Saccharomyces cerevisiae, the humble yeast, is really the hero of winemaking. Without it, there would be no wine, only sour grapes! It is Saccharomyces that fends off Hanseniaspora and other spoilers and completes the work of fermenting sugar to alcohol. In the process, the mighty Saccharomyces also transforms precursors into volatile aromas, and sacrifices its dying body to provide nutrients for malolactic bacteria, all while donating mannoproteins for mouthfeel. Though domesticated thousands of years ago, this yeast maintains enough genetic diversity to literally fill a catalogue! But how is a winemaker to choose? Does yeast strain even matter? There’s one way to find out… do an experiment!

Saccharomyces cerevisiae was first identified as the agent of fermentation by Louis Pasteur in 18601, however, this species has been around for a lot longer than that. Based on similarity in genetic sequences, it is thought that Saccharomyces cerevisiae as we know it arose from a single domestication event in ancient Mesopotamia, around the same time as the domestication of grape vines. From there, it spread around the world along with the vines that produce its food source. There are 5 distinct genetic lineages of Saccharomyces cerevisiae: West African, Malaysian, North American, Sake, and European. Wine yeast, whether commercial or ambient, all belong to the European group, indicating that technological usage is more important than geography in the genetic history of this organism. There do not appear to be separate genetic patterns for vineyard vs. winery strains of S. cerevisiae, suggesting the two populations mix freely, probably mediated by movements of insects and humans as well as dumping of pomace (1). 

Saccharomyces cerevisiae most likely originated from the sap and bark of oak trees and possibly the soil underneath those trees. It may be the offspring of Saccharomyces paradoxus and/or S. uvarum, as they share this native habitat (2). S. cerevisiae are only a minor member of the community of microbes found on grapes, often below the level of detection when these communities are surveyed. When fruit is damaged or seepage occurs, the conditions for growth improve, and Saccharomyces can be found in increasing numbers (1,2). In the winery, however, this yeast dominates, where it can be found in significant numbers on winery equipment and coating the walls themselves. The concentration of yeast cells found in the winery rivals the overall concentration of microbes found on grapes, often providing significant inoculum for both spontaneous and inoculated fermentations. When surveyed, it is common to find one or a few strains dominating a single winery, with limited geographic spread outside the winery (into the vineyard), while it is rare to find whole regions dominated by the same few strains. Resident strains do, however, shift from year to year (2).

from Wein-Plus glossary, Saccharomyces, by Megan, edited by Tischelmayer, 2018

 

The genetic code of domesticated Saccharomyces cerevisiae contains some clues to its secrets for dominance in wine fermentation. For example, the use of SO2 as an antimicrobial agent was likely introduced sometime between the Roman Empire and the Middle Ages. Around this time, there was a chromosomal rearrangement in one lineage of S. cerevisiae that led to a dominant allele for a sulfite pump, conferring a high level of SO2 resistance to the cells. This single innovation was so beneficial to the cells that have it that now 50% of modern isolates from wine carry this genetic signature. A similar bottleneck can be found for copper resistance genes arising around the time of copper introduction in the vineyard (1).

Despite its long history of domestication and these bottlenecks, Saccharomyces cerevisiae as a species includes a tremendous amount of genetic diversity that leads to a diverse set of responses to environmental stimuli, and results in real differences in fermentation kinetics, aromatic production, and other characteristics of the resulting wine. Regardless of the strain, all yeast convert sugar (glucose and fructose) to ethanol as a way of releasing energy for the cells to use to build cell structures and maintain order. This is done by a series of enzymatic reactions happening inside the cell. Each enzyme is encoded by a gene, which may have slight differences from the same gene in a different strain. These genetic differences may make that enzyme faster or slower, favor one substrate over another, or even nonfunctional. These differences add up over many thousands of enzyme reactions within the cell. Strains may also have genetic differences that make them more or less likely to respond to outside stimuli, such as a change in temperature or the presence of a competing microbe. For the cell, these may mean the difference between dominance and extinction in a given fermentation; for the winemaker, this may mean differences in aromatic profile or completion of the fermentation.

from Yeast and bacterial modulation of wine aroma flavor, Australian Journal of Grape and Wine Research, January 2005)

 

Much of the aroma and flavor of wine stems from the byproducts of fermentation and its side reactions, and genetic differences among strains leads to differences in the level of production of any given metabolite. Some known examples of enzyme derived aroma and flavor molecules include (2):

• The release of volatile terpenes, norisoprenoids, and thiols by the enzymatic cleavage of sugar molecules from non-volatile precursors 
• Production of acetic acid, acetoin, and succinic acid as byproducts of fermentation
• Release of acetate esters due to accumulation of Acetyl Co A 
• Release of higher alcohols as amino acids in the juice are broken down for nitrogen
• Production of aroma precursors from fatty acid production in cell walls based on environmental conditions (temperature) 
• Increase in glycerol production due to overexpression of glycerol-3-phosphate dehydrogenase. This also leads to accumulation of acetaldehyde, pyruvate, acetate 2,3, butanediol, succinate, and acetoin

A useful example of the impact of genetic variation is provided by Linda Bisson (2012)1 as she describes the superpowers of EC118 (aka Prise de Mousse). This strain is known as a steady fermenter, a good choice when it is essential that the fermentation finish with good kinetics. Among the reasons EC118 is so steady are the variants of transport proteins it houses in its cell membranes. This strain has a better peptide transporter than most, which means it does a better job of taking up organic nitrogen resources from the environment. The cell breaks down these amino acids by stripping nitrogen for its own use, leaving amino acid side chains that can be converted to higher alcohols and esters, i.e. aromas and flavors. EC118 also contains a gene for a fructose proton symporter, which means that it can transport fructose into the cell in the flow of protons. Many yeast have transport proteins that take up glucose preferentially, leaving leftover fructose (residual sugar) at the end of fermentation (1). This yeast is not perfect, however. It is also thought to produce relatively high levels of SO2 during alcoholic fermentation, which can inhibit malic acid bacteria.

The genetic diversity of Saccharomyces is such that no one strain accurately portrays the whole species (1). And it is this diversity that can become a valuable tool for the winemaker. A quick glance at the Scottlabs catalogue indicates there is a yeast strain for nearly every fermentation. Want to increase thiol release from percursors? There is a yeast for that! Want to break down proteins, produce fruit esters, stabilize color… there are yeast strains for that too! The diversity of commercial yeast strains that have been isolated, cultured and characterized can be overwhelming, but also exciting. Caution must be taken to choose appropriate strains for the situation, including temperature and alcohol tolerances, as well as nutrient needs. And, as Jackson (2014) points out, “the main characteristics of most commercial strains are known, but most of their other properties are not…It is once again up to the winemaker to do their own individual experimentation to determine what best suits their situations and preferences” (2).

---

References

(1) Bisson, L. F. Geographic Origin and Diversity of Wine Strains of Saccharomyces. American Journal of Enology and Viticulture 2012, 63 (2), 165–176. https://doi.org/10.5344/ajev.2012.11083.
(2) Jackson, R. S. Wine Science: Principles and Applications, 4 edition.; Academic Press: Amsterdam, 2014.

Summary

            Chardonnay is the most widely planted wine grape variety in Virginia¹ and is vinified in many different styles. Two elements that affect Chardonnay style are fermentation yeast and fermentation temperature. In this study, wine made with Ionys WF (Scottlabs) yeast, a yeast strain with properties of acid retention is compared to wine made with CY3079, a classic Chardonnay yeast in broad use. Each yeast strain was tested at cool (15°C) and warm (22°C) ambient temperatures. Ionys WF showed slower fermentation kinetics with greater acid retention. Warm fermentation with Ionys yeast had the lowest total SO₂ and lowest volatile acidity at the end of fermentation. In triangle tests comparing CY3079 fermentation at cold and warm temperature, the wines were found to be significantly different. Comparison of wines fermented at warm temperature with different yeasts were also different. Wine fermented with CY3079 at warm temperature received higher scores for body than those fermented cold or with Ionys yeast.

Introduction

Chardonnay is the most planted variety of wine grape in Virginia¹ and nearly every winery sells at least one Chardonnay product. Of all the grape varieties, Chardonnay winemaking may be the most varied. There are so many expressions of this grape! Some winemakers choose to treat the juice reductively while others intentionally oxygenate juice. Fermentation vessels include stainless steel, concrete eggs, neutral and new oak barrels. The choice of whether to allow or encourage malolactic fermentation is also different among cellars and among fruit sources. This study explores two other differences in Chardonnay winemaking: fermentation temperature and yeast strain. A primary concern in Chardonnay winemaking in Virginia is the retention of acidity and expression of minerality, so two yeast strains will be tested, one commonly used barrel fermenting yeast and a second that is thought to improve acid retention. This yeast is slow at cool fermentation temperatures, though, so each yeast will be tested at two ambient temperatures, one cooler fermentation in the barrel room (60°F0 and one warmer fermentation in the tank room (72°C). During a panel discussion on Chardonnay winemaking in Virginia at the VWA Technical Meeting in 2017, Jim Law of Linden Vineyards mentioned that he used to ferment his Chardonnay at cool temperatures, but in recent years has moved to a warmer fermentation temperature in this variety. This was further incentive that warm fermentation of Chardonnay should be examined.

Saccharomyces cerevisciae is the primary species of yeast used in commercial wine production. Despite a single domestication event in Mesopotamia around the same time as domestication of wine grapes themselves², it remains so genetically diverse that no one strain accurately portrays the whole species². The primary metabolic function of Saccharomyces cerevisciae in wine production is the metabolism of sugar into ethanol with the coupled release of energy. However, complex cellular machinery is also at work to build and maintain cellular components through alternative pathways. It is the work of these pathways that leads to the production of volatile aromas and flavors that gives wine its complexity and diversity³. At any given time, hundreds of substrates are taken up into the yeast cell, transformed by enzymatic pathways, with end products expelled into the environment or used by the yeast cell itself. Though most S. cerevisciae strains share most of these pathways in common, the extent to which any one pathway functions is determined by both genetic differences and environmental influences.³  It is the differences in the functioning and regulation of these enzymes that characterize the myriad of yeast strains that are commercially available for wine production.

Ever since Louis Pasteur first identified S. cerevisciae as the primary fermentative yeast in wine in 1860, many thousands of strains have been isolated, selectively bred and cultured for a wide range of characteristics. Manufacturers of enological products maintain vast collections of yeast strains and produce catalogues complete with descriptions of strain capabilities and tolerances. Regardless of the purpose for which a strain was initially bred, it is wise to heed this advice from Ronald Jackson:

“The main characteristics of most commercial strains are known, but most of their other properties are not, or if known they are buried in research papers not readily accessible to most winemakers. The local conditions often are crucial to feature expressions. It is again up to the winemaker to do their own individual experimentation to determine what best suits their situations and preferences.³”

In this study, two yeast strains will be explored for barrel fermentation of Virginia Chardonnay: CY3079 and Ionys WF.

CY3079 (Scottlabs) is a yeast strain commonly used for barrel fermentation of Chardonnay. Isolated from Burgundy, France, this strain is known for good fermentation kinetics in barrel. It has a temperature tolerance from 59-80°F and is highly recommended for sur lie aging as it autolyzes quickly to release polysaccharides for enhanced mouthfeel in the wine. This yeast enjoys widespread use in Virginia cellars.

According to the manufacturer’s literature, Lallemand Ionys WF yeast (Scottlabs) has been selectively bred to reduce alcohol production and retain acid during red wine fermentations. This yeast strain produces organic acids such as succinate during fermentation and has been shown to increase TA by 0.4 - 1.4 g/L tartaric equivalents. Matthieu Finot has been using this yeast in his barrel fermented Chardonnay for the acid-producing quality. A previous study (2016) did not show large differences in acid chemistry, but did show differences in fermentation kinetics (WF is much slower) as well as sensory differences with a preference for WF fermented wines.

One drawback to using Ionys yeast is that it was bred for fermentation of red wines, which are typically done at warmer temperature. The kinetics for Ionys fermentations at the cooler temperatures commonly used for white wines can be very long (up to four weeks in the previous study). Long fermentation can lead to juice oxidation and loss of reductive volatiles (like thiols) if fermentation takes too long to start and can run the risk of a stuck fermentation as yeast cells languish in a higher alcohol environment³. Also, the growth of epiphytic (non-Saccharomyces) yeast during a slow start may produce spoilage compounds such as acetic acid or ethyl acetate³. One approach to prevent these issues is to allow fermentations to be conducted at warmer temperature. However, fermentation temperature affects many more elements than just rate.

Gene expression in yeast changes in response to ambient temperature during fermentation. Principally, this is thought to allow cell membranes to stay intact in higher temperatures^4,5, however these changes also affect enzymatic pathways that produce aroma compounds, both positive and negative³. It is generally thought that fermentation at cooler temperature leads to wines that are fresher and fruitier. This has been confirmed by studies showing higher levels of fruity esters and fatty acid ethyl acetates produced and maintained at cooler temperatures.³ However, the effect of fermentation temperature on the sensory properties of the resulting wine may be a little more complex.

Killian and Ough (1979)⁶ measured the expression of several esters, compounds that contribute fruity and floral aromas to wine, at fermentation temperatures of 10°C and 15-20°C. They found that some fruity esters (isoamyl acetate, isobutyl acetate, ethyl butyrate, and N-hexyl acetate) were more likely to be expressed at lower temperatures while others (ethyl octanoate, 2-phenethyl acetate) were found in higher concentrations in the warmer fermentation. The esters produced in higher concentration at low temperatures have descriptors such as banana, pear, fruity, current, and apple while those produced more at higher temperatures have descriptors such as rose, honey, fruit, burned and beer. These authors attribute most of the difference in esters in the finished wine to the boiling point of the molecules, indicating loss of volatiles due to evaporation was likely to play a role.

Molina et al (2007)⁴ measured differences in gene expression during fermentation at 15°C and 28°C and correlated these to differences in concentration of volatiles, showing the differences were due to overall production not just evaporative loss. They used an artificial must containing only glucose, fructose, and amino acids as available substrates and were able to identify 23 volatile fermentation products, indicating yeast transformation of primary metabolites is essential to wine character. They did not find eight other commonly found volatile compounds thought to significantly contribute to wine aroma, indicating grape precursors are also essential⁴. They also found that ethyl esters were much higher in cooler fermentations (15°C vs. 28°C), but they, too, found positive volatiles that were produced in higher amounts in warmer temperatures. One of these, ethyl-2-methyl butanoate contributes fruity aromas of banana and pineapple, leading to a tropical fruit character to the wine.

Fermentation temperature is also thought to affect the production of thiols. Thiols are sulfur containing compounds⁷ that can have negative (rotten egg, cooked vegetable) or positive (passionfruit, guava, box tree) aromas. The positive attributes have become principle descriptors for Sauvignon Blanc, but many other varieties have been shown to contain thiols⁸, including Chardonnay⁹. At least one study has shown that warmer fermentation temperature produces a higher concentration of thiols¹⁰, though these may also be lost to evaporation at a faster rate in warmer fermentations³.

There are other sensory effects of cool fermentations. Often these have higher production of alcohol.  Alcohol itself has sensory characteristics (it adds body and can seem sweet in moderate concentrations) and it also affects the perception of other aromatic compounds, making them more volatile¹¹.  Slower fermentations have also been shown to have greater complexity due to the activity of non-Saccharomyces yeast, which, in addition to spoilage, can also contribute to highly desirable fruity-floral aromas.

The purpose of this trial is to compare wine chemistry and sensory characteristics of wine fermented with a “standard” yeast (CY3079) and a yeast selected to retain acid in Chardonnay (Ionys WF). Both strains were fermented in duplicate barrels at cool (60 °F ambient) and warm (72 °F ambient) locations.

Methods

With the exception of yeast inoculation and ambient temperature, all winemaking operations were the same among treatments. There were four treatments total:

1. CY3079 yeast at “cool” temperature (60°F ambient temp)
2. CY3079 yeast at “warm” temperature (72°F ambient temp)
3. Ionys WF yeast at “cool” temperature (60°F ambient temp)
4. Ionys WF yeast at “warm” temperature (72°F ambient temp)

Chardonnay grapes were harvested on August 29 and chilled overnight, then whole cluster pressed into tank with the addition 20 ppm SO₂.  After cold settling, on September 1, reverse osmosis was performed on the juice with 13% loss of volume/concentration.  Concentrated juice was racked into barrels of comparable age, cooper and dimension (Ana Selection J M 15 or T M 15) in replicate pairs so that there were two barrels per temperature/yeast pair. Juice chemistry was measured before and after RO operation. All barrels received juice from the same tank after reverse osmosis.

Inoculation differences: The initial dose of yeast at inoculation was an additional variable for this trial. Ionys WF is a yeast not usually used in white wines, and based on the experience of the winemaker, this yeast can be slow at lower temperatures. In order to encourage the completion of fermentation, this yeast was inoculated at a rate of 25 g/hL. CY3079 is a strong barrel fermenter that can produce off odors if overly vigorous and, in the experience of the winemaker, produces more pleasant wines when inoculated at 14 g/hL. The decision was made to use different inoculation doses to test the best quality wine produced from each yeast.

Barrels were inoculated with yeast rehydrated in 1.25X Goferm Evolution. For the purpose of the experiment, no acid additions were made. After inoculation, two barrels of each yeast group were placed in the cellar for fermentation (“cool” treatment) while two barrels were placed in the tank room for fermentation (“warm” treatment). It is estimated the cellar was 60°F and the tank room was 72°F during fermentation. Temperature was monitored during fermentation but not otherwise manipulated. Though both yeasts have high nutrient requirements and higher temperatures may increase nutrient utilization, YAN was measured to be 386 ppm, well above that requiring nutrient addition, therefore no nutrient additions were made. Wines were allowed to go through full malolactic fermentation without inoculation prior to the addition of 45 ppm SO₂. Wine was aged on lees with battonage once per week beginning when specific gravity dipped below 1 and continuing until sampling for the sensory session.

Sensory analysis was completed by a panel of ­­­29 wine producers. Wines were presented blind in randomly numbered glasses. Tasters were presented with three wines, two of one type and one of another, and asked to identify which wine was different (a triangle test). There were three tasting groups with the unique wine in the triangle test balanced between groups. Tasters were then asked to score each wine on a scale of 0 to 10 for fruit intensity, floral intensity, body and minerality. They were also given open ended questions to describe the wines. Two flights were presented this way, one comparing temperatures for CY3079 barrels and one comparing yeast groups at warm temperature. Results for the triangle test were analyzed using a one-tailed Z test. Descriptive scores were analyzed using repeated measures ANOVA.

A third flight was also presented with all four wines in randomly numbered glasses. The wines were presented in different order for each of three tasting groups. Tasters were asked to sort the wines into two groups of two and give descriptors for why they sorted the way they did. Then tasters were asked to rank the wines in order of preference. Preference rankings were analyzed using a Friedman’s test.

Results

Reverse osmosis treatment increased Brix and TA with little effect on pH (Table 1). The juice used for this trial began with high levels of YAN, indicating any sluggish fermentation was unlikely due to nutrients.

Fermentation kinetics for all barrels are shown in Figure 1. Table 2 summarizes some elements of fermentation kinetics for these wines. Replicate barrels for each treatment showed very similar fermentation kinetics. For each yeast group, fermentation at warm ambient temperature was faster and warmer than that in cool ambient temperature. Ionys fermentation in a warm environment had similar pace of the CY3079 fermentation in the cool environment, but with a warmer maximum temperature (21°C in Ionys warm vs. 17°C in CY cool). For each temperature regime, despite higher inoculation rates, Ionys fermented slower and cooler than CY3079 at the same ambient temperature. Ionys at cool temperature took the longest to complete primary fermentation, finishing near October 3 (a total of four weeks of fermentation). However, the barrels inoculated with CY3079 and fermented warm were the last to finish malolactic fermentation. They were not treated with SO₂ until January 29, putting them at risk for oxidation and accumulation of volatile acidity.

To investigate potential causes of slow malolactic fermentation in the warm CY3079 barrels, free and total SO₂ were measured at the end of alcoholic fermentation (Table 3). All total SO₂ levels were within a range that is expected to allow malolactic fermentation. Slow malolactic fermentation in these barrels may have been due to high nutrient demand caused by high fermentation temperature leaving little nutrient for malic acid bacteria. The CY3079 barrels fermented at warm temperature also had a higher alcohol than the other barrels (14% compared to 13.4% and below for the other barrels), which may have further inhibited malic acid bacteria. It is unexpected to have a higher level of alcohol in the wine with a warmer fermentation. This may be due to a chaptalization error.

The barrels inoculated with CY3079 and fermented at warm temperature ended malolactic fermentation with notably higher volatile acidity than the other treatments (Table 4), due to the extended period of malolactic fermentation. By contract, the barrels inoculated with Ionys WF and fermented at warm temperature had the lowest volatile acidity at the end of malolactic fermentation. These barrels also finished with the lowest total SO₂ and alcohol, and the highest TA. When compared, each Ionys barrel finished with higher acidity than each CY3079 barrel from the same temperature regime, indicating Ionys does help to retain acidity. For all wine chemistry measures, values were very similar in replicate barrels.

In a triangle test of cool and warm fermented wines inoculated with CY3079, 25 out of 31 respondents were able to distinguish which wine was different, indicating the wines were significantly different (Z=5.84, p< 3x10^-5). There were no significant differences in scores for fruit intensity (F=0.91, p=0.34), floral intensity (F=.002, p=0.96), or minerality (F=2.44, p=0.12). Cool fermentations had a small but significantly lower perception of body (F=4.18, p=0.05) with an average of 5.23 (SD=1.42) vs. 5.93 (SD=1.57) for the warmer fermentation. This may be due to the higher alcohol level in the warmer fermentation. When asked what distinguished the wines, several responses included references to “nutty” or oxidized character in the warmer fermentation wine, likely due to its prolonged malolactic fermentation.

In a triangle test of warm fermented wines inoculated with CY3079 vs Ionys WF, 24 out of 31 respondents were able to distinguish which wine was different, indicating the wines were significantly different (Z=5.02, p< 3x10^-5). There were no significant differences in scores for fruit intensity (F=0.82, p=0.37), floral intensity (F=.01, p=0.93), or minerality (F=1.66, p=0.2). Fermentation with CY3079 had a small but significantly higher perception of body (F=5.26, p=0.03) with an average of 5.98 (SD=1.23) vs. 5.18 (SD=1.43) for the fermentation with Ionys WF yeast. Once again, this may be due to the higher alcohol level in the warmer fermentation. Several respondents also mentioned a perception of acidity in the Ionys fermentation, which may have diminished the perception of body. Several also mentioned a lactic or oxidized character to the CY3079 fermentation.

When a tasting panel of wine producers was presented with all four wines in randomly numbered glasses and asked to sort them into two groups of two, they were most likely to be sorted by temperature (Figure 2), indicating that temperature was more important than yeast group in sensory characteristics of the wine. However, 9 responders sorted in a way that reflected neither yeast group nor temperature. There was no significant difference in preference among the wines (Q=1.81, p=0.61).

Conclusions

• Increasing the ambient temperature allowed Ionys WF fermentations to proceed more quickly, achieving the same fermentation time as CY3079 fermentations at cooler temperatures.
• Wines produced by Ionys WF yeast had higher total acidity and lactic acid at the end of fermentation than those produced with CY3079 yeast.
• The wine produced with CY3079 yeast at warm temperature had a prolonged malolactic fermentation. It is unclear why this fermentation did not progress well. This wine was perceived as nutty, lactic, and oxidized.

 

---

References

(1)        Wood, V.; Custer, S.; Watson, K.; Alper, D. Virginia 2018 Commercial Grape Report. 11.
(2)        Bisson, L. F. Geographic Origin and Diversity of Wine Strains of Saccharomyces. American Journal of Enology and Viticulture 2012, 63 (2), 165–176. https://doi.org/10.5344/ajev.2012.11083.
(3)        Jackson, R. S. Wine Science: Principles and Applications, 4th edition.; Academic Press: Amsterdam, 2014.
(4)        Molina, A. M.; Swiegers, J. H.; Varela, C.; Pretorius, I. S.; Agosin, E. Influence of Wine Fermentation Temperature on the Synthesis of Yeast-Derived Volatile Aroma Compounds. Appl Microbiol Biotechnol 2007, 77 (3), 675–687. https://doi.org/10.1007/s00253-007-1194-3.
(5)        Torija, M. J.; Beltran, G.; Novo, M.; Poblet, M.; Guillamón, J. M.; Mas, A.; Rozès, N. Effects of Fermentation Temperature and Saccharomyces Species on the Cell Fatty Acid Composition and Presence of Volatile Compounds in Wine. Int. J. Food Microbiol. 2003, 85 (1–2), 127–136. https://doi.org/10.1016/s0168-1605(02)00506-8.
(6)        Killian, E.; Ough, C. S. Fermentation Esters — Formation and Retention as Affected by Fermentation Temperature. Am J Enol Vitic. 1979, 30 (4), 301–305.
(7)        Waterhouse, A. L. Volatile Thiols. Waterhouse Lab.
(8)        Tominaga, T.; Baltenweck-Guyot, R.; Gachons, C. P. D.; Dubourdieu, D. Contribution of Volatile Thiols to the Aromas of White Wines Made From Several Vitis Vinifera Grape Varieties. Am J Enol Vitic. 2000, 51 (2), 178–181.
(9)        Capone, D.; Barker, A.; Osidacz Williamson, P.; Francis, I. The Role of Potent Thiols in Chardonnay Wine Aroma: Potent Thiols in Chardonnay Wine. Australian Journal of Grape and Wine Research 2017, 24. https://doi.org/10.1111/ajgw.12294.
(10)      Masneuf-Pomarède, I.; Mansour, C.; Murat, M.-L.; Tominaga, T.; Dubourdieu, D. Influence of Fermentation Temperature on Volatile Thiols Concentrations in Sauvignon Blanc Wines. Int. J. Food Microbiol. 2006, 108 (3), 385–390. https://doi.org/10.1016/j.ijfoodmicro.2006.01.001.
(11)      Sherman, E.; Greenwood, D. R.; Villas-Boâs, S. G.; Heymann, H.; Harbertson, J. F. Impact of Grape Maturity and Ethanol Concentration on Sensory Properties of Washington State Merlot Wines. Am J Enol Vitic. 2017, 68 (3), 344–356. https://doi.org/10.5344/ajev.2017.16076.

Summary

In 2011, the Virginia Wine Marketing Office began promoting Viognier as Virginia’s signature grape. ICV GRE (Scottlabs) yeast is popular for fermentation of Viognier due to its production of an aromatic profile that has become familiar in Virginia Viognier. However, winemakers report difficulty with stuck or sluggish fermentation in GRE inoculated Viognier fermentations. In this experiment, the aromatic profile of Viognier produced with ICV GRE was compared with that produced from the same juice using X-16 yeast (Laffort). Three barrels were inoculated for each yeast group. All three barrels inoculated with GRE yeast stuck with and average of 9.4 gh/L residual sugar while all three barrels inoculated with X-16 yeast finished completion (with residual sugar <0.4 g/L). The GRE barrels also had higher volatile acidity, likely as a result of yeast stress or longer time without SO₂. In a sensory panel, there were no significant differences in scores for fruit intensity, floral intensity or overall aromatic intensity. There was also no significant difference in preference between the yeast groups. Potential reasons for stuck fermentation are discussed.

Introduction

Viognier is a grape variety native to the Rhone region in France¹ that is widely planted in Virginia². This variety is known for aromas of stone fruit with tropical notes of pineapple and orange blossoms. Though the aromatics suggest sweetness, the wine is fermented as a dry table wine. In 2011, varietal Viognier was bottled by 76 of the then 196 wineries in the state³. In May of that that year, the Virginia Wine Board approved marketing of Viognier as Virginia’s signature grape³. As a delicate, aromatic white wine, winemaking choices such as harvest date, fermentation temperature, skin contact and yeast strain can make a significant impact on the resulting wine. In this study, the same Viognier juice was fermented with two different commercially available yeast strains marketed for aromatic white wines: ICV GRE (Scottlabs) and X-16 (Laffort).

Saccharomyces cerevisciae, the primary species of yeast used in commercial wine production, likely originated from a single domestication event in Mesopotamia around the same time as domestication of wine grapes themselves⁴. This line of fermentative yeast spread around the world along with the grape vines it had evolved to transform into wine. Though there have been a few genetic bottlenecks leading to rapid, widespread adoption of a single genetic type, (such as the evolution of sulfite resistance in the Middle Ages) this species remains very diverse, such that no one strain accurately portrays the whole species⁴.

The primary metabolic function of Saccharomyces cerevisciae in wine production is the metabolism of sugar into ethanol with the coupled release of energy. However, complex cellular machinery is also at work to build and maintain cellular components through alternative pathways. It is the work of these pathways that leads to the production of volatile aromas and flavors. At any given time, hundreds of substrates are taken up into the yeast cell, transformed by enzymatic pathways, with end products expelled into the environment or used within the cell. It is differences in the functioning and regulation of these enzymes that produce the myriad of commercially available yeast strains. Though most S. cerevisciae strains share most pathways in common, the extent to which any one pathway functions is determined by both genetic differences and environmental influences.⁵ Different strains have different number of copies or slightly different coding for enzymes in side pathways, leading to large differences in nutrient demand, environmental tolerances, and the amount of any one end product. For example, some strains of S. cerevisciae metabolize malic acid as part of the malo-ethanolic pathway while others have been shown to product malic acid⁵.  

It was Louis Pasteur who first identified S. cerevisciae as the primary fermentative yeast in wine in 1860. Since that time, many thousands of strains have been isolated, selectively bred and cultured for a wide range of characteristics and made available for commercial use. Some are selected for production of thiols (with high copy number of B-lyase genes) while others are tolerant to high levels of alcohol or acidity. Others boast the ability to help with color fixation or to restart stuck fermentations. Regardless of the source of the yeast strain, it is wise to heed this following advice from Ronald Jackson:

“The main characteristics of most commercial strains are know, but most of their other properties are not, or if known they are buried in research papers not readily accessible to most winemakers. The local conditions often are crucial to feature expressions. It is again up to the winemaker to do their own individual experimentation to determine what best suits their situations and preferences.”

In this study, wines made from two commercially available yeast strains were compared: ICV GRE (Scottlabs) and Zymaflore X-16 (Laffort). GRE yeast is a strain of Saccharomyces cerevisiae var. cerevisiae that is frequently used in Virginia for Viognier fermentations. Selected from the Cornas area of the Rhone Valley, this yeast is recommended for white, rose, and red wines. Product literature states that in “fruit-focused whites, such as Chenin Blanc, Riesling and Rhone whites, ICV GRE fermentations result in stable, fresh fruit characteristics such as melon and apricot while improving fore-mouth impact.⁶”  Product literature also specifically mentions this yeast for Riesling and Rousanne, however it is not one of the yeasts selected by Scottlabs for Viognier⁶.

GRE is currently the preferred Viognier yeast at Veritas Vineyards and Winery due to the aromatic profile of the wine it produces. However, the winemaker reports several instances when GRE has not completed fermentation, leaving higher than desired residual sugar (Emily Pelton, personal communication). Stuck and sluggish fermentation also sometimes leads to higher than desired levels of volatile acidity, leading the winery to explore an alternative yeast strain for Viognier.

X-16 (Laffort) is a strain of Saccharomyces recommended for production of modern white and Rosé wines. Product literature states it was bred for the “production of fermentative esters (white peach, yellow fruit) while retaining a sharp, clean aromatic profile” and promises “very high fermentative aroma production”, even from aromatically neutral grape varieties with high vine yield⁷. It is specifically recommended for Chardonnay, Chenin Blanc, Ugni Blanc, and Colombard⁷. Table 1 compares properties and recommendations for these two yeasts.

The purpose of this experiment was to compare fermentation kinetics and aromatic production when each yeast strain was used to ferment the same Viognier juice.

Methods

Grapes were harvested on August 29 and chilled overnight. On August 30, grapes were whole cluster pressed to a single tank with 70ppm SO₂ added at the press pan. The hard pressings were separated from free run and lighter pressings at cycle 20; only free run and lighter pressings were used for this experiment. Cinn Free (1.6 ml/hL) and Stab Micro M (15 g/hL) were added to the tank. Juice was cold settled for two days before racking off lees to a separate tank. Juice was then racked to two separate barrel lots for fermentation. Juice chemistry was determined after cold settling. All racking steps were conducted with care to remain as anaerobic as possible, including use of inert gas.

All winemaking steps were kept the same for both lots with the exception of the yeast strain used for inoculation. Yeast (25 g/hL) was rehydrated in Fermo Plus Energy Glu (6 g/hL). Fermentation was monitored daily using density and temperature. Wine was analyzed at the winery lab on September 26, when fermentation kinetics ceased to show density depletion. This corresponded to completed fermentation (residual sugar <1.0 g/L) for X-16, however GRE had not completed fermentation at that time. Sensory analysis at this stage led to the decision to rack the wine (on October 4). Further attempts to finish the fermentation of the GRE barrels were stopped on October 24 with the addition of 40 ppm SO₂ to all barrels. Three barrels per lot were analyzed for chemistry. A composite sample was used for sensory analysis.

Sensory analysis was completed by a panel of ­­­31 wine producers. Wines were presented blind in randomly numbered glasses. Panelists were presented with two wines, one of each type, and asked to identify which wine they preferred (a triangle test). Due to differences in residual sugar between wines, sensory analysis was done for aromatic traits only. Participants were instructed to smell but not taste the wines. There were three tasting groups with the unique wine in the triangle test balanced between groups. Participants were then asked to score each wine on a scale of 0 to 10 for fruit intensity, floral intensity and overall aromatic intensity. They were also given open ended questions to describe the wines. Results for the triangle test were analyzed using a two-tailed Z test. Descriptive scores were analyzed using repeated measures ANOVA.

Results

Juice for both treatment groups originated from the same settling tank. Juice chemistry is reported in Table 2. YAN was sufficient for fermentation according to the Scottlabs guidelines⁶, so no nutrient additions were made. Fermentation kinetics can be seen in Figure 1. Fermentations progressed along the same trajectory from September 2 through September 8, at which time the GRE fermentation slowed suddenly. After September 8, X-16 fermentation continued to progress to dryness while the GRE fermentation languished. Wine was analyzed at the winery lab on September 26, when fermentation kinetics ceased to show density depletion (Table 3).

Wine chemistry at the completion of density depletion can be seen in Table 3. Wine fermented with GRE had higher density than wine fermented with X-16. Three barrels from each lot were tested individually at that time to determine if all barrels showed the same level of completion. GRE barrels registered RS levels of 9.8, 9.9 and 10.1 g/L while all three X-16 barrels registered RS levels less than 0.1 g/L (ICV labs), indicating this is a consistent effect, not a single stuck barrel.  A composite of the finished wine (after SO₂ addition) was analyzed for each lot (Table 4). The difference in alcohol between the two lots is that which would be expected due to the differences in residual sugar, according to the formula⁸:

Potential Alcohol (%) = (glucose + fructose)(g/L)/16.83

Using this formula, a 0.56% difference in alcohol is predicted between the two wines. The difference is 0.62%.  The GRE has higher volatile acidity, which may be due to yeast stress near the end of fermentation.

In a paired preference test of wines produced by two different yeast strains, 10 participants preferred the wine produced by X-16 yeast while 15 preferred the wine produced by GRE. These preferences were not significantly different (Z=0.8, p= 0.42). There no significant differences in scores between the wines for fruit intensity (F=0.01, p=0.93), floral intensity (F=2.8, p=0.14) or overall aromatic intensity (F=0.36, p=0.55).

Why did GRE Stick?

There are several reasons a fermentation may become sluggish or stick⁹. These include:

• Nitrogen limitation
• Ethanol limitation
• Extremes in temperature
• pH extremes
• Fructose buildup due to yeast glucose preference
• Toxicity factors

In the case of GRE with Viognier at Veritas, all of the stated conditions for fermentation were met. Stress is cumulative, and having multiple parameters close to the limit can also cause fermentation to languish, however none of these parameters was particularly close. Though the initial temperature was colder than the specified range, fermentation began quickly and proceeded at a healthy pace.

When a fermentation proceeds normally and stops abruptly, it is usually indicative of an abrupt shock, such as temperature extreme¹⁰, however that was not the case here. Nutrient limitation can also present as a slow end to fermentation, however, these grapes provided adequate nitrogen. Overall amount of nitrogen may not be the only consideration. Dr. Nichola Hall points out that “focusing on nitrogen alone is not the whole story, it correlates well with fermentation rate and biomass production but not fermentation security, especially in the later stages of fermentation” (personal communication). Rather, Dr. Hall encourages focusing on proper rehydration with nutrient that also provides balanced vitamins and minerals (like Scottlabs Go Ferm Protect Evolution). She also points out that low nutrient requiring yeast don’t perform well in high nitrogen must, and vice versa.

Conclusions

• X-16 (Laffort) completed fermentation with normal kinetics while the fermentation inoculated with GRE became sluggish and eventually stopped with 10 g/L of residual sugar remaining.
• The extended end to fermentation likely led to the higher levels of acetic acid in the GRE fermented wine.
• There was not significant difference in preference between the wines. Scores for fruit intensity, floral intensity and overall aromatic intensity were not significantly different.

---

References

(1)        Viognier. Virginia Wine. https://www.virginiawine.org/varietals/viognier. Accessed 2/19/19

(2)        Wood, V.; Custer, S.; Watson, K.; Alper, D. VIRGINIA 2018 COMMERCIAL GRAPE REPORT. 11.

(3)        Wayne, N. The Best Virginia Viognier Wines. Washingtonian. November 22, 2011.

(4)        Bisson, L. F. Geographic Origin and Diversity of Wine Strains of Saccharomyces. American Journal of Enology and Viticulture 2012, 63 (2), 165–176.

(5)        Jackson, R. S. Wine Science: Principles and Applications, 4th edition.; Academic Press: Amsterdam, 2014.

(6)        Scottlabs. Fermentation Handbook; 2018.

(7)        Laffort. Xymaflore X-16. Laffort Product Information. https://lffort.com/en/product/zymaflore-x16/. Accessed 2/19/19

(8)        ETS. Predicting Potential Alcohol. ETS. https://www.etslabs.com/library/8. Accessed 2/19/19

(9)        Wann, G. Lecture 6A Stuck and Sluggish Fermentations. In Wine Production; 2014.

(10)      Bisson, L. F.; Butzke, C. E. Diagnosis and Rectification of Stuck and Sluggish Fermentations. Am J Enol Vitic. 2000, 51 (2), 168–177.

Summary

Long before the Scottlabs Catalogue provided a library of hundreds of yeast strains with known fermentation parameters conveniently available in freeze dried packaging, ambient yeast were doing the work of wine fermentation. There are many known pros and cons to ambient fermentation. The purpose of this experiment was to compare Chardonnay fermented using CY3079 with that using an ambient starter culture. The resulting chemistry of the two wines was nearly identical. The wines were able to be distinguished in a triangle test, with perception of acidity and freshness cited as a cause for discrimination.

 

Introduction

Long before the Scottlabs Catalogue provided a library of hundreds of yeast strains with known fermentation parameters conveniently available in freeze dried packaging, ambient yeast were doing the work of wine fermentation. Saccharomyces cerevisciae as a species is thought to have been domesticated in ancient Mesopotamia, around the same time vines themselves were domesticated¹. This line of fermentative yeast spread around the world along with the grape vines that provided its substrate for fermentation. It was Louis Pasteur who first named S. cerevisciae as the agent of fermentation in 1860¹. Ever since then, people have been isolating, culturing, selecting and preserving their favorite strains. Though there have been a few genetic bottlenecks leading to rapid, widespread adoption of a single genetic type (such as the evolution of copper resistance when this pesticide was introduced in the vineyard) this species remains very diverse, such that no one strain accurately portrays the whole species¹.  

The rise of molecular genetic techniques has provided much more detailed understanding of the microbes involved in fermentation, and with this information, we now have greater ability to understand the role of microbial diversity in the outcome of alcoholic fermentations. In recent years, studies have shown that non-Saccharomyces yeast play a large role in the beginning stages of fermentation that can be both positive (producing succinic acid and glycerol, liberating amino acids) and negative (producing acetic acid and ethyl acetate)². These yeast also compete with Saccharomyces for nutrients and can delay the onset of fermentation. Studies have also shown a diversity of Saccharomyces strains at work within a single fermentation, even when fermentations have been inoculated. However, this diversity is higher in non-inoculated fermentations^1,3. Overall microbial diversity has been shown to increase both complexity and intensity of aromatics in non-inoculated fermentations^1–3.

Ambient fermentations often start slower and take longer and may have additional nutrient needs due to higher levels of microbial activity. They may also languish near the end of fermentation. Commercial yeast have been bred as “strong fermenters” and winemakers have become accustomed to yeast that can continue to metabolize in the stressful conditions of the end of fermentation. Ambient yeast may or may not be as robust. Ambient fermentation is not recommended for re-starting a stuck fermentation, for use on compromised fruit or “difficult” fermentation conditions such as high Brix, high acid, or other chemistry that may lead yeast to not finish the fermentation. In the end, a balance between diversity and spoilage must be sought.

In practice, many wineries that do not use commercial yeast raise a starter culture that is used for inoculation. The idea is to select for a strain of Saccharomyces that is ambient to the grapes, the vineyard, or the winery. When using a starter culture, a small amount (a bucket or a keg) of crushed grapes or pressed juice is allowed to begin fermenting in a separate vessel. Fermentation is monitored until Brix depletion has reached roughly half and enough alcohol has built up to kill off spoilage organisms. This fermenting culture is then used to inoculate the larger batch. It is important to smell and taste the starter culture prior to addition to the larger lot. If available, a microscope can be used to check for adequate populations of budding Saccharomyces (round) cells. Lemon-shaped (apiculate) cells indicate Klockera, a potential spoilage organism. Addition of SO₂ or chitosan into the starter culture or the larger lot will select against non-Saccharomyces microbes. This lessens the risk of spoilage but also lessens diversity of aromatic compounds that can be produced. 

Winemakers that rely on spontaneous fermentations do so for a number of reasons. Some feel these provide greater aromatic complexity. Some feel this approach offers a better expression of the grape itself or the terroir in which it was grown. Others feel these better reflect yearly variations in character. However, there is also a greater risk of off odors, long lag periods, and spoilage with spontaneous fermentations. The purpose of this experiment was to compare Chardonnay fermented using CY3079 with that using an ambient starter culture.

 

Methods

A starter culture for ambient fermentation was prepared 4-5 days prior to harvest. Clean Chardonnay clusters were picked and crushed into a cleaned and sanitized 6-gallon bucket with a removable lid. A single SO₂ addition of 30 ppm was made. The container of crushed fruit was kept in the vineyard to limit exposure to commercial yeast in the winery and allow the native yeast fermentation to begin. The temperature of the starter was kept near 26°C by shading or sun exposure. The starter was monitored for Brix depletion and temperature daily, twice per day when fermentation began to move briskly. The starter was oxygenated on day 2 and 3. When Brix reached a level between 8 and 12, pomace in the bucket was strained and the fermenting juice was tasted to ensure a clean start to fermentation (no ethyl acetate or volatile acidity), then used for inoculation. 

Fruit was harvested on August 28 then chilled overnight prior to processing. Fruit was lightly crushed and pressed with 50 ppm SO₂ and 30 ml/ton Cinn Free. Juice was cold settled overnight then transferred to barrels of similar type (cooper, age, dimensions). Inoculated barrels received an addition of 15 g/hL CY3079 yeast rehydrated in 20 g/hL Superstart Blanc. The other set of barrels received an equal volume of the starter culture (after stirring). Fermentations were monitored twice daily for Brix depletion and temperature. Barrels were placed in the cellar, in the same ambient temperature environment with fermentation temperatures not exceeding 65°F. Sugar (15 g/L), acid (1 g/L tartaric acid, 0.3 g/L malic acid), and Fermaid O were added at 1/3 Brix depletion. After alcoholic fermentation was complete, each lot was allowed to go through malolactic fermentation. Malic depletion was monitored with paper chromatography. SO₂ (50 ppm) was added after completion of malolactic fermentation. Wine was aged on lees with stirring and SO₂ monitoring.

Sensory analysis was completed by a panel of 28 wine producers. Wines were presented blind in randomly numbered glasses. Tasters were presented with three wines, two of one type and one of another, and asked to identify which wine was different (a triangle test). There were three tasting groups with the unique wine in the triangle test balanced between groups. Tasters were then asked to score each wine on a scale of 0 to 10 for fruit intensity, complexity and Chardonnay varietal character. They were also given open ended questions to describe the wines. Results for the triangle test were analyzed using a one-tailed Z test. Descriptive scores were analyzed using repeated measures ANOVA.

 

Results

Both treatments received juice from the same press. Juice chemistry can be found in Table 1.  There was little difference in final acidity or alcohol production between the wines (Table 2). The ambient fermentation consumed all of the sugar while the inoculated fermentation left 2.2 g/L. Ambient fermentation resulted in less total sulfur, with the same resulting volatile acidity. 

In a triangle test of inoculated and ambient fermented wines, 15 out of 28 respondents were able to distinguish which wine was different, indicating the wines were significantly different (Z=2.07, p= 0.02). There were no significant differences in descriptive scores for fruit intensity (F=0.24, p=0.,63), complexity (F=0.13, p=0.72) or Chardonnay varietal character (F=0.14, p=0.71)(Figure 1). Responses from open ended questions indicate a perceived difference in acidity or freshness between the wines with the wine fermented with CY3079 perceived as fresher.

 

Conclusions

• The basic chemistry of inoculated and ambient fermented wines was very similar.
• Respondents were able to distinguish the wines in a triangle test, with perception of acidity the stated difference for several tasters, despite little difference in measured acidity.

---

References

(1)    Bisson, L. F. Geographic Origin and Diversity of Wine Strains of Saccharomyces. American Journal of Enology and Viticulture 2012, 63 (2), 165–176

(2)    Jackson, R. S. Wine Science: Principles and Applications, 4 edition.; Academic Press: Amsterdam, 2014.

(3)    Egli, C. M.; Edinger, W. D.; Mitrakul, C. M.; Henick‐Kling, T. Dynamics of Indigenous and Inoculated Yeast Populations and Their Effect on the Sensory Character of Riesling and Chardonnay Wines. Journal of Applied Microbiology 1998, 85 (5), 779–789.