. The survival of these bacteria after alcoholic fermentation should, however, be investigated further. The survival of high numbers of acetic acid bacteria after alcoholic fermentation might affect wine quality, since many of these bacteria are being ‘‘brought over’’ from the fermentation to the following winemaking processes (Fugelsang, 1997). During the process of cold soaking, the growth of acetic acid bacteria can also occur as has been proven by this study. During this process, the advantages of having a must with a low pH and high SO2 concentration have also been proven. The maintenance of a low temperature alone might, thus, not be sufficient to control the unwanted growth of these microorganisms. Similar results were obtained by Drysdale and Fleet (1989b). Joyeux et al. (1984a) concluded that a storage temperature of 10 to 15 C could be beneficial in stopping acetic acid bacteria from growing in wine. The addition of sufficient SO2 and the lowering of the maceration temperature and pH with acid additions should, thus, be considered when the winemaker wants to apply skin maceration prior to inoculation. 4.2. Identification of acetic acid bacteria There was in general a good correlation between the biochemical and physiological tests and the SDSPAGE numerical analysis of whole cell proteins. The two strains SI8 and SI18 clustered with A. pasteurianus and were ketogenesis-negative on glycerol, but were, however, also unable to overoxidize ethanol. Holt et al. (1994) also found that certain A. pasteurianus strains were unable to overoxidize ethanol. According to our results, A. aceti and A. pasteurianus were closely re-
63
lated and this is in accordance with Sievers et al. (1994). The results of 16S RNA sequencing showed that A. hansenii and A. liquefaciens were on the same phylogenetic level and this is also indicated by our protein results (Sievers et al., 1994). Some of the acetic acid bacteria isolated from the same fermenting musts also showed a high correlation (r 0.86) on the basis of whole-cell protein fingerprinting. In conclusion, this study clearly indicated that acetic acid bacteria can survive and even grow in grape must during the alcoholic fermentation and that the pH and molecular SO2 concentration of must and wine are important factors influencing this ability. Furthermore, winemaking practices such as cold soaking will also influence the survival or growth of acetic acid bacteria. A low pH and sufficient SO2 concentrations should, however, help to prevent the unwanted growth of acetic acid bacteria. G. oxydans dominated in the fresh must, with Acetobacter species, especially A. pasteurianus and A. liquefaciens, occurring later during alcoholic fermentation. It thus seems as if the South African scenario for the occurrence of these bacteria differs from that of other wine producing countries. The survival and occurrence of acetic acid bacteria in must and wine had been studied before, but this is the first study we know of that investigated the growth of these bacteria in different fermentations during different harvesting seasons. Clearly, more research is needed on this group of microorganisms, which can profoundly influence the quality of grape juice and wine.
Acknowledgements The authors would like to thank the following people for the intellectual and technical support: Prof. I.S. Pretorius, Dr. M. du Toit, Prof. L.M.T. Dicks, L.P. Ellis, M. Gey Van Pittius, J. Bayly and C. Font-Sala.
References Adlercreutz, P., Mattiasson, B., 1984. Oxygen supply to immobilized cells: 4. Use of p-benzoquinone as an oxygen substitute. Appl. Microbiol. Biotechnol. 20, 296 – 302. Blackwood, A.-C., Guimberteau, G., Penaud, E., 1969. Sur les bacte´ries ace´tiques isole´es de raisins. C. R. Seances Acad. Sci., Ser. D 269, 802 – 804.
64
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
De Ley, J., Gillis, M., Swings, J., 1984. Family VI. Acetobactaceae. Bergey’s Manual of Systematic Bacteriology, vol. 1, 9th edn. Williams and Wilkens, MD, USA, pp. 267 – 274. Drysdale, G.S., Fleet, G.H., 1985. Acetic acid bacteria in some Australian wines. Food Technol. Aust. 37, 17 – 20. Drysdale, G.S., Fleet, G.H., 1988. Acetic acid bacteria in winemaking: a review. Am. J. Enol. Vitic. 39, 143 – 154. Drysdale, G.S., Fleet, G.H., 1989a. The growth and survival of acetic acid bacteria in wines at different concentrations of oxygen. Am. J. Enol. Vitic. 40, 99 – 105. Drysdale, G.S., Fleet, G.H., 1989b. The effect of acetic acid bacteria upon the growth and metabolism of yeast during the fermentation of grape juice. J. Appl. Bacteriol. 67, 471 – 481. Dupuy, P., Maugenet, J., 1963. Me´tabolisme de l’acide lactique par Acetobacter rancens. Ann. Technol. Agric. 12, 5 – 14. Fugelsang, K.C., 1997. Wine Microbiology. Chapman & Hall, New York. Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T., 1994. Genus Acetobacter and Gluconobacter. Bergey’s Manual of Determinative Bacteriology, 9th edn. Williams and Wilkens, MD, USA, pp. 71 & pp. 84. Joyeux, A., Lafon-Lafourcade, S., Ribe´reau-Gayon, P., 1984a. Evolution of acetic acid bacteria during fermentation and storage of wine. Appl. Environ. Microbiol. 48, 153 – 156.
Joyeux, A., Lafon-Lafourcade, S., Ribe´reau-Gayon, P., 1984b. Metabolism of acetic acid bacteria in grape must: consequences on alcoholic and malolactic fermentation. Sci. Aliments 4, 247 – 255. Pot, B., Vandamme, P., Kersters, K., 1994. Analysis of electrophoretic whole-organism protein fingerprints. In: Goodfellow, M., O’Donnell, A.G. (Eds.), Chemical Methods in Prokaryotic Systematics. Wiley, Chichester, UK, pp. 493 – 521. Sievers, M., Ludwig, W., Teuber, M., 1994. Phylogenetic positioning of Acetobacter, Gluconobacter, Rhodopila, and Acidiphilium species as a branch of acidophilic bacteria in the asubclass of Proteobacteria based on 16S ribosomal DNA sequences. Syst. Appl. Microbiol. 17, 189 – 196. Splittstoesser, D.F., Churney, J.J., 1992. The incidence of sorbic acid-resistant Gluconobacters and yeasts on grapes grown in New York State. Am. J. Enol. Vitic. 43, 290 – 293. Sponholz, W.-R., 1993. Wine spoilage by microorganisms. In: Fleet, G.H. (Ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland, pp. 395 – 420. Swings, J., De Ley, J., 1981. The genera Acetobacter and Gluconobacter. In: Starr, M.P. (Ed.), The Prokaryotes. Springer-Verlag, Berlin, Germany, pp. 771 – 778.
The enumeration and identification of acetic acid bacteria from South African red wine fermentations W.J. Du Toit a, M.G. Lambrechts a,b,* a
Department of Viticulture and Oenology, University of Stellenbosch, Z.A.-7600, Stellenbosch, South Africa b Institute for Wine Biotechnology, University of Stellenbosch, Z.A.-7600, Stellenbosch, South Africa Received 29 May 2001; received in revised form 15 August 2001; accepted 6 October 2001
Abstract Acetic acid bacteria are microorganisms that can profoundly influence the quality of wine. Surprisingly, little research has been done on these microorganisms in the winemaking field. The object of this study was to investigate the occurrence of acetic acid bacteria in South African red wine fermentations and to identify the dominant species occurring. Acetic acid bacteria were isolated and enumerated from small-scale and commercial red must fermentations in 1998 and 1999, respectively. The initial occurrence of acetic acid bacteria in the must was shown to vary with cell numbers ranging from 106 – 107 to 104 – 105 cfu/ml for the 1998 and 1999 musts, respectively. The acetic acid bacteria decreased to 102 – 103 cfu/ml in musts having a low pH ( V 3.6), whereas in some musts having a high pH ( 3.7), the cell numbers increased during fermentation. During the process of cold soaking, the cell numbers of acetic acid bacteria also increased until inoculation with commercial wine yeast. Gluconobacter oxydans dominated in the fresh must and Acetobacter pasteurianus and A. liquefaciens during fermentation. This study showed that A. liquefaciens and A. hansenii were present in significant numbers, which has not been reported before. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Acetic acid bacteria; Red wine fermentations; pH
1. Introduction Acetic acid bacteria are divided into the genera Gluconobacter, Acetobacter and Frateuria (Holt et al., 1994). Of these, Gluconobacter oxydans, Acetobacter aceti, A. pasteurianus, A. liquefaciens and A. hansenii are normally associated with grapes and wine. The main species observed on unspoiled grapes and must is G. oxydans (Blackwood et al., 1969; Joyeux et * Corresponding author. Distell, Adam Tas Road, P.O. Box 46, Stellenbosch 7600, South Africa. Tel.: +27-21-809-7322; fax: +2721-886-5414. E-mail address: [email protected] ( M.G. Lambrechts).
al., 1984a). This species prefers a sugary rich environment and usually dies off during alcoholic fermentations. Acetobacter species prefer ethanol as carbon source (De Ley et al., 1984) and usually dominate during the later stages of fermentation and in wine (Drysdale and Fleet, 1985; Joyeux et al., 1984a). Acetic acid bacteria, although able to produce high amounts of acetic acid from ethanol, have not been of a major winemaking concern because of their requirement of oxygen for growth. The anaerobic conditions that generally prevail in wine and the presence of free sulfur dioxide are usually considered to be an adequate means of avoiding acetic acid bacterial spoilage. Therefore, surprisingly little research has been done
0168-1605/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 0 5 ( 0 1 ) 0 0 7 1 5 - 2
58
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
on the role of acetic acid bacteria in the winemaking process. However, in recent years, the ability of acetic acid bacteria to affect wine quality has been the subject of renewed interest. Joyeux et al. (1984a) have shown that acetic acid bacteria are able to survive and even grow during the winemaking process. Furthermore, acetic acid bacteria can produce other compounds, apart from acetic acid, that can influence wine quality (Drysdale and Fleet, 1989a). The objectives of this study were to investigate the occurrence of acetic acid bacteria in different South African red wine fermentations and to identify the dominant acetic acid bacteria strains. This study forms part of a research program to determine the origin of volatile acidity in South African red wines.
2. Materials and methods 2.1. Fermentations monitored Acetic acid bacteria were isolated and enumerated from red wine fermentations during the 1998 and 1999 harvesting seasons. During the 1998 season, six small-scale fermentations of Cabernet Sauvignon grapes were monitored. These fermentations consisted of 110-kg grapes each and were conducted in duplicate at 15, 22 and 30 C. At each fermentation temperature, the grapes were divided into two batches prior to yeast inoculation and only one half of the grapes were sulfured (50 mg/kg). The different musts were fermented on the skins until 0 B. The occurrence of acetic acid bacteria was determined in the must at 22, 18, 12, 6 and 0 B during the fermentation. During the 1999 harvesting season, bacteria were sampled from fermenting red musts at six commercial wine cellars in the Stellenbosch, Paarl, Franschhoek and Durbanville areas. Sampling was done prior to yeast inoculation, during the fermentation (between 10 and 14 B) and at the end of fermentation (between 0 and 5 B). Two of the cellars applied the technique of cold soaking prior to yeast inoculation. During cold soaking, which is a type of skin maceration, the must and skins were kept together between 15 and 18 C for 3 days. Five samples were taken from these musts.
2.2. Isolation and identification of acetic acid bacteria Acetic acid bacteria were isolated by plating 100 ml of a dilution series of juice or fermenting juice onto GYC medium [glucose (5% m/v), yeast extract (1% m/v), CaCO3 (3% m/v) and agar (2% m/v)] and Mannitol medium [mannitol (2.5% m/v), yeast extract (1% m/v) and agar (1.5% m/v)]. Growth of yeast and lactic acid bacteria (which normally exhibit weak growth on these types of media) were eliminated by adding 50 mg/l pimaricin (Delvocid, Gist-Brocades) and 50 mg/l nisin, respectively, to the media. Each dilution was plated out in triplicate and the plates incubated at 30 C for 6 days. Representative colonies were isolated and Gram staining and catalase tests were performed on each colony. Representative Gram ( ) rods and catalase (+) colonies were stored on GYC slants at 4 C and transferred monthly or kept at 80 C in 40% glycerol until identification. One hundred and fifteen acetic acid bacterial isolates were identified up to species level with the following biochemical and physiological tests: the ability to overoxidize ethanol, growth on sodium acetate, ketogenesis of glycerol, growth on dulcitol and production of a brown pigment on GYC medium (Drysdale and Fleet, 1988). Thirty-three of these isolates, chosen randomly, were also identified using numerical analysis of whole cell proteins by SDSPAGE (Pot et al., 1994). These 33 strains were grown on GYC agar for 70 h at 30 C. Cell-free extracts were prepared by washing the bacteria from the GYC agar with NaPBS buffer [consisting of 0.115% (m/v) Na2HPO4.12H2O, 0.023% (m/v) NaH2PO4.2H2O and 0.8% (m/v) NaCl, pH 7.3]. The bacteria were then centrifuged for 10 min at 10 000 rpm, the supernatant was discarded, the pellet was washed twice with 30 ml NaPBS buffer, and finally, 70 mg of the bacteria was placed in an eppendorf tube to which 1.26 ml of Sample Treatment Buffer [0.75% (m/v) Tris, C2H6OS (5% (v/v), pH 6.8], without sodium dodecyl sulphate (SDS), was added. The samples were vortexed and 0.14 ml of SDS [20% (m/v)] was added and vortexed. The mixture was then heated to 95 C for 10 min, cooled and centrifuged at 10 000 rpm for 5 min. The supernatant, which contained the extracted proteins, was removed and stored at 80 C. The denatured fraction was used for SDS-PAGE analysis. Registration of the protein electropherograms, normalization of
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
59
Table 1 Composition of musts used to monitor the evolution of acetic acid bacteria during red wine fermentations Year and area
Cultivar
Sugar concentration (B)a
Titratable acidity (g/l)
pH
SO2 dosage (mg/kg)b
1998 Stellenbosch (US) 1999 Stellenbosch (SI) Stellenbosch (SII) Stellenbosch (SIII) Paarl (PI) Franschhoek (FI) Durbanville (DI)
Cabernet Sauvignon Pinotage Merlot Merlot Cabernet franc Merlot Merlot
22.0 24.3 24.6 25.2 24.0 25.2 23.4
4.9 5.7 6.8 5.4 4.0 5.8 6.2
3.76 3.50 3.41 3.75 3.75 3.71 3.40
50 50 50 40 30 40 50
a b
B: Degree Balling where 1 B = 10 g/l sugar. Amount added at crushing.
at the higher fermentation temperatures. The number of acetic acid bacteria decreased from 106 –107 cfu/ml prior to yeast inoculation to 103 – 104 and 102 – 103 cfu/ml in the middle and at the end of the fermentation, respectively. During the 1999 harvesting season, fermenting musts from six commercial cellars were monitored for the occurrence and growth of acetic acid bacteria. The area, cultivar and general composition of the musts are shown in Table 1. In Fig. 1, the occurrence of these bacteria during fermentation is shown. It is clear that the acetic acid bacteria numbers decreased drastically from the beginning to the middle of fermentation in the low pH musts (pH < 3.6), but this did not happen in the higher pH musts (pH < 3.7). In the two high pH musts, the acetic acid bacteria numbers were slightly lower or even higher at the end of fermentation. In the low pH musts, these numbers were,
the densitometric traces, grouping of strains by the Pearson product moment correlation coefficient (r) and UPGMA cluster analysis were performed (Pot et al., 1994) using the Gelcompar version 4 software.
3. Results 3.1. Occurrence of acetic acid bacteria in red wine fermentations During the 1998 season, the growth of acetic acid bacteria was followed in Cabernet Sauvignon must. The composition of the must is shown in Table 1. The numbers of acetic acid bacteria as well as the dominant species during fermentation at the different temperatures are given in Table 2. The addition of sulfur dioxide influenced the acetic acid bacteria, especially
Table 2 Levels (cfu/ml) and species of acetic acid bacteria in Cabernet Sauvignon must fermented at 15, 22 and 30 C during 1998 Fermentation stage (B) 22 18 12
6 0
30 (C)
22 (C)
SO2
+ SO2 6
3.4 10 , G. oxydans 1.2 106, G. oxydans 6.0 103, A. aceti; 6.0 103, A. liquefaciens 1.0 103, A. pasteurianus 1.9 102, A. pasteurianus
15 (C)
SO2 6
1.2 10 , G. oxydans 3.2 103, G. oxydans 4.4 103, A. pasteurianus
1.4 103, A. pasteurianus 3.6 103, A. pasteurianus
+ SO2 6
8.5 10 , G. oxydans 3.2 106, G. oxydans 2.0 104, A. pasteurianus; 1.1 104, A. aceti 3.6 103, A. pasteurianus 3.2 102, A. pasteurianus
SO2 6
+ SO2 7
1.6 10 , G. oxydans 1.6 106, G. oxydans 7 103, A. pasteurianus
1.6 10 , G. oxydans 1.4 106, G. oxydans 3.1 104, A. aceti
1.4 106, G. oxydans 1.1 106, G. oxydans 3.0 104, A. pasteurianus
9.3 103, A. pasteurianus 1.0 103, A. pasteurianus
7.0 103, A. pasteurianus 1.3 103, A. pasteurianus
3.3 104, A. pasteurianus 4.2 103, A. pasteurianus
60
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
The growth of acetic acid bacteria in musts that underwent cold soaking prior to yeast inoculation is shown in Fig. 2. Both these winemakers incubated the must between 15 and 18 C during cold soaking and both added 40 to 50 mg/kg SO2 just after the grapes were crushed. During cold soaking, the population of acetic acid bacteria increased in the SIII must from 103 to almost 105 cfu/ml. After yeast inoculation, the bacteria quickly died off during fermentation, with a slight increase in cell numbers occurring at the end of fermentation. In the SI must, the acetic acid bacteria numbers remained almost the same during skin maceration and followed the same pattern as that of the SIII must during fermentation. 3.2. Identification of the isolated acetic acid bacteria
Fig. 1. Growth of acetic acid bacteria in different commercial red wine fermentations during the 1999 harvesting season. The pH values of the different musts are included.
however, between 102 and 103 cfu/ml lower than in the beginning of fermentation. G. oxydans dominated again in the freshly pressed must, except in the FI fermentation where only A. pasteurianus was detected (Table 3). During the course of fermentation, A. pasteurianus and A. liquefaciens dominated, with A. aceti and A. hansenii occurring irregularly and at lower cell numbers.
All Gram-negative, catalase-positive rods were preliminarily identified as acetic acid bacteria according to the biochemical and physiological tests described in the Materials and methods. The percentage of the 115 acetic acid bacterial strains belonging to G. oxydans, A. pasteurianus, A. hansenii, A. aceti and A. liquefaciens is indicated in Table 4. The identification of 33 of the above acetic acid bacteria was confirmed by the whole cell protein gel electrophoresis. A dendrogram of the clustering of the identified species is shown in Fig. 3.
Table 3 Occurrence of acetic acid bacterial species in commercial red wine fermentations during the 1999 harvesting season Area
Fermentation stage Beginning
Middle
End
SI SII
G. oxydans G. oxydans, A. liquefaciens
A. pasteurianus A. liquefaciens, A. hansenii
SIII
G. oxydans
PI FI
G. oxydans, A. pasteurianus A. pasteurianus
A. pasteurianus G. oxydans, A. liquefaciens, A. aceti A. pasteurianus, A. liquefaciens A. pasteurianus A. pasteurianus, A. liquefaciens
DI
G. oxydans
A. pasteurianus, A. liquefaciens, A. hansenii A. liquefaciens
A. liquefaciens
A. pasteurianus A. pasteurianus
Strains indicated in bold dominated at that stage of the fermentation.
Fig. 2. Effect of cold soaking and yeast inoculation on acetic acid bacterial growth. The grape must were incubated at between 15 and 18 C during the time of maceration.
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
61
Table 4 Biochemical and physiological tests used for the identification of acetic acid bacterial isolates Biochemical/ physiological test
EtOH oxidation Growth on sodium acetate Ketogenises from glycerol Growth on dulcitol Production of brown pigment
Number of isolates Protein cluster giving a (+) result for this test
I. II. III. IV. V. A. hansenii A. liquefaciens A. pasteurianus A. aceti G. oxydans
Percent of isolates 4 identified as strain
12
57
6
21
+ + + +
+ + +
+ +
+ + +
+
84 80 48 5 14
Five clusters with the SDS-PAGE numerical analysis of whole cell proteins were obtained. Cluster I, belonging to A. hansenii, also grew on dulcitol as carbon source. This is one of the tests described in Drysdale and Fleet (1988) to distinguish this Acetobacter strain from others. Cluster II, belonging to A. liquefaciens,
+
produced a brown pigment on GYC medium and Cluster III (A. pasteurianus) was ketogenesis-negative. These are tests used to identify these two species, respectively. These two clusters contained by far the most numerous species. Cluster IV belonged to A. aceti and Cluster V to G. oxydans, the latter being unable to
Fig. 3. Dendrogram showing the clustering analysis of acetic acid bacteria based on protein patterns of the red wine isolates and reference strains of Gluconobacter and Acetobacter spp. Grouping was performed by the unweighted average pair-group method. Reference strains were obtained from the Deutsche Sammlung von Mikroorganismen, Braunschweig, Germany.
62
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
overoxidize ethanol and grow on sodium acetate. These acetic acid bacteria are grouped into five, i.e. G. oxydans, A. pasteurianus, A. liquefaciens, A. hansenii and A. aceti well-defined clusters at r 0.68, which were related to each other at r 0.58. A large variation in strains is observed within each cluster if one considers the low r value of 0.68.
4. Discussion Acetic acid bacteria had long been believed to play little, if any, role during winemaking operations due to their aerobic nature (Drysdale and Fleet, 1988). Winemaking in general is an anaerobic process and the growth and survival of these bacteria under this and other unfavourable conditions like high ethanol concentrations, low pH and high SO2 concentrations seems unlikely. Some studies have, however, shown that acetic acid bacteria can survive during fermentation and the following operations in the winemaking process, such as malolactic fermentation and during maturation of the wine (Joyeux et al., 1984a; Drysdale and Fleet, 1985). Acetic acid bacteria had been shown to contribute significantly to volatile acidity in must and wine, and the production of acetic acid by these bacteria may thus also contribute to sluggish or stuck fermentations (Joyeux et al., 1984b; Drysdale and Fleet, 1989b). 4.1. Occurrence of acetic acid bacteria Acetic acid bacteria occurred at high numbers on grapes and the freshly pressed must in 1998. These numbers are higher than those expected from healthy grapes, but spoiled grapes can easily harbour these counts. The only species isolated at this stage was G. oxydans, which normally occurs on unspoiled grapes (Splittstoesser and Churney, 1992; Sponholz, 1993). The survival of this strain during the early stages of fermentation can only be explained by its low ethanol tolerance (De Ley et al., 1984). The higher toxicity of SO2 at 30 C could be explained by the bacteria being metabolically more active at this temperature, thus accumulating and metabolizing the SO2 faster. In the middle of the fermentation, A. aceti and A. liquefaciens occurred, especially in the fermentations containing no SO2. In the fermentations where SO2 had
been added, A. pasteurianus dominated and this strain also dominated towards the end of the fermentation. This could be due to the fact that the A. aceti and A. liquefaciens strains that occurred in these fermentations were more SO2-sensitive than the A. pasteurianus strain. Although the SO2 should not have had any effect on the bacteria at this stage of fermentation due to acetaldehyde binding it, the initial population of A. aceti and A. liquefaciens before fermentation commenced might have been affected more by the SO2 than was the case with A. pasteurianus. The dominance of A. pasteurianus towards the end of fermentation might also be due to the higher ethanol tolerance of this strain (De Ley et al., 1984). During the 1999 season, the bacterial counts were lower than in the 1998 fermentations. This difference in acetic acid bacteria counts was also observed in base wines for brandy distillation in a study undertaken at the Institute for Wine Biotechnology during these two seasons (unpublished data). G. oxydans dominated in most of the fresh musts, but an A. pasteurianus strain was found to dominate the FI must. It has not been reported before that an Acetobacter strain can dominate in a fresh, unspoilt must. The more ethanoltolerant Acetobacter strains dominated again the G. oxydans strains during fermentation. As in the 1998 season, A. pasteurianus dominated in some of the fermentations, with A. liquefaciens dominating others. This is a different scenario from that observed in French and Australian wineries where A. aceti and A. pasteurianus seemed to dominate (Joyeux et al., 1984a; Drysdale and Fleet, 1985). The growth of acetic acid bacteria during fermentation also correlated with the initial pH of the must. In the low pH musts, the decrease in acetic acid bacterial counts was more pronounced than in the high pH must (refer to Table 1). At a lower pH value, more of the free SO2 exists in the molecular form, which is the active toxic form towards bacteria. Furthermore, ethanol toxicity is known to increase at a lower pH (Dupuy and Maugenet, 1963). In one of the high pH musts, FI, the acetic acid bacterial counts were even higher at the end of the fermentation than at the beginning, indicating that acetic acid bacteria can grow during the alcoholic fermentation. The production of ethanol by the yeast has been shown to stimulate the growth of some Acetobacter strains that prefer ethanol over glucose as a carbon source (Swings and De Ley,
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
1981). This also raises the question if these bacteria are as aerobic as thought before, seeing that they are able to survive until the end of alcoholic fermentation. Acetic acid bacteria have been shown to be able to use quinones instead of oxygen as electron acceptors (Adlercreutz and Mattiasson, 1984). The growth of acetic acid bacteria in the different commercial fermentations also confirms the advantages of having a must with a low pH and the addition of SO2 to the must prior to alcoholic fermentation. However, in the SIII fermentation, which had a high pH, the acetic acid bacterial numbers also decreased drastically after yeast inoculation (Fig. 2). The sensitivity of acetic acid bacteria to pH, SO2 and ethanol is therefore also strain-dependent. The survival of these bacteria after alcoholic fermentation should, however, be investigated further. The survival of high numbers of acetic acid bacteria after alcoholic fermentation might affect wine quality, since many of these bacteria are being ‘‘brought over’’ from the fermentation to the following winemaking processes (Fugelsang, 1997). During the process of cold soaking, the growth of acetic acid bacteria can also occur as has been proven by this study. During this process, the advantages of having a must with a low pH and high SO2 concentration have also been proven. The maintenance of a low temperature alone might, thus, not be sufficient to control the unwanted growth of these microorganisms. Similar results were obtained by Drysdale and Fleet (1989b). Joyeux et al. (1984a) concluded that a storage temperature of 10 to 15 C could be beneficial in stopping acetic acid bacteria from growing in wine. The addition of sufficient SO2 and the lowering of the maceration temperature and pH with acid additions should, thus, be considered when the winemaker wants to apply skin maceration prior to inoculation. 4.2. Identification of acetic acid bacteria There was in general a good correlation between the biochemical and physiological tests and the SDSPAGE numerical analysis of whole cell proteins. The two strains SI8 and SI18 clustered with A. pasteurianus and were ketogenesis-negative on glycerol, but were, however, also unable to overoxidize ethanol. Holt et al. (1994) also found that certain A. pasteurianus strains were unable to overoxidize ethanol. According to our results, A. aceti and A. pasteurianus were closely re-
63
lated and this is in accordance with Sievers et al. (1994). The results of 16S RNA sequencing showed that A. hansenii and A. liquefaciens were on the same phylogenetic level and this is also indicated by our protein results (Sievers et al., 1994). Some of the acetic acid bacteria isolated from the same fermenting musts also showed a high correlation (r 0.86) on the basis of whole-cell protein fingerprinting. In conclusion, this study clearly indicated that acetic acid bacteria can survive and even grow in grape must during the alcoholic fermentation and that the pH and molecular SO2 concentration of must and wine are important factors influencing this ability. Furthermore, winemaking practices such as cold soaking will also influence the survival or growth of acetic acid bacteria. A low pH and sufficient SO2 concentrations should, however, help to prevent the unwanted growth of acetic acid bacteria. G. oxydans dominated in the fresh must, with Acetobacter species, especially A. pasteurianus and A. liquefaciens, occurring later during alcoholic fermentation. It thus seems as if the South African scenario for the occurrence of these bacteria differs from that of other wine producing countries. The survival and occurrence of acetic acid bacteria in must and wine had been studied before, but this is the first study we know of that investigated the growth of these bacteria in different fermentations during different harvesting seasons. Clearly, more research is needed on this group of microorganisms, which can profoundly influence the quality of grape juice and wine.
Acknowledgements The authors would like to thank the following people for the intellectual and technical support: Prof. I.S. Pretorius, Dr. M. du Toit, Prof. L.M.T. Dicks, L.P. Ellis, M. Gey Van Pittius, J. Bayly and C. Font-Sala.
References Adlercreutz, P., Mattiasson, B., 1984. Oxygen supply to immobilized cells: 4. Use of p-benzoquinone as an oxygen substitute. Appl. Microbiol. Biotechnol. 20, 296 – 302. Blackwood, A.-C., Guimberteau, G., Penaud, E., 1969. Sur les bacte´ries ace´tiques isole´es de raisins. C. R. Seances Acad. Sci., Ser. D 269, 802 – 804.
64
W.J. Du Toit, M.G. Lambrechts / International Journal of Food Microbiology 74 (2002) 57–64
De Ley, J., Gillis, M., Swings, J., 1984. Family VI. Acetobactaceae. Bergey’s Manual of Systematic Bacteriology, vol. 1, 9th edn. Williams and Wilkens, MD, USA, pp. 267 – 274. Drysdale, G.S., Fleet, G.H., 1985. Acetic acid bacteria in some Australian wines. Food Technol. Aust. 37, 17 – 20. Drysdale, G.S., Fleet, G.H., 1988. Acetic acid bacteria in winemaking: a review. Am. J. Enol. Vitic. 39, 143 – 154. Drysdale, G.S., Fleet, G.H., 1989a. The growth and survival of acetic acid bacteria in wines at different concentrations of oxygen. Am. J. Enol. Vitic. 40, 99 – 105. Drysdale, G.S., Fleet, G.H., 1989b. The effect of acetic acid bacteria upon the growth and metabolism of yeast during the fermentation of grape juice. J. Appl. Bacteriol. 67, 471 – 481. Dupuy, P., Maugenet, J., 1963. Me´tabolisme de l’acide lactique par Acetobacter rancens. Ann. Technol. Agric. 12, 5 – 14. Fugelsang, K.C., 1997. Wine Microbiology. Chapman & Hall, New York. Holt, J.G., Krieg, N.R., Sneath, P.H.A., Staley, J.T., Williams, S.T., 1994. Genus Acetobacter and Gluconobacter. Bergey’s Manual of Determinative Bacteriology, 9th edn. Williams and Wilkens, MD, USA, pp. 71 & pp. 84. Joyeux, A., Lafon-Lafourcade, S., Ribe´reau-Gayon, P., 1984a. Evolution of acetic acid bacteria during fermentation and storage of wine. Appl. Environ. Microbiol. 48, 153 – 156.
Joyeux, A., Lafon-Lafourcade, S., Ribe´reau-Gayon, P., 1984b. Metabolism of acetic acid bacteria in grape must: consequences on alcoholic and malolactic fermentation. Sci. Aliments 4, 247 – 255. Pot, B., Vandamme, P., Kersters, K., 1994. Analysis of electrophoretic whole-organism protein fingerprints. In: Goodfellow, M., O’Donnell, A.G. (Eds.), Chemical Methods in Prokaryotic Systematics. Wiley, Chichester, UK, pp. 493 – 521. Sievers, M., Ludwig, W., Teuber, M., 1994. Phylogenetic positioning of Acetobacter, Gluconobacter, Rhodopila, and Acidiphilium species as a branch of acidophilic bacteria in the asubclass of Proteobacteria based on 16S ribosomal DNA sequences. Syst. Appl. Microbiol. 17, 189 – 196. Splittstoesser, D.F., Churney, J.J., 1992. The incidence of sorbic acid-resistant Gluconobacters and yeasts on grapes grown in New York State. Am. J. Enol. Vitic. 43, 290 – 293. Sponholz, W.-R., 1993. Wine spoilage by microorganisms. In: Fleet, G.H. (Ed.), Wine Microbiology and Biotechnology. Harwood Academic Publishers, Chur, Switzerland, pp. 395 – 420. Swings, J., De Ley, J., 1981. The genera Acetobacter and Gluconobacter. In: Starr, M.P. (Ed.), The Prokaryotes. Springer-Verlag, Berlin, Germany, pp. 771 – 778.