UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effect of mixed Saccharomyces strain fermentation on Pinot noir wine Terrell, Emily Elizabeth 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_spring_terrell_emily.pdf [ 1.59MB ]
Metadata
JSON: 24-1.0069142.json
JSON-LD: 24-1.0069142-ld.json
RDF/XML (Pretty): 24-1.0069142-rdf.xml
RDF/JSON: 24-1.0069142-rdf.json
Turtle: 24-1.0069142-turtle.txt
N-Triples: 24-1.0069142-rdf-ntriples.txt
Original Record: 24-1.0069142-source.json
Full Text
24-1.0069142-fulltext.txt
Citation
24-1.0069142.ris

Full Text

EFFECT OF MIXED SACCHAROMYCES STRAIN FERMENTATION ON PINOT NOIR WINE  by EMILY ELIZABETH TERRELL B.Sc., B.A., B.M., University of Washington, 2006  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Food Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  February 2010 © Emily Elizabeth Terrell, 2010  ABSTRACT Pinot noir has traditionally been fermented by native winery flora; however, the recent practice of single yeast strain inoculation may simplify wine aromas and flavours. Thus, a need exists for yeasts that create the complexity of naturally fermented wines and provide the consistency of commercial strains. It has been suggested that mixed Saccharomyces strains may be capable of this dual role. In this study, three novel Burgundian S. cerevisiae strains were characterized for enological equivalency against five industrial strains recommended for Pinot noir. The volatile compounds produced by these strains along with four Burgundian strain mixtures were quantified in Pinot noir wine to evaluate the hypothesis that mixed S. cerevisiae strains contribute to the complexity of naturally fermented wine. The Burgundian strains were enologically equivalent to the industrial strains in terms of killer phenotype, fermentation kinetics, production of ethanol, glycerol, and acetic acid, conversion of sugar to ethanol, ethanol tolerance, foam production, sulfur dioxide production, and compatibility with malolactic fermentation. The concentrations of most of the 25 compounds quantified in the headspace of Pinot noir wines fermented at 22 °C and 27 °C significantly differed among yeast strains and between temperatures. Principal component analysis revealed different patterns of volatile production among the industrial, individual Burgundian and mixed Burgundian yeast strains. Mixed Burgundian strains produced greater amounts of most of the higher alcohols than individual Burgundian strains. Additionally, individual and mixed Burgundian strains produced greater amounts of ethyl esters than most industrial strains, but did not differ ii  from one another. In contrast, the pattern of acetate ester production differed between individual Burgundian and mixed Burgundian strains in a fermentation temperaturedependent manner. Cluster analysis revealed that differences in the patterns of volatile production among industrial, individual Burgundian and mixed Burgundian yeast strains extended over the complete volatile profile. Furthermore, cluster analysis of the averaged profiles showed greater overall similarity between the industrial and the individual Burgundian strains than the mixed Burgundian strains. Fermenting Pinot noir with mixed Burgundian yeast strains resulted in unique patterns of volatile production, which holds promise for mixed Saccharomyces products that yield wines of greater complexity.  iii  TABLE OF CONTENTS ABSTRACT....................................................................................................................ii TABLE OF CONTENTS ...............................................................................................iv LIST OF TABLES .........................................................................................................xi LIST OF FIGURES......................................................................................................xiii ACKNOWLEDGEMENTS........................................................................................... xv DEDICATION ............................................................................................................xvii 1 INTRODUCTION........................................................................................................1 1.1 An introduction to Pinot noir..................................................................................1 1.2 Wine fermentation technology: A brief history ......................................................1 1.3 Desirable technological traits in wine yeast strains.................................................4 1.3.1 Desirable fermentation properties....................................................................4 1.3.2 Additional beneficial technological wine yeast traits .......................................5 1.3.2.1 Killer positive phenotype..........................................................................6 1.3.2.2 Sulfur dioxide tolerance and low sulfur dioxide binding ...........................7 1.4 Desirable qualitative traits in wine yeast strains .....................................................7 1.4.1 High glycerol production is a desirable trait of wine yeast strains....................8 1.4.2 Low acetic acid production is a desirable trait of wine yeast strains...............10 1.4.3 Low acetaldehyde production is a desirable trait of wine yeast strains ...........11 1.4.4 Low production of higher alcohols is desirable in wine yeast strains .............11 iv  1.4.5 Acetate and ethyl ester production are important characteristics of wine yeast strains .........................................................................................13 1.4.6 Minimal production of sulfur compounds is desirable in wine yeast strains ...14 1.5 Desirable yeast-related aroma compounds specific to Pinot noir wine ..................16 1.6 Strategies for wine yeast optimization: Mixed strain fermentations ......................18 1.7 Proposed Research...............................................................................................19 2 MATERIALS AND METHODS ................................................................................20 2.1 Yeast and bacterial strains employed....................................................................21 2.2 Selected ratios of mixed S. cerevisiae strains .......................................................21 2.3 Media and culture conditions ...............................................................................22 2.4 Analytical methods ..............................................................................................23 2.4.1 Quantification of compounds with high pressure liquid chromatography.......23 2.4.2 Identification and quantification of volatile compounds using gas chromatography-mass spectrometry (GC-MS) .......................................23 2.5 Genetic and phenotypic characterization ..............................................................25 2.5.1 Genetic fingerprinting ...................................................................................25 2.5.2 Monitoring mixed strain population dynamics during fermentation ...............26 2.5.3 Killer factor typing........................................................................................27 2.6 Enological characterization ..................................................................................27 2.6.1 Model fermentations .....................................................................................27 2.6.2 Assessment of fermentation kinetics..............................................................28 v  2.6.3 Quantification of ethanol, glycerol, and acetic acid .......................................28 2.6.4 Growth phenotype assay ...............................................................................29 2.6.5 Ethanol tolerance assay .................................................................................29 2.6.6 Foam production assay..................................................................................30 2.6.7 Sulfur dioxide assay ......................................................................................30 2.6.8 Compatibility with malolactic fermentation...................................................30 2.7 Statistical analyses ...............................................................................................31 3 RESULTS ..................................................................................................................33 3.1 Genetic and phenotypic characterization ..............................................................33 3.1.1 Genetic fingerprinting differentiated eight out of nine wine yeast strains.......33 3.1.2 Mixed strain populations did not maintain their inoculated ratios during fermentation......................................................................................34 3.1.3 Burgundian strains A1, A2 and A3 are killer positive ....................................36 3.2 Enological characterization of yeast strains ..........................................................37 3.2.1 Fermentation kinetics were similar for all wine yeast strains .........................37 3.2.2 Ethanol production did not vary among yeast strains .....................................39 3.2.3 Conversion of sugar to ethanol by wine yeast strains .....................................40 3.2.4 Glycerol production varied significantly across wine yeast strain and fermentation temperature.......................................................................41 3.2.5 Acetic acid production significantly varied across wine yeast strain and fermentation temperature.......................................................................43 vi  3.2.6 Growth phenotypes varied among wine yeast strains .....................................45 3.2.7 Ethanol tolerance significantly varied across wine yeast strain and fermentation temperature.......................................................................48 3.2.8 Foam production significantly varied across wine yeast strain and fermentation temperature ..............................................................................49 3.2.9 Sulfur dioxide production significantly varied across wine yeast strain and fermentation temperature.......................................................................51 3.2.10 Compatibility with malolactic bacteria varied among wine yeast strains ......53 3.3 Production of volatile compounds ........................................................................56 3.3.1 Production of higher alcohols varied according to yeast strain and fermentation temperature .............................................................................59 3.3.1.1 Production of 1,3-butanediol ..................................................................64 3.3.1.2 Production of 2,3-butanediol ..................................................................64 3.3.1.3 Production of 2-methyl-1-butanol...........................................................65 3.3.1.4 Production of 3-methyl-1-butanol...........................................................66 3.3.1.5 Production of butanol .............................................................................67 3.3.1.6 Production of 1-hexanol .........................................................................68 3.3.1.7 Production of isobutanol.........................................................................69 3.3.1.8 Production of phenylethanol...................................................................69 3.3.1.9 Production of propanol ...........................................................................70 3.3.2 Production of ethyl esters varied according to yeast strain and fermentation temperature .............................................................................71 vii  3.3.2.1 Production of ethyl butanoate .................................................................76 3.3.2.2 Production of ethyl decanoate.................................................................76 3.3.2.3 Production of ethyl hexanoate ................................................................77 3.3.2.4 Production of ethyl lactate......................................................................78 3.3.2.5 Production of ethyl laurate......................................................................79 3.3.2.6 Production of ethyl octanoate .................................................................80 3.3.2.7 Production of ethyl palmitate..................................................................80 3.3.3 Production of acetate esters varied according to yeast strain selection and fermentation temperature.......................................................................81 3.3.3.1 Production of ethyl acetate .....................................................................85 3.3.3.2 Production of hexyl acetate.....................................................................85 3.3.3.3 Production of isoamyl acetate.................................................................86 3.3.3.4 Production of isobutyl acetate.................................................................87 3.3.3.5 Production of methyl acetate ..................................................................87 3.3.4 Production of aldehydes varied according to yeast strain selection and fermentation temperature.......................................................................88 3.3.4.1 Production of acetaldehyde.....................................................................90 3.3.4.2 Production of benzaldehyde....................................................................91 3.3.5 Production of acetic acid and 1,1-diethoxyacetal varied according to yeast strain selection and fermentation temperature......................................92 3.3.5.1 Production of acetic acid ........................................................................94 viii  3.3.5.2 Production of 1,1-diethoxyacetal ............................................................95 3.4 Principal component analysis revealed different patterns of volatile compound production in wines fermented at 22 °C and 27 °C .............................96 3.4.1 Individual Burgundian and mixed Burgundian yeast strains showed different patterns of volatile alcohol production .......................................... 100 3.4.2 Individual and mixed Burgundian yeast strains produced greater amounts of volatile ethyl esters than most industrial yeast strains during fermentation.................................................................................... 104 3.4.3 Individual Burgundian and mixed Burgundian yeast strains showed different patterns of volatile acetate ester production during fermentation ... 106 3.5 Cluster analysis revealed patterns of similarity and differences in volatile compound production among yeast strains............................................ 108 3.5.1 Differences in volatile compound production between fermentation temperatures outweighed differences between yeast strains in some cases ............................................................................................. 110 3.6 Individual Burgundian strains were more responsive to changes in fermentation temperature than industrial or mixed Burgundian strains based on volatiles produced during fermentation ............................................................. 111 3.7 Cluster analysis of collective industrial, individual Burgundian and mixed Burgundian yeast strains based on averaged volatile compound profiles revealed the relative distance between yeast strain groups .................... 113 4 DISCUSSION .......................................................................................................... 116 4.1 Novel Burgundian strains A1, A2 and A3 are enologically competent compared to industrial yeast strains................................................................... 116  ix  4.2 Mixed Burgundian strains did not collectively differ from pure Burgundian strains over primary enological parameters ....................................................... 121 4.3 Mixed Burgundian strains did not maintain their inoculated strain ratios throughout fermentation.................................................................................... 122 4.4 Pinot noir fermented with industrial, individual Burgundian and mixed Burgundian yeast strains differed in patterns of volatile compound production ....................................................................................... 123 4.4.1 Differences in the pattern of volatile alcohol production between individual Burgundian and mixed Burgundian yeast strains may have sensory implications................................................................................... 123 4.4.2 Differences in the pattern of volatile acetate ester production between individual Burgundian and mixed Burgundian yeast strains vary with temperature................................................................................. 125 4.4.3 Individual and mixed Burgundian yeast strains produced greater amounts of ethyl esters than industrial yeast strains .................................... 127 4.5 Further investigation is needed to correlate volatile compound quantification with sensory experience.............................................................. 128 4.6 Cluster analysis revealed overall differences in volatile compound production among industrial, individual Burgundian and mixed Burgundian yeast strains ................................................................................... 130 4.7 Wine yeast strains differ in their metabolic response to fermentation temperature .................................................................................. 131 5 CONCLUSIONS ...................................................................................................... 133 REFERENCES............................................................................................................ 135 APPENDIX A Complete GC-MS volatile compound dataset for all biological replicates........................................................................... 143 x  LIST OF TABLES Table 1. The mixed Burgundian S. cerevisiae strain ratios used in this study .................21 Table 2. Killer phenotypes of S. cerevisiae strains used in this study .............................37 Table 3. Ethanol concentrations (% v/v) of Pinot noir wine following fermentation at 22 °C and 27 °C with the industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains..................................40 Table 4. Conversion of sugar to ethanol by the individual and mixed S. cerevisiae strains during fermentations in Pinot noir grape must at 22 °C and 27 °C ........41 Table 5. Glycerol production (g/L) by yeast strains in Pinot noir wine ...........................43 Table 6. Acetic acid production (g/L) by yeast strains in Pinot noir wine .......................45 Table 7. Final optical densities (λ=600) observed during aerobic growth of eight individual S. cerevisiae strains in Chardonnay grape must .....................47 Table 8. Final ethanol concentrations of Pinot noir wine produced by individual and mixed Burgundian S. cerevisiae strains fermenting grape must supplemented with equimolar amounts of glucose and fructose ..............49 Table 9. Foam production by eight individual strains of S. cerevisiae during fermentation in Pinot noir must .......................................................................51 Table 10. Sulfur dioxide production (mg/L) by eight individual strains of S. cerevisiae following fermentation in synthetic must ..................................53 Table 11. Malic acid consumption (g/L) and lactic acid production (g/L) by O. oeni during malolactic fermentation in Pinot noir fermented with eight individual yeast strains .........................................................................56 Table 12. Quantifiable volatile compounds identified in Pinot noir wine fermented by eight individual and four mixed Burgundian S. cerevisiae strains ............58 Table 13. Quantifiable volatile compounds identified in Pinot noir wine fermented by eight individual and four mixed Burgundian S. cerevisiae strains (arranged by compound class). ......................................................................59 Table 14. Higher alcohols produced by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C............63  xi  Table 15. Ethyl esters (mg/L) produced by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C ........................................................................................75 Table 16. Acetate esters (mg/L) produced by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C ........................................................................................84 Table 17. Aldehydes (mg/L) produced by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C ........................................................................................90 Table 18. Acetic acid and 1,1-diethoxyacetal (mg/L) produced by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C........................................................................94 Table 19. Details of the vectors visible in Figure 19, Panel A ........................................99 Table 20. Fermentation temperature-mediated changes in volatiles produced by each industrial, individual Burgundian and mixed Burgundian yeast strain .................................................................................................. 112 Table 21. Sensory effects of ethyl esters quantified in this study.................................. 127 Table 22. Concentrations (mg/L) of all volatile compounds quantified in Pinot noir wines (n=3) fermented at 22 °C by industrial yeast strains AMH, AWRI 796, BGY, RA17 and RC212; individual Burgundian strains A1, A2 and A3 and mixed Burgundian strains M1, M2, M3 and M4.................. 144 Table 23. Concentrations (mg/L) of all volatile compounds quantified in Pinot noir wines (n=3) fermented at 27 °C by by industrial yeast strains AMH, AWRI 796, BGY, RA17 and RC212; individual Burgundian strains A1, A2 and A3 and mixed Burgundian strains M1, M2, M3 and M4.................. 149  xii  LIST OF FIGURES Figure 1. Genetic fingerprints of nine S. cerevisiae strains .............................................33 Figure 2. Mixed Burgundian strain population dynamics during fermentation in Pinot noir must..........................................................................................35 Figure 3. Killer phenotype assay of novel Burgundian strains A1, A2 and A3 ...............36 Figure 4. Fermentation kinetics of individual and mixed S. cerevisiae strains in Pinot noir must..........................................................................................38 Figure 5. Ethanol concentrations of Pinot noir wine fermented at 22 °C and 27 °C with eight pure and four mixed S. cerevisiae strains (n=3)..............................39 Figure 6. Glycerol production by S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C (n=3). ..............................................................................42 Figure 7. Acetic acid production by S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C (n=3). ..............................................................................44 Figure 8. Growth phenotypes of the eight pure strains of S. cerevisiae (n=3) grown aerobically in 150 µL of Calona Chardonnay grape must.....................46 Figure 9. Ethanol production by eight pure and four mixed strains of S. cerevisiae following fermentation in Pinot noir must supplemented to 33% (w/v) sugars with equimolar amounts of glucose and fructose ...........48 Figure 10. Foam production by eight individual strains of S. cerevisiae during fermentation in 10 mL of Pinot noir must (n=3) ...............................50 Figure 11. Sulfur dioxide production (mg/L) by eight individual S. cerevisiae strains in synthetic wine fermented at 22 °C and 27 °C (n=3) .......................52 Figure 12. Consumption of malic acid (g/L) and production of lactic acid (g/L) during malolactic fermentation in Pinot noir following alcoholic fermentation by eight individual S. cerevisiae strains (n=3). .........................54 Figure 13. A representative chromatogram obtained by GC-MS headspace analysis of Pinot noir wine ..........................................................................57 Figure 14. Concentrations (mg/L) of higher alcohols produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains.................................60  xiii  Figure 15. Concentrations (mg/L) of ethyl esters produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains.................................72 Figure 16. Concentrations (mg/L) of acetate esters produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains.................................82 Figure 17. Concentrations (mg/L) of aldehydes produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains.................................89 Figure 18. Concentrations (mg/L) of acetic acid and 1,1-diethoxyacetal produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains...92 Figure 19. PCA plots of the complete volatile compound profiles of wines fermented at 22 °C and 27 °C ......................................................................97 Figure 20. PCA plots of the volatile alcohol content of wine samples fermented at 22 °C .................................................................................... 101 Figure 21. PCA plots of the volatile alcohol content of wine samples fermented at 27 °C .................................................................................... 103 Figure 22. PCA plots of the volatile ethyl esters in wine samples fermented at 22 °C and 27 °C .................................................................................... 105 Figure 23. PCA plots of the volatile acetate esters in wine samples fermented at 22 °C and 27 °C .................................................................................... 107 Figure 24. Cluster analysis of yeast strains based on volatile compound production during fermentation in Pinot noir at 22 °C and 27 °C ............... 109 Figure 25. Cluster analysis of yeast strains based on volatile compound production during fermentation in Pinot noir at 22 °C and 27 °C ............... 110 Figure 26. Cluster analysis of industrial, individual Burgundian and mixed Burgundian yeast strains based on volatile compound production during fermentation in Pinot noir at 22 °C and 27 °C ................................ 113 Figure 27. Cluster analysis of industrial, Burgundian and mixed Burgundian yeast strains fermented at 22 °C and 27 °C based on averaged volatile compound profiles. ....................................................................... 115  xiv  ACKNOWLEDGEMENTS I would like to acknowledge and express my deepest gratitude to the individuals and organizations that assisted me in bringing this research to fruition, whether through academic supervision, donation of research materials, financing, or personal support. First, I would like to extend great thanks to my supervisor, Dr. Hennie J.J. van Vuuren for the countless hours he has spent guiding me in my research, editing my thesis, and sharing his knowledge of wine. I am much indebted to the members of my supervisory committee, Dr. Margaret Cliff and Dr. David Kitts. Dr. Cliff was extremely generous with her time and provided me with seemingly infinite support, advice and encouragement as I immersed myself in the statistical analyses of my volatile compound data. Dr. Kitts has also shown me a tremendous amount of support and was one the first people to make me feel welcome when I arrived at UBC. Thank you also to my defense committee chair, Dr. Eunice LiChan and my external examiner Dr. John Smit for generously offering their time. I would like to acknowledge Dr. David McArthur for improving my wine education by extending opportunities to me, including my wine science teaching assistantship and many invitations to private trade tastings and events. I would also like to extend my sincere thanks to Sandhill winemaker Howard Soon for the generous donation of hundreds of liters of grape must for this and other research projects. Great wine can only come from great grapes, and the same can undoubtedly be said about wine research. Thank you. A special thank you is also in order to the organizations and individuals who providing my funding during this project, including Dr. Hennie van Vuuren, the xv  University of British Columbia, the American Society for Enology and Viticulture, the American Wine Society and the Washington Association of Wine Grape Growers. Thank you to all of my colleagues at the Wine Research Centre, and especially to Calvin Adams, Dr. Chris Walkey, Dr. Zongli Luo, Lina Madilao and Mike Anderson, who have made my time in the lab so much more fruitful and enjoyable than it might otherwise have been. Thank you also to Rick Dorfer for his support and encouragement during my graduate studies. Finally, thank you to my parents, Doug and Joan Terrell for their constant support and encouragement. They have given me the skills and desire to endlessly pursue and achieve goals of all kinds. Thank you.  xvi  DEDICATION To the growers, makers, drinkers, and lovers of the Heartbreak Grape and its elusive child, Pinot noir wine.  xvii  1 INTRODUCTION 1.1 An introduction to Pinot noir Pinot noir, or red Burgundy, has long been touted as one of the most complex and revered red wine varietals in the world by consumers and winemakers alike. Winemakers enjoy the challenges presented by the delicate, thin-skinned Pinot noir Vitis vinifera grape, while connoisseurs appreciate the breadth and depth of great Burgundian red wines. Characterized as a medium-bodied red wine with red fruits, earthiness, and a pleasant acidity, Pinot noir remains one of the great favourites of both the old and the new world, where it is increasingly making a name for itself in cool climate regions such as British Columbia, Oregon, coastal California, and New Zealand. As Pinot noir’s popularity grows, there is an increasing demand for fermentation products designed to promote the development of varietal-specific aroma and flavour compounds. There is also an interest in exploring traditional fermentation practices in an effort to recapture the complexity that some wine aficionados believe has deteriorated following the widespread use of commercial yeast starter cultures (Hennie van Vuuren, 2008, personal communication). Recent advances in wine research and biotechnology may provide the best of both worlds: Yeast products that perform consistently yet produce premium Pinot noirs redolent of the finest old world techniques.  1.2 Wine fermentation technology: A brief history The inadvertent application of the wine yeast Saccharomyces cerevisiae by early Greek, Roman, and Egyptian winemakers represents one of the first applications of  1  biotechnology and the beginning of a millennia-long winemaking tradition. The method of allowing grapes to ferment without the deliberate addition of microorganisms has persisted for thousands of years and has contributed to the expression of terroir, or the regional specificity that can be evident in a wine’s flavour and aroma. This laissez-faire approach to fermentation is known as spontaneous or natural fermentation, and is still practiced today, particularly by the premium red wine producers of Burgundy. Although some of the finest wines in the world are a result of spontaneous fermentation, this method is not without its drawbacks. Because the winemaker is relying on the natural flora of the fruit and winery, results are often inconsistent and unpredictable, which can result in wines that are poor or exceptional. While France and the rest of Europe battled the Phylloxera epidemic in the mid19th century with grafting, the relatively new wine industry in California flourished. In 1880, California established a State Board of Viticulture Commissioners, which provided funding and established the current wine research department at the University of California Davis. Despite these early successes, prohibition soon followed and nearly eradicated the California wine industry. Although all research on alcoholic beverages was technically halted, interest in establishing a university yeast strain collection remained, and gave rise to “Burgundy” and “Champagne” S. cerevisiae strains, as well as the ubiquitous Montrachet strain (Kunkee and Cooke 1980). When prohibition ended, no tradition of spontaneous fermentation existed in California, so winemakers readily accepted the university’s recommendation to inoculate using pure starter cultures from the strain collection, which solidified Montrachet as a primary enological strain (Kunkee and Cooke 1980). The trend of pure strain fermentation inoculation continued, especially  2  after the development of commercial yeast cakes (Thoukis et al. 1963), and, by the mid1960s, active dry wine yeast. The consistent performance, ready-to-use convenience and long shelf life of active dry yeast products led to their rapid adoption by the new world, and their tentative acceptance in some parts of the old world (Reed and Nagodawithana 1988). Although active dry wine yeast was initially limited to a handful of strains, strain selection and commercialization have grown dramatically over the last few decades and addressed some early limitations, such as Montrachet’s tendency to produce sulfur compounds. Today, yeast companies abound in nearly every major winemaking region of the world and offer over 100 unique wine yeast strains. This increasing market has provided the impetus for the improvement of wine strains by elucidating which yeast phenotypes, and recently, genotypes are related to the production of aromatically superior wine. This search for premium fermentation products has led in two primary directions; some scientists have embraced modern biotechnology by developing genetically modified yeast strains (Husnik et al. 2007), while others have tried to capture the beneficial aspects of traditional spontaneous fermentation by pursuing mixed strain and mixed species yeast products (Clemente-Jimenez et al. 2005, Rojas et al. 2003, Swiegers et al. 2007). In either case, novel yeast products must not only possess unique genetic traits or provide exceptional complexity, but must also meet the wine yeast phenotypic criteria that have been described during the past 50 years of wine research. These enological traits have been grouped into two classes in the literature; technological traits that influence the efficiency of the fermentation process and qualitative traits that affect the chemical composition and the sensory profile of the finished wine (Zambonelli 1998).  3  1.3 Desirable technological traits in wine yeast strains As in all industries, the timely production of a consistently high quality product is of critical importance in the wine industry. The choice of wine yeast can either realize the full potential of the grapes or cause endless problems with stuck fermentations, demanding nutrient requirements and difficult fermentation behavior. Most S. cerevisiae strains isolated from wine environments tend to possess the required technological traits for a successful fermentation (Ranieri and Pretorius 2000). However, because this is not universally true, it is still necessary to confirm strain phenotypes when selecting potential new wine yeasts (Ranieri and Pretorius 2000).  1.3.1 Desirable fermentation properties The fermentation properties of wine yeast strains represent one subset of technological traits. Desirable fermentation properties include rapid initiation of fermentation, low nitrogen requirements, high fermentation efficiency, high osmotic stress tolerance, growth at high and low temperatures, moderate biomass formation and high ethanol tolerance (Pretorius 2000). In general, these traits promote reliable fermentation in grape musts of varying composition. Although often taken for granted, these traits have recently become increasingly important as warmer growing seasons have increased the rate of sugar accumulation in grape berries but not the rate of physiological ripening in warm viticultural areas (Jones et al. 2005). If forced to choose between ideal sugar content and physiological ripeness (a combination of phenolic ripeness, colour content, and flavour profile), winemakers will generally harvest based on physiological maturity to avoid risking low colour extraction and harsh green astringency. In such  4  years, the sugar content of the grape must can increase significantly, sometimes hovering above 27° Brix, which was recently the case at a winery in British Columbia’s Okanagan Valley (Stephanie Leinemann, 2008, personal communication). Wine yeasts must possess exceptionally high levels of osmotic and ethanol tolerance in order to ferment such grape musts to dryness, making these fermentation properties critical indicators of a wine yeast strain’s success in the changing world of winemaking.  1.3.2 Additional beneficial technological wine yeast traits Additional technological traits beyond fermentation properties are also desirable in wine yeast strains. These include genetic stability, capacity for genetic marking, killer phenotype, low foam production, flocculation, high sulfur dioxide tolerance, low sulfur dioxide binding, compact sediment formation, resistance to desiccation and proteolytic activity (Pretorius 2000). Some of these traits are useful from an active dry yeast production perspective such as resistance to desiccation, which can increase product viability and genetic stability and markers, which can make strain production more reliable and allow straightforward strain identification. Other traits are more relevant to winemaking. For example, low foam production is of critical importance in the winery due to the limited availability of stainless steel tank space during vintage, while flocculation (the tendency of yeast cells to clump together) and compact sediment formation make it easier to remove yeast sediment without filtration, which can modify the chemical composition of wine (Arriagada-Carrazanaa et al. 2005). Similarly, proteolytic activity could allow yeast strains to metabolize precipitated proteins that  5  cause wine haze and gummy polysaccharides, reducing the need for fining and filtration (Lagace and Bisson 1990).  1.3.2.1 Killer positive phenotype. The killer phenotype can be of particular importance to wine fermentation (Pérez et al. 2001). S. cerevisiae wine strains can be classified as killer positive (K), neutral (N), or sensitive (S) strains, based on their production of, or susceptibility to, the K2 killer toxin, which is encoded as a precursor protein in the double stranded RNA genome of a persisting cytoplasmic yeast virus (Tipper and Schmitt 1991) and is active in the pH range of wine (Shimizu et al. 1985). The K2 killer toxin is secreted following posttranslational modification; it kills sensitive S. cerevisiae cells by binding to a cell wall receptor and acting as a cation-specific ionophore, thereby disrupting plasma membrane function and eventually causing the leakage of small molecules from the cell (Magliani et al. 1997). This is concerning in the context of winemaking, because if as little as 0.1% of the yeast inoculum is killer positive and the remainder sensitive, the killer positive fraction may kill the sensitive strain and dominate the fermentation, which can result in problematic stuck or prolonged fermentations (Jacobs and van Vuuren 1991). The killer phenotypes of native S. cerevisiae wine strains differs by region, with some areas containing only killer positive strains and other only sensitive strains (Gutiérrez et al. 2001, Kapsopoulou et al. 2008). Nonetheless, in a killer positive region, grape must could contain enough native killer positive S. cerevisiae cells to influence a fermentation inoculated with a sensitive strain (Jacobs and van Vuuren 1991). Therefore, a killer positive phenotype is seen as an important technological trait in wine strains of S. cerevisiae.  6  1.3.2.2 Sulfur dioxide tolerance and low sulfur dioxide binding. Sulfur dioxide tolerance and low sulfur dioxide binding are also important attributes of wine yeast strains of S. cerevisiae. The protective practice of burning sulfur candles in empty wine vessels may have existed as early as Egyptian and Roman times (Hammond and Carr 1976). Current practices utilize sulfur dioxide as an antimicrobial and antioxidant compound; it is considered critical to the production of ageable wines (Ough 1992). Wine yeast strains can often tolerate around 250 mg/L total sulfur dioxide (Regadon et al.1997) which is an important characteristic since sulfur dioxide is used in various capacities from the grape harvest through to bottle stabilization. Because sulfur dioxide is only active in its unbound form, it is preferable to select wine yeast strains that do not tend to bind large amounts of sulfur dioxide, leaving it free to inhibit bacterial and nonSaccharomyces yeast species, as well as quench precursors and products of oxidation (Pretorius 2000).  1.4 Desirable qualitative traits in wine yeast strains Qualitative wine yeast traits directly affect the aroma or flavour of the finished wine. This encompasses the synthesis and liberation of a number of compounds, including glycerol, acetic acid, higher alcohols, esters and sulfur compounds (Lambrechts and Pretorius 2000, Pretorius 2000). These compounds are currently understood to varying degrees in terms of their synthetic pathways, desirable concentrations, sensory impact and influence on other volatiles. Nonetheless, the link between chemistry and sensory studies is growing (Ebeler and Thorngate 2009), which will likely make the chemical study of wine more relevant.  7  1.4.1 High glycerol production is a desirable trait of wine yeast strains High levels of glycerol production are acknowledged across the literature as being beneficial to wine yeast strains (Eustace and Thornton 1987, Pretorius 2000, Remize et al. 1999). However, the relationship that glycerol has to wine quality is not well defined. Glycerol is the third major fermentation end product after ethanol and carbon dioxide and is usually described as being important to the perceived sweetness, body, and mouthfeel of wine. However, these sensory claims have not been confirmed in the literature, particularly for red wine. Glycerol in wine typically ranges from 1-10 g/L (Ough et al. 1972), although its concentration has been found to vary according to geographic region, fermentation practices, and must composition (Scanes et al. 1998). Taste thresholds, described as a perceived sweetness, were found to be 5.2 g/L in a white table wine (Noble and Bursick 1984) and 13 g/L in a red table wine (Hinreiner et al. 1955). Based on these values, it is possible that glycerol contributes to the perceived sweetness of some white wines, but unlikely that it plays a role in the flavour of most red wines, which typically contain less than 13 g/L of glycerol. Rankine and Bridson (1971) first suggested that glycerol could play a role in the viscosity and mouthfeel of wine based on its oily, heavy character. However, the viscosity threshold value of glycerol in wine was later determined to be 25.8 g/L (Noble and Bursick 1984), well above the usual glycerol content of table wines (Ough et al. 1972). It is possible that glycerol is at least partially responsible for the viscosity of icewines and botrytized wines, where glycerol content is elevated and can approach the threshold value of 25 g/L (Nurgel et al. 2004).  8  Although glycerol may not always provide the sweetness and body that is often attributed to it in the literature, its production in yeast cells increases osmotolerance and shunts carbon away from ethanol production (Remize et al. 1999). Both of these attributes are looked upon favourably in the context of the challenges of modern winemaking. This fact, in addition to the numerous industrial applications of glycerol, has spurred an interest in the genetic modification of S. cerevisiae to overproduce glycerol during fermentation (Pretorius 2000, Remize et al. 1999, Taherzadeh et al.  2002). Although Remize et al. (1999) successfully upregulated glycerol production in industrial and laboratory strains of S. cerevisiae, additional metabolic changes occurred that made the resulting strain enologically unsuitable, such as an increase in acetate production. Eglinton et al. (2002) were able to block the overproduction of acetate of a glycerol overproducting yeast strain; however, many differences still existed in the production of other aroma- and flavour-active compounds that have yet to be examined with sensory studies. The alternate approach of breeding increased glycerol production into yeast strains (Eustace and Thornton 1987) has yielded somewhat more optimistic results in wine (Prior et al. 2000). Although the Chardonnay wine that showed improved body due to high levels of glycerol had poorer aroma and overall quality than the control, Prior et al. (2000) concluded that further breeding and selection could yield a strain with high glycerol production and a negligible impact on the overall balance of the wine. Of course, the genetic diversity found within the wild S. cerevisiae wine strain population could also yield such a phenotype.  9  1.4.2 Low acetic acid production is a desirable trait of wine yeast strains Acetic acid and the other volatile acids of wine collectively comprise a wine’s volatile acidity. When volatile acidity levels rise above the sensory threshold, the wine is considered faulty and often smells distinctly of vinegar. Acetic acid is the most prevalent and most important compound in volatile acidity composition, and is often individually assayed in the literature to determine volatile acidity effects. Excessive acetic acid in wine is most often linked to bacterial infection caused by poor fermentation practices. However, S. cerevisiae strains also synthesize acetic acid, resulting in a concentration of 0.2–0.3 g/L in most finished wines (Ribéreau-Gayon et al. 2000a). Depending on the style of wine, acetic acid levels ranging from 0.4–1.5 g/L are considered undesirable in wine (Davis et al. 1985), and many countries have imposed legal upper limits. Although the sensory threshold of acetic acid (> 0.7 g/L) is above the amounts typically produced by S. cerevisiae (Zoecklein et al. 1995), acetic acid combines with ethanol in aerobic conditions to yield ethyl acetate, an acetate ester with a low sensory threshold that is largely responsible for the aroma of wine oxidation. The level of acetic acid production by S. cerevisiae is related to the genetic background of the yeast as well as to the fermentation environment, where stresses such as excessive must clarification can contribute to increased production. However, as with several compounds in wine, acetic acid in small quantities has been demonstrated to contribute to wine’s complexity (Zoecklein et al. 1995). Therefore, wine yeast strains should ideally produce low, consistent levels of acetic acid with minimal fluctuation due to fermentation conditions.  10  1.4.3 Low acetaldehyde production is a desirable trait of wine yeast strains Acetaldehyde is the immediate precursor to ethanol synthesis during the fermentative metabolism of S. cerevisiae. As the biosynthetic needs of the cell change during growth and fermentation, NADH can be directed away from the reduction of acetaldehyde to ethanol. Acetaldehyde accumulates in the yeast and is eventually released into the fermenting must, which can result in a sour green apple aroma in the wine if the concentration of acetaldehyde exceeds its sensory threshold of 100 mg/L. Despite this accumulation, most of the acetaldehyde is transported back into the cell and transformed into ethanol prior to the wine reaching dryness (Lambrechts and Pretorius 2000). However, the acetaldehyde concentration of the finished wine has been found to vary considerably based on yeast strain selection (Fleet and Heard 1993). Because of the potential for a negative effect on wine aroma, in addition to the sensory association of acetaldehyde with wine oxidative spoilage, low acetaldehyde production is preferred among wine yeast strains.  1.4.4 Low production of higher alcohols is desirable in wine yeast strains The higher alcohols of wine are defined as those having more than two carbon atoms, and therefore a higher molecular weight and boiling point than ethanol. Although the metabolic function of these molecules is not well understood in S. cerevisiae (Boulton et al. 1995), they are important yeast-derived contributors to wine flavour and aroma. The most wine-relevant higher alcohols include propanol, butanol, isobutanol, active amyl alcohol (2-methyl-1-butanol), isoamyl alcohol (3-methyl-1-butanol), hexanol, tyrosol, tryptophol and phenylethyl alcohol (Lambrechts and Pretorius 2000). Typical  11  concentrations in wine range from 100 mg/L to greater than 500 mg/L (Nykänen 1986). At concentrations above 400 mg/L higher alcohols are thought to negatively influence wine quality due to the harsh aroma they impart (Rapp and Mandery 1986). However, at concentrations below 300 mg/L, higher alcohols can contribute positively to the fruity complexity of wine (Lambrechts and Pretorius 2000), making low levels of production a desirable characteristic of wine yeast strains. Higher alcohols are also important precursors of some aroma-active esters (Soles et al. 1982); thus, some level of higher alcohol production during fermentation may be critical to the development of esterderived wine aromas. Despite the relative abundance of higher alcohols compared to other compound classes, their direct effect on wine flavour and aroma is relatively small (Webb and Keppner 1961). Taste threshold studies showed that while quantitative differences of isoamyl alcohol in wine were discernable to some tasters, differences in 1-propanol and isobutyl alcohol were not discernable (Rankine 1967). However, as Lambrechts and Pretorius (2000) suggest, this does not take into account potential synergistic effects between compounds. Higher alcohols can be synthesized in yeast de novo from a sugar substrate (Äyräpää 1968), or can be produced through the Ehrlich pathway via the degradation of the branched-chain amino acids leucine, isoleucine, and valine. Several studies have recently focused on the genetic modification of higher alcohol synthetic pathways with varying success as related to wine (reviewed by Swiegers et al. 2005). However, production of higher alcohols is known to vary based on the genetic background of wine yeast strains (reviewed in Lambrechts and Pretorius 2000), yeast nutrient availability (Hernández-Orte et al. 2002), fermentation temperature (Ough and Amerine 1967), and  12  other winemaking parameters (Fleet and Heard 1993) suggesting that higher alcohol production can be controlled using current winemaking technology.  1.4.5 Acetate and ethyl ester production are important characteristics of wine yeast strains Esters are one of the largest volatile compound classes found in wine and one of the most influential in terms of wine flavour and aroma, despite their often minute quantities (Saerens et al. 2009). Many esters arise directly from S. cerevisiae metabolism; however, some are liberated from the grape during the fermentation process, resulting in grape varietal-specific ester profiles (Swiegers et al. 2005). There are two main classes of esters present in wine: The acetate esters, which are formed from higher alcohol precursors, and the ethyl esters, which are formed from medium-chain fatty acid precursors. The acetate esters have been represented in the literature as the more important class of wine esters (Thurston et al. 1981); however, Saerens et al. (2009) claims that this has been due to their relative abundance in wine and subsequent popularity as a research topic rather than their activity as aromatic compounds. Major acetate esters include ethyl acetate (fruity, solvent-like aroma), isoamyl acetate (banana aroma), isobutyl acetate (fruity aroma) and phenyl ethyl acetate (roses, honey aroma) (Saerens et al. 2009), while major ethyl esters include ethyl butanoate (ethereal/fruity pineapple/banana aroma), ethyl hexanoate (fruity pineapple/banana aroma), ethyl octanoate (soapy/candlewax aroma), ethyl decanoate (oily/fruity/floral aroma) and ethyl dodecanoate (oily/fruity/floral aroma) (Clarke and Bakker 2004, Lambrechts and Pretorius 2000). The role that acetate and ethyl esters play in wine aroma composition is  13  dependent on the type of wine produced, despite the fact that many esters are present at levels above their detection thresholds. Ferreira et al. (1995) found that acetate and ethyl esters created tree fruit and tropical fruit notes in white wines, but were not responsible for the intensity of fruit aromas present in red wines and only played a modulatory role in red wine aroma quality. Ester formation and liberation by S. cerevisiae during fermentation depends on a number of physiological and environmental factors. These include the yeast strain used, the fermentation temperature, the vinification methods, the presence of insoluble materials, the skin contact time, the cultivar used, the must pH and the application of sulfur dioxide (Lambrechts and Pretorius 2000). While these factors encompass the immediate production of esters, the resulting finished wine aroma is also dependent on age-mediated ester profile changes, as well as ester distribution between the yeast cells and the wine (Marais and Pool 1980). Because differences in ester production exist between yeast strains, research has begun with the dual goal of understanding the genetic control of ester production and creating ester-optimized yeast strains for wine production (reviewed in Swiegers et al. 2005). However, because the link between the complete ester profiles of wine and the corresponding sensory attributes has not yet been made, such research currently remains in the early stages of development.  1.4.6 Minimal production of sulfur compounds is desirable in wine yeast strains The production of sulfur compounds by S. cerevisiae during fermentation is generally considered to be detrimental to wine quality. This is especially true for the  14  production of hydrogen sulfide, a common problem in wine, which has a very low detection threshold of 50-80 µg/L and imparts an aroma of rotten eggs (Monk 1986). Major sulfur-derived compound classes present in wine include thiols, sulfides, polysulfides, thioesters, mercaptans and heterocyclic compounds (Mestres et al. 2000). The sensory descriptors of most of these compounds are negative, and include onion, garlic, rubber, cabbage and sulfurous (Mestres et al. 2000). However, three thiols have recently been identified as critical to the characteristic aromas of Sauvignon Blanc, including 4-mercapto-4-methylpenan-2-one (4MMP, cat urine, box tree/blackcurrant, broom), 3-mercaptohexan-1-ol (3MH, passionfruit, grapefruit) and 3-mercaptohexyl acetate (3MHA, Riesling-type note, passionfruit, box tree) (Dubourdieu et al. 2006). Rather than originating as sulfur-containing amino acids or sulfur-containing pesticides, these beneficial thiols are liberated from non-volatile cysteine-bound conjugates present in the grape during fermentation (Darriet et al. 1995). Although known to be particularly important to Sauvignon Blanc, these thiols have also been identified in wines made from other grape varietals, including Riesling, Semillon, Merlot and Cabernet Sauvignon (Tominaga et al. 2000), which could suggest a more universal role in wine aroma. Grape must is often deficient in the sulfur-containing amino acids cysteine and methionine. In order to accommodate this, S. cerevisiae responds by synthesizing these compounds from grape-derived inorganic sulfur sources (Park et al. 2000). This conversion begins with the sulfate reduction pathway, which produces hydrogen sulfide which then combines with O-acetyl serine and O-acetyl homoserine to form cysteine and methionine. However, in the absence of adequate nitrogen, hydrogen sulfide accumulates and can diffuse into the fermenting must, resulting in the characteristic rotten egg odour  15  (Giudici and Kunkee 1994). As a highly reactive species, hydrogen sulfide can also act in the cell to produce other noxious sulfur compounds, such as mercaptans, which can lead to the production of polysulfides, increasing the overall negative contribution of sulfur compounds to wine aroma (Vermeulen et al. 2005). Because the formation of sulfur compounds is closely linked to S. cerevisiae metabolism, it is also affected by yeast strain. Mendes-Ferreira et al. (2002) recently found that commercial wine yeast strains varied in sulfate reductase activity, and was able to classify strains as “non-producers,” “must-composition-dependent producers,” and “invariable producers” of hydrogen sulfide. This has fuelled an interest in characterizing and modifying the genetics of sulfur metabolism of wine yeast strains to minimize hydrogen sulfide production (reviewed in Swiegers and Pretorius 2007). Although good fermentation practices are often adequate in limiting production of hydrogen sulfide and other undesirable sulfur compounds, production of sulfur compounds is still a concern and should be considered when selecting novel wine yeast strains for industrial production.  1.5 Desirable yeast-related aroma compounds specific to Pinot noir wine Despite the popularity of Pinot noir, the literature has not been particularly successful in elucidating the key aromatic compounds responsible for its varietal profile. Several studies have used chemical and sensory analysis to profile Pinot noir wines (Fang and Qian 2005, Girard et al. 1997, Girard et al. 2001, Guinard and Cliff 1987, MirandaLopez et al. 1992, Moio and Etievant 1995, Schreier et al. 1980); however, only a few  16  have made conclusions about the importance of individual aromatic constituents in this varietal. In 1995, Moio and Etievant identified four important odorants in Pinot noir wine using gas chromatography/olfactometry on a volatile fraction obtained by adsorption chromatography of the total organic extracts of 13 wines from various appellations of Burgundy. The identified compounds were ethyl anthranilate (ethyl 2-aminobenzoate), ethyl cinnamate (ethyl 3-phenyl-2-propenoate), ethyl 2,3-dihydrocinnamate (ethyl 3-phenylpropanoate) and methyl anthranilate (methyl 2-aminobenzoate). These compounds corresponded to sweet fruity aromas (ethyl anthranilate), sweet cherry plum cinnamic aromas (ethyl cinnamate), fruity balsamic aromas (ethyl 2,3-dihydrocinnamate) and fruity grape aromas, respectively. The detection thresholds of these compounds are recognized as being very low; Etievant et al. (1983) previously reported the odour threshold of ethyl cinnamate to be 0.04 mg/L in Muscat wine. An additional study by Aubry et al. (1997) quantified these compounds in 33 Pinot noir wines. Ethyl anthranilate was found to be present at an average concentration of 2.4 µg/L, ethyl cinnamate at 0.8 µg/L, ethyl 2,3-dihydrocinnamate at 1.6 µg/L and methyl anthranilate at 0.2 µg/L. These values did not exceed the known thresholds of ethyl cinnamate and methyl anthranilate in water, 16 µg/L and 3 µg/L respectively (Etievant et al. 1983, Hirvi and Honkanen 1982); however Aubry et al. (1997) suggested that the thresholds may be much lower in wine. Aside from these esters, the compounds ethyl and methyl vanillate, acetovanillone, 3-methylthio-1-propanol, 2-phenylethanol, benzyl alcohol and  17  3-methylbutanoic, hexanoic, octanoic, and decanoic acids have been identified in the literature as potentially important in Pinot noir (Fang and Qian 2005). Recent work by Fang and Qian (2005) identified 2-phenylethanol and 3-methyl-1-butanol as important contributors to Pinot noir aroma based on aroma extract dilution analysis. However, their conclusion and the conclusion expressed for many types of wine in the literature was that the characteristic aroma of Pinot noir is most likely due to the interactions of numerous compounds synthesized or liberated during fermentation.  1.6 Strategies for wine yeast optimization: Mixed strain fermentations The primary strategies for the improvement of wine yeast fermentation products include the isolation and selection of novel wine yeast strains, the genetic engineering of wine yeast strains, and, increasingly, the investigation and characterization of mixed strain and mixed species inocula in an attempt to capture the benefits of spontaneous fermentation in a controlled environment. Much work has recently focused on the contribution of non-Saccharomyces yeast species from the genera Candida, Kloeckera, Hanseniaspora, Zygosaccharomyces, Schizosaccharomyces, Torulaspora, Brettanomyces, Saccharomycodes, Pichia and Williopsis to wine production (reviewed by Jolly et al. 2006). However, only a few studies have focused on the effects of deliberately mixing S. cerevisiae strains during fermentation. The most notable example of mixed S. cerevisiae strain fermentation products in the Australian Wine Research Institute’s development of the Alchemy I and II mixed S. cerevisiae strain yeast products, which promote beneficial volatile thiol production in  18  Sauvignon Blanc. By mixing yeast strains known to differ in their liberation of specific thiols, Swiegers et al. (2007) were able to demonstrate that the volatile thiol content and the sensory profiles of wines differed in mixed S. cerevisiae culture beyond the effects of each individual yeast strain, suggesting a synergistic effect. The Alchemy product line represents the optimization of this strategy and has demonstrated the relevance of yeast strain selection and composition to wine quality. Another study by Howell et al. (2006) confirmed that the unique volatile profiles created by mixing S. cerevisiae strains during wine fermentation cannot be replicated by fermenting each strain individually and then blending the resulting wine, which has contributed to the scientific credibility of mixed yeast products. This result also supports the previous finding that co-cultured yeasts are able to interact by sharing metabolites that are diffused into the medium (Cheraiti et al. 2005), which explains how mixed S. cerevisiae strains are capable of creating novel wine volatile profiles that are more than the sum of their parts. Aside from the particular case of the thiols of Sauvignon Blanc, the literature has yet to demonstrate that mixed S. cerevisiae strain fermentations are advantageous from a sensory perspective. However, as research continues to elucidate the chemistry and genetics of characteristic varietal aromas, mixed S. cerevisiae strain fermentations may represent a novel niche in the technological development of the wine industry.  1.7 Proposed Research The importance of yeast to wine aroma has been well established (Lambrechts and Pretorius 2000), as has the need for yeast strains that perform consistently and  19  improve the complexity and sensory profile of modern Pinot noir. Recent studies have shown that mixed S. cerevisiae strains may represent the ideal system for the creation of such products (Cheraiti et al. 2005, Howell et al. 2006, Swiegers et al. 2007). This research is guided by the hypothesis that fermenting Pinot noir grape must with mixed ratios of novel S. cerevisiae strains isolated in a premium wine region of Burgundy will result in unique volatile wine profiles that may be more complex than Pinot noir fermented by industrial or individual Burgundian strains of S. cerevisiae. The first objective of this study is to enologically characterize three novel S. cerevisiae strains isolated in Burgundy though a comparison to six industrial Pinot noir strains. Strains will be compared over a variety of relevant parameters, including genetic fingerprinting, killer phenotype, fermentation kinetics, ethanol production, conversion of sugar to ethanol, glycerol production, acetic acid production, growth phenotype, ethanol tolerance, foam production, sulfur dioxide production and compatibility with malolactic bacteria. If the individual Burgundian strains are enologically equivalent to the industrial strains, the second objective of this study is to determine the metabolic effects of mixed S. cerevisiae strain fermentations. This will be evaluated by fermenting Pinot noir grape must with industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains and evaluating the headspace of the finished wines with gas chromatography-mass spectrometry volatile compound analysis. Each compound will be quantified and evaluated for statistically significant differences among yeast strains and between fermentation temperatures. Patterns of volatile compound production will also be evaluated with multivariate statistics.  20  2 MATERIALS AND METHODS 2.1 Yeast and bacterial strains employed Industrial S. cerevisiae wine strains Enoferm Assmanshausen (AMH), Enoferm Burgundy (BGY), Lalvin RA17 (RA17) and Lalvin Bourgorouge RC212 (RC212) were purchased as active dry yeast from Lallemand Inc. (Rexdale, Canada). Industrial S. cerevisiae strains Australian Wine Research Institute 796 (AWRI 796) and Maurivin B (MB) were obtained as agar slants from Mauri Yeast Australia (Sydney, Australia). Novel S. cerevisiae strains A1, A2 and A3 were isolated from a vineyard in Burgundy, France, by Dr. Hennie van Vuuren. Saccharomyces bayanus wine strain EC1118 and S. cerevisiae wine strain UCD 522 (Montrachet) were obtained from freezer stocks maintained by the van Vuuren laboratory. Oenococcus oeni strain Lalvin 31 was purchased from Lallemand Inc. (Rexdale, Canada).  2.2 Selected ratios of mixed S. cerevisiae strains The Burgundian S. cerevisiae strains A1, A2 and A3 were mixed in the ratios shown in Table 1 to study the effects of mixing Burgundian S. cerevisiae strains on the complexity of the wine produced.  Table 1. The mixed Burgundian S. cerevisiae strain ratios used in this study (shown as A1:A2:A3). Mixture Name  M1  M2  Ratio (A1:A2:A3)  1:1:1' 1:2:3'  M3  M4  3:2:1'  1:3:2'  21  2.3 Media and culture conditions All S. cerevisiae strains were maintained as freezer stocks at -80 °C in 15% glycerol/yeast peptone dextrose (YPD) and cultured in Difco YPD broth (Becton, Dickinson and Co., Sparks, USA) according to standard procedures (Ausubel et al. 1999). Lyophilized O. oeni was rehydrated in 50 mL of sterile distilled water for 15 minutes and used directly for the malolactic fermentation compatibility study. To assess the killer phenotype of the S. cerevisiae strains, killer assay medium was formulated by buffering YPD agar with 50 mM dibasic phosphate and adjusting the pH to 4.2 with citric acid prior to autoclaving. Filter sterile (0.22 µm) methylene blue was added at a rate of 0.0015% w/v (adapted from Van Vuuren and Wingfield 1986). Free run Pinot noir must (2008) was obtained from Calona Vineyards in Kelowna, BC, Canada, and was used to conduct model fermentations and assess foam production. The Pinot noir must had soluble solids of 25.2° Brix, a pH of 3.77, 5.62 g/L titratable acidity, and 244 mg N/L yeast available nitrogen (YAN). To determine ethanol tolerance of yeasts, this must was supplemented with dextrose and fructose (Fisher Scientific, Ottawa, Ontario, Canada) to 16.5% glucose and 16.5% fructose, respectively. Due to its clarity and light colour, filter sterilized (0.22 µm) 2008 Calona Vineyards Chardonnay grape must (Kelowna, Canada) was used to characterize growth kinetics of S. cerevisiae strains. The Chardonnay must had soluble solids of 27 Brix, a pH of 3.46, 5.76 g/L titratable acidity, and 121 mg N/L YAN. A synthetic must medium used to quantify production of sulfur dioxide was modified from Denayrolles et al. (1995) by adding 1.0 mL/L of Tween 80 and adjusting the pH to 3.5 with KOH pellets.  22  2.4 Analytical methods 2.4.1 Quantification of compounds with high pressure liquid chromatography Ethanol, glycerol, glucose, fructose, and acetic acid were quantified with an Agilent 1100 series HPLC (Agilent Technologies, Palo Alto, USA) using a Supelcogel C-61OH 30 cm x 7.8 mm column (Sigma-Aldrich, Oakville, USA) maintained at 50 °C over a 22 minute isocratic run of 0.1% phosphoric acid solution at 0.75 mL/min (Adams and van Vuuren 2010). An Agilent G1362A refractive index detector with positive polarity and a 35 °C optical unit temperature was used to monitor compounds, which were later quantified based on standard curves using Agilent LC-MS ChemStation revision A.09.03 software. Malic and lactic acids were quantified with an Agilent 1100 series HPLC using a Macherey-Nagel Nucleogel 300 x 7.8 mm ION 300 OA column (Macherey-Nagel, Düren, Germany) maintained at 50 °C over a 30 minute isocratic run of 0.00425 M sulfuric acid solution at 0.45 mL/min. An Agilent G1362A refractive index detector with positive polarity and an optical unit temperature of 40 °C was used to monitor compounds, which were later quantified based on standard curves using Agilent LC-MS ChemStation revision A.09.03 software.  2.4.2 Identification and quantification of volatile compounds using gas chromatography-mass spectrometry (GC-MS) GC-MS headspace analysis was used to analyze Pinot noir wine samples according to the method of Danzer et al. (1999), without solid phase microextraction (SPME). A 10 mL wine sample was transferred into a 20 mL headspace vial containing  23  3 g of NaCl and placed into the autosampler. Samples were equilibrated at 85 °C for 10 minutes with agitation; 3-octanol at a concentration of 0.296 mg/L was used as the internal standard. An Agilent 6890N GC interfaced to a 5973N Mass Selective Detector along with a 60 m x 0.25 mm ID, 0.25 µm thickness DBWAX fused silica open tubular column (J&W Scientific, Folstom, USA) were used to detect and quantify volatile compounds in ultra high purity helium carrier gas flowing at a constant rate of 1.3 mL/minute; 1 mL of sample headspace was injected. The headspace sample valve was held at 100 °C while the transfer line was maintained at 110 °C. The GC oven was initially held at 40 °C for 5 minutes, then gradually increased to 100 °C at a rate of 5 °C/minute, and then to 200 °C at a rate of 3 °C/minute. The oven was held at 200 °C for 1 minute, and then was raised to 240 °C at a rate of 20 °C/minute. The MS was operated in scan mode (35-400 AMU) and the analysis was conducted in triplicate. Enhanced Chemstation software (MSD Chemstation Build75, Agilent Technologies, Palo Alto, USA) was used to analyze the data and compounds were identified by matching with the Wiley7Nist05 library (Wiley and Sons, Hoboken, USA). Compounds were considered quantifiable when their corresponding peaks had a signal to noise ratio greater than 10 and they were consistently integrable; compounds were detectable when their signal to noise ratio was greater than three. Compounds were quantified based on the internal response factor of a standard of known concentration that was dissolved into a base wine of 12% ethanol and 1 mM tartaric acid and normalized against the internal standard for each run.  24  2.5 Genetic and phenotypic characterization 2.5.1 Genetic fingerprinting S. cerevisiae strains were grown overnight in 5 mL YPD broth at 30 °C. Cells were harvested and genomic DNA was extracted as outlined in Short Protocols in Molecular Biology (Ausubel et al. 1999). Each strain was genetically fingerprinted using a polymerase chain reaction (PCR) method capable of differentiating wine yeast strains based on the amplification patterns of chromosomal regions between the repetitive δ sequences that are often associated with the Ty1 transposon (Schuller et al. 2004). An MJ Research Peltier Thermal Cycler 200 (Waltham, USA) was used to amplify chromosomal regions using the primers δ12 (5’-TCAACAATGGAATCCCAAC-3’) and δ2 (5’-GTGGATTTT TATTCCAAC-3’) (Schuller et al. 2004). A 50 µL reaction mixture was prepared, which contained 10 ng of DNA template, 5x GC buffer, 1 U iProof DNA polymerase (BioRad, Mississauga, Canada), 0.5% v/v DMSO, 0.2 mM of each dNTP, and 25 pmol of each primer. After the initial denaturation at 98 °C for 3 minutes, the reaction mixture was cycled 30 times according to the following program: 98 °C for 10 seconds, 55 °C for 1 minute, and 72 °C for 1 minute, which was followed by a final elongation period at 72 °C for 10 minutes. The PCR products were separated by gel electrophoresis on a 1.5% agarose gel and visualized with SYBR Safe DNA gel stain (Invitrogen Inc., Burlington, Canada). Strains BGY and MB were examined further by using a multiplex PCR protocol modified from López et al. (2003), which differentiates S. cerevisiae strains based on the number and location of introns in the COX1 gene using primers 3L (5’-GCTTTAATTG  25  GWGGWTTTGG-3’), 3R (5’-ATTGTCATACCATTTGTYCTYAT-3’), 4L (5’-GAAG TAGCAGGWGGWGGWGA-3’) and 5R (5’-GTTAGCTAAGGCWACWCCWGT-3’). A 50 µL reaction mixture containing 7.5 ng of DNA template, 5x GC buffer, 1 U iProof DNA polymerase, 2.5% v/v glycerol, 5% v/v DMSO, 0.2 µg BSA, 0.2 mM of each dNTP, and 15 pmol of each primer was used to amplify COX1 introns on an MJ Research Peltier Thermal Cycler 200. Initial denaturation took place at 95 °C for 5 minutes, followed by 35 cycles of 95 °C for 1.5 minutes, 52 °C for 2.5 minutes, and 72 °C for 3.5 minutes. A final elongation at 72 °C for 10 minutes completed the reaction.  2.5.2 Monitoring mixed strain population dynamics during fermentation Mixed Burgundian strain populations of the mixtures M1, M2, M3, and M4 were monitored based on the genetic fingerprints of individual Burgundian strains A1, A2 and A3, which were visualized through colony PCR. Cells were harvested by centrifugation at the midpoint (9% v/v ethanol) and end (13.5% v/v ethanol) of fermentation. Cells were resuspended and serially diluted in sterile MilliQ water before being grown up on YPD agar plates at 30 °C for 3 days and subsequently stored at 4 °C until analysis could be completed. The genetic fingerprints of 45 colonies from each replicate at each timepoint and temperature were assessed via colony PCR (Howell et al. 2005). Colony PCR was conducted in a similar manner to the method described in 2.5.1; however, rather than including DNA template, a small amount of colony was included in the reaction mixture and the initial denaturation period at 98 °C was increased to 10 minutes in order to lyse the cells.  26  2.5.3 Killer factor typing The killer factor phenotype was assessed in the individual Burgundian S. cerevisiae strains A1, A2 and A3, as information was available from yeast product manufacturers regarding the killer type of each of the industrial strains. S. cerevisiae strains A1, A2, A3, AMH, and UCD522, as well as the S. bayanus strain EC1118 were grown on YPD-agar plates for 72 hours at 30 °C. Three colonies of sensitive strain AMH were picked and resuspended in sterile MilliQ water to give 5 x 108 cells/mL; 300 µL of this suspension was spread as a lawn on a plate containing killer assay medium and allowed to dry. Colonies from each of the other strains were swabbed and spread as a thick line on top of the killer lawn. The plate was then incubated at 18 °C for 5 days (adapted from Van Vuuren and Wingfield 1986).  2.6 Enological characterization 2.6.1 Model fermentations S. cerevisiae freezer stocks were used to inoculate 5 mL cultures of YPD, which were grown overnight in a rotary wheel to stationary phase at 30 °C. Flasks containing 50 mL of YPD were subsequently inoculated at a rate of 5 x 105 cells/mL and grown aerobically in a shaker bath (180 rpm) for 24 hours at 30 °C. Cells were harvested by centrifugation (5,000 x g for 5 minutes), washed with sterile MilliQ water, and resuspended in fermentation medium at a density of 5 x 108 cells/mL. Fermentations were inoculated in biological triplicate at a rate of 2 x 106 cells/mL. In the case of the mixed strain fermentations, yeast strains were not combined prior to inoculation. All fermentations were conducted in media bottles topped with disinfected (70% ethanol)  27  rubber bungs and water-filled capped gas locks to ensure anaerobic conditions. Sampling occurred anaerobically and aseptically by piercing the rubber bungs with 5-inch hypodermic needles (Air-Tite, Virginia Beach, USA) attached to 3 mL syringes (Becton Dickinson, Franklin Lakes, USA) and extracting approximately 1 mL of sample. The primary experimental fermentations were conducted in triplicate in 900 mL of Calona Vineyards Pinot noir must at 22 °C and 27 °C, respectively, and were used to assess ethanol, glycerol, and acetic acid production, mixed strain population dynamics, fermentation kinetics, and production of volatile compounds. After sugars were depleted, 100 mg/L of potassium metabisulfite was added to the wine to protect against oxidation and samples were stored at 4 °C until the volatile compound analysis could be completed. Additional fermentations were required for a number of assays and are described in the methods that follow.  2.6.2 Assessment of fermentation kinetics Fermentation kinetics were monitored by tracking ethanol production and CO2 evolution throughout the fermentation process; CO2 evolution was quantified by recording fermentation weight loss. Sampling occurred twice daily in the early stage of fermentation, daily in the mid stage, and every two days in the end stage.  2.6.3 Quantification of ethanol, glycerol, and acetic acid Ethanol, glycerol, and acetic acid were quantified with the high pressure liquid chromatography (HPLC) method detailed in 2.4. Ethanol and glycerol were monitored throughout the fermentation at the timepoints described in 2.6.2, while acetic acid  28  production was quantified at the end of fermentation. Fermentation samples were vortexed, centrifuged, and filter sterilized (0.22 µm) before compounds were analyzed.  2.6.4 Growth phenotype assay The growth phenotypes of the S. cerevisiae strains were assayed in a Bioscreen C Growth Chamber (Thermo-Labsystems, Franklin, USA) in filter sterile (0.22 µm) Calona Vineyards Chardonnay grape must due to its clarity relative to Pinot noir. The S. cerevisiae strains were grown to stationary phase in 5 mL cultures of YPD at 30 °C in a rotary wheel. Cells were harvested by centrifugation (5,000 x g for 5 minutes) and were resuspended in Chardonnay must. Chardonnay must was then inoculated at a rate of 5 x 105 cells/mL and 150 µL aliquots were transferred by technical triplicate into a 100-well Bioscreen C optical plate (Thermo-Labsystems, Franklin, USA). This plate was placed in the Growth Chamber and grown for 96 hours with continuous shaking. The optical density (600 nm) was measured automatically each hour, and the data was compiled using the affiliated Biolink-dos software. The log of the optical density was then plotted against time to evaluate growth kinetics at 22 °C and 27 °C.  2.6.5 Ethanol tolerance assay The ethanol tolerance of the various S. cerevisiae strains was assessed by fermenting each strain in biological triplicate in 200 mL of Calona Vineyards Pinot noir must supplemented to 33% sugar with equimolar amounts of glucose and fructose at  29  22 °C and 27 °C. Fermentations were sampled initially and after 21 days, at which point no further fermentative activity was present. Glucose, fructose, and ethanol were quantified using the HPLC method presented in section 2.4.1.  2.6.6 Foam production assay Foam production was assessed in the S. cerevisiae strains at 22 °C and 27 °C using an assay modified from Regodón et al. (1997). Yeast cells were cultured in preparation for fermentation as outlined in the Model Fermentations section, and were then inoculated into 18 x 150 mm test tubes containing 10 mL of Calona Pinot noir grape must. Foam height was monitored three times per day and the maximum height achieved was measured and recorded in millimeters.  2.6.7 Sulfur dioxide assay Sulfur dioxide production by S. cerevisiae strains was assayed following fermentation in biological triplicate in 200 mL of synthetic must at 22 °C and 27 °C. Sulfur dioxide was quantified in technical triplicate according to manufacturer protocols using the “Total SO2” UV test kit from R-Biopharm (Darmstadt, Germany). Unfermented synthetic must was also assayed to ensure that it was free from sulfite contamination.  2.6.8 Compatibility with malolactic fermentation Fermentations investigating the malolactic compatibility of the S. cerevisiae strains were conducted in biological triplicate in 400 mL of Calona Vineyards Pinot noir must. The alcoholic portion of the fermentation was completed at 22 °C following  30  inoculation with the S. cerevisiae strains, while the malolactic fermentation was conducted at 20 °C following inoculation with O. oeni strain MBR 31. Samples were collected at 3-4 day intervals for 18 days, at which point the fermentation was no longer active. Malic and lactic acids were quantified using the HPLC method described under Analytical Methodology.  2.7 Statistical analyses Statistical analyses were conducted with Microsoft Excel 2007 (Microsoft, Redmond, USA) and Minitab 15 statistical software (Minitab Inc., State College, USA). Differences in the production of volatile compounds, ethanol, glycerol, acetic acid and sulfur dioxide were assessed among individual and mixed Burgundian S. cerevisiae strains with one-way analysis of variance (ANOVA) with replication and Fisher’s protected least significant difference (LSD). Differences in compound production between fermentation temperatures were evaluated with Student’s t-tests. Results were deemed significant at the 95% confidence level. Principal component analysis (PCA) utilizing a correlation matrix was used as a reductive strategy to examine patterns of volatile compound production among the twelve S. cerevisiae strains and mixtures. PCA was conducted on all yeast strain replicates (n=3) based on the complete volatile profiles of all replicates at both fermentation temperatures concurrently. PCA was also used to examine patterns of volatile alcohol, ethyl ester, and acetate ester production at 22 °C and 27 °C independently. Cluster analysis was conducted using the average linkage method with Euclidean distance in order to cluster yeast strains based on patterns of volatile compound  31  production during fermentation at 22 °C and 27 °C. This method was also applied to the averaged volatile profiles of industrial, individual Burgundian, and mixed Burgundian strains in order to examine overall differences among strain subsets.  32  3 RESULTS 3.1 Genetic and phenotypic characterization 3.1.1 Genetic fingerprinting differentiated eight out of nine wine yeast strains PCR-based genetic fingerprinting based on differences in the chromosomal regions between δ sequences (Schuller et al. 2004) successfully differentiated eight out of nine pure strains of S. cerevisiae (Figure 1, Panel A). Divergent banding patterns were most pronounced in the 250 to 500 base pair range, where differences were evident in all of the different strains. Individual Burgundian strains A1, A2 and A3 showed differences in this region, in addition to a novel 1700 base pair band in strain A3, which made this a  2000 bp 1500 bp 1000 bp 750 bp  Ctl  MB  B  BGY  AWRI 796  MB  BGY  RA17  RC212  AMH  A3  A2  A  A1  viable method for monitoring yeast strain populations during mixed strain fermentation.  2000 bp 1500 bp 1000 bp 750 bp  500 bp  500 bp  250 bp  250 bp  Figure 1. Genetic fingerprints of nine S. cerevisiae strains. (A) δ sequence typing of all strains. (B) COX1 intron amplification of industrial strains BGY and MB alongside a negative control lacking template DNA.  33  While most of the industrial control strains differed greatly, two strains, BGY and MB, gave identical banding patterns with this method. An additional method based on the number and location of the COX1 introns was undertaken (López et al. 2003) that also yielded identical results (Figure 1, Panel B). Based on the results from these two methods, along with the identical growth kinetics and fermentative metabolism of these two strains (not shown), it was concluded that BGY and MB were genetically identical. As a result, strain MB was excluded from further analysis.  3.1.2 Mixed strain populations did not maintain their inoculated ratios during fermentation Mixed Burgundian yeast strain populations during fermentation were quantified using colony PCR and genetic δ sequence fingerprinting at the fermentation midpoint (approximately 6% v/v ethanol) and after sugars were depleted (approximately 13.5% v/v ethanol) (Figure 2). At each timepoint, 45 colonies were fingerprinted for each of three biological replicates. None of the mixed Burgundian strain fermentations maintained their initial inoculated ratios throughout the fermentation process. When strains A1, A2 and A3 were inoculated in equal proportions (1:1:1, mixture M1), strain A3 represented over 50% of the colonies assayed during fermentation at 27 °C and at the midpoint of fermentation at 22 °C. When strains A1, A2 and A3 were inoculated in a ratio of 1:2:3 (M2), A3 represented over 50% of the colonies assayed at both temperatures and timepoints, whereas strain A2 was consistently below its inoculated rate of 33%.  34  D  80 60  A1  40  A2  20  A3  % Colonies  % Colonies  A  0  A2  20  A3 M1 M2 M3 M4 S. cerevisiae Strain Mixture  E  80 60 A1  40  A2  20  A3  80 60 A1  40  A2  20  A3  0 M1  % Total Inoculum  A1  40  0  0  C  60  M1 M2 M3 M4 S. cerevisiae Strain Mixture  % Colonies  % Colonies  B  80  M2  M3  M1  M4  M2  M3  M4  S. cerevisiae Strain Mixture  S. cerevisiae Strain Mixture 80 60 A1  40  A2  20  A3 0 M1  M2  M3  M4  S. cerevisiae Strain Mixture  Figure 2. Mixed Burgundian strain population dynamics (mean values (n=3) +/- standard deviation) during fermentation in Pinot noir must. (A) Mixed strain populations at the fermentation midpoint (5.65% ethanol) at 22 °C. (B) Mixed strain populations in the finished wine (13.5% ethanol) fermented at 22 °C. (C) Initial inoculation ratios of mixed strain fermentations M1 (1:1:1), M2 (1:2:3), M3 (3:2:1), and M4 (1:3:2). (D) Mixed strain populations at the fermentation midpoint (6% ethanol) at 27 °C. (E) Mixed strain populations in the finished wine (13.5% ethanol) fermented at 27 °C. Burgundian mixture M3 (3:2:1) was not overtaken by A3, although this strain consistently surpassed its 16% inoculum at all timepoints and temperatures. As a result, strains A1 and A2 fell below their inoculum rate of 50 and 33%, respectively, yet these two strains were present at levels comparable to those of A3, resulting in a more balanced 35  mixed Burgundian strain fermentation. Strain A3 was also not as predominant in Burgundian mixture M4 (1:3:2) as it was in other Burgundian mixtures. Burgundian mixture M4 showed a balance of strains A2 and A3, particularly at 22 °C. Strain A1 remained at levels below 25% of the colonies assayed, which was consistent with its inoculation rate of 16%.  3.1.3 Burgundian strains A1, A2 and A3 are killer positive The killer phenotype was assayed for the three individual Burgundian yeast strains. Control strains EC1118 and UCD522 were included as killer-positive and sensitive controls, respectively. All strains were swabbed thickly over a lawn of the sensitive strain AMH on a plate containing a killer assay medium. After growth, strains EC1118, A1, A2 and A3 were surrounded by a blue-ringed zone showing inhibition of the sensitive lawn strain AMH, indicating the production of active killer toxin (Figure 3).  Figure 3. Killer phenotype assay of novel Burgundian strains A1, A2 and A3. Control strains EC1118 and UCD522 were used as killer-positive and sensitive strains. Strains were plated onto a lawn (5 x 108 cells/mL) of sensitive strain AMH and grown at 18 °C. The blue-ringed zones of inhibition surrounding the swabs indicated a killer-positive result. 36  Methylene blue stains dead cells, which confirmed that the zones of inhibition were due to the death of the sensitive strain AMH, rather than its inhibition by the secretion of pheromone. The industrial yeast strains were not assayed because information on their killer phenotype was readily available from yeast manufacturers. The killer phenotypes of all strains in this study are shown in Table 2. Table 2. Killer phenotypes of S. cerevisiae strains used in this study. Yeast Strain Control, EC1118 Control, UCD 522 Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3  Killer type Killer positive* Sensitive* Sensitive* Killer positive** Sensitive* Sensitive* Neutral* Killer positive Killer positive Killer positive  *Indicates that killer phenotype information was obtained from Lallemand Wine Inc. ** Indicates that killer phenotype information was obtained from Maurivin Yeast  3.2 Enological characterization of yeast strains 3.2.1 Fermentation kinetics were similar for all wine yeast strains The fermentation kinetics of the industrial, individual Burgundian and mixed Burgundian yeast strains were assessed by quantifying CO2 evolution and ethanol production throughout fermentation at 22 °C and 27 °C in Pinot noir must (Figure 4, Panels A-D). Slight kinetic differences existed among the S. cerevisiae strains, particularly during fermentation at 27 °C (Figure 4, Panels C and D). Industrial strain 37  RA17 reached a higher concentration of ethanol more quickly than other strains, while the ethanol production of industrial strain BGY was slower than that of other strains, although in both cases, the final ethanol concentration was similar to all other strains. These kinetic variations were also detected in the evolution of carbon dioxide (Figure 4, Panel D).  A  C  14  14  8 6 4 2 0 0  2  4  6  8  10  12  8 6 4 2 0 0  Time (Days)  B  2  AMH AWRI 796 BGY  100  RA17 RC212 A1 A2 A3 M1  80 60 40  M2 M3 M4  20 0  4  6  8  10  Time (Days)  D  120  Carbon dioxide (g)  10  14  Carbon dioxide (g)  Ethanol % (v/v)  10  AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  12 Ethanol % (v/v)  AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  12  120 AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  100 80 60 40 20 0  0  2  4  6  8  Time (Days)  10  12  14  0  2  4  6  8  10  Time (Days)  Figure 4. Fermentation kinetics of individual and mixed S. cerevisiae strains in Pinot noir must. Mean values are shown (n=3). (A) Ethanol production during fermentation at 22 °C. (B) CO2 evolution during fermentation at 22 °C. (C) Ethanol production during fermentation at 27 °C. (D) CO2 evolution during fermentation at 27 °C.  38  3.2.2 Ethanol production did not vary among yeast strains Ethanol production by industrial, individual Burgundian and mixed Burgundian yeast strains was quantified in triplicate in Pinot noir wine fermented at 22 °C and 27 °C (Figure 5).  Figure 5. Ethanol concentrations of Pinot noir wine fermented at 22 °C and 27 °C with eight pure and four mixed S. cerevisiae strains. Mean values are shown (n=3); error bars indicate +/- standard deviation. Ethanol production was also assessed statistically across yeast strain and temperature (Table 3). No significant (p≤0.05) differences existed among yeast strains in the final ethanol concentrations produced at either temperature, nor did differences in ethanol production exist in single yeast strains across the two temperatures. Ethanol production at 22 °C ranged from 13.40 % (v/v) in wine fermented by strain RA17 to 13.77% (v/v) in wine fermented by strain A3. When the fermentation temperature was  39  increased to 27 °C, ethanol production ranged from 13.40% (v/v) in wine fermented by Burgundian mixture M4 to 13.59% (v/v) in wine fermented by Burgundian mixture M3. Table 3. Ethanol concentrationsa (% v/v) of Pinot noir wine following fermentation at 22 °C and 27 °C with the industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains. Strain and temperature effects are shown; p values indicate the temperature effect for each strain. Yeast Strain  22 °Cb  27 °C  pc  Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3 Mixed Burgundian (1:1:1), M1 Mixed Burgundian (1:2:3), M2 Mixed Burgundian (3:2:1), M3 Mixed Burgundian (1:3:2), M4  13.53 13.45 13.64 13.40 13.43 13.52 13.53 13.77 13.58 13.47 13.52 13.52  13.48 13.39 13.56 13.56 13.52 13.50 13.53 13.50 13.59 13.44 13.40 13.55  ns ns ns ns ns ns ns ns ns ns ns ns  a  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns indicates no significant difference between the fermentation temperatures for each yeast strain b  3.2.3 Conversion of sugar to ethanol by wine yeast strains The sugar to ethanol conversion factors for the industrial, individual Burgundian and mixed Burgundian yeast strains were calculated by dividing the total ethanol produced (% m/v) by the total glucose and fructose consumed (% m/v) during fermentation at 22 °C and 27 °C (Table 4). No significant differences in conversion factors existed among yeast strains; however, strain A3 had a significantly (p ≤ 0.001) greater conversion factor at 22 °C than at 27 °C. The conversion factors at 22 °C ranged  40  from 0.464 in industrial strain RA17 to 0.482 in Burgundian strain A3. At 27 °C, the conversion factors ranged from 0.465 in industrial strain AWRI 796 and mixture M3, to 0.472 in mixture M1. Table 4. Conversion of sugar to ethanola by the individual and mixed S. cerevisiae strains during fermentations in Pinot noir grape must at 22 °C and 27 °C. Strain and temperature effects are shown; p values indicate the temperature effect for each strain. Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3 Mixed Burgundian (1:1:1), M1 Mixed Burgundian (1:2:3), M2 Mixed Burgundian (3:2:1), M3 Mixed Burgundian (1:3:2), M4  22 °Cb 0.469 0.471 0.472 0.464 0.469 0.476 0.471 0.482 0.472 0.465 0.466 0.470  27 °C 0.469 0.465 0.471 0.471 0.469 0.470 0.469 0.469 0.472 0.467 0.465 0.470  pc ns ns ns ns ns ns ns *** ns ns ns ns  a  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns and *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.001 b  3.2.4 Glycerol production varied significantly across wine yeast strain and fermentation temperature The glycerol production of industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains were quantified in Pinot noir wine fermented at 22 °C and 27 °C (Figure 6). Significant differences in glycerol production were found between yeast strains and fermentation temperatures (Table 5). In all samples, glycerol production  41  was significantly (p≤0.05) reduced in wine fermented at 22 °C compared to wine fermented at 27 °C.  Figure 6. Glycerol production by S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Mean values are shown (n=3); error bars indicate +/- standard deviation. Significant differences also existed at both fermentation temperatures across yeast strains. At 22 °C, glycerol production ranged from 7.94 g/L in wine fermented by industrial strain RC212 to 9.34 g/L in wine fermented by industrial strain AWRI 796 (Table 5). The glycerol production of the individual and mixed Burgundian S. cerevisiae strains fit into the middle of the spectrum and differed significantly from the strains that comprised the extremes of glycerol production, such as RC212 and AWRI 796. The only exception to this pattern was the individual Burgundian strain A2, which produced significantly more glycerol than all other strains except AWRI 796. This pattern of glycerol production was similar at 27 °C, where production ranged from 9.12 g/L in wine fermented by industrial strain RC212 to 10.64 g/L in wine fermented by industrial strain  42  AWRI 796. At 27 °C, the glycerol content in wines fermented by the individual and mixed Burgundian S. cerevisiae strains was in the middle of the overall distribution, with the exception of the relatively low producer, individual Burgundian strain A1. Table 5. Glycerol productiona (g/L) by yeast strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects are shown; p values indicate the temperature effect for each strain. Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3 Mixed Burgundian (1:1:1), M1 Mixed Burgundian (1:2:3), M2 Mixed Burgundian (3:2:1), M3 Mixed Burgundian (1:3:2), M4  22 °Cb 8.79d 9.34e 8.80d 8.43bc 7.94a 8.49bcd 9.19e 8.44bc 8.37b 8.46bc 8.48bcd 8.72cd  27 °C 10.02d 10.64e 9.65bcd 9.31ab 9.12a 9.25a 9.63bc 9.41ab 9.33ab 9.85cd 9.37ab 9.65bcd  pc *** * ** * ** ** ** *** *** ** ** **  a  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01 and 0.001 b  3.2.5 Acetic acid production significantly varied across wine yeast strain and fermentation temperature Acetic acid production by the industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains was quantified in Pinot noir wine fermented at 22 °C and 27 °C (Figure 7). Significant (p≤0.05) differences existed among yeast strains and across fermentation temperature (Table 6). In all cases, acetic acid production increased with  43  fermentation temperature; this increase was significant (p≤0.05) in all strains and mixtures except in industrial strain RC212 and in Burgundian mixture M2.  Figure 7. Acetic acid production by S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Mean values are shown (n=3); error bars indicate +/- standard deviation.  Acetic acid production at 22 °C ranged from 0.119 g/L in wine fermented by individual Burgundian strain A1 to 0.337 g/L in wine fermented by industrial strain BGY. Similarly, at 27 °C, wine fermented by strain A1 contained the least acetic acid at 0.142 g/L, while wine fermented by BGY contained the most acetic acid at 0.411 g/L. The majority of yeast strains and Burgundian mixtures significantly differed from one another in acetic acid production at 22 °C; the only exceptions being the similar acetic acid profiles of wines fermented by industrial strain AWRI 796 and Burgundian strain A3, and Burgundian mixtures M1 and M3. Fewer significant differences were evident during 44  fermentation at 27 °C, which was likely due to the abundance of wine samples containing 0.16–0.20 g/L acetic acid.  Table 6. Acetic acid productiona (g/L) by yeast strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects are shown; p values indicate the temperature effect for each strain. Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3 Mixed Burgundian (1:1:1), M1 Mixed Burgundian (1:2:3), M2 Mixed Burgundian (3:2:1), M3 Mixed Burgundian (1:3:2), M4  22 °Cb 0.124b 0.132c 0.337j 0.224h 0.246i 0.119a 0.209g 0.135c 0.147d 0.160e 0.149d 0.177f  27 °C 0.187cde 0.179bcd 0.411h 0.292g 0.250f 0.142a 0.240f 0.167b 0.187cde 0.194de 0.171bc 0.200e  pc *** * ** *** ns ** * * ** ns ** **  a  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01 and 0.001 b  3.2.6 Growth phenotypes varied among wine yeast strains The growth characteristics of the eight individual S. cerevisiae strains were assessed aerobically in 150 µL of Calona Chardonnay over a period of 96 hours at 22 °C and 27 °C (Figure 8). Overall growth patterns were similar for all of the strains except A1, which showed an abrupt and inconsistent decrease in optical density as a result of its visible flocculation at high cell densities.  45  At 22 °C, the individual Burgundian strains A1 and A2 showed rapid growth with a relatively short lag phase, as did the industrial strains RC212 and RA17. Individual Burgundian strain A3 and industrial strains AWRI 796 and BGY showed a longer lag phase and did not enter log phase until at least 10 hours had elapsed. Industrial strain AMH had a less linear log phase and, along with industrial strain BGY, did not reach as great an optical density (OD600) as other strains in the study. At 27 °C, the growth pattern was quite similar to that observed at 22 °C; however, individual Burgundian strain A3 more closely resembled the faster growth patterns of strains A1, A2, RC212 and RA17. The log phase of industrial strain AMH was also more linear and of shorter duration than at 22 °C. In contrast, the lag phase of industrial strain BGY increased in duration; however, the corresponding log phase was very brief.  A  B  0.4  0.4 0.2  0.2 AMH  -0.2  BGY  -0.4  RA17  -0.6  RC212 A1  -0.8  A2  -1  A3  AMH  0  AWRI 796 Log OD600  Log OD600  0  AWRI 796  -0.2  BGY  -0.4  RA17 RC212  -0.6  A1  -0.8  A2  -1  A3  -1.2  -1.2 0  20  40  60  Time (Hours)  80  100  0  20  40  60  80  100  Time (Hours)  Figure 8. Growth phenotypes of the eight pure strains of S. cerevisiae grown aerobically in 150 µL of Calona Chardonnay grape must. Mean values are shown (n=3). (A) Growth at 22 °C. (B) Growth at 27 °C. The scattering pattern visible in strain A1 was due to the visible flocculation of the strain during later stages of growth. The final OD values achieved by the eight S. cerevisiae strains were assessed for variance and are shown in Table 7. Because of its evident flocculation, individual 46  Burgundian strain A1 was excluded from this analysis. At 22 °C, final OD values ranged from 1.494 in industrial strain AMH to 1.940 in industrial strain RC212. AMH had a significantly (p≤0.05) lower optical density than all of the other strains, including industrial strain BGY, which had the next lowest final OD at 1.658. The final OD values of individual Burgundian strains A2 and A3, 1.819 and 1.861 respectively, were in the middle of the spectrum. Strain A2 differed significantly (p≤0.05) from all other strains except A3, while strain A3 only differed significantly from industrial strains AMH and BGY at 22 °C. At 27 °C, the final OD values of the S. cerevisiae strains ranged from 1.667 in industrial strain AMH to 1.992 in industrial strain RA17. Strain AMH had a significantly (p≤0.05) lower final OD than all of the other strains. Burgundian strain A2 had the next lowest OD and differed from all strains except industrial strains BGY and AWRI 796. In contrast, Burgundian strain A3 was at the higher end of the spectrum with an OD of 1.979 and differed significantly from strains AMH, BGY, and A2. Table 7. Final optical densitiesa (λ=600) observed during aerobic growth of eight individual S. cerevisiae strains in Chardonnay grape must. Strain and temperature effects are shown; p values indicate the temperature effect for each strain.  a  Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3  22 °Cb 1.494a 1.938d 1.658b 1.931d 1.940d n/a 1.819c 1.861cd  27 °°C 1.667a 1.914bc 1.841b 1.992c 1.989c n/a 1.831b 1.979c  pc ** ns ns *** * n/a ns **  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01 and 0.001 b  47  3.2.7 Ethanol tolerance significantly varied across wine yeast strain and fermentation temperature The relative ethanol tolerance of the industrial, individual Burgundian and mixed Burgundian strains was assessed by fermenting Pinot noir must supplemented to 33% (w/v) sugar with equimolar amounts of glucose and fructose (Figure 9).  Figure 9. Ethanol production by eight pure and four mixed strains of S. cerevisiae following fermentation in Pinot noir must supplemented to 33% (w/v) sugars with equimolar amounts of glucose and fructose. Mean values are shown (n=3); error bars indicate +/- standard deviation.  The ethanol production of all strains was significantly (p≤0.05) greater at 22 °C than at 27 °C, and ranged from 17.68% (v/v) in wine fermented by industrial strain AMH to 18.90% (v/v) in wine fermented by industrial strain RC212 (Table 8). At 27 °C, ethanol production was more variable than at 22 °C, and ranged from 15.42 % (v/v) in wine fermented by industrial strain AMH to 17.75% v/v in wine fermented by Burgundian mixtures M1 and M4. In this instance, M1 and M4 produced significantly (p≤0.05) more ethanol than all but one of the industrial strains and all of the individual 48  Burgundian strains that comprised the mixtures. This effect was not found at 22 °C. In all cases, the ethanol tolerance of the individual Burgundian strains A1, A2 and A3 fell in the middle of the observed range and did not collectively differ significantly (p≤0.05) from the industrial S. cerevisiae strains. Table 8. Final ethanol concentrationsa of Pinot noir wine produced by individual and mixed Burgundian S. cerevisiae strains fermenting grape must supplemented with equimolar amounts of glucose and fructose. Strain and temperature effects are shown; p values indicate the temperature effect for each strain. Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Individual Burgundian, A1 Individual Burgundian, A2 Individual Burgundian, A3 Mixed Burgundian (1:1:1), M1 Mixed Burgundian (1:2:3), M2 Mixed Burgundian (3:2:1), M3 Mixed Burgundian (1:3:2), M4  22 °Cb 17.68a 18.30bc 18.21b 18.56cd 18.90e 18.20b 18.73de 18.15b 18.55cd 18.40bc 18.57cd 18.50cd  27 °C 15.42a 17.66ef 16.32b 17.59e 17.58e 17.56de 17.35c 17.42cd 17.75f 17.64ef 17.72ef 17.75f  pc *** ** *** *** *** *** *** *** *** *** *** *  a  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  3.2.8 Foam production significantly varied across wine yeast strain and fermentation temperature The relative foam production of the eight individual S. cerevisiae strains was assayed by measuring the maximum foam height observed during fermentation of 10 mL of Calona Pinot noir in standard 18 x 150 mm test tubes (Regodón et al. 1997) (Figure 10). 49  Figure 10. Foam production by eight individual strains of S. cerevisiae during fermentation in 10 mL of Pinot noir must. Mean values are shown (n=3); error bars indicate +/- standard deviation.  Foam production was analyzed across the S. cerevisiae strains at 22 °C and 27 °C (Table 9). At 22 °C, foam production ranged from 9.7 mm in must fermented by the industrial strain AMH to 34 mm in must fermented by industrial strain BGY, which produced significantly (p≤0.05) more foam than all of the other strains at this temperature. At 27 °C, foam production ranged from 7.3 mm in must fermented by industrial strain AMH to 32.3 mm in must fermented by individual Burgundian strain A2. At this temperature, AMH produced significantly (p≤0.05) less foam than all of the other strains, and foam production was generally more variable across yeast strains than at 22 °C. Strains that produced significantly (p≤0.05) more foam at 27 °C than at 22 °C included industrial strain AWRI 796 (p≤0.05), industrial strain RA17 (p≤0.01), and individual Burgundian strain A2 (p≤0.001). Overall, at both temperatures, the collective  50  individual Burgundian strains did not differ significantly from the collective industrial strains. Table 9. Foam productiona (height in mm) by eight individual strains of S. cerevisiae during fermentation in Pinot noir must. Strain and temperature effects are shown; p values indicate the temperature effect for each strain.  a  Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Burgundian, A1 Burgundian, A2 Burgundian, A3  22 °Cb 9.7a 10.0a 34.0b 13.3a 16.3a 16.0a 15.7a 16.3a  27 °C 7.3a 18.0b 28.0de 20.3bc 24.0cd 28.3de 32.3e 20.3bc  pc ns * ns ** ns ns *** ns  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  3.2.9 Sulfur dioxide production significantly varied across wine yeast strain and fermentation temperature Sulfur dioxide production by eight individual strains of S. cerevisiae during fermentation was assessed by measuring the sulfur dioxide content of synthetic wine fermented at 22 °C and 27 °C. Sulfur dioxide measurements were conducted with a photometric assay in technical triplicate (Figure 11). Sulfur dioxide production ranged from 9.2 mg/L in wine fermented by industrial strain AWRI 796 to 36.5 mg/L in wine fermented by industrial strain RC212 when fermentation was conducted at 22 °C (Table 10). The wine fermented by individual Burgundian strains A1, A2 and A3 contained 26.1, 32.6, and 20.6 mg/L of sulfur dioxide, 51  respectively, placing these strains in the upper middle range of overall sulfur dioxide production.  Figure 11. Sulfur dioxide production (mg/L) by eight individual S. cerevisiae strains in synthetic wine fermented at 22 °C and 27 °C. Mean values are shown (n=3); error bars indicate +/- standard deviation.  Sulfur dioxide production ranged from 6.4 mg/L in wine fermented by industrial strain BGY to 50.3 mg/L in wine fermented by industrial strain RA17 when fermentation was conducted at 27 °C (Table 10). At this higher temperature, wine fermented by individual Burgundian strains A1, A2 and A3 contained 40.9, 30.4, and 18.2 mg/L of sulfur dioxide respectively, which again placed the individual Burgundian strains in the upper middle range of sulfur dioxide production. Two yeast strains differed significantly in sulfur dioxide production across temperature; industrial strain RA17, which increased from 24.8–50.3 mg/L and individual Burgundian strain A1, which increased from 26.1–40.9 mg/L. 52  Table 10. Sulfur dioxide productiona (mg/L) by eight pure strains of S. cerevisiae following fermentation in synthetic must. Strain and temperature effects are shown; p values indicate the temperature effect for each strain. Yeast Strain Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Burgundian, A1 Burgundian, A2 Burgundian, A3 a  22 °Cb 27 °C  pc  15.0ab 9.2a 9.9a 24.8c 36.5d 26.1c 32.6d 20.6bc  ns ns ns ** ns *** ns ns  12.5ab 12.1ab 6.4a 50.3f 28.4cd 40.9ef 30.4de 18.2bc  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  3.2.10 Compatibility with malolactic bacteria varied among wine yeast strains The compatibility of eight individual S. cerevisiae strains with malolactic bacteria was assessed by monitoring malolactic fermentation with O. oeni strain MBR 31 in Calona Pinot noir wine previously fermented with the S. cerevisiae strains. The resulting malic acid consumption and lactic acid production over time are shown in Figure 12. The wines produced with the various S. cerevisiae strains varied significantly (p≤0.05) in the concentration of malic and lactic acid present following alcoholic fermentation, as well as in the concentration of lactic acid following malolactic fermentation (Table 11). In contrast, the concentration of malic acid following malolactic fermentation did not vary significantly among yeast strains (Table 11). Industrial strains  53  AMH and BGY had apparently already undergone native malolactic fermentation prior to inoculation with O. oeni, as is evident in Figure 12. However, because of the comparative nature of this study, their data was included in the analysis.  A  B  6  6  AMH  AMH  AWRI 796  5  5  AWRI 796 BGY  Lactic Acid (g/L)  Malic Acid (g/L)  BGY RA17  4  RC212  3  A1 A2  2  A3  4  RA17 RC212  3  A1 A2  2  A3  1  1  0  0 0  4  7 11 Time (days)  15  18  0  4  7 11 Time (days)  15  18  Figure 12. Consumption of malic acid (g/L) (A) and production of lactic acid (g/L) (B) during malolactic fermentation in Pinot noir following alcoholic fermentation by eight individual S. cerevisiae strains. Mean values are shown (n=3); error bars indicate +/standard deviation.  The concentration of malic acid varied from 0.63 g/L in wine fermented by industrial strain AMH to 5.23 g/L in wine fermented by individual Burgundian strain A1 prior to O. oeni inoculation. Most of the wines differed significantly in their initial malic acid content (Table 11); however, wines fermented by industrial strains AMH and BGY did not differ significantly (p≤0.05) from one another, nor did wines fermented by industrial strains RA17 and RC212. The initial malic acid concentrations in the wines fermented by individual Burgundian strains A1 and A3 were toward the high end of the spectrum at 5.23 g/L and 4.84 g/L, respectively, while the wine fermented by A2 was more toward the low end at 4.19 g/L. 54  Because some native malolactic fermentation had occurred in wines fermented by industrial strains AMH and BGY, the initial concentration of lactic acid also differed significantly (p≤0.05) in the wines fermented by the various S. cerevisiae strains and ranged from 0.32 g/L in the wine fermented by Burgundian strain A2 to 2.91 g/L in the wine fermented by industrial strain AMH. Most wines differed significantly (p≤0.05) from one another in the initial lactic acid content; however, wine fermented by individual Burgundian strain A2 did not differ significantly from wine fermented by industrial strain RA17, and wine fermented by individual Burgundian strain A1 did not differ significantly from wine fermented by industrial strain AWRI 796. In terms of initial lactic acid concentration, the wines fermented by individual Burgundian strains represented the low-to-middle range of the spectrum. The final concentration of malic acid did not differ significantly in wines fermented with the various S. cerevisiae strains and only ranged from 0.61 g/L in several strains to 0.64 g/L in individual Burgundian strain A1. In contrast, the final concentration of lactic acid was variable and differed significantly (p≤0.05) among all of the wine samples, ranging from 1.82 g/L in wine fermented with industrial strain BGY to 3.62 g/L in wine fermented with industrial strain AWRI 796. The individual Burgundian strains A1 and A3 represented the higher portion of the spectrum, while A2 settled more toward the low end, in keeping with the pattern observed in the initial malic acid concentrations.  55  Table 11. Malic acid consumptiona (g/L) and lactic acid production (g/L) by O. oeni during malolactic fermentation in Pinot noir fermented with eight individual yeast strains. Yeast Strain  a  Industrial, AMH Industrial, AWRI 796 Industrial, BGY Industrial, RA17 Industrial, RC212 Burgundian, A1 Burgundian, A2 Burgundian, A3  Malic Acid (g/L) Initialb Final 0.63a 0.61 de 5.04 0.61 0.64a 0.63 4.46c 0.61 c 4.52 0.61 5.23e 0.64 b 4.19 0.61 4.84d 0.61  Lactic Acid (g/L) Initial Final 2.91f 2.89c c 0.44 3.62h 1.71e 1.82a 0.34a 2.99d 0.49d 3.11e 0.45c 3.53g 0.32a 2.75b 0.39b 3.22f  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different  b  3.3 Production of volatile compounds The volatile compounds in the headspace of wines fermented by the eight individual and four mixed Burgundian S. cerevisiae strains were analyzed using a GC-MS with enhanced Chemstation software, which revealed differences in volatile compound concentration based on yeast strain and fermentation temperature. The commercial Pinot noir wine fermented from 60 percent bioidentical Pinot noir must was obtained from Calona Vineyards and was also assessed for volatile compounds. While the results from this sample were excluded from statistical analysis due to its differing must composition and a lack of replicates, its volatile profile is included in the bar graphs (as CALPN) as a representative young industrial standard. Compounds were considered quantifiable if their corresponding peak had a signal to noise ratio greater than 10 and was consistently integrable; compounds were detectable if the signal to noise ratio was greater than three. A representative chromatogram with the quantifiable peaks numbered is shown in Figure 13; the quantifiable compounds and their retention times are listed in Table 12. 56  Figure 13. A representative chromatogram obtained by GC-MS headspace analysis of Pinot noir wine. Quantifiable peaks are numbered; their corresponding compound names and retention times are listed in Table 12.  57  Table 12. Quantifiable volatile compounds identified in Pinot noir wine fermented by eight individual and four mixed Burgundian S. cerevisiae strains. Volatile Compound Acetaldehyde Methyl acetate Ethyl acetate 1,1-diethoxyacetal Isobutyl acetate Ethyl butanoate Propanol Isobutanol Isoamyl acetate Butanol 2-methyl-1-butanol 3-methyl-1-butanol Ethyl hexanoate Hexyl acetate Ethyl lactate 1-hexanol Ethyl octanoate Acetic acid Benzaldehyde 2,3-butanediol 1,3-butanediol Ethyl decanoate Ethyl laurate Phenylethanol Ethyl palmitate  Class Aldehyde Acetate ester Acetate ester Acetal Acetate ester Ethyl ester Alcohol Alcohol Acetate ester Alcohol Alcohol Alcohol Ethyl ester Acetate ester Ethyl ester Alcohol Ethyl ester Acid Aldehyde Alcohol Alcohol Ethyl ester Ethyl ester Alcohol Ethyl ester  Peak Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25  Retention Time (Minutes) 3.973 4.959 5.872 6.031 9.169 9.836 10.051 11.906 12.327 13.437 15.223 15.316 15.838 17.018 19.167 19.586 22.169 22.453 24.982 25.64 26.876 29.134 36.216 38.215 49.274  Of the 25 quantifiable compounds, nine were identified as alcohols, seven were ethyl esters, five were acetate esters, two were aldehydes, one was an acid, and one was an acetal (Table 13).  58  Table 13. Quantifiable volatile compounds identified in Pinot noir wine fermented by eight individual and four mixed Burgundian S. cerevisiae strains (arranged by class). Compound  Class  1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  3.3.1 Production of higher alcohols varied according to yeast strain and fermentation temperature Higher alcohols were the most prevalent and most abundant compound class identified in the Pinot noir wine samples. Nine higher alcohols were quantified in wines  59  fermented by eight individual and four mixed Burgundian S. cerevisiae strains at 22 °C and 27 °C (Figure 14).  A 1,3-butanediol (mg/L)  12 10  ■ 22 °C ■ 27 °C  8 6 4 2 0  2,3-butanediol (mg/L)  B  Yeast Strain 3 2.5  ■ 22 °C ■ 27 °C  2 1.5 1 0.5 0  2-methyl-1-butanol (mg/L)  C  Yeast Strain 3 2.5  ■ 22 °C ■ 27 °C  2 1.5 1 0.5 0  Yeast Strain  Figure 14. Concentrations (mg/L) of higher alcohols produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains. Mean values are shown (n=3); error bars indicate +/- standard deviation. CALPN indicates the industrial Calona Pinot noir sample. 60  Figure 14 (continued)  D 3-methyl-1-butanol (mg/L)  14 12  ■ 22 °C ■ 27 °C  10 8 6 4 2 0  E  Yeast Strain 0.6  ■ 22 °C ■ 27 °C  Butanol (mg/L)  0.5 0.4 0.3 0.2 0.1 0  F  Yeast Strain 3.5  1- hexanol (mg/L)  3  ■ 22 °C ■ 27 °C  2.5 2 1.5 1 0.5 0  Yeast Strain  61  Figure 14 (continued)  G 140  ■ 22 °C ■ 27 °C  Isobutanol (mg/L)  120 100 80 60 40 20 0  H  Yeast Strain  Phenylethanol (mg/L)  1.2  ■ 22 °C ■ 27 °C  1 0.8 0.6 0.4 0.2 0  I  Yeast Strain 16  Propanol (mg/L)  14  ■ 22 °C ■ 27 °C  12 10 8 6 4 2 0 Yeast Strain  Variance in each alcohol was assessed across yeast strain and across fermentation temperature (Table 14).  62  Table 14. Higher alcohols produceda by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects by volatile are shown; p values indicate the temperature effect. Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  1,3-butanediol 22 °Cb 27 °C bcde 7.216e 3.150 2.651abcde 5.585cd 2.773abcde 4.336ab 5.219bcd 3.359de abcd 3.886a 2.392 a 4.226ab 2.112 abc 4.663abc 2.317 3.818a 2.239ab e 5.893d 3.610 e 5.801d 3.554 cde 5.734cd 3.234 e 5.187bcd 3.575  pc * ** ** * *** *** ** *** * * ** *  2,3-butanediol 22 °C 27 °C bcde 0.929 2.001e 0.854abcd 1.497cd 0.913bcde 1.213abc 1.000cde 1.405bcd 0.773abc 1.050a 0.669a 1.166ab abc 0.790 1.322abcd 0.748ab 1.057a e 1.110 1.524d de 1.087 1.549d bcde 0.977 1.510cd de 1.075 1.343abcd  p * * ** * ** *** ** *** ns * ** ns  2-methyl-1-butanol 22 °C 27 °C a 1.693 2.024a 2.089bcd 2.416bcd 2.433ef 2.481cd ab 1.951 2.392bcd bc 2.053 2.674d def 2.357 2.282abc cdef 2.315 2.140ab 2.090bcd 2.074a 2.306cdef 2.309abc 2.265cde 2.484cd 2.330cdef 2.278abc 2.580f 2.299abc  p ns ns ns ns ** ns ns ns ns * ns ns  Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  3-methyl-1-butanol 22 °C 27 °C a 9.309a 8.229 bcd 10.815bcd 9.840 de 11.003cd 10.949 b 11.004cd 9.486 bc 11.843d 9.605 10.888bcd 11.518e cde 9.745ab 10.736 bcd 9.360a 9.921 bcde 10.648bcd 10.693 bcde 11.197cd 10.492 cde 10.541abc 10.828 10.209abc 11.712e  p ns ns ns ns * ns * ns ns * ns ns  Butanol 22 °C 0.185cd 0.169bc 0.169bc 0.157ab 0.263f 0.147a 0.208e 0.189cde 0.184cd 0.194de 0.171bc 0.203de  p *** *** * ** *** *** * *** *** * *** *  1-hexanol 22 °C 27 °C ab 2.206 2.520 2.413abc 2.354 2.649cd 2.448 a 2.164 2.275 2.214ab 2.600 2.499bc 2.096 2.497abc 2.110 2.351abc 2.005 2.54bcd 2.171 abc 2.487 2.393 2.533bcd 2.151 2.841d 2.187  p ns ns ns ns ns * * * ns ns ns *  Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  Isobutanol 22 °C 31.380a 50.592bc 71.731d 53.548bc 72.158d 62.119cd 44.142b 43.416ab 58.954c 56.976c 60.020cd 53.698bc  p ns ns ns ns ** ** ns * ns ns ns ns  Phenylethanol 22 °C 27 °C 0.472a 0.827 cd 0.709 0.783 0.680bcd 0.829 0.565ab 0.745 0.570ab 0.732 0.652bc 0.672 0.623bc 0.618 bc 0.632 0.641 0.645bc 0.682 0.636bc 0.746 0.713cd 0.661 0.804d 0.673  p ** ns ns ns * ns ns ns ns ns ns ns  Propanol 22 °C 12.845f 11.198de 8.654ab 8.179a 8.221a 13.487f 9.555bc 8.815ab 10.984de 9.989c 11.775e 10.258cd  p ns ns ns ns ns ns ns * ns ns ns **  a  27 °C 28.825a 53.129b 72.157d 66.743cd 110.846f 87.520e 55.413b 51.732b 56.836bc 56.554bc 58.112bc 51.769b  27 °C 0.270cd 0.317e 0.212a 0.225ab 0.517f 0.227ab 0.280cd 0.290de 0.271cd 0.254bc 0.272cd 0.261cd  27 °C 11.804e 12.111e 7.456a 7.837ab 9.086cd 14.164f 9.217cd 7.767ab 9.908d 8.589bc 11.122e 8.380abc  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  63  3.3.1.1 Production of 1,3-butanediol. The alcohol 1,3-butanediol was quantified in the headspace of all Pinot noir wine samples and found to differ with yeast strain and fermentation temperature (Figure 14, Panel A and Table 14). In the headspace of wines fermented at 22 °C, 1,3-butanediol ranged from 2.112–3.610 mg/L in wines fermented by individual Burgundian strain A1 and Burgundian mixture M1, respectively. Statistical analysis revealed significant (p≤0.05) differences in 1,3-butanediol production, including the differentiation of the low producers A1 and A3 (2.112–2.239 mg/L) from the high producers AMH, M3, RA17, M2, M4 and M1 (3.150–3.610 mg/L). Additionally, mixed Burgundian strains M1, M2, and M4 contained significantly more 1,3-butanediol than all of the individual Burgundian strains (A1, A2 and A3). When the fermentation temperature was increased to 27 °C, the amount of 1,3-butanediol present in the headspace increased significantly in all strains and mixtures and ranged from 3.818–7.216 mg/L in wines fermented by individual Burgundian strain A3 and industrial strain AMH, respectively. Many significant (p≤0.05) differences existed among yeast strains based on the production of 1,3-butanediol at this temperature. For example, the low producers A3, RC212, A1 and BGY (3.818–4.336 mg/L) differed from the moderate producers M3, M2 and M1 (5.734–5.893 mg/L), which differed from the high producer AMH (7.216 mg/L). At this temperature, Burgundian mixtures M1 and M2 produced significantly more 1,3-butanediol than all of the individual Burgundian strains (A1, A2 and A3).  3.3.1.2 Production of 2,3-butanediol. The alcohol 2,3-butanediol was quantifiable in the headspace of all wines assessed at both fermentation temperatures (Figure 14, Panel B,  64  and Table 14). At 22 °C, the headspace concentration of 2,3-butanediol ranged from 0.669–1.11 mg/L in wines fermented by individual Burgundian strain A1 and Burgundian mixture M1, respectively. Some significant (p≤0.05) differences in 2,3-butanediol production existed among yeast strains at this temperature, including the differentiation of the low producing strains A1, A3, RC212 and A2 (0.669–0.790 mg/L), from the high producing strains M4, M2, and M1 (1.075–1.11 mg/L). When fermentation was conducted at 27 °C, the concentration of volatile 2,3-butanediol increased in all samples. This difference was significant in wines fermented by all strains except Burgundian mixtures M1 and M4. At 27 °C, concentrations of 2,3-butanediol ranged from 1.050–2.001 mg/L in wines fermented by industrial strains RC212 and AMH, respectively. Statistical analysis significantly (p≤0.05) differentiated the low producers RC212, A3 and A1 (1.050–1.166 mg/L), from the moderate producers AWRI 796, M3, M1 and M2 (1.497–1.549 mg/L), which were further differentiated from the high producer AMH (2.001 mg/L). The significant differences observed between the mixed and individual Burgundian strains at 22 °C were less pronounced at 27 °C, where Burgundian mixtures M1, M2, and M3 produced significantly more 2,3-butanediol than individual Burgundian strains A1 and A3, but not A2.  3.3.1.3 Production of 2-methyl-1-butanol. The compound 2-methyl-1-butanol was quantified in all wine samples fermented at both temperatures and found to differ based on yeast strain and, to a lesser extent, fermentation temperature (Figure 14, Panel C and Table 14). When fermentation was conducted at 22 °C, volatile 2-methyl-1-butanol  65  ranged from 1.693–2.580 mg/L in wines fermented by industrial strain AMH and Burgundian mixture M4, respectively. Statistical analysis revealed some significant (p≤0.05) differences between yeast strains, including the differentiation of the low producers AMH and RA17 (1.693–1.951 mg/L) and the high producers M1, A2, M3, A1 and BGY (2.306–2.580). The concentration of 2-methyl-1-butanol did not change significantly between wines fermented at 22 °C and wines fermented at 27 °C, except in wines fermented by industrial strain RC212 and mixed strain M2, where volatile 2-methyl-1-butanol content increased with fermentation temperature. When fermentation was conducted at 27 °C, 2-methyl-1-butanol ranged from 2.024–2.674 mg/L in wines fermented by industrial strains AMH and RC212, respectively. Statistical analysis detected significant (p≤0.05) differences in 2-methyl-1-butanol among yeast strains, which included the differentiation between low producers AMH, A3 and A2 (2.024–2.140 mg/L), and high producers BGY, M2 and RC212 (2.481–2.674 mg/L).  3.3.1.4 Production of 3-methyl-1-butanol. The volatile alcohol 3-methyl-1-butanol was quantified in all wine samples fermented at both temperatures, which revealed many significant (p≤0.05) differences among yeast strains (Figure 14, Panel D and Table 14). When fermentation was conducted at 22 °C, volatile 3-methyl-1-butanol ranged from 8.229–11.712 mg/L in wines fermented by industrial strain AMH and Burgundian mixture M4, respectively. Some significant (p≤0.05) differences in 3-methyl-1-butanol production existed among yeast strains, which differentiated the low producer AMH (8.229 mg/L) from the moderate producers RA17, RC212, AWRI 796 and A3  66  (9.486–9.921 mg/L), which were also differentiated from the high producers A1 and M4 (11.518–11.712 mg/L). When the fermentation temperature was increased to 27 °C, strains RC212 and mixture M2 produced significantly (p≤0.05) more volatile 3-methyl-1-butanol, while strain A2 produced significantly less 3-methyl-1-butanol compared to when fermentation was conducted at 22 °C. At 27 °C, the concentration of volatile 3-methyl-1-butanol ranged from 9.309–11.843 mg/L in wines fermented by industrial strains AMH and RC212, respectively. Fewer significant differences in the concentration of volatile 3-methyl-1-butanol were evident at this fermentation temperature; however, the low producers AMH, A3 and A2 (9.309–9.745 mg/L) differed significantly from the high producers BGY, RA17, M2 and RC212 (11.003–11.843 mg/L).  3.3.1.5 Production of butanol. Butanol was quantified in all wine samples fermented at 22 °C and 27 °C, which revealed many significant (p≤0.05) differences among yeast strains and between fermentation temperatures (Figure 14, Panel E and Table 14). In wines fermented at 22 °C, butanol ranged from 0.147–0.263 mg/L in wines fermented by individual Burgundian strain A1 and industrial strain RC212, respectively. Many significant (p≤0.05) differences in butanol production existed among yeast strains, which allowed the differentiation of the low producers A1 and RA17 (0.147–0.157 mg/L) from the moderate producers A3, M2, M4 and A2 (0.189–0.208 mg/L), and the high producer RC212 (0.263 mg/L). When fermentation was conducted at 27 °C, all wines contained significantly (p≤0.05) more butanol than when fermentation was conducted at 22 °C. When  67  fermentation was conducted at 27 °C, butanol ranged from 0.212–0.517 mg/L in wines fermented by industrial strains BGY and RC212, respectively. More significant (p≤0.05) differences occurred among yeast strains at this temperature. Therefore, the low producers BGY and RA17 (0.212–0.225 mg/L) differed from the moderate producers M4, AMH, M1, M3 and A2 (0.261–0.280 mg/L), which differed from the moderate-high producer AWRI 796 (0.317 mg/L), which differed from the highest producer RC212 (0.517 mg/L).  3.3.1.6 Production of 1-hexanol. The alcohol 1-hexanol was quantified and analyzed in wine samples fermented at both temperatures, which revealed significant (p≤0.05) differences in production among yeast strains (Figure 14, Panel F and Table 14). When fermentation was conducted at 22 °C, volatile 1-hexanol ranged from 2.164–2.841 mg/L in wines fermented by industrial strain RA17 and Burgundian mixture M4, respectively. Statistical analysis revealed significant (p≤0.05) differences among yeast strains, including the differentiation of the low producers RA17, AMH and RC212 (2.164–2.214 mg/L), from the high producers M1, BGY and M4 (2.540–2.841 mg/L). When fermentation was conducted at 27 °C, the concentration of 1-hexanol in wines fermented with Burgundian strains A1, A2 and A3 and mixture M4 decreased significantly compared to wines conducted at 22 °C. The concentration of 1-hexanol ranged from 2.010–2.600 mg/L in wine fermented at 27 °C by individual Burgundian strain A3 and industrial strain RC212, respectively. No significant differences in 1-hexanol production existed among yeast strains at this fermentation temperature.  68  3.3.1.7 Production of isobutanol. Isobutanol was quantified in the headspace of wine samples fermented at 22 °C and 27 °C and varied significantly (p≤0.05) among yeast strains and between fermentation temperatures (Figure 14, Panel G and Table 14). When fermentation was conducted at 22 °C, isobutanol ranged from 31.380–72.158 mg/L in wines fermented by industrial strains AMH and RC212, respectively. Statistical analysis significantly (p≤0.05) differentiated the low producers AMH and A3 (31.380–43.416 mg/L) from the moderate producers M2 and M1 (56.976–58.954 mg/L), which differed from the high producers BGY and RC212 (71.731–72.158 mg/L). When the fermentation temperature was increased to 27 °C, isobutanol ranged from 28.825–110.846 mg/L in wines fermented by industrial strains AMH and RC212, respectively. Industrial strain RC212 and Burgundian strains A1 and A3 produced significantly more isobutanol when fermentation was conducted at the higher temperature. Among the yeast strains, statistical analysis differentiated the low isobutanol producer AMH (28.825 mg/L) from the medium producers A3, M4, A2, M2, M1 and M3 (51.732–58.112 mg/L), which also differed from the high producers BGY, A1 and RC212 (72.157–110.846 mg/L). Because of the wide range of isobutanol production observed at this fermentation temperature, the high isobutanol producers BGY, A1 and RC212 also significantly differed from each other.  3.3.1.8 Production of phenylethanol. Phenylethanol was quantified in all wine samples fermented at 22 °C and 27 °C and varied across yeast strain and fermentation temperature (Figure 14, Panel H and Table 14). When fermentation was conducted at 22 °C, phenylethanol ranged from 0.472–0.804 mg/L in wines fermented by industrial strain  69  AMH and Burgundian mixture M4, respectively. Several significant (p≤0.05) differences between yeast strains were evident, including differences between the low producer AMH (0.472 mg/L), the moderate producers A2, A3, M2, M1 and A1 (0.623–0.652 mg/L), and the high producer M4 (0.804 mg/L). Production of volatile phenylethanol was similar when fermentation was conducted at 27 °C and only differed significantly (p≤0.05) between the two fermentation temperatures in wines fermented by industrial strains AMH and RC212. When fermentation was conducted at 27 °C, phenylethanol ranged from 0.620–0.830 mg/L in wines fermented by individual Burgundian strain A2 and industrial strains AMH and BGY, respectively. No significant (p≤0.05) differences in phenylethanol production existed between yeast strains when fermentation was conducted at this temperature.  3.3.1.9 Production of propanol. Propanol was quantified in the headspace of wine samples fermented at 22 °C and 27 °C by the eight pure and four mixed yeast strains and found to vary across yeast strain and temperature (Figure 14, Panel I and Table 14). When fermentation was conducted at 22 °C, propanol ranged from 8.179–13.487 mg/L in wines fermented by industrial strain RA17 and individual Burgundian strain A1, respectively. Many significant (p≤0.05) differences were evident between strains, including the differentiation of the low producers RA17, RC212, BGY and A3 (8.179– 8.815 mg/L), the low-moderate producer M2 (9.989 mg/L), the moderate producers M1, AWRI 796 and M3 (10.984–11.775 mg/L), and the high producers AMH and A1 (12.845–13.487 mg/L).  70  When the fermentation temperature was increased to 27 °C, wines fermented by Burgundian strain A3 and mixture M4 contained significantly less propanol than when fermentation was conducted at 22 °C. Following fermentation at 27 °C, propanol ranged from 7.456–14.164 mg/L in wines fermented by industrial strain BGY and individual Burgundian strain A1, respectively. Statistical analysis revealed significant differences in propanol production between the low producers BGY, A3 and RA17 (7.456–7.837 mg/L), the moderate producers RC212, A2 and M1 (9.086–9.908 mg/L), the moderatehigh producers AMH and AWRI 796 (11.804–12.111 mg/L), and the high producer A1 (14.164 mg/L).  3.3.2 Production of ethyl esters varied according to yeast strain and fermentation temperature Ethyl esters represented the second most numerous compound class identified in the volatile compound analysis of Pinot noir wine fermented at 22 °C and 27 °C. Seven identified ethyl esters differed in the headspace of the Pinot noir wine samples (Figure 15).  71  A 4  Ethyl butanoate (mg/L)  3.5 3  ■ 22 °C ■ 27 °C  2.5 2 1.5 1 0.5 0  B 0.045 Ethyl decanoate (mg/L)  0.04 0.035  ■ 22 °C ■ 27 °C  0.03 0.025 0.02 0.015 0.01 0.005 0  C 0.08  Ethyl hexanoate (mg/L)  0.07 0.06  ■ 22 °C ■ 27 °C  0.05 0.04 0.03 0.02 0.01 0  Figure 15. Concentrations (mg/L) of ethyl esters produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains. Mean values are shown (n=3); error bars indicate +/- standard deviation. CALPN indicates the industrial Calona Pinot noir sample.  72  Figure 15 (continued)  D 25 Ethyl lactate (mg/L)  20  ■ 22 °C ■ 27 °C  15 10 5 0  E 3 Ethyl lactate (mg/L)  2.5  ■ 22 °C ■ 27 °C  2 1.5 1 0.5 0  F 0.012  Ethyl laurate (mg/L)  0.01  ■ 22 °C ■ 27 °C  0.008 0.006 0.004 0.002 0  73  Figure 15 (continued)  G Ethyl octanoate (mg/L)  0.08 0.07 0.06  ■ 22 °C ■ 27 °C  0.05 0.04 0.03 0.02 0.01 0  H 0.18 Ethyl palmitate (mg/L)  0.16 0.14  ■ 22 °C ■ 27 °C  0.12 0.1 0.08 0.06 0.04 0.02 0  Variation in the production of ethyl esters was assessed across yeast strain and fermentation temperature (Table 15).  74  Table 15. Ethyl esters (mg/L) produceda by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects by volatile are shown; p values indicate the temperature effect. Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  Ethyl butanoate 22 °Cb 27 °C ab 1.707a 2.689 cde 2.327bc 3.232 a 2.193abc 2.368 bcd 2.180abc 2.862 cde 2.116abc 3.181 3.131e 3.262de a 2.107ab 2.286 e 2.893de 3.387 de 3.080e 3.275 e 2.626cde 3.316 de 3.117e 3.252 2.494bcd 2.826bc  Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4 Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4 a  pc *** ns ns *** * ns ns * ns ns ns ns  Ethyl decanoate 22 °C 27 °C d 0.031 0.020ab cd 0.025 0.024abc bc 0.021 0.017a bc 0.022 0.024abc c 0.024 0.023abc 0.025cd 0.034e ab 0.015 0.031cde c 0.023 0.026bcde c 0.024 0.030cde c 0.024 0.024abc cd 0.025 0.033de 0.013a 0.025bcd  Ethyl lactate 22 °C 27 °C 0.682b 0.548a ab 0.882cd 0.847 c 0.671b 1.773 0.726b 0.631ab ab 0.939d 0.753 b 0.936d 0.973 ab 0.671b 0.749 ab 0.719b 0.724 ab nqa 0.817 ab 0.799bc 0.780 0.766bc 0.857ab ab 0.698b 0.902  p * ns ns ns ** ns ns ns *** ns ns ns  Ethyl laurate 22 °C 27 °C 0.006def 0.004a 0.007ef 0.008bcd bcd 0.004 0.004a 0.004abc 0.005ab 0.005cd 0.006ab f 0.007 0.009cd ab 0.002 0.007bc abc 0.004 0.006ab cd 0.005 0.009d cd 0.004 0.007bcd 0.005cde 0.009d 0.002a 0.006b  Ethyl palmitate 22 °C 27 °C 0.054a 0.045a ef 0.133d 0.118 0.083b 0.072abc abc 0.101bc 0.068 def 0.143d 0.103 f 0.145d 0.123 ab 0.101bc 0.056 cde 0.101bc 0.090 ef 0.139d 0.105 0.120cd 0.108ef ef 0.127cd 0.114 bcd 0.090b 0.077  p ns ns ns * ns ns * ns ns ns ns ns  p ** ns ns ns ns * *** ns ns ns ns ns  Ethyl hexanoate 22 °C 27 °C d 0.060 0.040ab cd 0.059 0.043bcd ab 0.046 0.036a abc 0.051 0.041abc abc 0.052 0.040abc 0.054cd 0.050d 0.046ab 0.046cd 0.053bcd 0.041abc 0.054cd 0.048d 0.053bcd 0.042abc 0.054cd 0.050d 0.045a 0.041abc  p *** * * ** ns ns ns ** ns ns ns ns  p ** ns ns ns ns ** ** * * ns * *  Ethyl octanoate 22 °C 27 °C 0.068d 0.035a 0.053bc 0.040ab 0.046abc 0.033a 0.047abc 0.040ab 0.046abc 0.035a 0.057cd 0.052cd 0.043ab 0.052cd 0.055bc 0.040ab 0.057cd 0.048bcd 0.056cd 0.040ab 0.057cd 0.055d 0.036a 0.042abc  p *** ns ns ns ns ns *** * ns ns ns ns  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  75  3.3.2.1 Production of ethyl butanoate. Ethyl butanoate was quantified in Pinot noir wine fermented at 22 °C and 27 °C by the eight pure and four mixed S. cerevisiae strains (Figure 15, Panel A and Table 15). When the fermentation was conducted at 22 °C, ethyl butanoate ranged from 2.286–3.387 mg/L in wines fermented by individual Burgundian strains A2 and A3, respectively. Statistical analysis revealed significant (p≤0.05) differences in the production of ethyl butanoate, including the difference between low producers A2, AMH and BGY (2.286–2.689 mg/L), and the high producers AWRI 796, M3, A1, M1, M2 and A3 (3.232–3.387 mg/L). When the fermentation temperature was increased to 27 °C, strains AMH, RA17, RC212, and A3 produced significantly (p≤0.05) less ethyl butanoate than at 22 °C. At 27 °C, volatile ethyl butanoate ranged from 1.707–3.131 mg/L in wines fermented by industrial strain AMH and individual Burgundian strain A1. At this temperature, statistical analysis revealed significant (p≤0.05) differences among yeast strains, including the differences between the low producers AMH, A2, RC212, RA17 and BGY (1.707–2.193 mg/L) and the high producers A3, M1, M3 and A1 (2.893–3.131 mg/L).  3.3.2.2 Production of ethyl decanoate. The concentration of ethyl decanoate in Pinot noir wine fermented at 22 °C and 27 °C by eight pure and four mixed S. cerevisiae strains varied according to yeast strain and fermentation temperature (Figure 15, Panel B and Table 15). When fermentation was conducted at 22 °C, ethyl decanoate ranged from 0.013–0.031 mg/L in wines fermented by Burgundian mixture M4 and industrial strain AMH, respectively. Statistical analysis revealed some significant (p≤0.05) differences among yeast strains, including the differences between the low producers M4 and A2  76  (0.013–0.015 mg/L) and the high producers AWRI 796, A1, M3 and AMH (0.025–0.031 mg/L). When the fermentation temperature was increased to 27 °C, strain AMH produced significantly less ethyl decanoate, while strains A1 and A2 produced significantly more compared to fermentation at 22 °C. Ethyl decanoate production at this temperature ranged from 0.017–0.034 mg/L in wines fermented by industrial strain BGY and individual Burgundian strain A1, respectively. More significant (p≤0.05) differences in ethyl decanoate production existed among yeast strains at 27 °C, which allowed the differentiation of the low producers BGY, AMH, RC212, AWRI 796 and RA17 (0.017–0.024 mg/L) from the high producers M3 and A1 (0.033–0.034 mg/L).  3.3.2.3 Production of ethyl hexanoate. The concentration of ethyl hexanoate in Pinot noir wine fermented by the eight pure and four mixed strains of S. cerevisiae following fermentation at 22 °C and 27 °C varied according to yeast strain and fermentation temperature (Figure 15, Panel C and Table 15). When fermentation was conducted at 22 °C, ethyl hexanoate ranged from 0.045–0.060 mg/L in wines fermented by Burgundian mixture M4 and industrial strain AMH, respectively. Statistical analysis revealed significant (p≤0.05) differences in ethyl hexanoate production among yeast strains, which included the differentiation of the low producers M4, BGY and A2 (0.045–0.046 mg/L) from the high producer AMH (0.060 mg/L). When the fermentation temperature was increased to 27 °C, wines fermented by strains AMH, AWRI 796, BGY, RA17, and A3 contained significantly (p≤0.05) less ethyl hexanoate than they did following fermentation at 22 °C. At 27 °C, ethyl hexanoate  77  ranged from 0.036 mg/L in wine fermented by industrial strain BGY to 0.050 mg/L in wines fermented by Burgundian mixture M3 or individual Burgundian strain A1. Statistical analysis revealed many statistically significant (p≤0.05) differences among yeast strains, including the differentiation of low producers BGY, AMH, RC212, RA17, A3, M4 and M2 (0.036–0.042 mg/L) from high producers M1, A1 and M3 (0.048–0.050 mg/L).  3.3.2.4 Production of ethyl lactate. Ethyl lactate was quantified in Pinot noir wines fermented by the eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C and found to vary according to yeast strain and fermentation temperature (Figure 15, Panels D-E and Table 15). When fermentation was conducted at 22 °C, ethyl lactate ranged from 0.548–1.773 mg/L in wines fermented by industrial strains AMH and BGY, respectively. Statistical analysis revealed significant (p≤0.05) differences among select yeast strains. The low producer AMH (0.548 mg/L) differed significantly from the moderate producer A1 (0.973 mg/L), which differed from the high producer BGY (1.773 mg/L). The remaining strains fell between low and moderate ethyl lactate production and only differed significantly from the high producer BGY. When the fermentation temperature was increased to 27 °C, strain AMH and strain RC212 produced significantly more ethyl lactate while mixture M1 produced significantly less. Following fermentation at 27 °C, ethyl lactate ranged from below quantification to 0.939 mg/L in wines fermented by Burgundian mixture M1 and industrial strain RC212, respectively. Statistical analysis revealed significant (p≤0.05) differences among yeast strains, which included the differentiation of the low producer  78  M1 (not quantifiable) from the moderate producers BGY, A2, AMH, M4, A3 and RA17 (0.671–0.726 mg/L), which were differentiable from the high producers A1 and RC212 (0.936–0.939 mg/L).  3.3.2.5 Production of ethyl laurate. Ethyl laurate was quantified in Pinot noir wine fermented by the eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C, which varied according to yeast strain and fermentation temperature (Figure 15, Panel F and Table 15). When fermentation was conducted at 22 °C, ethyl laurate ranged from 0.002–0.007 mg/L in wines fermented by Burgundian mixture M4 and individual Burgundian strain A1, respectively. Statistical analysis revealed many significant differences among yeast strains, including the differentiation of the low producers A2, A3 and RA17 (0.002–0.004 mg/L), from the high producers AMH, AWRI 796 and A1 (0.006–0.007 mg/L). When the fermentation temperature was increased to 27 °C, volatile ethyl laurate production increased significantly in strains A1, A2, A3, M1, M3 and M4 and decreased significantly in strain AMH. At 27 °C, volatile ethyl laurate production ranged from 0.004–0.008 mg/L in wines fermented by industrial strain BGY and Burgundian mixture M1, respectively. Statistical analysis revealed many significant (p≤0.05) differences among yeast strains, including the differentiation of the low producers AMH and BGY (0.004 mg/L) from the high producers M2, AWRI 796, A1, M1 and M3 (0.007–0.009 mg/L).  79  3.3.2.6 Production of ethyl octanoate. The concentration of ethyl octanoate in Pinot noir wines fermented by eight pure and four mixed strains of S. cerevisiae at 22 °C and 27 °C varied according to yeast strain and temperature (Figure 15, Panel G and Table 15). When fermentation was conducted at 22 °C, ethyl octanoate ranged from 0.036–0.068 mg/L in wines fermented by Burgundian mixture M4 and industrial strain AMH, respectively. Statistical analysis revealed many significant (p≤0.05) differences based on yeast strain, which included the differentiation of the low producers M4 and A2 (0.036–0.043 mg/L) from the high producer AMH (0.068 mg/L). When the fermentation temperature was increased to 27 °C, the production of ethyl octanoate significantly (p≤0.05) decreased in strain AMH and strain A3 but significantly increased in strain A2. At this fermentation temperature, ethyl octanoate ranged from 0.033–0.055 mg/L in wines fermented by industrial strain BGY and Burgundian mixture M3, respectively. Statistical analysis revealed many significant differences among yeast strains, including the differentiation between low producers BGY, AMH, RC212, AWRI 796 and RA17 (0.033–0.040 mg/L) from high producers A1, A2 and M3 (0.052–0.055 mg/L).  3.3.2.7 Production of ethyl palmitate. Ethyl palmitate was quantified in Pinot noir wine fermented by eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C and varied according to yeast strain and fermentation temperature (Figure 15, Panel H and Table 15). Following fermentation at 22 °C, ethyl palmitate ranged from 0.045–0.123 mg/L in wines fermented by industrial strain AMH and individual Burgundian strain A1, respectively. Statistical analysis revealed many significant (p≤0.05) differences in ethyl  80  palmitate production, including differences between the low producers AMH, BGY and RA17 (0.045–0.072 mg/L) from the high producers RC212, M1, M2, M3, AWRI 796 and A1 (0.103–0.123 mg/L). When the fermentation temperature was increased to 27 °C, strains RA17 and A2 produced significantly (p≤0.05) more ethyl palmitate. At this temperature, ethyl palmitate ranged from 0.054–0.145 mg/L in wines fermented by industrial strain AMH and individual Burgundian strain A1, respectively. Statistical analysis revealed many significant differences in ethyl palmitate production between yeast strains, including the differentiation of the low producer AMH (0.054 mg/L) from the moderate producers BGY and M4 (0.083–0.089 mg/L), which were also differentiable from the high producers AWRI 796, M1, RC212 and A1 (0.133–0.145 mg/L).  3.3.3 Production of acetate esters varied according to yeast strain selection and fermentation temperature Acetate esters were the third most numerous compound class identified in the volatile compound analysis of Pinot noir wine fermented at 22 °C and 27 °C. Five distinct acetate esters were quantifiable at the two fermentation temperatures (Figure 16).  81  A 16  Ethyl acetate (mg/L)  14 12  ■ 22 °C ■ 27 °C  10 8 6 4 2 0  B 0.04  Hexyl acetate (mg/L)  0.035 0.03  ■ 22 °C ■ 27 °C  0.025 0.02 0.015 0.01 0.005 0  C 0.35 Isoamyl acetate (mg/L)  0.3  ■ 22 °C ■ 27 °C  0.25 0.2 0.15 0.1 0.05 0  Figure 16. Concentrations (mg/L) of acetate esters produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains. Mean values are shown (n=3); error bars indicate +/- standard deviation. CALPN indicates the industrial Calona Pinot noir sample.  82  Figure 16 (continued)  D Isobutyl acetate (mg/L)  0.0035 0.003 0.0025  ■ 22 °C ■ 27 °C  0.002 0.0015 0.001 0.0005 0  E  Methyl acetate (mg/L)  2.5 2  ■ 22 °C ■ 27 °C  1.5 1 0.5 0  Variance in the production of each acetate ester was assessed across yeast strain and across fermentation temperature (Table 16).  83  Table 16. Acetate esters (mg/L) produceda by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects by volatile are shown; p values indicate the temperature effect in each yeast strain. Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  Ethyl acetate 22 °Cb 27 °C bc 8.968ab 10.039 10.191bcd 9.889bc bc 9.636abc 10.083 bc 10.111bcd 9.944 a 8.615a 8.824 c 11.271de 10.467 ab 10.342bcd 9.335 9.342ab 10.077bc c 11.471de 10.613 c 10.994cde 10.850 c 11.807e 10.599 c 9.751abc 10.360  Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  Isobutyl acetate 22 °C 27 °C a 0.00071a 0.00091 ab 0.00137b 0.00105 de 0.00187cd 0.00148 bcd 0.00194cde 0.00125 e 0.00202cde 0.00158 e 0.00286f 0.00155 0.00165bc 0.00096a abc 0.00159bc 0.00111 cde 0.00215de 0.00133 cde 0.00176bcd 0.00133 de 0.00235e 0.00140 ab 0.00166bc 0.00104  a  pc ns ** ns * ns ns * ns ns * ns ns p ** ns ns ns ** ns ** ns ** * ** ns  Hexyl acetate 22 °C 27 °C 0.029 0.017a 0.022 0.020ab 0.016 0.018a 0.027 0.027abcd 0.022 0.016a 0.024 0.036d 0.015 0.033cd 0.025 0.025abcd 0.027 0.030bcd 0.027 0.020ab 0.027 0.032cd 0.016 0.024abc  p ns ns ns ns ns ns *** ns ns ns ns ns  Methyl acetate 22 °C 27 °C a 0.889 0.784a cd 1.245 1.041bc bc 1.105 1.077bc ab 0.996 0.932ab d 1.304 1.126cd d 1.311 1.175cde 0.946a 1.017bc cd 1.196 1.006bc cd 1.183 1.252de cd 1.198 1.335e cd 1.192 1.333e cd 1.172 1.065bc  p ns ns ns ns ns ns ns * ns ns ns ns  Isoamyl acetate 22 °C 27 °C abcd 0.203 0.148a 0.188abc 0.213bcd 0.171a 0.184abc bcd 0.228 0.255def abc 0.190 0.178ab cd 0.235 0.320g a 0.167 0.267defg 0.211abcd 0.222bcd 0.234cd 0.291efg d 0.237 0.225bcd d 0.241 0.305fg ab 0.182 0.241cde  p ns ns ns ns ns * *** ns ns ns ns ns  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  84  3.3.3.1 Production of ethyl acetate. Ethyl acetate was quantified in Pinot noir wine fermented by eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C and found to vary according to yeast strain and fermentation temperature (Figure 16, Panel A and Table 16). When fermentation was conducted at 22 °C, volatile ethyl acetate production ranged from 8.824–10.850 mg/L in wines fermented by industrial strain RC212 and Burgundian mixture M2, respectively. Statistical analysis revealed significant (p≤0.05) differences in ethyl acetate production among yeast strains, including the differentiation of the low producers RC212 and A2 (8.824–9.335 mg/L) from the high producers M4, A1, M3, M1, and M2 (10.360–10.850 mg/L). When the fermentation temperature was increased to 27 °C, strains AWRI 796, RA17, A2, and mixture M2 produced significantly (p≤0.05) more ethyl acetate than at 22 °C. At this temperature, ethyl acetate ranged from 8.615–11.807 mg/L in wines fermented by industrial strain RC212 and Burgundian mixture M3, respectively. Statistical analysis revealed many differences in ethyl acetate production among yeast strains, including the differentiation of the low producers RC212, AMH and A3 (8.615–9.342 mg/L) from the high producers A1, M1 and M3 (11.271–11.807 mg/L).  3.3.3.2 Production of hexyl acetate. Hexyl acetate was quantified in Pinot noir wine fermented at 22 °C and 27 °C by eight pure and four mixed strains and found to vary according to yeast strain and fermentation temperature (Figure 16, Panel B and Table 16). At 22 °C, hexyl acetate ranged from 0.015–0.029 mg/L in wines fermented by individual Burgundian strain A2 and industrial strain AMH, respectively. Significant differences did not exist among yeast strains at this temperature.  85  When the fermentation temperature was increased to 27 °C, strain A2 produced significantly more hexyl acetate than at 22 °C. At this temperature, hexyl acetate ranged from 0.016–0.036 mg/L in wines fermented by industrial strain RC212 and individual Burgundian strain A1, respectively. Statistical analysis revealed significant (p≤0.05) differences in hexyl acetate production among yeast strains, including the differences between the low producers RC212, AMH, BGY, AWRI 796 and M2 (0.016–0.020 mg/L) and the high producers M3, A2 and A1 (0.032–0.036 mg/L).  3.3.3.3 Production of isoamyl acetate. Isoamyl acetate was quantified in Pinot noir wine fermented at 22 °C and 27 °C by eight pure and four mixed S. cerevisiae strains and found to vary according to yeast strain and fermentation temperature (Figure 16, Panel C and Table 16). At 22 °C, isoamyl acetate ranged from 0.167–0.241 mg/L in wines fermented by individual Burgundian strain A2 and Burgundian mixture M3, respectively. Many significant (p≤0.05) differences existed among yeast strains at this fermentation temperature, including differences between the low producers A2, BGY and M4 (0.167–0.182 mg/L), and the high producers M2 and M3 (0.237–0.241 mg/L). When the fermentation temperature was increased to 27 °C, strains A1 and A2 produced significantly more isoamyl acetate than when fermentation was conducted at 22 °C. At 27 °C, isoamyl acetate ranged from 0.148–0.320 mg/L in wines fermented by industrial strain AMH and individual Burgundian strain A1, respectively. The significant (p≤0.05) differences among yeast strains at this temperature included the differences between the low producers AMH, RC212 and BGY (0.148–0.184 mg/L) and the high producers A2, M1, M3 and A1 (0.267–0.320 mg/L).  86  3.3.3.4 Production of isobutyl acetate. Very small amounts of isobutyl acetate were quantified in Pinot noir wine fermented by eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C and found to differ according to yeast strain and fermentation temperature (Figure 16, Panel D and Table 16). At 22 °C, isobutyl acetate ranged from 0.00091–0.00158 mg/L in wines fermented by industrial strains AMH and RC212, respectively. Many significant (p≤0.05) differences in isobutyl acetate production existed among yeast strains, including the difference between the low producers AMH, A2, M4 and AWRI 796 (0.00091–0.00105 mg/L) and the high producers BGY, RC212, M3 and A1 (0.00187–0.00286 mg/L). When the fermentation temperature was increased to 27 °C, strains RC212, A2, M1, M2, and M3 produced significantly more isobutyl acetate compared to fermentation at 22 °C while strain AMH produced significantly less. At 27 °C, isobutyl acetate ranged from 0.00071–0.00286 mg/L in wines fermented by industrial strain AMH and individual Burgundian strain A1, respectively. Many significant differences in isobutyl acetate production existed between yeast strains, including the difference between the very low producer AMH (0.00071 mg/L), the low producer AWRI 796 (0.00137 mg/L), the moderate producers RA17, RC212, M1 and M3 (0.00194–0.00235 mg/L), and the high producer A1 (0.00286 mg/L).  3.3.3.5 Production of methyl acetate. Methyl acetate was quantified in Pinot noir wine fermented by eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C and found to vary according to yeast strain and fermentation temperature (Figure 16, Panel E  87  and Table 16). At 22 °C, methyl acetate ranged from 0.889–1.311 mg/L in wines fermented by industrial strain AMH and individual Burgundian strain A1, respectively. Some significant (p≤0.05) differences in methyl acetate production existed between yeast strains at this temperature, including the differences between the low producers AMH, A2 and RA17 (0.889–0.996 mg/L) and the high producers M4, M1, M3, A3, M2, AWRI 796, RC212 and A1 (1.172–1.311 mg/L). When the fermentation temperature was increased to 27 °C, strain A3 produced significantly less methyl acetate than at 22 °C, while other strains did not differ in methyl acetate production between the two temperatures. When fermentation was conducted at 27 °C, methyl acetate ranged from 0.784–1.335 mg/L in wines fermented by industrial strain AMH and Burgundian mixture M2, respectively. Many significant differences in methyl acetate production existed between yeast strains, including the difference between the low producers AMH and RA17 (0.784–0.932 mg/L) and the high producers A1, M1, M2 and M3 (1.175–1.335 mg/L).  3.3.4 Production of aldehydes varied according to yeast strain selection and fermentation temperature Two aldehydes, acetaldehyde and benzaldehyde, were quantified in the headspace analysis of Pinot noir wines fermented by eight pure and four mixed S. cerevisiae strains at 22 °C and 27 °C; the concentration of aldehydes varied according to yeast strain and fermentation temperature (Figure 17).  88  A 2.5 Acetaldehyde (mg/L)  2  ■ 22 °C ■ 27 °C  1.5 1 0.5 0  B 0.08 Benzaldehyde (mg/L)  0.07 0.06  ■ 22 °C ■ 27 °C  0.05 0.04 0.03 0.02 0.01 0  Figure 17. Concentrations (mg/L) of aldehydes produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains. Mean values are shown (n=3); error bars indicate +/- standard deviation. CALPN indicates the industrial Calona Pinot noir sample. Both compounds varied significantly across yeast strain and across fermentation temperature (Table 17).  89  Table 17. Aldehydes (mg/L) produceda by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects by volatile are shown; p values indicate the temperature effect. Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4 a  Acetaldehyde 22 °Cb 27 °C cd  1.026 0.504a 0.670a 1.339e 0.871bc 0.839b 0.979bcd 0.901bcd 0.923bcd 1.009cd 1.049d 0.937bcd  bcd  0.967 0.614a 0.690ab 1.357ef 0.803ab 0.947bcd 1.156cde 0.737ab 1.377ef 1.538f 1.245def 0.939bc  p  c  ns * ns ns ns ns ns ns *** ns ns ns  Benzaldehyde 22 °C 27 °C ab  0.029 0.056fg 0.045de 0.063g 0.050ef 0.034bc 0.026a 0.025a 0.039cd 0.037cd 0.043de 0.050ef  abcd  0.056 0.061bcdef 0.056abcd 0.068f 0.063def 0.055abc 0.053ab 0.052a 0.064ef 0.062cdef 0.057abcde 0.059abcde  p * ns * ns ** *** ** *** *** ** * ns  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, ***indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  3.3.4.1 Production of acetaldehyde. Acetaldehyde was quantified in Pinot noir wine fermented at 22 °C and 27 °C by the eight pure and four mixed S. cerevisiae strains and found to vary according to yeast strain and fermentation temperature (Figure 17, Panel A and Table 17). At 22 °C, acetaldehyde ranged from 0.504–1.339 mg/L in wines fermented by industrial strains AWRI 796 and RA17, respectively. Some significant differences in acetaldehyde production occurred among yeast strains, including the difference between the low producers AWRI 796 and BGY (0.504–0.670 mg/L), the moderate producers A1, RC212, A3, M1, M4 and A2 (0.839–0.979 mg/L) and the high producer RA17 (1.339 mg/L).  90  When the fermentation temperature was increased to 27 °C, acetaldehyde production increased significantly across temperature in strains AWRI 796 and mixture M1. At 27 °C, acetaldehyde ranged from 0.614–1.538 mg/L in wines fermented by industrial strain AWRI 796 and Burgundian mixture M2, respectively. Some significant differences in acetaldehyde production occurred among yeast strains, including the difference between the low producers AWRI 796, BGY, A3 and RC212 (0.614–0.803 mg/L) and the high producers M3, RA17, M1 and M2 (1.245–1.538 mg/L).  3.3.4.2 Production of benzaldehyde. Benzaldehyde was quantified in Pinot noir wine following fermentation at 22 °C and 27 °C and found to vary according to yeast strain and fermentation temperature (Figure 17, Panel B and Table 17). At 22 °C, benzaldehyde ranged from 0.025–0.063 mg/L in wines fermented by individual Burgundian strain A3 and industrial strain RA17, respectively. Significant (p≤0.05) differences in benzaldehyde production existed among yeast strains, including the differences between the low producers A3, A2 and AMH (0.025–0.029 mg/L), the moderate producers M2, M1 and M3 (0.037–0.043 mg/L), and the high producers AWRI 796 and RA17 (0.056–0.063 mg/L). Strains AMH, BGY, RC212, A1, A2, A3, M1, M2 and M3 produced significantly more benzaldehyde during fermentation at 27 °C compared to fermentation at 22 °C. At 27 °C, benzaldehyde ranged from 0.052–0.068 mg/L in wines fermented by individual Burgundian strain A3 and industrial strain RA17, respectively. Some significant differences in benzaldehyde production existed among yeast strains, including differences  91  between the low producers A3 and A2 (0.052–0.053 mg/L) and the high producers RC212, M1 and RA17 (0.063–0.068 mg/L).  3.3.5 Production of acetic acid and 1,1-diethoxyacetal varied according to yeast strain selection and fermentation temperature Two additional compounds that did not fit into larger classifications, acetic acid and 1,1-diethoxyacetal, were quantified in Pinot noir wine fermented at 22 °C and 27 °C with eight pure and four mixed S. cerevisiae strains and found to vary according to yeast strain and fermentation temperature (Figure 18).  A 1.2  Acetic acid (mg/L)  1  ■ 22 °C ■ 27 °C  0.8 0.6 0.4 0.2 0  Figure 18. Concentrations (mg/L) of acetic acid (A) and 1,1-diethoxyacetal (B) produced by eight individual and four mixed Burgundian S. cerevisiae yeast strains. Mean values are shown (n=3); error bars indicate +/- standard deviation. CALPN indicates the industrial Calona Pinot noir sample.  92  Figure 18 (continued)  B 1,1-diethoxyacetal (mg/L)  7 6  ■ 22 °C ■ 27 °C  5 4 3 2 1 0  The differences in the acetic acid and 1,1-diethoxyacetal content of wines fermented by eight individual and four mixed Burgundian strains of S. cerevisiae at 22 °C and 27 °C are shown in Table 18.  93  Table 18. Acetic acid and 1,1-diethoxyacetal (mg/L) produceda by eight individual and four mixed Burgundian S. cerevisiae strains in Pinot noir wine fermented at 22 °C and 27 °C. Strain and temperature effects by volatile are shown; p values indicate the temperature effect. Yeast Strain AMH AWRI 796 BGY RA17 RC212 A1 A2 A3 M1 M2 M3 M4  Acetic acid 22 °Cb 27 °C a 0.387 0.427ab 0.375a 0.394a b 0.662 0.658cd 0.864d 0.450a 0.480a 0.693cd 0.586bc 0.332a a 0.347a 0.439 0.356a 0.847d 0.445a 0.726cd 0.415a 0.533abc a 0.342 0.586bc 0.383a 0.629bc  c  p ns ** ns ns ns ns ns ns ns ** ** *  1,1-diethoxyacetal 22 °C 27 °C b 2.833 3.357d 2.049a 2.364b a 1.885 2.808c 4.677e 4.827f 3.210bcd 2.750c 3.303bcd 3.569d 3.362bcd 3.336d 3.169bc 2.373b 3.294bcd nqa 3.780cd nqa 3.909d 4.163e 3.601cd 3.498d  p ns ns ns ns ns ns ns ns ns ns ns ns  a  The mean values of the biological replicates of each yeast strain are shown (n=3) Yeast strain means with different superscripts are significantly (p≤0.05) different at each temperature c ns, *, **, *** indicate significance between the fermentation temperatures for each yeast strain at not significant and p≤ 0.05, 0.01and 0.001 b  3.3.5.1 Production of acetic acid. Acetic acid was quantified in Pinot noir wine following fermentation at 22 °C and 27 °C and found to vary according to yeast strain and fermentation temperature (Figure 18, Panel A and Table 18). When fermentation was conducted at 22 °C, acetic acid ranged from 0.332–0.662 mg/L in wines fermented by individual Burgundian strain A1 and industrial strain BGY, respectively. Additionally, industrial strain BGY produced significantly (p≤0.05) more acetic acid than all other strains and Burgundian mixtures, which did not differ significantly from one another. When the fermentation temperature was increased to 27 °C, mixtures M2, M3 and M4 produced significantly more acetic acid than when fermentation was conducted at  94  22 °C, while strain AWRI 796 produced significantly less acetic acid. At 27 °C, acetic acid ranged from 0.347–0.864 mg/L in wines fermented by individual Burgundian strain A2 and industrial strain RA17, respectively. A greater number of significant (p≤0.05) differences existed among yeast strains at this temperature, including differences between the low producers AWRI 796 and AMH (0.375–0.427 mg/L) and the high producers A3 and RA17 (0.847–0.864 mg/L).  3.3.5.2 Production of 1,1-diethoxyacetal. The compound 1,1-diethoxyacetal was quantified in Pinot noir wine following fermentation at 22 °C and 27 °C and found to vary according to yeast strain and fermentation temperature (Figure 18, Panel B and Table 18). When fermentation was conducted at 22 °C, 1,1-diethoxyacetal ranged from 1.885–4.677 mg/L in wines fermented by industrial strains BGY and RA17, respectively. Significant (p≤0.05) differences in 1,1-diethoxyacetal production existed among yeast strains and included the differences between the low producers BGY and AWRI 796 (1.885–2.049 mg/L), the moderate producers AMH, A3, RC212, M1, A1 and A2 (2.833–3.362 mg/L), and the high producer RA17 (4.677 mg/L). When the fermentation temperature was increased to 27 °C, none of the yeast strains differed significantly (p≤0.05) in 1,1-diethoxyacetal production compared to fermentation at 22 °C. At 27 °C, 1,1- diethoxyacetal production was below quantifiable levels in Burgundian mixtures M1 and M2, while industrial strain RA17 produced a maximum value of 4.927 mg/L. Significant differences in 1,1-diethoxyacetal production existed between yeast strains and included the differences between the lowest producers M1 and M2 (not quantifiable), the low producers AWRI 796 and A3 (2.364–2.373  95  mg/L), the low-moderate producers RC212 and BGY (2.750–2.808 mg/L), the moderate producers A2, AMH, M4 and A1 (3.336–3.569 mg/L), the high producer M3 (4.163 mg/L), and the very high producer RA17 (4.827 mg/L).  3.4 Principal component analysis revealed different patterns of volatile compound production in wines fermented at 22 °C and 27 °C Principal component analysis (PCA) of all replicates revealed different patterns of volatile compound production at the two fermentation temperatures (Figure 19, Panels A and B). Three eigenvalues were retained, which explained 28.4, 23.2, and 14.9 percent of the variance, respectively. The pattern of volatile compound production based on fermentation temperature was most evident in the second factor, which was represented in Figure 19 by the Y-axis in Panel A and the X-axis in Panel B. The similar sample groupings according to fermentation temperature in Panels A and B demonstrated that these patterns of volatile compound production were consistent in the third dimension.  96  23.2% Variation 7  A  + PC II  22 °C 27 °C  5 3 1  − PC I -7  -5  -3  -1-1  + PC I 28.4% Variation 1  3  5  7  -3 -5 -7  − PC II  14.9% Variation 7  B  + PC III  22 °C 27 °C  5 3 1  − PC II -7  -5  -3  -1-1  + PC II 23.2% Variation 1  3  5  7  -3 -5 -7  − PC III  Figure 19. PCA plots of the complete volatile compound profiles of wines fermented at 22 °C and 27 °C. (A) The first and second principal components. (B) The second and third principal components. Samples fermented at 22 °C are indicated with ●, while samples fermented at 27 °C are shown with ▲. The blue ellipses (visual aids only) enclose samples fermented at 22 °C, while the red ellipses enclose samples fermented at 27 °C. Each vector represents a single quantified volatile compound. Vectors were scaled by a factor of seven to appear on the same scale as the samples.  97  The details of the different patterns of volatile compound production based on fermentation temperature were revealed by assigning numbers 1-25 to the vectors in Figure 19, Panel A beginning on the + PC I-axis and moving counterclockwise through the quadrants (Table 19). Overall, wine samples fermented at 22 °C contained greater amounts of the compounds ethyl butanoate, ethyl octanoate, ethyl hexanoate, and 1,1-diethoxyacetal, while wine samples fermented at 27 °C tended to contain more of all of the other volatile compounds except hexyl acetate, propanol, ethyl lactate, and 1-hexanol, which did not show production patterns based on fermentation temperature. In order to isolate the volatile compound production differences between yeast strains from the differences between fermentation temperatures, subsequent PCA considered the results from the two fermentation temperatures separately.  98  Table 19. Details of the vectors visible in Figure 19, Panel A. Numbers were assigned beginning at the + PC I-axis and moving in a counterclockwise direction. Each vector represents a single volatile compound. Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25  Compound Hexyl acetate Propanol Ethyl butanoate Ethyl octanoate Ethyl hexanoate 1,1-diethoxyacetal Ethyl lactate 1-hexanol 2-methyl-1-butanol Phenylethanol 3-methyl-1-butanol Isobutanol Butanol Benzaldehyde Acetic acid 2,3-butanediol 1,3-butanediol Ethyl palmitate Isobutyl acetate Methyl acetate Acetaldehyde Ethyl laurate Ethyl acetate Isoamyl acetate Ethyl decanoate  Class Acetate ester Alcohol Ethyl ester Ethyl ester Ethyl ester Acetal Ethyl ester Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Aldehyde Acid Alcohol Alcohol Ethyl ester Acetate ester Acetate ester Aldehyde Ethyl ester Acetate ester Acetate ester Ethyl ester  Relevant Loading + PC I + PC I + PC II + PC II + PC II + PC II − PC I − PC I − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II − PC II  Temperature of Greater Production Inconclusive Inconclusive 22 °C 22 °C 22 °C 22 °C Inconclusive Inconclusive 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C 27 °C  99  3.4.1 Individual Burgundian and mixed Burgundian yeast strains showed different patterns of volatile alcohol production The PCA plots of the volatile alcohol content of Pinot noir wines fermented at 22 °C and 27 °C revealed different patterns of volatile alcohol production between the mixed and the individual Burgundian yeast strains at both fermentation temperatures (Figure 20 and Figure 21). In the PCA plot of Pinot noir fermented at 22 °C, three eigenvalues were retained, which explained 43.8, 22.2, and 16.4 percent of the total variance, respectively. Wines fermented by the individual Burgundian and mixed Burgundian strains showed discrete groupings on the resulting PCA plots that were consistent with the third dimension (Figure 20, Panels A and B), indicating a different pattern of volatile alcohol production between the two groups. As is evident in Panel A, wines fermented by mixed Burgundian strains tended to produce more 1,3-butanediol, 2,3-butanediol, and propanol than wines fermented with individual Burgundian strains. However, as shown in Panel B, wines fermented with individual Burgundian strains produced more 1-hexanol, 2-methyl-1butanol, and 3-methyl-1-butanol.  100  22.2% Variation 6  A  + PC II  ●  Industrial 22 °C  ▲ Individual Burgundian 22 °C ■ Mixed Burgundian 22 °C  4 2  − PC I -6  + PC I  0  -4  -2  C0 B A -2  2  4  43.8% Variation  6  -4 -6  − PC II  16.4% Variation 5  B  + PC III  ●  Industrial 22 °C  ▲ Individual Burgundian 22 °C ■ Mixed Burgundian 22 °C  3  1 D E F  − PC II -5  -3  -1  -1  1  + PC II 3  22.2% Variation  5  -3  -5  − PC III  Figure 20. PCA plots of the volatile alcohol content of wine samples fermented at 22 °C. (A) The first two principal components. Vectors A, B, and C represent 1,3-butanediol, 2,3-butanediol, and propanol, respectively. (B) The second and third principal components. Vectors D, E, and F represent 1-hexanol, 3-methyl-1-butanol, and 2-methyl1-butanol. Industrial strains are indicated with ●, individual Burgundian strains with ▲, and mixed Burgundian strains with ■. Red ellipses (visual aids only) indicate the primary cluster of individual Burgundian strains while green ellipses indicate the primary cluster of mixed Burgundian strains. Each vector represents a single quantified volatile alcohol. Vectors were scaled by a factor of three to appear on the same scale as the samples. 101  Three eigenvalues were retained in the PCA of the volatile alcohols in Pinot noir fermented at 27 °C, which explained 39.7, 33.0, and 11.8 percent of the total variance, respectively. Discrete groupings for the majority of the samples fermented by individual Burgundian and mixed Burgundian strains were evident in the PCA plots (Figure 21). Pinot noir fermented by mixed Burgundian strains typically contained greater amounts of most of the volatile alcohols than Pinot noir fermented by the individual Burgundian strains (Figure 21, Panels A and B). However, in contrast to the results when fermentation was conducted at 22 °C, two out of the three individual Burgundian strains did not produce greater amounts of any of the volatile alcohols than the mixed Burgundian strains at 27 °C. The production of volatile alcohols by the industrial strains at the two fermentation temperatures represented another interesting trend. In wines fermented at 22 °C, the majority of the industrial strains were represented near the individual Burgundian strains A1, A2 and A3 (Figure 20). However, when wines were fermented at 27 °C, the majority of the industrial strains were closer in proximity to the mixed Burgundian strains (Figure 21). This suggested that the individual Burgundian, mixed Burgundian, and industrial yeast strains responded in different ways to changes in fermentation temperature.  102  33% Variation 6  A  + PC II  ●  Industrial 27 °C  ▲ Individual Burgundian 27 °C ■ Mixed Burgundian 27 °C  4 2  − PC I -6  + PC I  0 -4  -2  0  2  4  39.7% Variation  6  -2 -4 -6  − PC II  11.8% Variation 5  B  + PC III  4  ●  Industrial 27 °C  ▲ Individual Burgundian 27 °C ■ Mixed Burgundian 27 °C  3 2 1  − PC II -5  + PC II  0 -3  -1  -1  1  3  33% Variation 5  -2 -3 -4 -5  − PC III  Figure 21. PCA plots of the volatile alcohol content of wine samples fermented at 27 °C. (A) The first two principal components. (B) The second and third principal components. Industrial strains are indicated with ●, individual Burgundian strains with ▲, and mixed Burgundian strains with ■. Red ellipses (visual aids only) indicate the primary cluster of individual Burgundian strains while green ellipses indicate the primary cluster of mixed Burgundian strains. Each vector represents a single quantified volatile alcohol. Vectors were scaled by a factor of three to appear on the same scale as the samples. 103  3.4.2 Individual and mixed Burgundian yeast strains produced greater amounts of volatile ethyl esters than most industrial yeast strains during fermentation PCA revealed that individual and mixed Burgundian yeast strains showed a different pattern of volatile ethyl ester production than the industrial yeast strains (Figure 22). Two eigenvalues were retained for the analysis at each fermentation temperature. At 22 °C, this explained 60 and 19.2 percent of the total variance, while at 27 °C, this explained 67.1 and 15.6 percent of the total variance. At both temperatures, the majority of the wines fermented by individual and mixed Burgundian yeast strains grouped apart from the wines fermented by the industrial strains (Figure 22, Panels A and B). In both cases, the directions of the PCA plot vectors indicated that the individual and mixed Burgundian strains produced more ethyl esters during fermentation than the industrial strains. The wines fermented by the individual and mixed Burgundian strains did not form discrete clusters, suggesting that few collective differences in the pattern of volatile ethyl ester production existed; however, individual yeast strains varied in patterns of volatile ethyl ester production.  104  19.2% Variation 6  ●  + PC II  A  Industrial 22 °C  ▲ Individua l Burgundian 22 °C ■ Mixed Burgundian 22 °C  4  2  − PC I -6  + PC I  0 -4  -2  0  2  60% Va riation  4  6  ●  Industrial 27 °C  -2  -4  − PC II  -6  15.6% Variation 6 + PC II  B  ▲ Individua l Burgundian 27 °C ■ Mixed Burgundian 27 °C  4  2  − PC I -6  + PC I  0 -4  -2  0  2  4  67.1% Va riation  6  -2  -4  -6  − PC II  Figure 22. PCA plots of the volatile ethyl esters in wine samples fermented at 22 °C and 27 °C. (A) PCA plot of wines fermented at 22 °C. (B) PCA plot of wines fermented at 27 °C. Industrial strains are indicated with ●, individual Burgundian strains with ▲, and mixed Burgundian strains with ■. Blue ellipses (visual aids only) indicate the primary cluster of wines fermented by industrial strains. Each vector represents a single quantified volatile alcohol. Vectors were scaled by a factor of three to appear on the same scale as the samples. 105  3.4.3 Individual Burgundian and mixed Burgundian yeast strains showed different patterns of volatile acetate ester production during fermentation PCA revealed different patterns of volatile acetate ester production between individual Burgundian and mixed Burgundian yeast strains following fermentation in Pinot noir at 22 °C and 27 °C (Figure 23). Two eigenvalues were retained for the analysis at each fermentation temperature. At 22 °C, this explained 60.2 and 23.2 percent of the total variance, respectively (Figure 23, Panel A), while at 27 °C, this explained 73.7 and 17.2 percent of the total variance (Figure 23, Panel B). At both fermentation temperatures, the majority of the wines fermented by the individual Burgundian and the mixed Burgundian strains appeared in discrete regions of the PCA plot (Figure 23, Panels A and B). However, at 22 °C, the wines fermented by the individual Burgundian strains were distributed according to the relative abundance of methyl acetate and isobutyl acetate (indicated as D and E in Figure 23), which resulted in a less defined grouping than seen for other compound classes. In general, wines fermented at 22 °C with mixed Burgundian yeast strains contained more hexyl acetate, isoamyl acetate, and ethyl acetate than wines fermented at by individual Burgundian strains. However, this trend was reversed when the fermentation temperature was increased to 27 °C. In this case, wines fermented by individual Burgundian strains tended to contain more hexyl acetate and isoamyl acetate than wines fermented by mixed Burgundian strains, while wines fermented by mixed Burgundian strains tended to contain more methyl acetate and ethyl acetate than wines fermented by individual Burgundian strains.  106  23.2% Va riation 5  + PC II  A  ●  Industrial 22 °C  ▲ Individual Burgundian 22 °C 3  ■ Mixed Burgundian 22 °C  D E 1  − PC I -5  + PC I 60.2% Variation  B -3  -1  -1  1  C A  3  5  -3  -5  − PC II  17.2% Va riation  B  5  ●  + PC II  Industrial 27 °C  ▲ Individual Burgundian 27 °C ■ Mixed Burgundian 27 °C  3 A B  1  − PC I -5  + PC I 73.7% Variation  E -3  -1  -1  -3  -5  1  C  3  5  D  − PC II  Figure 23. PCA plots of the volatile acetate esters in wine samples fermented at 22 °C and 27 °C. (A) PCA plot of wines fermented at 22 °C. (B) PCA plot of wines fermented at 27 °C. Vectors A, B, C, D and E represent hexyl acetate, isoamyl acetate, ethyl acetate, methyl acetate, and isobutyl acetate. Industrial strains are indicated with ●, individual Burgundian strains with ▲, and mixed Burugundian strains with ■. Red ellipses (visual aids only) indicate the primary cluster of individual Burgundian strains while green ellipses indicate the primary cluster of mixed Burgundian strains. Each vector represents a single acetate ester. Vectors were scaled by a factor of three to appear on the same scale as the samples. 107  3.5 Cluster analysis revealed patterns of similarity and differences in volatile compound production among yeast strains Cluster analysis revealed similarities and differences among individual and mixed yeast strains based on volatile compound production during fermentation in Pinot noir (Figure 24). When fermentation was conducted at 22 °C, yeast strains AWRI 796, RA17 and M4 clustered together based on similar volatile compound production, as did strains A1, M1, M2 and M3 (Figure 24, Panel A). Smaller clusters contained strains A2 and A3, and BGY and RC212, while strain AMH was individually set apart. Relationships between these clusters indicated that the cluster containing AWRI 796, RA17 and M4 was most similar to the cluster containing A1, M1, M2 and M3. Together, these two clusters were most similar to the cluster containing individual Burgundian strains A2 and A3, followed by the cluster containing BGY and RC212, and then the single strain AMH, which differed the most from all other yeast strains in terms of volatile compound production. When fermentation was conducted at 27 °C, yeast strains AWRI 796, A2, A3 and M4 clustered together based on similar volatile compound production, as did strains M1, M2 and M3. At this temperature, smaller clusters were formed between strains BGY and RA17, and A1 and RC212, while strain AMH remained apart at this fermentation temperature. The cluster containing strains AWRI 796, A2, A3 and M4 was the most similar to the cluster containing M1, M2 and M3. These two clusters were collectively most similar to the cluster containing BGY and RA17, followed by strain AMH, and finally the cluster containing strains A1 and RC212. While strain AMH differed from the three most similar clusters by a similar distance to that observed during fermentation at  108  22 °C, strains A1 and RC212 differed from all other strains by a much larger distance than previously observed at 22 °C.  A  Distance  25.91  17.27  8.64  0.00  6 79 I R AW  4 M  17 RA  1 A  1 M  3 M  2 M  Yeast Strain  2 A  3 A  Y H 12 BG RC2 AM  B  Distance  44.37  29.58  14.79  0.00  6 79 I R W A  2 A  3 A  4 M  1 M  7 3 H M BGY A1 R AM Yeast Strain 2 M  1 A  12 C2 R  Figure 24. Cluster analysis of yeast strains based on volatile compound production during fermentation in Pinot noir at 22 °C (A) and 27 °C (B).  109  3.5.1 Differences in volatile compound production between fermentation temperatures outweighed differences between yeast strains in some cases Cluster analysis of yeast strain volatile compound production at 22 °C and 27 °C revealed that fermentation temperature-derived differences in volatile compound production outweighed differences between yeast strains in some cases (Figure 25). This was particularly evident in individual Burgundian strains A1, A2 and A3, and industrial strains RA17 and RC212. Rather than clustering with their genetically identical counterparts, these strains clustered separately based on fermentation temperature. This was not the case for mixed Burgundian strains M1, M2, M3, and M4, and industrial strains AWRI 796, BGY, and AMH, which clustered very near their counterparts regardless of fermentation temperature.  Distance  44.53  29.69  14.84  0.00 °C °C °C °C ° C °C ° C °C ° C °C ° C °C ° C °C ° C °C ° C °C °C °C °C °C °C °C 7 22 22 27 22 27 27 27 27 - 27 22 22 22 22 22 22 27 22 22 27 27 22 27 27 2 6- 6- 4- 2- 7- 3- 4- 1- 2- 3 1- 3- 2- 1- 2- 3- Y- Y- 2- 7- H- H- 1- 279 79 M A A1 A M M M M M M M A A A BG BG 21 A1 M M A 21 I R R RI RC R A A RC W W A A Yeast Strain and Fermentation Temperature  Figure 25. Cluster analysis of yeast strains based on volatile compound production during fermentation in Pinot noir at 22 °C and 27 °C. 110  3.6 Individual Burgundian strains were more responsive to changes in fermentation temperature than industrial or mixed Burgundian strains based on volatiles produced during fermentation Many more compounds differed significantly across temperature in wines fermented by individual Burgundian strains A1, A2 and A3 than in wines fermented by most industrial or mixed Burgundian strains (Table 20). Additionally, compounds that differed over fermentation temperature tended to differ at a higher significance level in wines fermented by the individual Burgundian strains than in wines fermented by other strains or Burgundian mixtures. Thus, pure Burgundian strains appeared more vulnerable to fermentation temperature-mediated changes in volatile compound production than most industrial or mixed strains.  111  Table 20. Fermentation temperature-mediated changes in volatiles produced by each industrial, individual Burgundian and mixed Burgundian yeast strain. The symbols *, **, and *** indicate significance at the p≤0.05, p≤0.01, and p≤0.001 levels. Blanks indicate that no significant difference existed. Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class AMH AWRI 796 BGY RA17 RC212 Alcohol * ** ** * *** Alcohol * * ** * ** Alcohol ** Alcohol * Alcohol *** *** * ** *** Alcohol Alcohol ** Alcohol ** * Alcohol Ethyl ester *** *** * Ethyl ester ** Ethyl ester *** * * ** Ethyl ester * ** Ethyl ester ** Ethyl ester *** Ethyl ester * Acetate ester Acetate ester Acetate ester Acetate ester * Acetate ester Aldehyde * Aldehyde * * ** Acid ** Acetal * *  A1 *** ***  *** * **  A2 ** ** * * *  A3 *** ***  M1 *  *** * *  ***  M2 * * * * *  M3 ** **  M4 *  ***  * *  * * *  **  *** **  **  * ***  ** *** * * *** *** ***  ***  **  * *  ** * *** *  *** *  *  *  **  **  *  * **  *  *** *** ***  ** ** ***  112  3.7 Cluster analysis of collective industrial, individual Burgundian and mixed Burgundian yeast strains based on averaged volatile compound profiles revealed the relative distance between yeast strain groups Cluster analysis conducted on the averaged volatile compound profiles of collective industrial, individual Burgundian and mixed Burgundian yeast strains revealed that industrial and individual Burgundian averages were most similar in volatile compound production and clustered together, while the mixed strain average was the least similar, and was set apart at a greater relative distance. This was true when fermentation was conducted at 22 °C and at 27 °C (Figure 26, Panels A and B).  A  Distance  6.70  4.46  2.23  0.00  Industrial  Individual Burgundian  Mixed Burgundian  Yeast Strain Group  Figure 26. Cluster analysis of industrial, individual Burgundian and mixed Burgundian yeast strains based on volatile compound production during fermentation in Pinot noir at 22 °C (A) and 27 °C (B).  113  Figure 26 (continued)  B  Distance Distance  9.95 9.95  6.63 6.63  3.32 3.32  0.00 0.00  Industrial Industrial  Individual Burgundian Individual Burgundian  Mixed Burgundian Mixed Burgundian  YeastStrain Strain Group Yeast  Applying cluster analysis to the averaged volatile profiles of the industrial, individual Burgundian and mixed Burgundian yeast strains from both fermentation temperatures revealed differences in cluster patterns. Industrial and individual Burgundian strains clustered by fermentation temperature, while mixed Burgundian strains clustered by strain similarity (Figure 27). Therefore, industrial and individual Burgundian strains fermented at 22 °C clustered together, as did mixed Burgundian strains fermented at 22 °C and 27 °C, and industrial and individual Burgundian strains fermented at 27 °C. Industrial and individual Burgundian strains fermented at 22 °C were collectively most similar to mixed Burgundian strains fermented at both temperatures, while these two clusters were differentiated from the industrial and individual Burgundian strains fermented at 27 °C by a greater distance. 114  Distance  12.33  8.22  4.11  0.00  Yeast Strain Group  Figure 27. Cluster analysis of industrial, Burgundian and mixed Burgundian yeast strains fermented at 22 °C and 27 °C based on averaged volatile compound profiles.  115  4 DISCUSSION 4.1 Novel Burgundian strains A1, A2 and A3 are enologically competent compared to industrial yeast strains A number of physiological traits have been identified as being advantageous to yeast starter cultures since the initial release of wine yeast products in 1965. These include traits that have long histories of desirability, such as the properties of low foam production and low sulfide formation, and traits that have only recently been identified as beneficial, such as low ethanol production and genetic marking capabilities (Pretorius 2000). A number of these traits were examined and compared to five commercially available industrial Pinot noir yeast strains in an effort to assess the enological competency and desirability of three novel Burgundian wine yeast strains, which were studied individually and in mixed culture. The primary traits that were assessed included genetic fingerprinting, killer phenotype, fermentation kinetics, ethanol production, sugar to ethanol conversion, glycerol production, acetic acid production, growth phenotype, ethanol tolerance, foam production, sulfur dioxide production, compatibility with malolactic fermentation and production of volatile compounds. Some traits were virtually indistinguishable among all of the yeast strains assayed, such as the fermentation kinetics and ethanol production at 22 °C and at 27 °C, with the exception of the reduced rate of ethanol production in industrial strain BGY at 27 °C. All strains were also virtually identical in the sugar to ethanol conversion factor; however, strain A3 had a significantly (p≤0.001) lower conversion factor at 27 °C compared to  116  22 °C. Therefore, the Burgundian strains were found to be equivalent to the industrial strains in terms of fermentation kinetics, ethanol production, and sugar to ethanol conversion factor. Other enological traits showed marked differences among yeast strains. These included glycerol production, acetic acid production, growth phenotype, ethanol tolerance, foam production, sulfur dioxide production, and malolactic compatibility. However, although statistically significant differences existed among these traits, not all differences were relevant from a sensory perspective. For example, glycerol production was in the 8-10.5 g/L range for all yeast strains at both fermentation temperatures and many statistically significant differences existed among strains and across temperatures. According to the literature, the glycerol content in wine ranges from 1–10 g/L in table wine and contributes to perceived sweetness and viscosity (Ough et al. 1972, Radler and Schütz 1982). Noble and Bursick (1984) found that the amount of glycerol required for trained Panelists to detect a difference in perceived sweetness in a white base wine was 5.2 g/L, while a concentration of 25.8 g/L was required for Panelists to detect a difference in viscosity. Additionally, Hinreiner and colleagues (1955) found that the difference threshold of glycerol in red wine was 13.0 g/L, well above the established natural concentration range in table wine. Thus, it is unlikely that a total glycerol range of 8-10.5 g/L would have implications on the winemaking methods employed or the finished wine, resulting in functional equivalency for all assayed yeast strains. A similar pattern emerged in acetic acid production, which ranged from 0.12–0.41 g/L for the yeast strains in this study. The established range of acetic acid in finished  117  wine is 0.2–0.3 g/L (Ribéreau-Gayon et al. 2000a), while the aroma detection threshold ranges from 0.6–0.9 g/L in red wine (Eglinton and Henschke 1999), and the legal limit in the United States is 1.2 g/L. The only yeast strain in this study that exceeded the published acetic acid level was BGY, which produced 0. 337 g/L acetic acid during fermentation at 22 °C and 0.411 g/L acetic acid during fermentation at 27 °C. However, the acetic acid production of all strains fell below the aroma threshold value and well below the legal limit, rendering them functionally equivalent, despite statistically significant differences in the amounts produced. In contrast to glycerol and acetic acid production, the growth phenotypes of the yeast strains offered more meaningful insight into their industrial application. Two parameters were considered; the overall growth pattern and the final optical density after four days of incubation. The most aberrant strain in terms of overall growth pattern was Burgundian strain A1, which showed a complete scattering of optical density measurements upon reaching stationary phase. This was due to its visible flocculation in dense cell culture, which could prove problematic if it were to be commercialized and offered as an active dried yeast product due to lyophilizing difficulties. However, flocculation is also a desirable wine yeast trait (Pretorius 2000), as it allows easier exclusion of yeast sediment during racking, particularly in premium unfiltered wines. Another notable growth pattern phenotype was a visibly longer lag phase at 27 °C in industrial strain BGY. This is industrially undesirable, as it gives other viable microorganisms in the grape must the chance to replicate and in some cases, dominate the fermentation. Most of the final optical densities attained were similar, although industrial strain AMH only reached approximately 75% of the final cell density of the other strains.  118  Based on growth phenotype, the Burgundian strains were enologically competent, as long as the flocculation of strain A1 does not prove to be a negative factor for commercialization. Although statistically significant differences in ethanol tolerance existed among the yeast strains, the range of ethanol tolerance for most strains fell within a single percentage point (18.15–18.90% (v/v) ethanol at 22 °C, 17.35–17.75% (v/v) ethanol at 27 °C). However, industrial strains AMH and BGY were markedly less ethanol tolerant than the other yeast strains. Strain AMH produced 17.68% (v/v) ethanol at 22 °C, and 15.42% (v/v) ethanol at 27 °C, while BGY produced 18.30% (v/v) at 22 °C, but only 16.33% (v/v) ethanol at 27 °C. The high values of ethanol production were likely due to the composition of the sugar-supplemented Pinot noir must, which is known to influence final alcohol concentrations (Casey and Ingledew 1986). At both temperatures, the individual Burgundian strains represented the moderate range of ethanol tolerance, which indicated enological equivalency to industrial strains. Foam production is also known to depend largely on grape must composition (Edwards et al. 1982); however, as with the ethanol tolerance assay, a comparative measure of foam production was appropriate for this study. Although two of the Burgundian strains, A1 and A2, produced foam at the high end of the range, they did not significantly differ from industrial strain BGY, or, in the case of A1, industrial strain RC212. While further investigation in several grape musts may be necessary to ensure that strain A1 and A2’s foaming tendencies are not industrially problematic, the results described here indicate that all Burgundian strains fall within the range established by industrial strains.  119  Sulfur dioxide production is not often considered a primary wine yeast selection factor; therefore, little literature exists on the subject. However, it can prove important due to the sensitivity displayed by the malolactic bacterium O. oeni. An abundance of sulfur dioxide produced during alcoholic fermentation can result in inhibited or stuck malolactic fermentations, which can restrict winemaking decisions and limit the balance of the finished wine. Typically, S. cerevisiae strains produce 10–30 mg/L of sulfur dioxide during fermentation; however, sulfur dioxide formation also depends on the fermentation conditions and the chemical composition of the grape must (Eschenbruch 1974). In this study, sulfur dioxide ranged from 10–50 mg/L, although this was likely influenced by the use of nutrient-poor synthetic grape must, which was necessary to ensure an absence of initial sulfites. The Burgundian strains produced much more sulfur dioxide than three of the industrial strains, AMH, AWRI 796, and BGY, but nearly the same amount as industrial strains RA17 and RC212. Most O. oeni strains can tolerate 15 mg/L free sulfur dioxide and 60–100 mg/L of total sulfur dioxide. Thus, considering the sulfur binding capacity of yeast cells and grape must proteins, it is unlikely that the amounts of sulfur dioxide produced by these yeast strains would negatively impact malolactic fermentation. The influence of yeast strain selection on subsequent malolactic fermentation was assayed more directly in Pinot noir must by inoculating wine samples with O. oeni strain MBR 31 (Lalvin) and analyzing the subsequent malolactic fermentation in terms of malic acid consumption and lactic acid production. In this assay, the wine samples fermented by the Burgundian strains completed malolactic fermentation as expected, while the wine samples fermented with the very low sulfur dioxide producing strains AMH and BGY  120  had steady low levels of malic and lactic acid, suggesting that native malolactic fermentation had already occurred, perhaps in combination with yeast-mediated metabolism of a portion of the malic acid. In either case, the Burgundian strains were equivalent to the majority of the industrial strains, making them just as enologically suitable for subsequent malolactic fermentation. Based on the assayed parameters, it was concluded that the novel Burgundian strains A1, A2 and A3 are enologically suitable compared to industrial strains for the fermentation of Pinot noir grape must.  4.2 Mixed Burgundian strains did not collectively differ from pure Burgundian strains over primary enological parameters The enological equivalency of mixed Burgundian strains was assessed during fermentation in 200 mL Pinot noir. The behaviour of individual Burgundian strains was assumed representative of mixed Burgundian strains for experiments that utilized other systems and media, such as the assays for sulfur dioxide production, foam production, and growth phenotype. Assayed parameters included fermentation kinetics, ethanol production, sugar to ethanol conversion factor, glycerol production, acetic acid production and ethanol tolerance. In all of these parameters, mixed Burgundian strains were found to be functionally and often statistically equivalent to pure Burgundian strain fermentations. The one exception to this equivalency was a slight but statistically significant increase in ethanol tolerance at both fermentation temperatures, particularly in mixtures M1 and M4. One hypothesis for this is the idea that the increased genetic diversity that comes with  121  mixed strain fermentation may provide increased metabolic solutions to fermentative challenges, such as osmotic or ethanol stress.  4.3 Mixed Burgundian strains did not maintain their inoculated strain ratios throughout fermentation Colony PCR was used in conjunction with delta sequence typing to monitor mixed strain ratios at the midpoint and at the end of fermentation. Four different strain ratios were assessed in this study because some strains of S. cerevisiae inevitably dominate or out-grow other strains. However, relative strain abundance was still a problem. Strain A3 was consistently more abundant than strains A1 and A2 at the fermentation midpoint in Burgundian mixtures M1, M2, and M4, and at the end of fermentation in M2 and M4. Thus, mixtures M3 appeared to be more representative of truly mixed strain fermentations due to the lower inoculation rate of A3. Additionally, strain A2 rarely maintained its inoculated ratio. Even in M4, where A2 comprised a full half of the inoculum (50%), A2 was never represented in more than 42% of the fingerprinted colonies. The tendency of A3 abundance to increase and A2 abundance to decrease from inoculated levels was amplified by increasing the fermentation temperature from 22 to 27 °C. If a mixed Burgundian yeast product were to be commercialized, the growth kinetics of the three source strains A1, A2 and A3 would need to be more thoroughly examined both individually and in combination in order to optimize performance. A functional limitation but a commercial benefit to such a product would undoubtedly be the inability of the consumer to successfully recreate the necessary strain ratios by  122  reculturing the product. This is the case with Alchemy I and II, the mixed strain yeasts for Sauvignon Blanc that is currently produced by Anchor Yeast.  4.4 Pinot noir fermented with industrial, individual Burgundian, and mixed Burgundian yeast strains differed in patterns of volatile compound production Volatile compound analysis was conducted with GC-MS following alcoholic fermentation and identified the same 25 quantifiable volatile compounds in all wine samples; however, while wine fermented with mixed Burgundian strains did not contain any novel volatile compounds, quantitative differences did exist across the industrial, individual Burgundian, and mixed Burgundian yeast strains. It is well established that the production of volatile esters and alcohols varies with fermentation temperature (Aragon et al. 1998, Killian and Ough 1979). This was confirmed in this study and dictated the separate PCA analyses of volatile compound data for the two fermentation temperatures. PCA analysis of the volatile alcohols and acetate esters revealed differences in the patterns of compound production between the individual Burgundian and the mixed Burgundian strains. Although the individual Burgundian and the mixed Burgundian strains showed similar patterns of ethyl ester production, these patterns differed from that of the industrial strains, which could represent an important distinction.  4.4.1 Differences in the pattern of volatile alcohol production between individual Burgundian and mixed Burgundian yeast strains may have sensory implications At 22 °C, differences in the pattern of volatile alcohol production showed that mixed Burgundian strains tended to produce more 1,3-butanediol, 2,3-butanediol, and  123  propanol, while individual Burgundian strains tended to produce more 1-hexanol, 3methyl-1-butanol, and 2-methyl-1-butanol. The viscous and slightly bitter compound 2,3-butanediol is produced in large quantities during fermentation as a result of the reduction of acetoin, and is thought to contribute to the sensory profile of wine (Romano 1997). However, this contribution remains unclear due to the high detection threshold (150 mg/L) of 2,3-butanediol (Dubois 1994). The contribution of propanol to wine flavour and aroma is also disputed due to its high threshold (Rankine 1967). S. cerevisiae strains are known to vary in propanol production, which has been shown to be inversely correlated to hydrogen sulfide production (Giudici et al. 1993). Because of this, increased propanol production by mixed Burgundian yeast strains could protect against flaws related to sulfur compound metabolism. In contrast to 2,3-butanediol and propanol, the higher alcohols 1-hexanol, 3-methyl-1-butanol, and 2-methyl-1-butanol are known as fusel oils, which contribute to wine quality at low concentrations and detract from it at high concentrations (Clarke and Bakker 2004). The compound 1-hexanol is known to contribute herbaceous notes, while 2-methyl-1-butanol and 3-methyl-1-butanol ideally contribute a pleasant earthy pungency and fruity-winey notes in their lower concentration ranges (Amerine and Roessler 1983, Ribéreau-Gayon et al. 2000b). Additionally, 2-methyl-1-butanol and 3-methyl-1-butanol are the alcohol precursors to the fruity esters 2-methylbutyl acetate and isoamyl acetate, respectively. In the context of this study, it is difficult to determine whether the increased volatile concentrations of these fusel oils in wines fermented at 22 °C with the individual  124  Burgundian strains represents a positive, negative, or negligible impact on the sensory properties of the wine. When Pinot noir was fermented at 27 °C, wines produced with mixed Burgundian strains tended to contain greater quantities of all of the quantifiable volatile alcohols except isobutanol, butanol, and propanol. The production of higher alcohols is known to vary with fermentation temperature; however, the direction of this correlation is not consistent for all compounds (Zoecklein et al. 1995). Without utilizing liquid extraction to quantify these compounds in the wine rather than in the headspace, it is difficult to draw conclusions as to the benefit or detriment of an overall increase in volatile alcohol production. Nevertheless, the differences in volatile alcohol production in wines fermented with mixed Burgundian strains compared to wines fermented with individual Burgundian strains could indicate the future potential of mixed strain yeast products.  4.4.2 Differences in the pattern of volatile acetate ester production between individual Burgundian and mixed Burgundian yeast strains vary with temperature Ethyl and acetate esters are often regarded as being among the most important classes of volatile compounds for wine flavour and aroma. In this study, Pinot noir fermented by mixed Burgundian yeast strains at 22 °C contained more hexyl acetate, isoamyl acetate, and ethyl acetate than wines fermented by individual Burgundian strains. However, in Pinot noir fermented at 27 °C, wines fermented by mixed Burgundian yeast strains contained more methyl acetate than wines fermented by individual Burgundian yeast strains, which contained more hexyl acetate and isoamyl acetate. However, at both fermentation temperatures, the individual Burgundian and mixed Burgundian yeast  125  strains formed discrete groups following PCA, confirming overall differences in the patterns of acetate ester production. Hexyl acetate and isoamyl acetate contribute to the apple and banana notes of wine, respectively, while methyl acetate and ethyl acetate are known to add fruity notes (Clarke and Bakker 2004). Although high levels of ethyl acetate are identified as a common wine fault, lower levels may be beneficial to the fruitiness and complexity of wine (Amerine and Roessler 1983). The tendency of hexyl and isoamyl acetates to predominate in wines fermented by mixed Burgundian strains at 22 °C and in wines fermented by individual Burgundian strains at 27 °C could be due to a variety of factors. In wine, the volatility of each individual compound is influenced by the concentration and volatility of all other compounds. Thus, while headspace analysis is an excellent method for accurately sampling the “nose” of a wine, differences in a compound’s headspace concentration may or may not be due to differences in the actual concentration or biosynthesis of the compound. However, because the sensory experience of wine comes primarily from the nose, it can be argued that the headspace concentration is the relevant measurement in attempting to establish links between relative compound biosynthesis and wine quality. The fact that the concentration of a greater number of volatile compounds differed significantly across fermentation temperature in wines fermented by individual Burgundian strains compared to wines fermented by mixed Burgundian strains could be responsible for the difference in the production pattern of hexyl and isoamyl acetate.  126  4.4.3 Individual and mixed Burgundian yeast strains produced greater amounts of ethyl esters than industrial yeast strains Although individual and mixed Burgundian yeast strains did not differ in their production of ethyl esters according to PCA, they did produce greater concentrations of almost all of quantified volatile ethyl esters than the industrial strains at both fermentation temperatures. As previously emphasized, esters are very influential in the composition of wine aroma; the sensory contributions of the ethyl esters quantified in this study are shown in Table 21.  Table 21. Sensory effects of ethyl esters quantified in this study (Clarke and Bakker 2004). Ethyl ester Ethyl butanoate Ethyl hexanoate Ethyl lactate Ethyl octanoate Ethyl decanoate Ethy laurate Ethyl palmitate  Odour/Flavour Fruity: pineapple/banana Fruity: pineapple/banana Negligable Soapy/Candlewax Oily/Fruity/Floral Oily/Fruity/Floral None identified  By producing greater quantities of these odour active compounds, individual and mixed Burgundian and mixed strains may have the potential to create Pinot noir wines of greater complexity. Although ester production is often limited in red winemaking with higher fermentation temperatures, most of the volatile compounds identified in the literature as being particularly important to the aroma and flavour of Pinot noir are esters (Moio and Etievant 1995). Therefore, a yeast strain’s tendency to produce greater amounts of volatile esters may be an advantage to modern Pinot noir winemaking.  127  4.5 Further investigation is needed to correlate volatile compound quantification with sensory experience Although differences in the patterns of volatile compound production were found between wines fermented by industrial, individual Burgundian and mixed Burgundian yeast strains, it is difficult to determine if these chemical differences are relevant from a sensory perspective. The relationship between analytical and sensory data represents one of the most challenging areas of current wine research. Despite repeated attempts to make this link, little progress has been made in the literature, particularly with red grape varietals such as Pinot noir. Such attempts have been more successful with the white grape varietal, Sauvignon Blanc, largely because of the volatile thiols that are responsible for the characteristic aromas of boxwood, passionfruit, and guava (Dubourdieu et al. 2006). Because S. cerevisiae is responsible for cleaving these thiols from odourless grapederived precursors (Darriet et al. 1995), Swiegers and colleagues (2007) were able to take advantage of the natural differences in the thiol cleaving capacity of the various commercial yeast strains, which, in combination, produced dramatically different Sauvignon Blanc wine and gave rise to the first mixed S. cerevisiae yeast products, which are sold under the Alchemy label by Anchor Yeast. In the case of Pinot noir, many volatile compounds have been identified as being important to the characteristic Pinot noir aroma, yet so far the chemical creation of that aroma has remained elusive. Of the 25 volatile compounds quantified in this study, 16 are known to be important to wine aroma (Swiegers et al. 2005). Additionally, the quantified compounds phenylethanol and 3-methyl-1-butanol have been implicated specifically in  128  Pinot noir aroma, along with the detectable compounds hexanoic, octanoic, and decanoic acids (Fang and Qian 2006). Neither phenylethanol nor 3-methyl-1-butanol showed important quantitative differences between the mixed strain fermentations and the Burgundian or industrial strain fermentations. The esters ethyl anthranilate, ethyl cinnamate, ethyl 2,3-dihydrocinnamate, and methyl anthranilate are likely the most notable compounds in the literature related to Pinot noir aroma, but were not detected in this study, probably due to their extremely low concentrations (Moio and Etievant 1995). Sensory analysis in conjunction with analytical chemistry offers the possibility of linking chemical identity to sensory experience. Olfactory GC-MS offers another combined approach that has the potential to lend meaning to analytical analysis as well as possibly improving the detection sensitivity of odour-active compounds present in minute quantities. A more straightforward way to lend analytical data sensory relevance is the consideration of the known sensory thresholds of the various chemical compounds. However, because this study utilized headspace analysis, the quantified concentrations applied only to the volatile portion of each compound, which necessitates the use of odour thresholds. These thresholds primarily exist along the air/water interface, rendering them minimally useful in the context of this study. Because much of wine’s complexity is derived from the aging process, it is possible that sensory differences could become apparent among the wines fermented by industrial, individual Burgundian and mixed Burgundian yeast strains after a period of time. There are large differences between the industrial processes of winemaking and the laboratory-scale model system used here, particularly in terms of surface area and potential oxygen exposure, rendering bulk aging in the laboratory impossible. However,  129  if a modern winery were to use some of the strains in this study to ferment and subsequently age Pinot noir, the chemical differences detected here by GC-MS might result in diverse changes during the aging process, which could manifest as differences in wine flavour and aroma, potentially creating a link between chemical and sensory data across the aging process in an industrial setting.  4.6 Cluster analysis revealed overall differences in volatile compound production among industrial, individual Burgundian and mixed Burgundian yeast strains Cluster analysis revealed that the differences in the patterns of volatile compound production observed between the industrial, individual Burgundian and mixed Burgundian yeast strains existed not only within volatile compound classes but also across the complete volatile profiles of the wine samples. As seen with PCA, the clustering was not consistent for all yeast strains. However, the general pattern of clustering among industrial, individual Burgundian and mixed Burgundian strains was apparent at both temperatures (Figure 24). Furthermore, when the volatile profiles of the industrial, individual Burgundian and mixed Burgundian strains were averaged, cluster analysis revealed that the pattern of volatile production in mixed Burgundian strains was most distant from the patterns of volatile compound production of the individual Burgundian and industrial strains (Figure 26). This supported the initial hypothesis that mixed Saccharomyces strain fermentations may result in unique wine volatile profiles.  130  4.7 Wine yeast strains differ in their metabolic response to fermentation temperature Cluster analysis revealed general differences in how yeast strains respond to temperature. Strains AWRI 796, M2, and M3 clustered in similar locations at 22 °C and 27 °C, and clustered consistently near their counterpart when data from both fermentation temperatures was analyzed simultaneously, suggesting that temperature had a small influence on volatile compound production. In contrast, strains A1 and RC212 clustered very differently at the two fermentation temperatures, and clustered far apart when the data were analyzed simultaneously, which demonstrated that temperature had a greater influence on volatile compound production in these strains (Figure 24 and Figure 25). This was largely supported by the tabulation of the significant differences in individual volatile compound production for each yeast strain (Table 20). The one notable exception to this was the strain AMH, which had many significant differences in volatile concentrations between fermentation temperatures yet consistently clustered in the same location. However, although temperature did affect the volatile profile of wine fermented by AMH, these effects were apparently not greater than the differences in volatile production between AMH and the other yeast strains, which appeared to be quite large based on clustering. These patterns of temperature response were found for industrial, individual Burgundian and mixed Burgundian yeast strains. Cluster analysis was conducted on the averaged volatile profiles of industrial, individual Burgundian, and mixed Burgundian yeast strains fermented at 22 °C and 27 °C, respectively, and demonstrated that the differences in volatile compound production between the two temperatures outweighed  131  the differences between yeast strains for industrial and individual Burgundian strains, but not for mixed Burgundian strains (Figure 27). This could indicate that mixed Burgundian strains are more consistent in volatile compound production between fermentation temperatures, which is supported by the fact that the production of fewer volatile compounds differed between fermentation temperatures in mixed Burgundian strains compared to individual Burgundian strains (Table 20). Alternatively, this finding could be due to the differences between the volatile profiles resulting from mixed Burgundian strains and the volatile profiles resulting from the industrial and the individual Burgundian strains, which supports the original hypothesis that mixed Saccharomyces strains might yield unique volatile wine profiles.  132  5 CONCLUSIONS The novel Burgundian S. cerevisiae strains A1, A2 and A3 were shown to be enologically equivalent to industrial strains AMH, AWRI 796, BGY, RA17 and RC212. Each of the novel Burgundian strains had a unique genetic fingerprint that allowed their differentiation during fermentation. Additionally, strains A1, A2 and A3 were killer positive, completed alcoholic fermentation, and were compatible with malolactic bacteria. Although many significant differences existed among yeast strains in terms of the conversion of sugar to ethanol and the production of glycerol, acetic acid, foam and sulfur dioxide, the individual Burgundian strains fell within the range associated with the industrial strains, which rendered them enologically equivalent for the purposes of this study. The GC-MS headspace analysis of Pinot noir fermented at 22 °C and 27 °C by the industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains allowed the quantification of 25 volatile higher alcohols, ethyl esters, acetate esters, aldehydes, acids and acetals in all wines. Many of these compounds significantly (p≤0.05) differed among yeast strains and between fermentation temperatures. PCA of the volatile compounds produced in Pinot noir during fermentation revealed differences in the pattern of volatile alcohol and acetate ester production between individual Burgundian and mixed Burgundian strains. Mixed Burgundian strains tended to produce greater amounts of higher alcohols than individual Burgundian strains during fermentation at 22 °C and 27 °C. Mixed Burgundian strains produced greater amounts of some acetate esters during fermentation at 22 °C; however, this pattern was not consistent when fermentation was conducted at 27 °C. Although the individual  133  Burgundian and mixed Burgundian S. cerevisiae strains did not differ from one another in ethyl ester production, they collectively produced greater amounts of ethyl esters than the industrial strains, which is likely to be an important factor for the production of premium Pinot noir. Cluster analysis over the complete volatile profiles of Pinot noir wines fermented at 22 °C and 27 °C delineated most strains into clusters containing either industrial, individual Burgundian, or mixed Burgundian yeast strains, which indicated that differences in the patterns of volatile production extended over the complete volatile profile. When wines fermented at the two temperatures were included in the same analysis, clustering revealed that the volatile profiles of individual Burgundian strains and some industrial strains were more variable over fermentation temperature than mixed Burgundian strains, which was supported by the significant differences in individual compound production over temperature. When the volatile profiles of wines fermented by industrial, individual Burgundian and mixed Burgundian S. cerevisiae strains were averaged, the mixed Burgundian strains as a group were separated by the greatest distance from industrial and individual Burgundian strain groups over averaged volatile compound profiles, which further supported the hypothesis that mixed Saccharomyces strain fermentation would result in unique volatile compound profiles in Pinot noir. Potential may exist for a mixed Burgundian strain yeast product based on the enological equivalency of the individual Burgundian strains along with the unique volatile production profile and consistent fermentation behavior of mixed Burgundian strains.  134  REFERENCES Adams, C., and H.J.J. van Vuuren. 2010. The timing of diammonium phosphate addition to fermenting grape must affects the production of ethyl carbamate in wine. American Journal of Enology and Viticulture 61: 125-9. Amerine, M.A. and E.B. Roessler. 1983. Wines: Their Sensory Evaluation. 2nd ed. W.H.Freeman & Co Ltd, New York. Aragon, P., J. Atienza, and M.D. Climent. 1998. Influence of clarification, yeast type, and fermentation temperature on the organic acid and higher alcohols of Malvasia and Muscatel wines. American Journal of Enology and Viticulture 49: 211-9. Arriagada-Carrazanaa, J.P., C. Sáez-Navarretea, and E. Bordeu. 2005. Membrane filtration effects on aromatic and phenolic quality of Cabernet Sauvignon wines. Journal of Food Engineering 68: 363-8. Aubry, V., P.X. Etievant, C. Ginies, and R. Henry. 1997. Quantitative determination of potent flavor compounds in Burgundy Pinot noir wines using a stable isotope dilution assay. Journal of Agricultural and Food Chemistry 45: 2120-3. Ausubel, F.M., R. Brent, R.E. Kingston, D.D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl. 1999. Short Protocols in Molecular Biology 3rd ed. Wiley & Sons, New York. Äyräpää, T. 1968. Formation of higher alcohols by various yeasts. Journal of the Institute of Brewing 74: 169-79. Boulton, R.B., V.L. Singleton, L.F. Bisson, and R.E. Kunkee. 1995. Principles and Practices of Winemaking. Chapman Hall, New York. Casey, G.P., and W.M. Ingledew. 1986. Ethanol tolerance in yeasts. Critical Reviews in Microbiology 13: 219-80. Cheraiti, N., S. Guezenec, and J.M. Salmon. 2005. Redox interactions between Saccharomyces cerevisiae and Saccharomyces uvarum in mixed culture under enological conditions. Applied and Environmental Microbiology 71: 255-60. Clarke, R.J. and J. Bakker. 2004. Wine Flavour Chemistry. Blackwell Publishing, Oxford. Clemente-Jimenez, J.M., L. Mingorance-Cazorla, S. Martínez-Rodríguez, F.J. Las HerasVázquez, and F. Rodríguez-Vico. 2005. Influence of sequential yeast mixtures on wine fermentation. International Journal of Food Microbiology 98: 301-8. Darriet, P., T. Tominga, V. Lavigne, J. Boidron, and D. Dubourdieu. 1995. Identification of a powerful aromatic compound of Vitis vinifera L. var. Sauvignon wines: 4-Mercapto-4methylpentan-2-one. Flavour and Fragrance 10: 385-92. 135  Davis, C.R., D.J. Wibowo, R.E. Eschenbruch, T.H. Lee, and G.H. Fleet. 1985. Practical implications of malolactic fermentation: A review. American Journal of Enology and Viticulture 36: 290-301. Denayrolles, M., M. Aigle, and A. Lonvaud-Funel. 1995. Functional expression in Saccharomyces cerevisiae of the Lactococcus lactis mleS gene encoding the malolactic enzyme. FEMS Microbiology Letters 125: 37-43. Dubois, P. 1994. Les aromes des vins et leurs defauts. Revue Francaise d'Oenologie. 145: 27-40. Dubourdieu, D., T. Tominaga, I. Masneuf, C. Peyrot Des Gachons, and M.L. Murat. 2006. The role of yeast in grape flavour development during fermentation: The example of Sauvignon blanc. American Journal of Enology and Viticulture 57: 81-8. Ebeler, S.E., and J.H. Thorngate. 2009. Wine chemistry and flavour: Looking into the crystal glass. Journal of Agricultural and Food Chemistry 57: 8098-108. Edwards, M., R. Eschenbruch, and P.C. Molan. 1982. Foaming in winemaking. I. A technique for the measurement of foaming in winemaking. European Journal of Applied Microbiology and Biotechnology 16: 105-9. Eglinton, J., and P.A. Henschke. 1999. The occurrence of volatile acidity in Australian wines. Australian and New Zealand Grapegrower and Winemaker 426a: 7-12. Eglinton, J.M., A.J. Heinrich, A.P. Pollnitz, P. Langridge, P.A. Henschke, and M.B. Lopes. 2002. Decreasing acetic acid accumulation by a glycerol overproducing strain of Saccharomyces cerevisiae by deleting the ALD6 aldehyde dehydrogenase gene. Yeast 19: 295301. Eschenbruch, R. 1974. Sulfite and sulfide formation during winemaking—a review. American Journal of Enology and Viticulture 25: 157-61. Etievant, P., S. Issanchou, and C.L. Bayonove. 1983. The flavour of Muscat wine: The sensory contribution of some volatile compounds. Journal of the Science of Food and Agriculture 34: 497-504. Eustace, R., and R.J. Thornton. 1987. Selective hybridization of wine yeasts for higher yields of glycerol. Canadian Journal of Microbiology 33: 112-7. Fang, Y., and M.C. Qian. 2006. Quantification of Selected Aroma-Active Compounds in Pinot noir Wines from Different Grape Maturities. Journal of Agricultural and Food Chemistry 22: 8567-73.  136  Ferreira, V., P. Fernández, C. Peña, A. Escudero, and J.F. Cacho. 1995. Investigation on the role played by fermentation esters in the aroma of young Spanish wines by multivariate analysis. Journal of the Science of Food and Agriculture 67: 381-92. Fleet, G.H.H., and G.M. Heard. 1993. Yeasts: growth during fermentation. In Wine Microbiology and Biotechnology. G.H. Fleet (Ed.), pp. 27-54. Harwood Academic Publishers, Chur, Switzerland. Girard, B., T.G. Kopp, A.G. Reynolds, and M. Cliff. 1997. Influence of vinification treatments on aroma constituents and sensory descriptors of Pinot noir wines. American Journal of Enology and Viticulture 48: 198-206. Girard, B., D. Yuksel, M.A. Cliff, P. Delaquis, and A.G. Reynolds. 2001. Vinification effects on the sensory, colour, and GC profiles of Pinot noir wines from British Columbia. Food Research International 34: 483-99. Giudici, P., and R.E. Kunkee. 1994. The effect of nitrogen deficiency and sulfur-containing amino acids on the reduction of sulfate to hydrogen sulfide by wine yeasts. American Journal of Enology and Viticulture 45: 107-12. Giudici, P., C. Zambonelli, and R.E. Kunkee. 1993. Increased production of n-propanol in wine by yeast strains having an impaired ability to form hydrogen sulfide. American Journal of Enology and Viticulture 44: 17-21. Guinard, J.X., and M. Cliff. 1987. Descriptive analysis of Pinot noir wines from Carneros, Napa, and Sonoma. American Journal of Enology and Viticulture 38: 211-5. Gutiérrez, A.R., S. Epifanio, P. Garijo, R. López, and P. Santamaría. 2001. Killer yeasts: Incidence in the ecology of spontaneous fermentation. American Journal of Enology and Viticulture 52: 352-6. Hammond, S.M. and J.G. Carr. The antimicrobial activity of SO2 with particular reference to fermented and non-fermented fruit juices. In Inhibition and Inactivation of Vegetative Microbes F.A. Skinner and W.B. Hugo (Ed.), pp. 89-110. Academic Press, London. Hernández-Orte, P., J.F. Cacho, and V. Ferreira. 2002. Relationship between varietal amino acid profile of grapes and wine aromatic composition. Experiments with model solutions and chemometric study. Journal of Agricultural and Food Chemistry 50: 2891-9. Hinreiner, E., F. Filipello, H.W. Berg, and A.D. Webb. 1955. Evaluation of thresholds and minimum difference concentrations for various constituents of wines. IV. Detectable differences in wine. Food Technology 9: 489-90. Hirvi, T., and E. Honkanen. 1982. The volatiles of two new strawberry cultivars, “Annelie” and “Alaska pioneer”. Zeitschrift für Lebensmittel-Untersuchung und -Forschung 175: 113-6.  137  Howell, K.S., D. Cozzolino, E.J. Bartowsky, G.H. Fleet, and P.A. Henschke. 2006. Metabolic profiling as a tool for revealing Saccharomyces interactions during wine fermentation. FEMS Yeast Research 6: 91-101. Husnik, J., P.A. Delaquis, M.A. Cliff, and H.J.J. van Vuuren. 2007. Functional analysis of the malolactic wine yeast ML01. American Journal of Enology and Viticulture 58: 42-52. Jacobs, C.J. and H.J.J. van Vuuren. 1991. Effects of different killer yeasts on wine fermentations. American Journal of Enology and Viticulture 42: 295-300. Jolly, N.P., O.P.H. Augustyn, and I.S. Pretorius. 2006. The role and use of non-Saccharomyces yeasts in wine production. South African Journal of Enology and Viticulture 27: 15-39. Jones, G.V., M.A. White, O.R. Cooper, and K. Storchmann. 2005. Climate change and global wine quality. Climatic Change 73: 319-43. Kapsopoulou, K., E. Barouti, A. Makrioniti, and K. Kostaki. 2008. Occurrence of Saccharomyces cerevisiae killer yeasts in wine-producing areas of Greece. World Journal of Microbiology and Biotechnology 24: 1967-71. Killian, E., and C.S. Ough. 1979. Fermentation esters—formation and retention as affected by fermentation temperature. American Journal of Enology and Viticulture 30: 301-5. Kunkee, R.E., and G.M. Cooke. 1980. 100 years of wine microbiology. California Agriculture 34: 11-3. Lagace, L.S., and L.F. Bisson. 1990. Survey of wine yeast proteases for effectiveness of wine haze reduction. American Journal of Enology and Viticulture 41: 147-55. Lambrechts, M.G., and I.S. Pretorius. 2000. Yeast and its importance to wine aroma—a review. South African Journal of Enology and Viticulture 21: 97-129. López, V., M.T. Fernández-Espinar, E. Barrio, D. Ramón, and A. Querol. 2003. A new PCRbased method for monitoring inoculated wine fermentations. International Journal of Food Microbiology 81: 63-71. Magliani, W., S. Conti, M. Gerloni, D. Bertolotti, and L. Polonelli. 1997. Yeast killer systems. Clinical Microbiology Reviews 10: 369-400. Marais, J., and H.J. Pool. 1980. Effect of storage time and temperature on the volatile composition and quality of dry white table wines. Vitis 19: 151-64. Mendes-Ferreira, A., A. Mendes-Faia, and C. Leao. 2002. Survey of hydrogen sulphide production by wine yeasts. Journal of Food Protection 65: 1033-7.  138  Mestres, M., O. Busto, and J. Guasch. 2000. Analysis of organic sulfur compounds in wine aroma. Journal of Chromatography 881: 569-81. Miranda-Lopez, R., L.M. Libbey, B.T. Watson, and M.R. McDaniel. 1992. Identification of additional odor-active compounds in Pinot noir wines. American Journal of Enology and Viticulture 43: 90-2. Moio, L., and P.X. Etievant. 1995. Ethyl anthranilate, ethyl cinnamate, 2,3-dihydrocinnamate, and methyl anthranilate: Four important odorants identified in Pinot noir wines of Burgundy. American Journal of Enology and Viticulture 46: 392-8. Monk, P.R. 1986. Formation, utilization and excretion of hydrogen sulphide by wine yeast. Australian Wine Industry Journal 1: 10-6. Noble, A.C., and G.F. Bursick. 1984. The contribution of glycerol to perceived viscosity and sweetness in white wine. American Journal of Enology and Viticulture 35: 110-2. Nurgel, C., G. Pickering, and D. Inglis. 2004. Sensory and chemical characteristics of Canadian icewines. Journal of the Science of Food and Agriculture 84: 1675-84. Nykänen, L. 1986. Formation and occurrence of flavour compounds in wine and distilled alcoholic beverages. American Journal of Enology and Viticulture 37: 84-96. Ough, C.S., and M.A. Amerine. 1967. Studies with controlled fermentation. X. Effect of fermentation temperature on some volatile compounds in wine. American Journal of Enology and Viticulture 18: 157-64. Ough, C.S. 1992. Winemaking Basics. The Haworth Press, Binghamton. Ough, C.S., G. Fong, and M.A. Amerine. 1972. Glycerol in wine: Determination and some factors affecting. American Journal of Enology and Viticulture 23: 1-5. Park, S.K., R.B. Boulton, and A.C. Noble. 2000. Formation of hydrogen sulfide and glutathione during fermentation of white grape musts. American Journal of Enology and Viticulture 51: 91-7. Peréz, F., M. Ramírez, and J.A. Regodón. 2001. Influence of killer strains of Saccharomyces cerevisiae on wine fermentation. Antonie van Leeuwenhoek 79: 393-9. Pretorius, I.S. 2000. Tailoring wine yeast for the new millennium: Novel approaches to the ancient art of winemaking. Yeast 16: 675-729. Prior, B.A., T.H. Toh, N. Jolly, C. Baccari, and R.K. Mortimer. 2000. Impact of yeast breeding for elevated glycerol production on fermentative activity and metabolite formation in Chardonnay wine. South African Journal of Enology and Viticulture 21: 92-9.  139  Radler, F., and H. Schütz. 1982. Glycerol production of various strains of Saccharomyces. American Journal of Enology and Viticulture 33: 36-40. Ranieri, S., and I.S. Pretorius. 2000. Selection and improvement of wine yeasts. Annals of Microbiology 50: 15-31. Rankine, B.C. 1967. Formation of higher alcohols by wine yeasts, and relationship to taste threshold. Journal of the Science of Food and Agriculture 18: 583-9. Rankine, B.C., and D.A. Bridson. 1971. Glycerol in Australian wines and factors influencing its formation. American Journal of Enology and Viticulture 22: 6-12. Rapp, A., and H. Mandry. 1986. Wine aroma. Experientia 42: 873-84. Reed, G., and T.W. Nagodawithana. 1988. Technology of yeast usage in winemaking. American Journal of Enology and Viticulture 39: 83-9. Regodón, J.A., F. Peréz, M.E. Valdés, C. De Miguel, and M. Ramírez. 1997. A simple and effective procedure for selection of wine yeast strains. Food Microbiology 14: 247-54. Remize, F., J.L. Roustan, J.M. Sablayrolles, P. Barre, and S. Dequin. 1999. Glycerol overproduction by engineered Saccharomyces cerevisiae wine yeast strains leads to substantial changes in by-product formation and to a stimulation of fermentation rate in stationary phase. Applied and Environmental Microbiology 65: 143-9. Ribéreau-Gayon, P., Y. Glories, A. Maujean, and D. Dubourdieu. 2000a. Handbook of Enology, Volume I, Wine and Wine Making. John Wiley & Sons, Chichester. Ribéreau-Gayon, P., Y. Glories, A. Maujean, and D. Dubourdieu. 2000b. Handbook of Enology, Volume II, The Chemistry of Wine, Stabilization and Treatments. John Wiley & Sons, Chichester. Rojas, V., J.V. Gil, F. Piñaga, and P. Manzanares. 2003. Acetate ester formation in wine by mixed cultures in laboratory fermentations. International Journal of Food Microbiology 86: 181-8. Romano, P. 1997. Metabolic characteristics of wine strains during spontaneous and inoculated fermentation. Food Technology and Biotechnology 35: 255-60. Saerens, S.M.G., F.R. Delvaux, K.J. Verstrepen, and J.M. Thevelein. 2009. Production and biological function of volatile esters in Saccharomyces cerevisiae. Microbial Biotechnology 1-13. Scanes, K.T., S. Hohmann, and B.A. Prior. 1998. Glycerol production by the yeast Saccharomyces cerevisiae and its relevance to wine: A review. South African Journal of Enology and Viticulture 19: 17-24. 140  Schreier, P., F. Drawert, and K.O. Abraham. 1080. Identification and determination of volatile constituents in Burgundy Pinot noir wines. Food Science and Technology 13: 318-21. Schuller, D., E. Valero, S. Dequin, and M. Casal. 2004. Survey of molecular methods for the typing of wine yeast strains. FEMS Microbiology Letters 231: 19-26. Shimizu, K., T. Adachi, K. Kitano, T. Shimazaki, A. Totsuka, S. Hara, and H.H. Dittrich. 1985. Killer properties of wine yeasts and characterization of killer wine yeasts. Journal of Fermentation Technology 63: 421-9. Soles, R.M., C.S. Ough, and R.E. Kunkee. 1982. Ester concentration differences in wine fermented by various species and strains of yeast. American Journal of Enology and Viticulture 33: 94-8. Swiegers, H., E. King, B. Travis, L. Francis, and I.S. Pretorius. 2007. Enhancement of Sauvignon blanc wine aroma through yeast combinations. Wynboer www.wynboer.co.za/recentarticles/200712sauvignon.php3 Swiegers, H., and I.S. Pretorius. 2007. Modulation of volatile sulfur compounds by wine yeast. Applied Microbiology and Biotechnology 74: 954-60. Swiegers, J.H., E.J. Bartowsky, P.A. Henschke, and I.S. Pretorius. 2005. Yeast and bacterial modulation of wine aroma and flavour. Australian Journal of Grape and Wine Research 11: 139-73. Taherzadeh, M.J., L. Adler, and G. Lidén. 2002. Strategies for enhancing fermentative production of glycerol—a review. Enzyme and Microbial Technology 31: 53-66. Thoukis G, Reed G, Bouthilet RJ. 1963. Production and use of compressed yeast for winery fermentation. American Journal of Enology and Viticulture 14: 148-54. Thurston, P.A., R. Taylor, and J. Ahvenainen. 1981. Effects of linoleic acid supplements on the synthesis by yeast of lipids and acetate esters. Journal of the Institute of Brewing 87: 92-5. Tipper, D.J., and M.J. Schmitt. 1991. Yeast dsRNA viruses: Replication and killer phenotypes. Molecular Microbiology 5: 2331-8. Tominaga, T., R. Baltenweck-Guyot, C. Peyrot de Gachons, and D. Dubourdieu. 2000. Contribution of volatile thiols to the aromas of white wines made from several Vitis vinifera grape varieties. American Journal of Enology and Viticulture 51: 178-81. Van Vuuren, H.J.J., and B.D. Wingfield. 1986. Killer yeasts—the cause of stuck fermentations in a wine cellar. South African Journal of Enology and Viticulture 7: 113-8.  141  Vermeulen, C., L. Gijs, and S. Collin. 2005. Sensorial contribution and formation pathways of thiols in foods: A review. Food Reviews International 21: 69-137. Webb, A.D., and R.E. Kepner. 1961. Fusel oil analysis by means of gas-liquid partition chromatography. American Journal of Enology and Viticulture 12: 51-9. Zambonelli, C. 1998. Microbiologia e Biotecnologia dei Vini. Edagricole, Bologna. Zoecklein, B.W., K.C. Fugelsang, B.H. Gump, and F.S. Nury. 1995. Wine Analysis & Production. Chapman & Hall, New York.  142  APPENDIX A Complete GC-MS volatile compound dataset for all biological replicates (Table 22 and Table 23)  143  Table 22. Concentrations (mg/L) of all volatile compounds quantified in Pinot noir wines (n=3) fermented at 22 °C by industrial yeast strains AMH, AWRI 796, BGY, RA17 and RC212; individual Burgundian strains A1, A2 and A3 and mixed Burgundian strains M1, M2, M3 and M4. Volatile Compound  Class  Strain AMH-1  Strain AMH-2  Strain AMH-3  AMH Average  Standard Deviation  AWRI 796-1  AWRI 796-2  AWRI 796-3  AWRI 796 Average  Standard Deviation  1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  1.925 0.636 1.564 7.721 0.170 2.057 29.664 0.408 11.884 2.506 0.031 0.055 0.515 0.006 0.065 0.037 9.319 0.026 0.185 0.001 0.844 0.949 0.028 0.278 2.534  2.351 0.747 1.657 8.136 0.187 2.133 33.288 0.481 13.294 2.708 0.032 0.060 0.546 0.006 0.069 0.046 10.355 0.031 0.213 0.001 0.883 1.064 0.021 0.296 2.601  5.161 1.400 1.862 8.847 0.198 2.432 31.260 0.529 13.379 2.862 0.030 0.064 0.583 0.005 0.069 0.052 10.465 0.030 0.211 0.001 0.944 1.063 0.040 0.582 3.392  3.150 0.929 1.693 8.229 0.185 2.206 31.380 0.472 12.845 2.689 0.031 0.060 0.548 0.006 0.068 0.045 10.039 0.029 0.203 0.001 0.889 1.026 0.029 0.387 2.833  1.758 0.413 0.152 0.570 0.014 0.198 1.817 0.061 0.840 0.179 0.001 0.004 0.034 0.000 0.002 0.008 0.632 0.003 0.016 0.000 0.050 0.066 0.010 0.171 0.477  2.570 0.866 2.035 9.773 0.166 2.343 53.555 0.740 12.355 3.815 0.032 0.068 0.891 0.009 0.064 0.153 11.096 0.032 0.239 0.001 1.387 0.542 0.060 0.452 2.179  2.585 0.799 2.094 9.798 0.168 2.441 50.567 0.675 10.494 2.980 0.024 0.056 0.811 0.006 0.050 0.108 9.272 0.018 0.168 0.001 1.209 0.465 0.053 0.333 1.935  2.813 0.902 2.148 9.992 0.174 2.463 47.860 0.714 10.793 2.915 0.020 0.053 0.843 0.005 0.046 0.093 9.342 0.015 0.157 0.001 1.140 0.506 0.054 0.399 2.030  2.651 0.854 2.089 9.840 0.169 2.413 50.592 0.709 11.198 3.232 0.025 0.059 0.847 0.007 0.053 0.118 9.889 0.022 0.188 0.001 1.245 0.504 0.056 0.394 2.049  0.137 0.052 0.056 0.120 0.004 0.064 2.848 0.033 0.999 0.502 0.006 0.008 0.040 0.002 0.009 0.031 1.033 0.009 0.044 0.000 0.128 0.038 0.004 0.060 0.123  144  Table 22 (continued)  Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain BGY-1 2.634 0.858 2.544 11.465 0.173 2.783 83.415 0.718 8.381 2.366 0.019 0.045 1.885 0.004 0.040 0.075 10.151 0.014 0.161 0.001 1.149 0.699 0.043 0.522 1.706  Strain BGY-2 3.114 0.969 2.072 9.353 0.160 2.282 52.409 0.544 9.096 2.331 0.027 0.049 0.902 0.005 0.055 0.065 9.621 0.021 0.186 0.001 0.968 0.706 0.046 0.699 1.698  Strain BGY-3 2.610 0.921 2.693 12.084 0.176 2.896 79.601 0.784 8.533 2.419 0.018 0.045 2.539 0.003 0.042 0.076 10.530 0.014 0.168 0.002 1.207 0.618 0.045 0.770 2.260  BGY Average 2.773 0.913 2.433 10.949 0.169 2.649 71.731 0.680 8.654 2.368 0.021 0.046 1.773 0.004 0.046 0.072 10.083 0.016 0.171 0.001 1.105 0.670 0.045 0.662 1.885  Standard Deviation 0.285 0.056 0.325 1.432 0.009 0.327 16.908 0.124 0.377 0.044 0.005 0.002 0.824 0.001 0.008 0.006 0.457 0.004 0.013 0.000 0.125 0.049 0.002 0.128 0.322  Strain RA17-1 3.820 1.055 2.003 9.640 0.161 2.219 59.408 0.592 8.293 2.858 0.018 0.051 0.628 0.003 0.043 0.068 9.782 0.028 0.230 0.001 1.000 1.341 0.068 0.369 5.119  Strain RA17-2 3.091 0.942 1.790 8.813 0.141 1.967 52.855 0.456 7.614 2.810 0.027 0.051 0.541 0.005 0.051 0.063 9.750 0.027 0.221 0.001 0.987 1.358 0.056 0.362 3.948  Strain RA17-3 3.184 1.009 2.072 10.059 0.168 2.318 48.875 0.649 8.677 2.939 0.022 0.052 0.729 0.004 0.048 0.073 10.363 0.027 0.233 0.001 1.006 1.336 0.064 0.621 4.995  RA17 Average 3.359 1.000 1.951 9.486 0.157 2.164 53.548 0.565 8.179 2.862 0.022 0.051 0.631 0.004 0.047 0.068 9.944 0.027 0.228 0.001 0.996 1.339 0.063 0.450 4.677  Standard Deviation 0.397 0.057 0.147 0.634 0.014 0.181 5.319 0.099 0.539 0.065 0.004 0.001 0.094 0.001 0.004 0.005 0.345 0.001 0.006 0.000 0.010 0.012 0.007 0.147 0.643  145  Table 22 (continued) Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain A2-1 2.204 0.782 2.431 11.230 0.226 2.664 45.943 0.695 9.813 2.347 0.014 0.047 0.860 0.002 0.043 0.066 9.426 0.016 0.176 0.001 1.004 0.912 0.026 0.431 3.563  Strain A2-2 2.515 0.825 2.328 10.800 0.207 2.477 43.031 0.614 9.575 2.292 0.015 0.045 0.761 0.002 0.042 0.060 9.429 0.015 0.170 0.001 0.946 1.013 0.024 0.489 3.278  Strain A2-3 2.250 0.768 2.194 10.209 0.193 2.356 43.535 0.563 9.306 2.228 0.017 0.046 0.626 0.003 0.044 0.042 9.195 0.014 0.157 0.001 0.894 1.022 0.026 0.400 3.268  A2 Average 2.317 0.790 2.315 10.736 0.208 2.497 44.142 0.623 9.555 2.286 0.015 0.046 0.749 0.002 0.043 0.056 9.335 0.015 0.167 0.001 0.946 0.979 0.026 0.439 3.362  Standard Deviation 0.168 0.030 0.119 0.513 0.016 0.155 1.556 0.066 0.254 0.059 0.002 0.001 0.117 0.000 0.001 0.012 0.134 0.001 0.010 0.000 0.055 0.061 0.001 0.045 0.168  Strain A3-1 2.454 0.779 1.997 9.363 0.182 2.174 41.984 0.574 8.607 3.253 0.018 0.049 0.711 0.003 0.048 0.109 9.641 0.025 0.205 0.001 1.146 0.773 0.026 0.342 2.862  Strain A3-2 2.052 0.708 2.038 9.789 0.183 2.369 41.396 0.624 8.703 3.355 0.026 0.055 0.671 0.004 0.058 0.068 9.958 0.026 0.211 0.001 1.142 0.985 0.023 0.361 3.101  Strain A3-3 2.254 0.767 2.255 10.706 0.204 2.532 46.958 0.710 9.218 3.589 0.023 0.056 0.795 0.004 0.059 0.093 10.739 0.025 0.220 0.001 1.308 0.951 0.026 0.370 3.585  A3 Average 2.239 0.748 2.090 9.921 0.189 2.351 43.416 0.632 8.815 3.387 0.023 0.053 0.724 0.004 0.055 0.090 10.077 0.025 0.211 0.001 1.196 0.901 0.025 0.356 3.169  Standard Deviation 0.201 0.038 0.139 0.687 0.012 0.179 3.056 0.069 0.328 0.172 0.004 0.004 0.064 0.001 0.006 0.021 0.565 0.001 0.008 0.000 0.095 0.114 0.002 0.015 0.368  146  Table 22 (continued) Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain M1-1 4.045 1.223 2.186 10.232 0.177 2.392 54.043 0.649 10.684 3.308 0.025 0.054 0.787 0.005 0.057 0.102 10.569 0.029 0.245 0.001 1.170 0.925 0.042 0.481 3.248  Strain M1-2 3.612 1.129 2.563 11.762 0.191 2.865 59.799 0.692 11.072 3.048 0.020 0.050 0.869 0.004 0.051 0.099 10.322 0.018 0.195 0.001 1.194 0.838 0.041 0.495 3.034  Strain M1-3 3.178 0.980 2.172 10.094 0.185 2.363 63.187 0.596 11.206 3.476 0.027 0.058 0.796 0.006 0.063 0.113 10.972 0.034 0.261 0.001 1.187 1.007 0.033 0.359 3.615  M1 Average 3.610 1.110 2.306 10.693 0.184 2.540 58.954 0.645 10.984 3.275 0.024 0.054 0.817 0.005 0.057 0.105 10.613 0.027 0.234 0.001 1.183 0.923 0.039 0.445 3.294  Standard Deviation 0.434 0.123 0.222 0.926 0.007 0.282 4.623 0.048 0.271 0.216 0.004 0.004 0.045 0.001 0.006 0.007 0.328 0.008 0.034 0.000 0.012 0.084 0.005 0.075 0.294  Strain M2-1 3.437 1.031 2.191 10.234 0.183 2.395 62.599 0.562 9.607 3.316 0.026 0.054 0.705 0.005 0.056 0.098 10.807 0.029 0.241 0.001 1.227 0.913 0.039 0.453 3.221  Strain M2-2 3.540 1.102 2.344 10.805 0.202 2.561 55.217 0.672 10.233 3.429 0.023 0.054 0.839 0.004 0.058 0.114 11.116 0.028 0.244 0.001 1.235 1.099 0.037 0.392 4.203  Strain M2-3 3.688 1.128 2.270 10.479 0.197 2.515 53.856 0.676 10.167 3.219 0.022 0.052 0.798 0.004 0.055 0.111 10.672 0.025 0.227 0.001 1.136 1.016 0.036 0.400 3.927  M2 Average 3.554 1.087 2.265 10.492 0.194 2.487 56.976 0.636 9.989 3.316 0.024 0.053 0.780 0.004 0.056 0.108 10.850 0.027 0.237 0.001 1.198 1.009 0.037 0.415 3.780  Standard Deviation 0.126 0.051 0.076 0.287 0.010 0.086 4.705 0.065 0.344 0.105 0.002 0.001 0.069 0.001 0.001 0.009 0.227 0.002 0.009 0.000 0.055 0.093 0.002 0.033 0.507  147  Table 22 (continued) Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain M3-1 3.883 1.129 2.251 10.605 0.176 2.446 50.396 0.755 12.193 3.469 0.026 0.057 0.916 0.006 0.059 0.138 11.258 0.033 0.266 0.002 1.242 1.175 0.047 0.400 4.077  Strain M3-2 3.229 0.937 2.547 11.658 0.178 2.820 63.683 0.697 11.778 2.959 0.020 0.050 0.885 0.004 0.051 0.094 10.051 0.018 0.199 0.001 1.189 0.974 0.038 0.358 3.185  Strain M3-3 2.863 0.866 2.197 10.242 0.160 2.341 65.920 0.687 11.375 3.331 0.029 0.056 0.774 0.006 0.062 0.109 10.501 0.031 0.259 0.001 1.146 0.999 0.044 0.269 4.472  M3 Average 3.324 0.977 2.330 10.828 0.171 2.533 60.020 0.713 11.775 3.252 0.054 0.857 0.005 0.057 0.114 10.599 0.027 0.241 0.001 1.192 1.049 0.043 0.342 3.909  Standard Deviation 0.517 0.136 0.188 0.736 0.010 0.252 8.392 0.037 0.409 0.264 0.004 0.004 0.074 0.001 0.006 0.022 0.610 0.008 0.037 0.000 0.048 0.110 0.005 0.067 0.659  Strain M4-1 4.481 1.225 2.309 10.799 0.197 2.602 46.095 0.605 10.767 3.251 0.025 0.056 0.795 0.005 0.060 0.118 11.375 0.029 0.255 0.001 1.176 1.166 0.055 0.523 4.286  Strain M4-2 3.495 1.076 2.397 10.850 0.192 2.638 51.067 0.761 9.614 2.601 0.013 0.042 0.806 0.002 0.034 0.074 9.429 0.015 0.165 0.001 1.063 0.838 0.050 0.353 3.223  Strain M4-3 3.391 1.036 2.623 11.880 0.204 2.847 58.138 0.809 10.419 2.926 0.012 0.046 0.861 0.002 0.036 0.069 10.843 0.016 0.189 0.001 1.233 0.941 0.048 0.374 4.031  M4 Average 3.575 1.075 2.580 11.712 0.203 2.841 53.698 0.804 10.258 2.826 0.013 0.045 0.902 0.002 0.036 0.077 10.360 0.016 0.182 0.001 1.172 0.937 0.050 0.383 3.601  Standard Deviation 0.602 0.100 0.162 0.610 0.006 0.132 6.052 0.107 0.591 0.325 0.007 0.008 0.035 0.002 0.015 0.027 1.006 0.008 0.047 0.000 0.087 0.168 0.003 0.093 0.555  148  Table 23. Concentrations (mg/L) of all volatile compounds quantified in Pinot noir wines (n=3) fermented at 27 °C by by industrial yeast strains AMH, AWRI 796, BGY, RA17 and RC212; individual Burgundian strains A1, A2 and A3 and mixed Burgundian strains M1, M2, M3 and M4. Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain AMH-1 7.327 2.049 2.173 9.917 0.278 2.765 30.347 0.927 11.872 1.646 0.017 0.039 0.746 0.003 0.033 0.059 8.719 0.012 0.123 0.001 0.850 0.969 0.057 0.580 3.303  Strain AMH-2 8.266 2.245 1.776 8.364 0.266 2.136 25.713 0.708 12.167 1.855 0.024 0.041 0.614 0.004 0.041 0.051 9.957 0.027 0.205 0.001 0.736 1.049 0.051 0.635 3.415  Strain AMH-3 6.096 1.721 2.126 9.661 0.266 2.663 30.439 0.850 11.397 1.624 0.018 0.040 0.689 0.003 0.032 0.053 8.252 0.012 0.117 0.001 0.769 0.889 0.061 0.543 3.356  AMH Average 7.216 2.001 2.024 9.309 0.270 2.520 28.825 0.827 11.804 1.707 0.020 0.040 0.682 0.004 0.035 0.054 8.968 0.017 0.148 0.001 0.784 0.967 0.056 0.586 3.357  Standard Deviation 1.088 0.265 0.217 0.833 0.007 0.338 2.703 0.111 0.388 0.128 0.004 0.001 0.066 0.001 0.005 0.004 0.881 0.009 0.049 0.000 0.058 0.080 0.005 0.046 0.056  AWRI 796-1 4.912 1.273 2.039 9.279 0.316 1.904 50.335 0.592 11.351 2.746 0.029 0.046 0.690 0.009 0.048 0.114 10.172 0.028 0.247 0.002 0.942 0.589 0.060 0.562 2.407  AWRI 796-2 5.098 1.384 2.586 11.449 0.315 2.595 63.878 0.853 12.576 2.184 0.019 0.045 1.008 0.006 0.037 0.128 10.290 0.017 0.199 0.001 1.124 0.586 0.062 0.613 2.413  AWRI 796-3 6.740 1.835 2.629 11.735 0.321 2.565 45.184 0.907 12.423 2.055 0.022 0.039 0.947 0.008 0.036 0.158 10.128 0.016 0.193 0.001 1.058 0.670 0.062 0.584 2.279  AWRI 796 Average 5.585 1.497 2.416 10.815 0.317 2.354 53.129 0.783 12.111 2.327 0.024 0.043 0.882 0.008 0.040 0.133 10.191 0.020 0.213 0.001 1.041 0.614 0.061 0.586 2.364  Standard Deviation 1.006 0.298 0.329 1.343 0.003 0.391 9.656 0.169 0.667 0.368 0.005 0.004 0.169 0.002 0.007 0.022 0.083 0.007 0.029 0.000 0.092 0.047 0.001 0.026 0.076  149  Table 23 (continued)  Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain BGY-1 4.587 1.236 2.487 10.985 0.201 2.441 84.591 0.832 6.707 1.956 0.012 0.031 0.645 0.003 0.024 0.079 8.738 0.013 0.152 0.002 1.003 0.711 0.053 0.634 2.658  Strain BGY-2 4.403 1.248 2.337 10.465 0.232 2.327 60.803 0.841 8.400 2.535 0.023 0.039 0.719 0.004 0.042 0.087 10.698 0.028 0.239 0.002 1.079 0.687 0.052 0.967 2.791  Strain BGY-3 4.019 1.157 2.621 11.569 0.203 2.576 71.157 0.815 7.267 2.091 0.017 0.037 0.650 0.004 0.033 0.082 9.485 0.013 0.162 0.002 1.153 0.669 0.064 0.943 2.983  BGY Average 4.336 1.213 2.481 11.003 0.212 2.448 72.157 0.829 7.456 2.193 0.017 0.036 0.671 0.004 0.033 0.083 9.636 0.018 0.184 0.002 1.077 0.690 0.056 0.847 2.808  Standard Deviation 0.290 0.049 0.142 0.552 0.017 0.124 11.927 0.013 0.862 0.303 0.005 0.004 0.041 0.001 0.009 0.004 0.989 0.009 0.047 0.000 0.075 0.021 0.007 0.186 0.164  Strain RA17-1 4.528 1.191 2.189 10.166 0.214 2.050 76.258 0.585 7.515 2.237 0.029 0.044 0.604 0.007 0.046 0.091 10.115 0.035 0.288 0.002 0.902 1.355 0.065 0.384 4.932  Strain RA17-2 6.022 1.668 2.814 12.789 0.249 2.747 67.286 0.952 8.354 2.055 0.019 0.039 0.855 0.004 0.033 0.120 9.810 0.016 0.199 0.002 1.020 1.406 0.069 0.662 4.661  Strain RA17-3 5.165 1.368 2.196 10.163 0.215 2.053 57.081 0.704 7.719 2.259 0.024 0.040 0.729 0.005 0.041 0.095 10.462 0.031 0.279 0.002 0.874 1.327 0.070 0.843 4.885  RA17 Average 5.219 1.405 2.392 11.004 0.225 2.275 66.743 0.745 7.837 2.180 0.024 0.041 0.726 0.005 0.040 0.101 10.111 0.027 0.255 0.002 0.932 1.357 0.068 0.629 4.827  Standard Deviation 0.750 0.241 0.358 1.515 0.020 0.402 9.595 0.187 0.438 0.112 0.005 0.003 0.126 0.001 0.007 0.016 0.326 0.010 0.049 0.000 0.077 0.040 0.002 0.232 0.145  150  Table 23 (continued)  Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain RC212-1 3.828 1.034 2.790 12.292 0.484 2.701 114.341 0.726 8.937 1.937 0.019 0.038 0.908 0.004 0.030 0.114 8.284 0.012 0.158 0.002 1.142 0.754 0.064 0.372 2.635  Strain RC212-2 3.997 1.103 2.510 11.249 0.545 2.375 100.921 0.699 9.874 2.576 0.029 0.047 0.982 0.007 0.045 0.149 9.816 0.026 0.226 0.002 1.164 1.019 0.063 0.519 3.112  Strain RC212-3 3.832 1.015 2.726 12.001 0.523 2.728 117.376 0.772 8.457 1.838 0.020 0.035 0.926 0.006 0.028 0.166 7.752 0.011 0.150 0.002 1.071 0.638 0.062 0.391 2.508  RC212 Average 3.886 1.050 2.674 11.843 0.517 2.600 110.846 0.732 9.086 2.116 0.023 0.040 0.939 0.006 0.035 0.143 8.615 0.016 0.178 0.002 1.126 0.803 0.063 0.427 2.750  Standard Deviation 0.096 0.046 0.147 0.538 0.031 0.196 8.757 0.037 0.721 0.401 0.006 0.007 0.038 0.001 0.009 0.027 1.072 0.008 0.042 0.000 0.049 0.195 0.001 0.080 0.318  Strain A1-1 4.093 1.196 2.336 11.141 0.231 2.132 85.279 0.692 14.314 3.216 0.034 0.050 0.896 0.009 0.050 0.133 11.572 0.036 0.331 0.003 1.158 0.942 0.059 0.448 3.684  Strain A1-2 4.343 1.162 2.220 10.605 0.221 2.046 89.677 0.660 13.710 3.036 0.033 0.050 0.898 0.009 0.053 0.146 10.991 0.037 0.313 0.003 1.125 0.918 0.052 0.321 3.622  Strain A1-3 4.259 1.146 2.296 10.940 0.231 2.112 87.428 0.668 14.498 3.150 0.035 0.049 1.018 0.009 0.054 0.157 11.280 0.035 0.315 0.003 1.245 0.983 0.054 0.358 3.400  A1 Average 4.226 1.166 2.282 10.888 0.227 2.096 87.520 0.672 14.164 3.131 0.034 0.050 0.936 0.009 0.052 0.145 11.271 0.036 0.320 0.003 1.175 0.947 0.055 0.375 3.569  Standard Deviation 0.127 0.025 0.059 0.271 0.006 0.045 2.199 0.017 0.412 0.091 0.001 0.001 0.070 0.000 0.002 0.012 0.291 0.001 0.010 0.000 0.062 0.033 0.004 0.065 0.149  151  Table 23 (continued)  Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain A2-1 3.993 1.194 2.165 9.840 0.304 2.141 63.587 0.610 9.425 2.083 0.030 0.045 0.694 0.006 0.053 0.085 10.131 0.033 0.270 0.002 1.031 1.074 0.046 0.398 3.250  Strain A2-2 4.996 1.418 2.069 9.367 0.257 2.056 53.581 0.577 8.455 2.017 0.030 0.045 0.655 0.006 0.052 0.095 9.973 0.032 0.252 0.002 0.941 1.022 0.058 0.585 3.496  Strain A2-3 5.062 1.367 2.189 10.048 0.280 2.137 48.726 0.667 9.792 2.223 0.032 0.049 0.667 0.008 0.051 0.123 10.948 0.034 0.280 0.002 1.077 1.377 0.057 0.626 3.275  A2 Average 4.663 1.322 2.140 9.745 0.280 2.110 55.413 0.618 9.217 2.107 0.031 0.046 0.671 0.007 0.052 0.101 10.342 0.033 0.267 0.002 1.017 1.156 0.053 0.533 3.336  Standard Deviation 0.599 0.117 0.063 0.349 0.023 0.048 7.578 0.046 0.691 0.105 0.001 0.002 0.020 0.001 0.001 0.020 0.523 0.001 0.014 0.000 0.069 0.192 0.007 0.121 0.135  Strain A3-1 3.720 1.032 2.025 9.162 0.276 2.010 49.778 0.619 7.745 2.683 0.025 0.040 0.672 0.005 0.038 0.086 8.868 0.023 0.212 0.001 0.924 0.749 0.051 0.345 2.309  Strain A3-2 3.765 1.067 2.124 9.598 0.305 2.092 53.534 0.669 8.117 2.927 0.025 0.042 0.731 0.006 0.039 0.102 9.455 0.025 0.222 0.002 1.059 0.841 0.058 0.325 2.447  Strain A3-3 3.974 1.075 2.073 9.325 0.287 1.914 51.847 0.636 7.443 3.069 0.027 0.042 0.754 0.006 0.042 0.116 9.706 0.026 0.233 0.002 1.035 0.623 0.047 0.372 2.367  A3 Average 3.818 1.057 2.074 9.360 0.290 2.005 51.732 0.641 7.767 2.893 0.026 0.041 0.719 0.006 0.040 0.101 9.342 0.025 0.222 0.002 1.006 0.737 0.052 0.347 2.373  Standard Deviation 0.136 0.023 0.050 0.220 0.015 0.089 1.881 0.026 0.338 0.196 0.001 0.001 0.042 0.000 0.002 0.015 0.430 0.002 0.011 0.000 0.072 0.110 0.005 0.024 0.069  152  Table 23 (continued)  Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain M1-1 6.907 1.865 2.316 10.863 0.280 2.125 55.770 0.769 10.919 3.260 0.031 0.049 nq 0.010 0.050 0.166 12.709 0.031 0.303 0.002 1.358 1.433 0.063 0.884 nq  Strain M1-2 5.305 1.339 2.352 10.661 0.266 2.216 59.531 0.606 9.313 2.924 0.026 0.046 nq 0.007 0.043 0.116 10.373 0.028 0.277 0.002 1.082 1.380 0.064 0.553 nq  Strain M1-3 5.479 1.373 2.268 10.468 0.267 2.180 55.479 0.679 9.531 3.068 0.033 0.049 nq 0.010 0.050 0.137 11.377 0.030 0.295 0.002 1.326 1.334 0.066 0.746 nq  M1 Average 5.893 1.524 2.309 10.648 0.271 2.171 56.836 0.682 9.908 3.080 0.030 0.048 nq 0.009 0.048 0.139 11.471 0.030 0.291 0.002 1.252 1.377 0.064 0.726 nq  Standard Deviation 0.879 0.295 0.042 0.198 0.008 0.046 2.260 0.082 0.871 0.169 0.004 0.002 nq 0.002 0.004 0.025 1.172 0.001 0.013 0.000 0.151 0.050 0.002 0.166 nq  Strain M2-1 5.442 1.481 2.589 11.564 0.241 2.493 58.064 0.738 8.479 2.263 0.020 0.038 0.798 0.006 0.035 0.124 10.123 0.014 0.189 0.001 1.188 1.211 0.060 0.804 nq  Strain M2-2 5.227 1.401 2.472 11.070 0.228 2.388 52.371 0.692 7.928 2.213 0.017 0.037 0.751 0.005 0.032 0.104 9.873 0.014 0.191 0.001 1.296 1.310 0.057 0.752 nq  Strain M2-3 6.883 1.800 2.425 11.110 0.294 2.327 60.003 0.809 9.487 3.440 0.034 0.050 0.856 0.011 0.052 0.136 13.159 0.031 0.298 0.002 1.551 2.127 0.071 1.053 nq  M2 Average 5.801 1.549 2.484 11.197 0.254 2.393 56.554 0.746 8.589 2.626 0.024 0.042 0.799 0.007 0.040 0.120 10.994 0.020 0.225 0.002 1.335 1.538 0.062 0.864 nq  Standard Deviation 0.901 0.211 0.085 0.274 0.035 0.084 3.967 0.059 0.791 0.694 0.009 0.007 0.052 0.003 0.011 0.016 1.829 0.010 0.062 0.001 0.186 0.503 0.007 0.161 nq  153  Table 23 (continued)  Volatile Compound 1,3-butanediol 2,3-butanediol 2-methyl-1-butanol 3-methyl-1-butanol Butanol 1-hexanol Isobutanol Phenylethanol Propanol Ethyl butanoate Ethyl decanoate Ethyl hexanoate Ethyl lactate Ethyl laurate Ethyl octanoate Ethyl palmitate Ethyl acetate Hexyl acetate Isoamyl acetate Isobutyl acetate Methyl acetate Acetaldehyde Benzaldehyde Acetic acid 1,1-diethoxyacetal  Class Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Alcohol Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Ethyl ester Acetate ester Acetate ester Acetate ester Acetate ester Acetate ester Aldehyde Aldehyde Acid Acetal  Strain M3-1 5.957 1.558 2.386 11.033 0.279 2.303 59.140 0.733 11.195 2.954 0.032 0.048 0.823 0.009 0.052 0.127 11.138 0.030 0.293 0.002 1.361 1.295 0.060 0.801 4.205  Strain M3-2 5.167 1.410 2.088 9.675 0.254 1.946 52.546 0.587 10.699 2.975 0.027 0.047 0.726 0.008 0.049 0.130 11.575 0.032 0.289 0.002 1.166 1.282 0.054 0.616 3.984  Strain M3-3 6.126 1.573 2.377 10.995 0.285 2.220 62.952 0.670 11.565 3.442 0.039 0.054 0.759 0.011 0.064 0.125 12.777 0.035 0.334 0.003 1.477 1.176 0.059 0.678 4.330  M3 Average 5.734 1.510 2.278 10.541 0.272 2.151 58.112 0.661 11.122 3.117 0.033 0.050 0.766 0.009 0.055 0.127 11.807 0.032 0.305 0.002 1.333 1.245 0.057 0.693 4.163  Standard Deviation 0.512 0.090 0.170 0.773 0.017 0.187 5.265 0.073 0.435 0.276 0.006 0.004 0.049 0.002 0.008 0.003 0.849 0.002 0.025 0.000 0.157 0.065 0.003 0.095 0.175  Strain M4-1 5.628 1.475 2.303 10.294 0.290 2.220 48.569 0.733 8.609 2.585 0.025 0.041 0.726 0.007 0.039 0.099 9.932 0.025 0.249 0.002 1.091 1.129 0.064 0.703 4.106  Strain M4-2 5.331 1.406 2.526 11.094 0.241 2.427 51.742 0.786 8.268 2.184 0.022 0.038 0.761 0.005 0.038 0.089 9.475 0.016 0.206 0.001 1.126 0.791 0.057 0.671 3.142  Strain M4-3 4.632 1.156 2.073 9.262 0.252 1.920 55.001 0.504 8.283 2.717 0.029 0.044 0.609 0.007 0.049 0.083 9.871 0.030 0.269 0.002 0.982 0.890 0.055 0.605 3.249  M4 Average 5.187 1.343 2.299 10.209 0.261 2.187 51.769 0.673 8.380 2.494 0.025 0.041 0.698 0.006 0.042 0.090 9.751 0.024 0.241 0.002 1.065 0.939 0.059 0.658 3.498  Standard Deviation 0.511 0.168 0.226 0.918 0.025 0.255 3.216 0.150 0.193 0.277 0.004 0.003 0.080 0.001 0.006 0.008 0.248 0.007 0.033 0.000 0.075 0.174 0.005 0.050 0.528  154  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0069142/manifest

Comment

Related Items