|
A Technical Guide for Wine Producers
|
|
RECENT ARTICLES | WYNBOER HOME
The use of non-Saccharomyces fructophilic yeasts for efficient fermentation of grape juice
Neil Jolly
Key words: glucose:fructose ratio, non-Saccharomyces yeasts
In South Africa, the glucose to fructose ratio (GFR) in grapes at harvest may vary from 0.96 to as low as 0.77. Wine fermentation with glucophilic yeast, starting at a GFR of less than one, has a greater chance of reaching a critical imbalance (GFR ≤ 0.1), with its accompanying fermentation problems. In this study the effect of GFR on fermentation was investigated in fructose spiked musts during laboratory-scale fermentations. This was followed by an investigation into the use of fructophilic yeasts to change the GFR at the onset of fermentation in a proactive approach to prevent stuck fermentations during small-scale wine production trials. Preliminary results indicate that a combination of non-Saccharomyces and S. cerevisiae yeasts leads to more efficient utilisation of grape sugars. INTRODUCTION The glucose to fructose ratio (GFR) in grapes at harvest can vary, but is generally accepted to be close to one (Amerine & Thoukis, 1958; Amerine et al. 1972; Bisson, 1993). During fermentation, the glucophilic wine yeast, Saccharomyces cerevisiae, can preferentially utilise glucose above fructose. Should the GFR decrease to 0.1, it can result in a stuck fermentation with a higher residual fructose than glucose concentration (Gafner et al., 2000). Inoculation at this stage with fructophilic non-Saccharomyces yeast can rectify the imbalance, whereupon S. cerevisiae can start fermenting again (Sütterlin et al., 2004). In South Africa, the GFR of some grapes at commercial harvest can vary between 0.96 to as low as 0.77 (Snyman, 2006; Jolly, unpublished data, 2006). This appears to be related to cultivar and area of production (G. Baumgarten, personal communication, 2005). The resultant fermentations with the glucophilic S. cerevisiae could therefore be at a greater risk of having their GFR dropping to below 0.1 with the accompanying risks of a stuck fermentation. Correction of this GFR imbalance with selected non-Saccharomyces yeasts as demonstrated by Sütterlin et al. (2004) requires that the yeast be ethanol tolerant, as well as be able to grow in an environment where nutrients other than sugar may be limiting.  Fig. 1. Decline in fructose concentration (average of two fermentations) in a 2006 Chardonnay must fermented by different yeast combinations.
In a proactive approach the GFR could be altered at the start of fermentation by fructophilic yeast. Here ethanol tolerance would not be required, but osmo-tolerance would be an added bonus, especially for the high sugar concentration in South African musts. Suppression of the fructophilic yeast by the fast growing S. cerevisiae is also of less importance. Therefore, in this study the effect of a lower GFR on the subsequent fermentation was investigated in fructose-spiked musts during laboratory-scale fermentations. This was followed by an investigation into the use of fructophilic non-Saccharomyces yeasts to change the GFR at the onset of fermentation in a proactive approach to prevent stuck fermentations during small-scale wine production trials. MATERIAL AND METHODS Yeast strains The yeast strains used in this study are shown in Table 1. Laboratory-scale fermentations A previously frozen, clarified base must was used to prepare two sets of four fermentation mediums, i.e. a normal must and three with increasing concentrations of added fructose as shown in Tables 2 and 3. A nitrogen source (Nutrivin Super, Anchor Bio-Technologies, South Africa) was added at a concentration of 0.5 g/L to the second set of fermentation mediums. The individual musts were aliquoted into 250 mL lots and sterilised by autoclaving. A selection of yeasts (Table 1) was used (2% inoculum, 24h culture, YPD broth [Biolab, Merck]) and the fermentation vessels were closed by tightly fitting fermentation caps. The fermentations were all done in duplicate at 20°C. Residual glucose and fructose concentrations were determined at completion of fermentation by the enzymatic method (Vinlab, Stellenbosch; and Koelenhof Wynlaboratoriumdienste (Wine laboratory services), Stellenbosch). Small-scale wine production Non-Saccharomyces yeasts (Table 1) were investigated in combination with a commercial S. cerevisiae (strain VIN 13, Anchor Bio-Technologies, South Africa) yeast for small-scale production of wine in aliquots of 18 L of freshly prepared Chardonnay and Chenin blanc must from the 2006 vintage (adjusted to 50 mg/L SO2). All the trials were done in duplicate. For the Chardonnay must, six non-Saccharomyces yeast cultures i.e. 46, 45, 48, 74, 110 and UCD, (propagated in YPD broth) were inoculated individually at 1 x 106 cells/mL. This was followed within one hour by active dried wine yeast (VIN 13) at 30 g/hL. Other wine-production treatments were according to the standard Nietvoorbij procedures for small-scale white wine production (Jolly et al., 2003a). During fermentation, 20 mL aliquots were removed under CO2 gas for glucose and fructose analyses (enzymatic method). Chenin blanc wine production followed the same method, with the exception that experimental active dried non-Saccharomyces yeasts i.e. C1-15 and 825 were used at 25 g/hL instead of wet cultures. Reference fermentations were inoculated with VIN 13 only. The wines were subjected to a descriptive sensory analysis five months after production as previously described (Jolly et al., 2003b). The judging criteria were “fruity aroma”, “butter aroma”, “body” and “general quality”; and “fruity aroma”, “guava aroma” and “general quality” for the Chardonnay and Chenin blanc wines, respectively. RESULTS AND DISCUSSION Sourcing of grape must with a specific low GFR is not practical for laboratory trials, therefore a laboratory protocol was needed whereby stuck fermentations due to GFR imbalances could be induced by using fructose-spiked grape must inoculated with S. cerevisiae references yeasts. Table 2 shows that the lower the GFR at the start of fermentation, the higher the total residual sugar (glucose + fructose) at the end of fermentation. As was expected, the fructose fraction was always higher than the glucose. The reference yeast Fermichamp, marketed as a “fructophilic” S. cerevisiae, did not perform very well in this evaluation. However, it did appear to utilise fructose better in the lower GFR musts. It also appeared that the cut-off GFR of 0.1, as shown by Gafner et al. (2000), does not always hold true, as some of the fermentations went to dryness even with a GFR below 0.1. From this experiment it appears that in some cases BDX is better able to utilise fructose than NT 112, although in South Africa NT 112 is generally considered to be a stronger fermenter than BDX. When the experiment was repeated in a second must, but with the addition of a complex nitrogen source i.e. Nutrivin super (containing inactivated yeast, di-ammonium phosphate and thiamine), the S. cerevisiae yeasts were better able to ferment to dryness (Table 3). Here NT 112 generally out performed BDX and Fermichamp, and it was only in the 0.80 GFR must (i.e. N3 + 30) that NT 112 was unable to utilise all the fructose. From this data it therefore appears that some of the problems regarding residual fructose may, in some instances, be prevented by the judicious use of complex nitrogen sources. Cellar-scale evaluation of fructophilic yeasts An initial screening of yeasts found in the ARC Infruitec-Nietvoorbij microbiology culture collection resulted in the selection of five yeasts that showed faster fructose utilisation than S. cerevisiae reference yeast strains NT 112 and BDX (data not shown). This selection consisted of two Candida stellata strains and one strain each of Zygosaccharomyces bailii, Candida pulcherrima and S. cerevisiae (strain 110) (Table 1). A sixth yeast, a Candida zemplinina strain, was also included. This is a newly described species that has been designated an extreme fructophilic yeast (Sipiczki, 2003; D. Mills, personal communication, 2005). The strain used in this investigation was originally isolated from fermentations of Botrytis-affected Semillon grapes (Mills et al., 2002).
During the small-scale (18 L) wine production trials using Chardonnay must, the six selected yeasts were co-inoculated individually with the commercial S. cerevisiae strain VIN 13. This was necessary as the fructophilic yeasts were only expected to play a role in the initial phase of fermentation, while the commercial S. cerevisiae wine yeast was desired for completion of fermentation and production of a specific wine style. The Chardonnay grape must had an initial GFR of 1.01 and was therefore not expected to present any fermentation problems. With one exception, all the glucose and fructose was utilised (data not shown). The exception was the S. cerevisiae strain 110 / S. cerevisiae strain VIN 13 combination where some residual fructose remained (Fig. 1). However, when the same data was presented in the form of GFR (Fig. 2), it appears that there were three types of fermentation. The first is the already mentioned S. cerevisiae / S. cerevisiae fermentation, where the GFR dropped to 0.1 (the ratio where fermentations are expected to cease) and then further to 0.03. This GFR pattern therefore represents an inefficient fermentation i.e. with residual fructose (Fig. 1). The second type of fermentation was the C. stellata strain 46 / S. cerevisiae fermentation where the GFR decreased to 0.14 and then started increasing again, in other words, the 0.1 mark was never reached. This GFR pattern can therefore be equated to an efficient fermentation. The third group, comprising the remaining five fermentations, all had their GFRs decreasing to below 0.1 before increasing again. However, as they all went to fructose dryness, these could be viewed as marginal fermentations. In this instance they were efficient, but it can be speculated that should other fermentation conditions have been stressful to the yeast e.g. nitrogen status of must, incorrect fermentation temperatures, etc. then these fermentations may have finished with residual fructose. The C. stellata strain 46 / S. cerevisiae strain VIN 13 combination, representing the most efficient fermentation, produced wine of a similar quality to the reference fermentation (Table 4). This supports the findings of Jolly et al. (2003b), who showed that although their C. stellata strain could produce unacceptably high levels of volatile acidity when fermenting on its own, it did not do so in co-inoculated fermentations with S. cerevisiae. The Chenin blanc must with an initial GFR of 1.04 was also not expected to present any fermentation problems. The two non-Saccharomyces yeasts used for these trials, C. pulcherrima strains 825 and C1-15, were originally selected for improved sensory quality in co-inoculated fermentations (Jolly et al., 2003a). During the small-scale wine fermentations the GFR (Fig. 3) followed the same pattern of change for an efficient fermentation as previously discussed (GFR did not drop below 0.1). In contrast, the S. cerevisiae only fermentations with VIN 13 and Fermichamp were less efficient (marginal), as their respective GFR’s dropped to below 0.1 before increasing. These experimentally dried non-Saccharomyces yeasts therefore not only enhance the sensory profile of the wine (Jolly et al., 2003a), but also contribute to an efficient utilisation of sugars. CONCLUSIONS From the data presented it appears that a grape must with a GFR below one can, in some instances, lead to incomplete fermentation. The judicious use of a complex nitrogen source may, in some instances, help to prevent this. However, the use of selected fructophilic, non-Saccharomyces yeasts co-inoculated with the standard wine yeast, at the start of fermentation, can lead to more efficient utilisation of sugars, ensuring a problem free fermentation. LITERATURE CITED Amerine, M.A. & Thoukis, G., 1958. The glucose-fructose ratio of California grapes. Vitis 1, 224-229. Amerine, M.A., Berg, H.W. & Cruess, W.V., 1972. Technology of wine making 3rd edition. The AVI Publishing Company, Inc., Westport. Bisson, L.F., 1993. Yeasts – metabolism of sugars. In Fleet, G.H. (ed.). Wine microbiology and biotechnology. Harwood Academic Publishers, Chur, Switzerland. Gafner, J., Hoffmann-Boller, Porret, N.A. & Pulver, D., 2000. Restarting sluggish and stuck fermentations. Paper: 2nd International Viticulture and Enology Congress, 8-10 November, Cape Town, South Africa. Jolly, N.P., Augustyn, O.P.H. & Pretorius, I.S., 2003a. The use of Candida pulcherrima in combination with Saccharomyces cerevisiae for the production of Chenin blanc wine. S. Afr. J. Enol. Vitic 24, 63-69. Jolly, N.P., Augustyn, O.P.H. & Pretorius, I.S., 2003b. The effect of non-Saccharomyces yeasts on fermentation and wine quality. S. Afr. J. Enol. Vitic 24, 55-62. Mills, D.A., Johannsen, E.A. & Cocolin, L., 2002. Yeast diversity and persistence in Botrytis-affected wine fermentations. Appl. Environ. Microbiol. 68, 4884-4893. Sipiczki, M., 2003. Candida zemplinina sp. Nov., an osmotolerant and psychrotolerant yeast that ferments sweet botrytised wines. Int. J. Syst. Evol. Microbiol. 53, 2079 83. Snyman, P., 2006. Die glukose:fruktose verhouding van wyndruiwe. Wineland April, 60-61 (Wynboer 200). Sütterlin, K.A., Hoffmann-Boller, P. & Gafner, J., 2004. Kurieren von gärstockungen mit der fructophilen weinhefe Zygosaccharomyces bailii. Poster: 7th International Symposium on Innovations in Enology, Intervitis Interfructa 2004, 10 11 May, Stuttgart-Killesberg, Germany. For further details, please contact Neil Jolly at e-mail: jollyn@arc.agric.za |
|
Wynboer is incorporated in WineLand, magazine of the SA wine producers.
Visit our sister sites:
|
| UP |
COPYRIGHT (C) 2000 WineLand |