Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Metabolic engineering and characterisation of the malolactice wine yeast ML01 Husnik, John Ivan 2006

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

Item Metadata

Download

Media
831-ubc_2007-267288.pdf [ 11.83MB ]
Metadata
JSON: 831-1.0100406.json
JSON-LD: 831-1.0100406-ld.json
RDF/XML (Pretty): 831-1.0100406-rdf.xml
RDF/JSON: 831-1.0100406-rdf.json
Turtle: 831-1.0100406-turtle.txt
N-Triples: 831-1.0100406-rdf-ntriples.txt
Original Record: 831-1.0100406-source.json
Full Text
831-1.0100406-fulltext.txt
Citation
831-1.0100406.ris

Full Text

METABOLIC ENGINEERING AND CHARACTERISATION OF THE MALOLACTIC WINE YEAST ML01 B y J O H N I V A N H U S N I K B . S c , University of Guelph, 1994 M . S c , University of Guelph, 2001 A THESIS S U B M I T T E D IN P A R T I A L F U L F I L L M E N T OF T H E R E Q U I R E M E N T S F O R T H E D E G R E E OF D O C T O R OF P H I L O S O P H Y In T H E F A C U L T Y OF G R A D U A T E S T U D I E S (Genetics) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A December 2006 ©John Ivan Husnik, 2006 A B S T R A C T Malolactic fermentation ( M L F ) is essential for deacidification of high acid grape must and the production of well-balanced wines. The bacterial M L F is unreliable and stuck M L F s often lead to spoilage o f wines and the production of biogenic amines. A genetically stable industrial strain of Saccharomyces cerevisiae was constructed by integrating a linear cassette containing the Schizosaccharomyces pombe malate permease gene (mael) and the Oenococcus oeni malolactic gene (mleA) under control of the S. cerevisiae PGK1 promoter and terminator sequences into the URA3 locus of an industrial wine yeast strain. The malolactic yeast strain, M L 0 1 , completes the M L F during the alcoholic fermentation in a variety of musts including a high acid Chardonnay must containing 9.2 g/L of malate. M L 0 1 cannot appreciably decarboxylate L-malic acid to L-lactic acid when present at levels below 1% of the total inoculum. M L 0 1 contains no antibiotic resistance marker genes or vector D N A sequences. Global gene expression patterns and analysis of the proteome showed that no metabolic pathway was affected by the introduction of the malolactic cassette. The presence of the malolactic cassette in the genome does not affect growth, ethanol production, fermentation kinetics or metabolism of M L 0 1 . Wines produced by the M L 0 1 yeast have lower volatile acidity and improved color properties compared to wines produced with the parental yeast and a bacterial M L F . G C / M S analysis of volatile compounds revealed that wine produced by M L 0 1 did not contain any compounds that were not detected in wine produced with the parental strain S92 or with S92 and malolactic bacteria. Moreover, M L 0 1 reduces the processing time after alcoholic fermentation and produces wine that is judged highest in overall quality by trained tasters. Analyses of the phenotype, D N A , R N A , and proteins demonstrate that the recombinant yeast M L 0 1 is substantially equivalent to the parental strain S92. M L 0 1 has been approved for use in Canada and has 'Generally Regarded A s Safe' status with the U S F D A . It is the first metabolically engineered yeast to be commercialised by the wine industry and is currently available in Canada, the U S A and Moldova. T A B L E OF C O N T E N T S A B S T R A C T i i T A B L E O F C O N T E N T S iv LIST OF T A B L E S x i LIST O F F I G U R E S xiv LIST OF A B B R E V I A T I O N S xvi P R E F A C E xxi A C K N O W L E D G E M E N T S xxi i 1 I N T R O D U C T I O N 1 1.1 Saccharomyces cerevisiae 2 1.2 Overview of winemaking 4 1.3 The malolactic fermentation 7 1.3.1 Wine acidity 7 1.3.2 Oenococcus ami and other lactic acid bacteria o f wine 8 1.3.3 Biochemistry of the malolactic fermentation 11 1.3.4 The effect of the malolactic fermentation on wine and the winemaking process 11 1.4 Genetic engineering of Saccharomyces cerevisiae to perform the M L F 17 1.5 Genetically engineered Saccharomyces cerevisiae strains in the wine industry.... 21 iv 1.6 Proposed Research 31 1.6.1 Significance of research 31 1.6.2 General hypothesis 31 1.6.3 Ma in objectives 32 2 M A T E R I A L S A N D M E T H O D S 33 2.1 Strains and plasmids employed in the genetic construction and characterisation of M L 0 1 33 2.2 Culture conditions 34 2.3 Genetic construction of malolactic wine yeast 35 2.3.1 Co-transformation of the malolactic cassette and pUT332 35 2.3.2 Screening of transformants for the integrated malolactic cassette 36 2.3.3 Loss of plasmid pUT322 37 2.4 Functionality of malolactic wine yeast 37 2.4.1 Malate decarboxylation and residual sugar concentrations of wine produced by malolactic clones 37 2.4.2 Functionality of active dry wine yeast M L 0 1 38 2.5 Genetic characterisation of M L 0 1 38 2.5.1 Chromosome karyotyping of M L 0 1 and S92 38 2.5.2 Southern blot analyses 39 2.5.3 Sequencing o f the malolactic cassette integrated into the genome o f M L 0 1 .. 40 2.5.4 Sequence analysis 42 2.5.5 Genetic stability of the malolactic cassette in the genome of M L 0 1 42 v 2.5.6 Global gene expression analyses 43 2.5.7 Confirmation of D N A microarray results by Real-Time P C R 45 2.5.8 Transcription of URA3 and transgenes mael and mleA 46 2.5.9 Analysis of the proteome of M L 0 1 47 2.6 Phenotypic characterisation of M L 0 1 49 2.6.1 Growth kinetics of M L 0 1 49 2.6.2 Utilization o f malate as sole carbon source by M L 0 1 and S92 50 2.6.3 Winemaking with M L 0 1 51 2.6.4 Analyses of must and wine 54 2.6.5 Sensory analysis 55 2.6.6 Analysis of volatile compounds in wine by gas chromatography/mass spectrometry 57 2.6.7 Quantification of ethyl carbamate in wine produced by M L 0 1 58 2.6.8 Effect o f residual M L 0 1 populations on M L F in wine fermented primarily with the parental strain S92 58 2.6.9 Post-fermentation viability of M L 0 1 59 2.7 Statistical analyses 59 3 R E S U L T S 61 3.1 Integration of the malolactic cassette into the genome of S92 61 3.2 Functionality of malolactic wine yeast 61 3.3 Genetic characteristics of M L 0 1 63 3.3.1 Confirmation of the identity of the parental strain 63 vi 3.3.2 Correct integration of the malolactic cassette into the genome o f M L 0 1 63 3.3.3 Genetic stability of the malolactic cassette in the genome of M L 0 1 66 3.3.4 M L 0 1 does not contain bla and Tn5ble antibiotic markers 66 3.3.5 Sequence of the malolactic cassette integrated into the genome of M L 0 1 68 3.3.6 Effect of the integrated malolactic cassette on the transcriptome of M L 0 1 .... 71 3.3.7 Effect of the integrated malolactic cassette on the proteome of M L 0 1 76 3.4 Phenotypic properties of M L 0 1 77 3.4.1 Growth kinetics 77 3.4.2 Utilization of malate as sole carbon source by M L 0 1 and S92 77 3.4.3 Malolactic fermentation in Chardonnay and Cabernet Sauvignon musts by M L 0 1 78 3.4.4 Sensory profile of Chardonnay wines produced by M L 0 1 84 3.4.5 Volatile compounds in wine produced by M L 0 1 88 3.4.6 Ethyl carbamate in wine produced by M L 0 1 90 3.4.7 Effect of increasing populations of M L 0 1 on M L F in wine conducted with the parental yeast 91 3.4.8 Post-fermentation viability of M L 0 1 93 4 D I S C U S S I O N 94 4.1 Integration of the malolactic cassette into the genome of S92 yielded the functional malolactic yeast M L 0 1 95 4.2 M L 0 1 completes the M L F during alcoholic fermentation in Chardonnay and Cabernet Sauvignon musts 97 4.3 Wines produced by M L 0 1 have improved physicochemical and organoleptic properties , 99 4.4 Integration of the malolactic cassette in S92 does not confer an advantage to M L 0 1 nor does it affect the production of A D Y or wine making 104 4.5 The malolactic cassette without any antibiotic resistance markers is integrated correctly and stably into the genome of M L 0 1 105 4.6 The malolactic cassette has a minimal effect on the transcriptome and proteome of M L 0 1 110 4.7 Ethical considerations concerning use of the genetically modified yeast M L 0 1 114 5 C O N C L U S I O N S 121 5.1 Future Directions 124 R E F E R E N C E S 125 A P P E N D I X A Schematic representation of pUT332 and probes used to confirm the absence o f antibiotic markers in M L 0 1 156 A P P E N D I X B Strategy for sequencing the malolactic cassette from the genome o f M L O l 157 A P P E N D I X C The University of British Columbia Clinical Research Ethics Board approval certificates for sensory analysis o f wines produced by M L 0 1 158 A P P E N D I X D Participant consent form 160 A P P E N D I X E P C R confirmation of the screening method 164 A P P E N D I X F Degradation o f malate and consumption o f glucose and fructose by malolactic wine yeast clones 165 A P P E N D I X G Confirmation o f the identity of the parental strain by P C R amplification of the d elements of the T y l retrotransposon 166 A P P E N D I X H Southern blot confirming integration of the malolactic cassette into the URA3 locus of S92 (PGK1 promoter probe) 167 A P P E N D I X I Ascospore formation by the M L 0 1 and S92 strains 168 A P P E N D I X J Detailed description of the D N A sequences that comprise the malolactic cassette 169 A P P E N D I X K Discrepancies between the integrated malolactic cassette and published sequences 172 A P P E N D I X L Confirmation o f D N A microarray data by real-time reverse transcription P C R 173 A P P E N D I X M Transcripts of mael, mleA and URA3 in the M L 0 1 yeast 174 A P P E N D I X N Vinification trials with M L 0 1 and S92 in Chardonnay must (2004 harvest) 175 ix A P P E N D I X O Physicochemical characteristics of Chardonnay wines (2004 harvest) produced by M L 0 1 , S92, and S92 plus O. ceni 176 A P P E N D I X P Correlation matrix 177 A P P E N D I X Q Health Canada approval to use M L 0 1 for the commercial production of wine in Canada 178 A P P E N D I X R U S Food and Drug Administraion G R A S notice 180 L I S T O F T A B L E S Table 1. Oenological targets for the genetic improvement of S. cerevisiae wine strains 22 Table 2. Genetic engineering of wine yeast and stratagies used for their modification 24 Table 3. Different strains used in the metabolic engineering and characterisation of the malolactic yeast M L 0 1 33 Table 4. Different plasmids used in the metabolic engineering and characterisation of malolactic clone M L 0 1 34 Table 5. Primers used for the sequencing of the malolactic cassette integrated into S. cerevisiae M L O l 41 Table 6. Primers used in semi-quantitative reverse transcriptase real-time P C R to confirm D N A microarray data 46 Table 7. Definition of sensory attributes for visual, olfactory and gustatory evaluations 56 Table 8. Effect of the integrated the malolactic cassette in the genome of S92 on global gene expression patterns in S. cerevisiae M L O l at 48 hours (> 2 fold change) 74 Table 9. Effect of the integrated malolactic cassette in the genome of S92 on global gene expression patterns in S. cerevisiae M L O l at 144 hours (> 2 fold change) 74 Table 10. Physicochemical and colour measurements of high-acid Chardonnay wines (2000 harvest) produced by M L O l , S92 and S92 plus O. ami 82 xi Table 11. Physicochemical and colour measurements of Cabernet Sauvignon wines produced by M L 0 1 , S92 and S92 plus O. ami 83 Table 12. F-values from analysis of variance of Chardonnay wines for sensory attributes (three wines, 14 judges, two replications) 85 Table 13. Concentration of volatile compounds in Chardonnay wines produced with M L 0 1 , S 9 2 , S 9 2 p l u s a « m 89 Table 14. The production of ethyl carbamate in Chardonnay wines produced with M L 0 1 , S92 and S92 with a bacterial M L F 91 Table 15. Degradation of malate (g/L) by individual colonies of malolactic clones inoculated into synthetic must containing 4.5 g/L of malate 165 Table 16. Consumption of glucose (g/L) by individual colonies of malolactic clones inoculated into synthetic must containing 100 g/L of glucose (and 100 g/L fructose) 165 Table 17. Consumption of fructose (g/L) by individual colonies of malolactic clones inoculated into synthetic must containing 100 g/L of fructose (and 100 g/L glucose) 165 Table 18. Comparison of the M L 0 1 and S92 strains capability to sporulate 168 Table 19. The source and description of the malolactic cassette D N A sequences 169 Table 20. Discrepancies found when comparing the genomic sequence o f the integrated locus and previously published sequences 172 x i i Table 21. Comparison of fold change data for ten genes as determined by D N A microarray and by real-time reverse transcription P C R 173 Table 22. Physicochemical and colour measurements of Chardonnay wines (2004 harvest) produced by M L O l , S92 and S92 plus O. ami 176 Table 23. Correlation matrix obtained from the sensory analysis of Chardonnay wines 177 x in L I S T O F F I G U R E S Figure 1. Schematic representation of the malolactic cassette integrated into the URA3 locus of S. cerevisiae S92 61 Figure 2. Test for L-lactic acid production in S92 cells transformed with the malolactic cassette and pUT332 62 Figure 3. M L 0 1 and S92 chromosomes as separated by pulsed field gel electrophoresis 63 Figure 4. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot analysis using a URA3 probe 64 Figure 5. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot analysis using a mleA probe 65 Figure 6. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot analysis using a mael probe 65 Figure 7. The phleomycin resistance gene is absent in the genome of M L 0 1 67 Figure 8. The ampicillin resistance gene and 970 bp of pUT332 non-Saccharomyces vector are absent in the genome of M L 0 1 68 Figure 9. The upstream and downstream sequences flanking the malolactic cassette in S. cerevisiae M L 0 1 are 100% identical 69 xiv Figure 10. A schematic representation of new open reading frames (ORFs) of more than 100 codons generated during construction of the malolactic cassette; four new ORFs primarily composed of S. cerevisiae sequences, were created 70 Figure 11. Malate degradation and lactate production by M L O l and S92 yeast strains in synthetic must (n=3) 72 Figure 12. Growth of M L O l and S92 yeast strains in synthetic must (n=3) 73 Figure 13. Ethanol production and C 0 2 loss of M L O l and S92 yeast strains in synthetic must (n=3) 75 Figure 14. M L O l and S92 cannot consume L-malate as a sole carbon source 78 Figure 15. Ethanol production by M L O l and S92 in high-acid Chardonnay grape must fermented at 20 °C was positively affected by introduction of the malolactic cassette into a URA3 locus in the industrial wine yeast S92 80 Figure 16. M L F by M L O l is completed in the first five days of the alcoholic fermentation in high-acid Chardonnay grape must (9.2 g/L) 81 Figure 17. Cobweb diagram showing the significantly different mean sensory attributes o f Chardonnay wines produced by M L O l , S92, and S92 plus O. ceni (n=26) 86 Figure 18. A principal component analysis plot showing the significantly different mean sensory attributes for Chardonnay wines produced by M L O l , S92, and S92 plus O. azni 87 xv Figure 19. M L F was not detected in wines containing an inoculum less than 1% of M L 0 1 yeast : 92 Figure 20. Post-fermentation viability of M L 0 1 is similar to that of S92 in Chardonnay wine 93 Figure 21. Plasmid map of pUT332 and schematic representation of the probes used in Southern blot experiments to confirm the absence of antibiotic markers in S. cerevisiae M L 0 1 156 Figure 22. Strategy used to sequence the malolactic cassette from the genome of S. cerevisiae M L 0 1 157 Figure 23. P C R confirmation of the screening method used to detect integration of the malolactic cassette into the URA3 locus of S92 164 Figure 24. Genetic patterns o f the M L 0 1 and the S92 yeast strains based on amplification of genomic D N A regions in between d elements of the T y l retrotransposon 166 Figure 25. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot analysis using a PGK1 promoter probe 167 Figure 26. The presence of the mael, mleA and URA3 transcripts in the M L 0 1 yeast during fermentation 174 Figure 27. Malate degradation and lactate and ethanol production by M L 0 1 and S92 in Chardonnay must (from fruit harvested in 2004)..... 175 xvi LIST OF ABBREVIATIONS aa Amino acid A D Y Active dry yeast amu Atomic mass unit A N O V A Analysis of variance A O A C Association of Official Analytical Chemists bp Base pair B C British Columbia B C E Before the Common Era °C Degree Celsius c D N A Complementary deoxyribonucleic acid C F U Colony forming units C I E L A B Color model of the International Commission on Illumination cm Centimetre Co. Company c R N A Ribonucleic acid derived from c D N A D N A Deoxyribonucleic acid D T T Dithiothreitol E C Ethyl carbamate E D T A Ethylenediamine tetraacetic acid F D A Food and Drug Administraion g gram g . Unit of acceleration (9.80665 m/s 2) xvn G C Gas chromatography G C / M S Gas chromatography-mass spectrometry G M Genetically modified G M O Genetically modified organism G O Gene ontology GPvAS Generally regarded as safe h L hectolitre ID Inner diameter kbp kilo base pair kg kilogram L Litre L A B Lactic acid bacteria L C M S / M S Liquid chromatography-tandem mass spectrometry log Logarithm L S D Least square difference m metre M Molarity u L microlitre um micrometre umax Maximum specific growth rate mg milligram min minute m L millilitre M L F Malolactic fermentation m M millimolar M M T S Methyl methanethiosulphonate M S Mass spectrometer N A D Nicotinamide adenine dinucleotide N C B I National Center for Biotechnology Information ng nanogram nt nucleotide O D Optical density O R F Open reading frame P A R C Pacific Agri-Food Research Centre P C A Principal component analysis P C R Polymerase chain reaction P D M Prise de Mousse P E S T Protein region that consists of proline, glutamic acid, serine, threonine and to a lesser extent aspartic acid P F G E Pulsed Field Gel Electrophoresis R N A Ribonucleic acid rpm Revolutions per minute R T - P C R Reverse transcriptase - polymerase chain reaction s second S C X Strong cation exchange S D S - P A G E Sodium dodecyl sulfate polyacrylamide gel electrophoresis xix S G D Saccharomyces genome database sp. Species S P M E Solid phase microextraction T A Titratable acidity T B E Buffer consisting of Tris base, boric acid, E D T A and water T C A Tricarboxylic acid T E Buffer consisting of Tris base, E D T A and water T O F M S Time-of-flight mass spectrometry U K United Kingdom U S A United States of America V volt v/v volume per volume Y A N C Yeast assimilable nitrogen concentration Y E G Medium consisting of yeast extract and dextrose Y P D Medium consisting of yeast extract, peptone, and dextrose xx P R E F A C E The following dissertation is prepared in the traditional format as described by the Faculty of Graduate Studies at the University of Brtitsh Columbia. It encompasses two different fields of study, oenology and molecular biology. Hence, some of the terms used by oenologists may not be familiar to molecular biologists and vice versa. In most cases the meaning of certain terms wi l l be evident to the reader. The remaining terms, such as "must" may be confusing to non-oenologists and therefore I have provided a short list o f definitions of such oenological terms. Br ix A measurement of the dissolved solids (primarily sugars) in fruit juices at 20°C. Lees The sediment from an alcoholic fermentation (primarily composed o f yeast). Must The unfermented or fermenting juice of grapes (or other fruits). Racking The process of removing wine off the lees to allow clarification and aid in stabilisation. xxi ACKNOWLEDGEMENTS I would like to acknowledge the contributions of the following people and organisations, each of whom played a critical role in the completion of this research and my development as a scientist. First my sincere thanks to Dr. Hennie J J . van Vuuren, my Research Supervisor, for his support, guidance in writing this thesis, financial assistance, and for the innumerable occasions that I have knocked on his door and was immediately welcomed. Working with Prof has provided me the opportunity to do cutting edge research in a very well equipped laboratory, and helped me to appreciate how fundamental science can partner effectively with industry. I truly appreciate the opportunities that have been presented to me as a member of his lab. I would also like to acknowledge the contributions of my Supervisory Committee: Dr. Brian El l is , Dr. Phi l Hieter and Dr. J im Kronstad. I appreciate the breadth of experience and perspectives which they brought to our meetings and I learned a lot from their advice and criticism. A special thank you to Dr. Ron Subden (University of Guelph), who was key to my early development as a scientist and M r . J.P. Rossi (Lesaffre International) for his commitment to the idea of a commercial malolactic wine yeast strain. I am also thankful for the productive collaboration with Drs. Jurgen Bauer, Margaret Cliff, Didier Colavizza, Pascal J. Delaquis, Zongli Luo, and Heinrich Volschenk, my co-authors on the manuscripts that have arisen thus far from this thesis. Also a thank you to my Wine Research Centre lab colleagues past and present, especially Dr. Joanna Coulon, Dr. Danie Erasmus, Dr. Zongli Luo, Dr. George van der Merwe and xxn Dr. Heinrich Volschenk for their friendship and scientific discussions. I would also like to thank my colleagues and friends at Lesaffre Development, especially J.P. Rossi^, Dr. Didier Colvizza, Dr. Jurgen Bauer, and Olivier Letalleur that welcomed me to their research centre in Marcq-en-Baroeul, France. I greatly appreciate the contributions of those who funded this research or a portion of my studies: Natural Sciences and Engineering Research Council o f Canada, Lesaffre International and the Canadian Vintners Association. Thank you also to the Genetics Graduate program and the U B C Faculty of Graduate Studies for their assistance with travel awards to enable me to present some of my results at international conferences. I would also like to thank Dr. Hugh Brock and Monica from the Genetics Graduate Program for their assistance over the years. And a final thank you to my parents, Ignac and Annie Husnik, and my sister, Angela, for their unconditional support. Their love and sincere happiness in my pursuits are, and have always been, a great source o f encouragement, fala. Also thank you to Bob and Marlene Rideout for their support and assistance. To Candice, thank you, thank you, and thank you - for the love, inspiration, motivation, and support you provided. I also greatly appreciate the time you have spent reviewing this dissertation, the help with the preliminary pages, and for all o f the suggestions and comments throughout the manuscript. The thought of trying to complete this thesis without you is unimaginable. * - deceased x x i n 1 INTRODUCTION Approximately 27 bil l ion litres of wine are produced annually worldwide (Pretorius and Bauer, 2002); the quality of wine depends on the grapes used in its production, the microorganisms involved in the alcoholic and malolactic fermentations, and the skills of the winemaker. The malolactic fermentation ( M L F ) is an indispensable tool for the deacidification of high acid grape must; it is also one of the most difficult steps to control in the winemaking process. Oenococcus ceni and other lactic acid bacteria ( L A B ) deacidify wine by converting L-malic acid to L-lactic acid and CO2, resulting in a decrease in titratable acidity and an increase in wine p H (Bousbouras and Kunkee, 1971). The decarboxylation of malate to lactate is catalyzed by the malolactic enzyme (L -mala te :NAD + carboxy lyase) without the production of any free intermediates (Caspritz and Radler, 1983; Naouri et al., 1990; Spettoli et al., 1984). The reduction o f acidity in wine is particularly important in cooler climates where L -mal ic acid can be present at concentrations up to 9 g/L. In spite of the use of commercially available bacterial malolactic starter cultures, stuck and sluggish M L F s are common in wines because growth of L A B can be inhibited by many factors including sulphur dioxide, low temperature, p H , ethanol, low nutrient content of wine, the presence of fatty acids and interactions with other microorganisms (see reviews Davis et al., 1985; Henick-Kling, 1995; van Vuuren and Dicks, 1993; Wibowo et al., 1985). For the entire duration o f the bacterial M L F , wine is at risk from microbial spoilage and oxidation since the addition of sulphur dioxide must be delayed and the temperature is often elevated in order to achieve a satisfactory M L F . Moreover, L A B can produce toxic substances such as biogenic amines and precursors of ethyl 1 carbamate that are of concern to consumers (Liu, 2002; Lonvaud-Funel, 2001; Marcobal et al., 2006). This research was undertaken to construct and characterise an industrial wine yeast strain that can avoid the negative aspects of the bacterial M L F by completing the M L F during the alcoholic fermentation. The resulting wine can be immediately sulphited and processed; thereby reducing the effects of chemical oxidation and the probability of microbial spoilage and the production of off-flavours. The construction of a malolactic yeast strain that can complete the M L F during the alcoholic fermentation w i l l be a significant addition to the repertoire of tools that winemakers can access to produce high-quality wines that should be free of toxic biogenic amines such as histamine and tyr amine. In this chapter the yeast Saccharomyces cerevisiae w i l l be briefly introduced, followed by an overview of winemaking, a detailed description of the M L F and the history of the genetic engineering o f S. cerevisiae to perform the M L F . Finally, a review of all the genetically engineered S. cerevisiae strains applicable to the wine industry wi l l be presented. 1.1 Saccharomyces cerevisiae The budding yeast S. cerevisiae is considered by many as a domesticated unicellular organism. For centuries, S. cerevisiae has been used by humans in the fermentation industries of baking, brewing, distilling and winemaking. In 1996, the laboratory strain of S. cerevisiae S288C, was the first eukaryotic organism to have its entire genome sequenced (Goffeau et al., 1996). S288C is largely derived from a strain, 2 E M 9 3 , that was originally isolated in 1938 from rotting figs (Mortimer and Johnston, 1986). The yeast genome contains approximately 5800 genes (-70% characterised as of September, 2006) (http://www.yeastgenome.org/) containing relatively few introns and little repetitive D N A . Haploid laboratory strains have approximately 12-13 M b of D N A confined to 16 linear chromosomes (Olson, 1991). Saccharomyces cerevisiae as haploid or diploid cells can reproduce vegetatively through budding. Haploids can be either "a" or " a " mating types; mating of opposite types wi l l yield a diploid cell (Herskowitz, 1988). Dip lo id cells can undergo sporulation (meiosis), induced by growth in poor carbon sources or nitrogen starvation, to produce four haploid spores formed within an ascus (Herskowitz, 1988). The recovery of the meiotic products allows advanced genetic analyses to be performed that are not possible in most eukaryotic organisms. S. cerevisiae can also be heterothallic or homothallic. Heterothallic strains have a fixed mating type and homothallic strains are able to switch mating types (Herskowitz, 1988). Most laboratory strains of S. cerevisiae are haploid or diploid, heterothallic, sporulate efficiently when diploid and contain multiple nutritional auxotrophic mutations. In contrast, industrial strains are predominantly diploid or anueuploid and occasionally polyploid, homothallic, prototrophic and have low sporulation efficiency combined with poor spore viability (Codon et al., 1998; Snow, 1983). Industrial strains are grown under commercial manufacturing conditions to be used as inocula in their respective industries, such as baking and winemaking. W i l d isolates are mostly homothallic diploids that are often heterozygous (Bisson, 2004). 3 Although many natural isolates of S. cerevisiae have been obtained from grapes, the actual origin of wine strains is the subject of some controversy. Some researchers claim that the principal source of wine yeast is the vineyard (Mortimer and Polsinelli , 1999; Torok et al., 1996); others believe that modern wine strains are the result of an association with artificial environments such as wineries (Martini, 1993; Vaughan-Martini and Martini , 1995). The controversy is due to the fact that it is relatively difficult to find S. cerevisiae on the surface of healthy, undamaged grapes (Martini, 1993), although it can be isolated in berries damaged by birds or insects which represent about 1 in 100 grapes (Landry et al., 2006). The population size of S. cerevisiae within any damaged fruit can range from 10 4 to 10 5 cells (Mortimer and Polsinelli, 1999). However, a recent investigation on the survival and development of an inoculated S. cerevisiae wine strain showed that this particular wine strain failed to colonise and was unable to out compete the epiphytic yeast present on damaged and undamaged grapes (Comitini and Ciani, 2006). Other environments that S. cerevisiae has been isolated from include soil associated with oak trees (Sniegowski et al., 2002), the Danube River (Slavikova and Vadkertiova, 1997) and occasionally from immunocompromised humans (Malgoire et al., 2005; Sethi and Mandell , 1988). Evidence also exists that insects and birds are important vectors for the dispersal of yeasts (Mortimer and Polsinelli , 1999; Phaff and Starmer, 1987). 1.2 Overview of winemaking Several lines of archaeological evidence suggest that wine was made as early as the seventh century B C E (Robinson, 1994). Molecular evidence for fermentation by S. 4 cerevisiae in wine has been obtained from wine jars discovered in Egypt dating back to 3150 B C E (Cavalieri et al., 2003). Alcoholic fermentation represents the oldest form of biotechnological application of microorganisms and from the early days of winemaking to the present, the basic principles have changed very little. Following the harvest of grapes (which can be performed manually or by machine), the berries are delivered to the winery where they are crushed and the stems removed. A t this point, modern winemakers generally add pectolytic enzymes to increase the volume of free juice and to assist with clarification (Moreno-Arribas and Polo, 2005). After crushing, the grape must can be used directly for fermentation (for red wines) or pressed to separate the juice from the skins (for white wines). The must may be fermented with selected wine yeast strains or left to be fermented by the resident microflora found on the surfaces o f grapes and in the winery. The inoculation of commercial yeast starter strains into must was started in the 1960s and by the mid-1980s became common practice in most of the world's wine regions (Moreno-Arribas and Polo, 2005; Reed and Nagodawithana, 1988). Approximately 150 different strains are now commercially available; these strains produce high quality wines with reproducible characteristics including a complete and rapid fermentation (Schuller and Casal, 2005). In uninoculated "spontaneous" fermentations there is a sequential growth pattern of indigenous yeasts. Yeast o f the genera Kloeckera, Hanseniaspora and Candida generally predominate in the early stages, followed by Metschnikowia and Pichia when ethanol concentrations reach 3-4% (Fleet and Heard, 1993). The final stages are ultimately dominated by the ethanol-tolerant S. cerevisiae, which rapidly ferments the present sugars (18-40%, depending on style and geography of wine being 5 produced) to produce ethanol and carbon dioxide (Querol et al., 2003). The high levels of ethanol, the low p H , osmotic stress and anaerobic conditions essentially eliminate the other less tolerant microorganisms. During the fermentation of red wines, the skins and other insoluble material w i l l form a cap on the surface that must be submerged or stirred in order to extract the skin cell components (Bisson, 2004). The cap can be submerged by either "punching down" or must from the bottom of the tank can be "pumped over" the cap. Red wines are also fermented at higher temperatures, 18-30 °C, and whites are commonly fermented at 12-18 °C. After the alcoholic fermentation, certain styles of wines are immediately treated with SO2 to prevent chemical oxidation and inhibit microbial activity. Most red wines and certain white wines are not sulphited until after they have undergone a M L F . After the M L F , the wine is clarified by racking, fining, centrifuging or filtering (Boulton, 1996). The winemaker may also choose to delay the clarification in order to extract desired flavours from the lees (yeast sediment) (Ough, 1992). During wine processing and storage, the SO2 levels are adjusted to prevent oxidation of wine and proliferation of spoilage microorganisms; protective SO2 levels are then monitored and maintained until bottling. After clarification, wines can be stored, at low temperatures with no air contact, in inert containers or oak barrels. Prior to the final step in the winemaking process, bottling, wines may be blended and tested for physical instabilities, such as the potential to form precipitates (Boulton, 1996). 6 1.3 The malolactic fermentation The M L F refers to the biological conversion of the dicarboxylic L-malic acid, into the monocarboxylic acid L-lactic acid, and CO2. In 1858, Pasteur's "Memoire sur la fermentation lactique^ revealed that the M L F was caused by l iving organisms; Muller-Thurgau, in 1891, showed that the organisms were bacteria (Bartowsky, 2005; Paul, 1996). In 1901, the equation for the conversion of L-malic acid to L-lactic acid and CO2 was independently revealed by Moslinger and Seifert (Bartowsky, 2005). In 1928, using Moslinger and Seifert's equation, Ferre showed that the bacteria transformed 1.0 g of malic acid into 0.671 g of lactic acid and 0.329 g of CO2 (Paul, 1996). Since this time, M L F has continued to attract considerable attention from researchers around the globe. The bacterial M L F is an important secondary fermentation that typically occurs after the alcoholic fermentation has been completed. The M L F deacidifies wine and is favoured in cool-climate regions (such as northern Europe, eastern United States, New Zealand and Canada) where the grapes at harvest tend to have naturally higher acid at harvest. Conversely, in warmer regions of the world, grapes usually have lower levels of acidity and the M L F is less desired (Beelman and Gallander, 1979). The difference in total fixed acids between cool and warm climates is partly the result o f respiratory catabolism of L-malic acid by the grape that is enhanced by warmer temperature during the ripening (Jackson and Schuster, 1987). 1.3.1 Wine acidity Grape juice contains a variety of organic acids, the dominant acids being tartaric, malic and citric acids. Tartaric and malic acids represent, on average, 90% of the 7 titratable acids prior to the fermentation (Boulton, 1996; Radler, 1993; Ribereau-Gayon et al., 2000a). Near maturity in warmer climates, tartaric acid is the predominant acid in grapes, accounting for 2.0-8.0 g/L, with malic acid accounting for 10-40% of the total acid fraction (Boulton, 1996; Ough, 1992). In cooler climates or in grapes picked at early maturity, the amounts of malic acid can exceed those of tartaric acid and may constitute as much as 60% of the organic acid fraction (Boulton, 1996; Ough, 1992). Mal i c acid is usually present in grapes at concentrations ranging from 2.0-6.0 g/L (Boulton, 1996), but can reach 9 g/L in cool viticultural areas. L -Ma l i c acid is an essential compound, with important cellular functions in metabolic pathways such as the tricarboxylic acid ( T C A ) cycle, glyoxylate cycle, and malate-aspartate shuttle; in grapes it is synthesized from glucose via pyruvate (Mathews and van Holde, 1990). Smaller amounts of citric acid and trace amounts of other acids o f the citric acid cycle are also present in the juice. 1.3.2 Oenococcus ceni and other lactic acid bacteria of wine Lactic acid bacteria ( L A B ) responsible for the M L F in wine are from the bacterial genera Lactobacillus, Pediococcus, Leuconostoc and Oenococcus (Dicks et al., 1995; London, 1976; van Vuuren and Dicks, 1993). The L A B can be isolated from the skins of healthy grape berries in low concentrations (<103 colony forming units/g) (Fleet, 1998; Lonvaud-Funel, 1999; Wibowo et al., 1985) and can also be found in the winery (Boulton, 1996). In grape must the L A B population varies from 10 2 cfu/mL to 10 4 cfu/mL, however, during alcoholic fermentation the L A B population w i l l decrease to a few cells per m L (Lonvaud-Funel, 1999; van Vuuren and Dicks, 1993). Moreover, the original diversity of the species diminishes and O. ceni typically remains as the 8 predominant L A B at the end o f alcoholic fermentation (Lonvaud-Funel, 1999; Tracey and Britz, 1989; van Vuuren and Dicks, 1993). In 1967, O. ami was first characterised and classified as Leuconostoc oenos, to differentiate it from other less acid-tolerant Leuconostoc species (Garvie, 1967). Leuconostoc oenos was later re-classified into the new genus, Oenococcus, o f which O. ceni is the only species (Dicks et al., 1995). The genomes of two O. ami strains, IOEB8413 and P S U 1 , have recently been sequenced (Klaenhammer et al., 2002; M i l l s et al., 2005), although neither of the sequences are currently accessible by the public. The reported size of the O. ceni genomes are 1.78 and 1.75 M b for PSU1 and IOEB8413, respectively (Mil l s et al., 2005). The use of multilocus sequence typing and phylogenetic analyses revealed that O. ceni seems to have a high level of genetic heterogeneity and may also have a panmictic (highly variable) population structure (de las Rivas et al., 2004). A number of O. ami genes have also been studied at the molecular level including genes associated with the M L F (Denayrolles et al., 1994; Labarre et al., 1996a; Labarre et al., 1996b; Volschenk et al., 1997b), genes involved with diacetyl metabolism (Garmyn et al., 1996), genes associated with amino acid metabolism (Coton et al., 1998a; Divo l et al., 2003; Marcobal et al., 2004; Tonon et al., 2001), and stress-related genes (Bourdineaud et al., 2003; Bourdineaud et al., 2004; Fortier et al., 2003; Jobin et al., 1999a; Jobin et al., 1999b; Jobin et al., 1997). A putative glucosyltransferase gene, dps, was also detected in O. ami isolated from a highly viscous wine (Walling et al., 2005). The M L F begins during the early stages of O. oeni growth (or the growth of other L A B ) and a significant rate of malate degradation is usually not observed until cell densities reach a concentration of 10 6 cfu/mL or more (Costello et al., 1983; Lafon-9 Lafourcade et al., 1983; Wibowo et al., 1985). L A B cannot grow with L-malic acid as a sole carbon source and they rely on residual sugars not fermented by the yeast (Liu et al., 1995a) or amino acids such as arginine to allow cell growth (L iu and Pilone, 1998). Although, O. ceni cannot grow on L -mal ic acid as a sole carbon source, the malolactic conversion confers an energetic advantage to the cell. The M L F provides cells with additional metabolic energy from the increased intracellular p H which produces an increased proton motive force (? p) across the cytoplasmic membrane (Cox and Henick-Kl ing , 1989; Cox and Henick-Kling, 1995). The proton gradient is created across the cell membrane after one molecule of lactate (after malate decarboxylation) leaves the cell accompanied by one H + (Cox and Henick-Kling, 1989). The resulting increase in the proton gradient can be used to drive transport processes and to produce A T P via the membrane-bound ATPase (Cox and Henick-Kling, 1989; Henick-Kling, 1995; Olsen et al., 1991). A t a p H of <4.5 and with limited amounts of sugar (as prevalent in wine conditions), the additional A T P can allow for increased growth yields (Garcia et al., 1992; Henick-Kling, 1993; Renault et al., 1988). This theory helps to explain a stimulatory effect observed during the early stages of growth (Kunkee, 1991). In addition to indirect generation of A T P , the M L F may allow O. ceni to take up nutrients by the chemiosmotic mechanism and help to maintain a suitable p H for enzymatic activity and cell growth (Cox and Henick-Kling, 1989). Once all o f the malic acid is decarboxylated, wine is sulphited and further activity from O. ceni is inhibited (as well as activity from other S02-sensitive bacteria and yeast). After the M L F , the risk of further microbial activity in the wine is considerably reduced 10 since the wine is essentially depleted of essential nutrients and fermentable substrates (Davis etal., 1988). 1.3.3 Biochemistry of the malolactic fermentation Decarboxylation of L-malic acid to L-lactic acid is catalyzed by L-mala te :NAD + carboxy lyase without the production of any free intermediates (Caspritz and Radler, 1983; Naouri et al., 1990; Spettoli et al., 1984). This enzyme has been termed the malolactic enzyme and it requires the cofactors M n 2 + and N A D + (Caspritz and Radler, 1983; Naouri et al., 1990; Spettoli et al., 1984). The malolactic enzyme was first purified from Lactobacillusplantarum (Lonvaud, 1975; Schutz and Radler, 1973). Characteristics of the malolactic enzyme from the various malolactic bacteria (£. plantarum, L. casei, L. murinas, L. mesenteroides, and O. ceni) show significant differences in regards to the enzyme affinity constants for L-malic acid, N A D + and M n 2 + , but they are quite similar in molecular weight (ranging form 60-70 kDa) (Battermann and Radler, 1990; Caspritz and Radler, 1983; Lonvaud-Funel and Desaad, 1982; Naouri et al., 1990; Schutz and Radler, 1973; Spettoli et al., 1984; Strasser de Saad et al., 1984). The malolactic enzyme is hypothesized to be active as a homodimer or homotetramer (Battermann and Radler, 1990; Labarre et al., 1996b) and it is inducible by L-malic acid in the presence of fermentable sugars (Nathan, 1961; Renault et al., 1989). 1.3.4 The effect of the malolactic fermentation on wine and the winemaking process The main effect of the M L F on wine is the reduction of the total acidity and the increase in p H as a result of the decarboxylation of L-malate to L-lactate and CO2. The 11 consequences of an elevated p H can leave some wines susceptible to the growth of less fastidious spoilage microorganisms whose growth is normally inhibited by low p H (Davis et al., 1988). Bacteriological stability in low-acid wines can be accomplished by the addition of acidulating agents (tartaric acid, citric acid, lactic acid or D(+)-malic acid) after the M L F (Boulton, 1996; Rankine, 1977). Another disadvantage of p H increase after the M L F is the relative loss of colour (up to 30%) in red wine (Kunkee, 1967). A s wine p H increases, anthocyanin pigments undergo structural changes to uncoloured forms (Jackson, 1994; Lonvaud-Funel, 1999; Ribereau-Gayon et al., 2000b). The most notable sensory difference after a M L F is the disappearance of the taste of malic acid ('tart') and the appearance of lactic acid ('soft'). The next most evident sensory characteristic of a M L F is the production of diacetyl, an aroma compound with a 'buttery' or 'butterscotch' flavour/aroma (Martineau et al., 1995a; Martineau et al., 1995b). Diacetyl is produced from the oxidative decarboxylation of a-acetolactate, an unstable intermediary compound formed during the reductive decarboxylation of pyruvic acid to 2,3-butanediol (Cogan, 1987; Ramos et al., 1995). Pyruvic acid is derived from the metabolism of citrate and sugar. Yeast cells are also able to synthesise diacetyl as an intermediary compound but it is generally further reduced to acetoin and 2,3-butanediol during the alcoholic fermentation (Martineau and Henick-Kling, 1995). O. oeni and other L A B in certain conditions are also capable of further metabolising diacetyl to acetoin and 2,3 butanediol, thereby reducing the buttery sensory characteristics (Ramos et al., 1995). Diacetyl in wine usually tends to be found at a concentration of 5-10 mg/1 (Lonvaud-Funel, 1999). 12 The impact of the M L F on the organoleptic qualities of wine beyond the primary role of deacidification and the production of diacetyl is not completely defined. Numerous studies have concluded that the M L F affects wine aroma, however, few compounds or mechanisms have been implicated in those changes. The M L F has been implicated in an increase of the fruity aroma of wine (probably due to the production of esters by wine L A B ) , a decrease in vegetative/grassy aromas (likely due to the catabolism of acetaldehyde) and it may improve the body, mouthfeel and after taste of wine (potentially through the production of polyols and polysaccharides) (Henick-Kling, 1993; L i u , 2002; L i u and Pilone, 2000). These observations suggest benefits to the winemaking process beyond the deacidification that results from the M L F . However, not all sensory changes attributed to the M L F can be considered advantageous (Davis et al., 1985). A substantial increase in acetic acid (the main component of volatile acidity) can accompany a M L F due to the metabolism of sugars by L A B , especially i f the yeast did not completely ferment the sugars or the L A B started to multiply prior to the completion of the alcoholic fermentation (Wibowo et al., 1985). O. ceni and other heterofermentative L A B {Leuconostoc sp. and certain Lactobacillus sp.) ferment hexoses via the phosphoketolase pathway to lactate, ethanol, CO2 and A T P (Liu, 2002). Acetic acid (and additional A T P ) is generated during hexose fermentation by converting acetyl phosphate to acetate (instead of ethanol) (Liu, 2002; Pilone et al., 1991). The heterofermentative L A B can also use oxygen and pyruvate as electron acceptors which further results in the production of more acetate and A T P (Liu , 2002). Homofermentative species of Lactobacillus and Pediococcus produce D-lactic acid and A T P through the Embden-Meyerhof-Parnas pathway (Liu, 2002; Sponholz, 1993). 13 Pentoses are thought to be metabolised by both heterofermentative and homofermentative wine L A B via the pentose phosphate pathway to produce A T P , lactate and acetate (Sponholz, 1993). Hence, the volatile acidity of wine can increase significantly after a bacterial M L F . Other sensory faults attributed to L A B are acrolein formation, development of a 'mousy' off-flavour and an increase in viscosity known as 'ropiness'. Acrolein is produced by L A B degradation of glycerol and is associated with an unpleasant bitterness (Sponholz, 1993). Mousy off-flavours (2-acetyltetrahydropyridine, 2-ethyltetrahydropyridine and 2-acetyl-l-pyrroline) are also produced by L A B , possibly by the catabolism of glucose and fructose, and the amino acids ornithine and lysine, in the presence o f ethanol (Costello and Henschke, 2002; Lonvaud-Funel, 1999). Ropiness is caused by L A B strains that can synthesize extracellular polysaccharides from residual sugars (Gindreau et al., 2001; Manca de Nadra and Strasser de Saad, 1995). In addition to the production of spoilage chemicals and cosmetic problems, L A B can produce harmful compounds such as biogenic amines and ethyl carbamate. More than twenty amines have been found in wine (Lehtonen, 1996), the most notable being histamine, cadaverine, phenylethylamine, putrescine and tyramine (Lonvaud-Funel, 2001; Zee et al., 1983). Biogenic amines are produced from their respective precursor amino acids by specific amino acid decarboxylases. The characterisation of a histidine decarboxylase gene from an O. ceni strain (Coton et al., 1998a; Coton et al., 1998b) and the tyrosine decarboxylase operon from a Lactobacillus brevis strain (Lucas et al., 2003; Moreno-Arribas and Lonvaud-Funel, 1999; Moreno-Arribas and Lonvaud-Funel, 2001) have been completed. The presence of biogenic amines in wine can be of great concern 14 for consumers since these molecules have been shown to produce undesirable physiological effects in susceptible individuals. For example, histamine is known to cause headaches and other allergenic symptoms such as hypotension, edema, palpitations, flushing, diarrhea, and vomiting (Santos, 1996; Soufleros et al., 1998; Wantle et al., 1994). Tyramine and phenyl ethyl amine have been associated with migraines and hypertension (Soufleros et al., 1998). It is important to note that alcohol, acetaldehyde, antidepressant drugs and other biogenic amines such as cadaverine and putrescine can potentiate the toxic effect of histamine, tyramine and phenyl ethyl amine (Straub et al., 1995; ten Brink et al., 1990). Given that wine consumers w i l l be exposed to toxic biogenic amines through a fermented alcoholic beverage that likely contains acetaldehyde, suggests that the negative effects of the biogenic amines wi l l in all cases be enhanced. Biogenic amines are also linked to carcinogenesis. Nitrosable secondary amines (dimethylamine, piperidine, pyrrolidine, spermidine, spermine) detected in wine can react with nitrous acid and its salts to form carcinogenic nitrosoamines (Santos, 1996; Shalaby, 1996). This may represent an additional risk for consumers. Arginine metabolism by certain strains of O. ceni and other L A B leads to the formation o f ethyl carbamate (urethane) precursors (Liu and Pilone, 1998). Ethyl carbamate is a known animal carcinogen and potential human carcinogen (Ough, 1976; Ough, 1993). Ethyl carbamate precursors in wine are urea (produced by yeast), and citrulline and carbamyl phosphate (produced by L A B ) . Ethyl carbamate is formed through the chemical reaction of these carbamylic compounds and ethanol. The metabolism of arginine in L A B involves three enzymes: arginine deiminase, ornithine transcarbamylase and carbamate kinase (Liu et al., 1995b). The genes involved in the 15 arginine deiminase pathway have been characterised at the molecular level in O. ceni (Tonon et al., 2001) and L. hilgardii (Arena et al., 2002). The legal limit for ethyl carbamate in table wine in Canada is 30 ug / L (Battaglia et al., 1990; Conacher et al., 1987), and in the U S A there is a voluntary limit of 15 ug/L (Canas et al., 1994; L i u and Pilone, 1998). However, there are preliminary data to suggest that wines purchased in Canada may commonly exceed the legal limit for ethyl carbamate in this country (van Vuuren, personal communication, 2006). Given its potential role as a carcinogen in humans, ethyl carbamate in wine could represent an additional health risk for habitual wine consumers. A l l L A B , to various degrees, are inhibited by ethanol, low p H , SO2, low temperature, fatty acids produced by yeasts, decreased nutrient content, competitive interactions with yeast and other L A B , and bacteriophage infections (see reviews by Alexandre et al., 2004; Davis et al., 1985; Henick-Kling, 1993; van Vuuren and Dicks, 1993; Wibowo et al., 1985). Thus, malolactic bacteria often grow poorly and unpredictably in wine, especially Chardonnay wines, and thereby complicate the management of the winemaking process. The M L F can also occur in bottled wines, resulting in off-flavours and trapped carbon dioxide. Even under favourable conditions (optimal p H , no ethanol) the specific growth rate o f O. ceni is low, 0.01 - 0.04 h"1 on glucose and 0.06 - 0.10 h~' on glucose and fructose (Maicas et al., 1999; Salou et al., 1994; Zhang and Lovitt, 2005). To overcome some of the difficulties associated with a M L F , selected starter cultures have been developed and commercialized (Maicas et al., 1999; Nielsen et al., 1996; Rodriguez et al., 1990). In some cases the use o f starter cultures w i l l reduce the time required to complete the M L F compared to a spontaneous 16 M L F . However, wine is still prone to oxidation and possible microbial spoilage because the malolactic starter culture is usually inoculated only after the conclusion of the alcoholic fermentation in order to avoid an increase in volatile acidity due to sugar metabolism by O. ceni. Despite the use of malolactic starter cultures, wineries still experience many problems in ensuring an efficient M L F in high-acid white wines. Indeed, a 'stuck' or 'sluggish' M L F that may take weeks or months to complete is detrimental to wine quality because the addition o f protective concentrations of SO2 are delayed and wine is exposed to the negative effects of oxidation and the possibility of microbial spoilage. To avoid the negative aspects of the bacterial M L F , winemakers can use blending, carbonate additions, precipitation of acids, dilution and carbonic maceration to reduce the acidity of wine or grape must. Although these methods can reduce the acidity o f wine, they are laborious and often result in poor quality wine. Alternative technologies include the use of Schizosaccharomycespombe (Gallander, 1977; Silva et al., 2003), high density cell suspensions of yeasts (Gao and Fleet, 1995), and immobilization o f O. ceni, Lactobacillus sp., or the malolactic enzyme on a variety of matrices (see reviews by Kourkoutas et al., 2004; Maicas, 2001; Zhang and Lovitt, 2006). Unfortunately these methods often result in wine of inferior quality and they are not applicable to production of quality wine on a commercial scale. 1.4 Genetic engineering of Saccharomyces cerevisiae to perform the M L F Wine microbiologists have been studying the problems associated with a bacterial M L F for many years; however, it was not until the advent of genetic engineering that a 17 possible solution to the M L F dilemma became available. Research concerning the construction of a malolactic wine yeast strain by cloning the malolactic gene of L A B and expressing it in S. cerevisiae has been progressing for over two decades. In 1984, a D N A fragment containing the malolactic enzyme from Lactobacillus delbruekii was first cloned into Escherichia coli (Williams et al., 1984). However, only a very low level of expression of the malolactic enzyme was obtained from the isolated clones (Williams et al., 1984). A D N A fragment containing the malolactic enzyme from O. ceni was cloned into E. coli as well , but due to instability problems with the cloned D N A , no further research was conducted (Lautensach and Subden, 1984). The malolactic gene (mleS) from Lactococcus lactis was also cloned and characterised (Ansanay et al., 1993; Denayrolles et al., 1994). The open reading frame (ORF) of the mleS is 1620 nucleotides, encoding a putative protein of 540 amino acids (59 kDa) (Ansanay et al., 1993; Denayrolles et al., 1994). A n alignment of the deduced protein sequence of the mleS with malic enzymes from different sources revealed highly conserved regions described as NAD-bind ing domains, a malate binding site, and other regions of unknown function (Denayrolles et al., 1994; Lonvaud-Funel, 1995). Expression of the mleS gene in E. coli and S. cerevisiae resulted in a weak M L F (Anasanay et al., 1993; Denayrolles et al., 1995). The first gene isolated and sequenced from the M L F system of L. lactis was the regulatory mleR gene (Renault et al., 1989). The product of this gene serves as a positive activator of the malolactic gene of L. lactis in the presence o f L-malic acid. The structural gene for the malolactic enzyme of O. ceni (mleA) was eventually sequenced from a 3.4 kb fragment that also contained a gene for a malate carrier protein (mleP) (Labarre et al., 1996b). The mleA gene encodes a protein with a theoretical mass 18 of 59.1 k D a and the mleA amino acid sequence has a 66% homology to the mleS amino acid sequence (Labarre et al., 1996b). The heterologous expression of the mleA in E. coli and S. cerevisiae resulted in low M L F activity (Labarre et al., 1996b). In addition to the mleA and mleP genes, a third O R F transcribed in the opposite direction was found upstream of the apparent M L F operon and encoded for a protein belonging to the LysR-type regulatory protein family (Labarre et al., 1996a). This protein seems to be similar to the activator protein mleR found in L. lactis (Labarre et al., 1996a). Commercial wine yeast strains of S. cerevisiae can metabolize malate to a very limited extent (10-20%) (Kuczynski and Radler, 1982). The basis of the inefficient malate degradation by S. cerevisiae is the lack of an active malate transporter (Grobler et al., 1995; Volschenk et al., 1997b) and the low substrate affinity of its NAD-dependent malic enzyme ( K m = 50 m M ) (Fuck et al., 1973) which is also subject to catabolite repression (Redzepovic et al., 2003). The malic enzyme of S. cerevisiae has been isolated and requires M n 2 + and N A D + or N A D P + as cofactors (Fuck et al., 1973; Osothsilp, 1987). Due to the absence of an active malate transport system in S. cerevisiae, previous attempts to construct recombinant yeast strains capable of M L F did not succeed. While S. cerevisiae cannot degrade malate efficiently, other K (-) yeasts like Schizosaccharomyces pombe can (Baranowski and Radler, 1984; Kuczynski and Radler, 1982; Rodriquez and Thornton, 1989). Yeast can be categorised into K(-) or K(+) groups depending on their ability to use L-malic acid and other T C A cycle intermediates as sole carbon or energy sources (Barnett and Kornberg, 1960; Barnett et al., 1990; Rodriguez and Thornton, 1990; Volschenk et al., 2003). The K(- ) group can use the T C A cycle intermediates only in the presence of an assimilable carbon source 19 such as glucose; whereas the K(+) group can utilise T C A cycle intermediates directly as sole carbon or energy sources (Barnett and Kornberg, 1960; Volschenk et al., 2003). The efficient S. pombe malate metabolism depends on three enzymes: a malate permease, a malic enzyme, and a mitochondrial dehydrogenase enzyme (Osothsilp and Subden, 1986a). Malate transport in S. pombe is constitutive, active, and not subject to glucose repression (Grobler et al., 1995; Rodriguez and Thornton, 1990; Sousa et al., 1992; Sousa et al., 1995). The malate permease is also a general dicarboxylic acid transporter, responsible for the transport of L-malic acid, succinate and malonic acid (Grobler et al., 1995), and it functions as a proton dicarboxylate symporter (Sousa et al., 1992). The optimum p H of the transport of malate in S. pombe is 3.5 (Osothsilp and Subden, 1986b). Under fermentative conditions the cytosolic malic enzyme is solely involved in the degradation of L-malic acid to ethanol and CO2, commonly known as the malo-ethanolic fermentation pathway (Magyar and Panyik, 1989; Mayer and Temperli, 1963; Taillandier and Strehaiano, 1991; Taillandier et al., 1988; Volschenk et al., 2003). Under aerobic conditions both the malic enzyme and malate dehydrogenase metabolise L-malic acid. However, the malate dehydrogenase only degrades approximately 10% of the malate in aerobic conditions (Osothsilp and Subden, 1986a; Subden et al., 1998). The creation of S. pombe mutants unable to transport malate and the subsequent cloning of a S. pombe Hindlll D N A fragment capable of complementing this mutation (Osothsilp and Subden, 1986a; Subden et al., 1998) was the first step that lead to the cloning of the malate transport gene. Grobler et al., (1995) used this Hindlll D N A fragment to subclone, sequence and characterise the malate transport (mael) gene. The structural gene o f the malate permease encodes an O R F of 1314 bp that translates into a 20 protein of 438 amino acids with a theoretical weight of approximately 49 k D a (Grobler et al., 1995). The mael gene is located on chromosome 1, is constitutively transcribed and is not subject to catabolite repression (Grobler et al., 1995). Analysis of the hydrophobic and hydrophilic composition of the amino acid sequence of the malate permease revealed typical motifs similar to other membrane transport-proteins. The maelp contains a hydrophillic amino- and carboxy-terminal as well as ten putative membrane-spanning domains separated by hydrophilic linkers. Several conserved elements were identified in the maelp, such as a leucine zipper motif, P E S T region and several N-l inked glycosylation and protein kinase C phosphorylation sites (Grobler et al., 1995). Volschenk et al. (1997), functionally co-expressed the S. pombe mael gene and the L. lactis mleS gene under the regulation o f the S. cerevisiae 3-phosphoglycerate kinase (PGK1) promoter and terminator sequences on multicopy plasmids in a laboratory strain of S. cerevisiae. The recombinant laboratory strain was able to efficiently decarboxylate 4.5 g/L L-malate to L-lactate and carbon dioxide in 4 days (Volschenk et al., 1997a; Volschenk et al., 1997b). 1.5 Genetically engineered Saccharomyces cerevisiae strains in the wine industry The possible targets for the genetic improvement of wine yeast strains are presented in Table 1. Several strategies and methods can be used to obtain these desired properties in wine yeast. The classic methods include selection o f clonal variants, mutagenesis and selection, and hybridisation via mating, rare-mating, cytoduction or spheroplast fusion (Bisson, 2004; Pretorius, 2000; Pretorius and Bauer, 2002). The recent application of recombinant D N A technologies have enabled a far more specific 21 and rational approach to improvements of industrial yeast strains. The capability to transform yeast based on chemical, electrical or biolistic methods, the development of a variety of vectors, and the publication of the yeast genome has led to major advances in the construction of wine yeast strains (Akada, 2002; Bisson, 2004; Pretorius, 2000; Pretorius and Bauer, 2002). Table 1. Oenological targets for the genetic improvement of S. cerevisiae wine strains (adapted from Bisson, 2004; Giudici et al., 2005; Pretorius, 2000; Pretorius and Hoj , 2005) Target Fermentation performance • Improve stress tolerance • Improve fermentation rate • Improve substrate utilisation • Improve nitrogen assimilation • Improve competitiveness • Increase range of growth temperatures • Reduce foam formation Improve sensory attributes • Biological adjustment of acidity • Increase glycerol production • Increase desirable esters • Liberate grape terpenoids • Optimised fusel oi l production • Increase autolysis flavour production Reduce off-character production • Decrease sulphur volatiles • Decrease acetate, volatile acidity • Decrease aldehydes • Decrease higher alcohols Reduce off-character production • Decrease phenolic derivatives Improve wine processing • Optimise flocculation and sedimentation • Improve protein and polysaccharide clarification • Tannin reduction 22 Target Improve biological control of wine spoilage microorganisms • Increase production/tolerance to SO2 • Express antimicrobial peptides or enzymes Improve wholesomeness • Decrease ethyl carbamate • Decrease bioamine formation • Increase production of resveratrol • Increase vitamin production • Increase pesticide and metal ion scavenging • Decrease ethanol concentration The current state of progress of genetically engineered wine yeast is illustrated in Table 2. Although significant advances have been accomplished in the last two decades only one genetically modified ( G M ) wine yeast is currently commercially available, it is the malolactic yeast reported in this dissertation. Furthermore, the urea-degrading yeast also produced by our group (Coulon et al., 2006) is ready for commercialisation. The majority of genetically engineered yeast strains reported in Table 2 are only at the 'proof of principle' stage and cannot be commercialised until several further requirements are met. 23 Table 2. Genetic engineering of wine yeast and stratagies used in their modification (adapted from Schuller and Casal, 2005) Target Protein Gene(s) Source Promoter/ Marker Plasmid/ Host Reference Terminator Integration Strain* Fermentation Performance -Stress Glycogen synthase GSY2 S. cerevisiae Native URA3 2u Laboratory (Perez-tolerance Torrado et al., 2002) -Killer factor K l killer toxin KIL-K1 S. cerevisiae ADH1/ADH1 LEU2 and CEN-based Laboratory (Boone et al., synthesis (Zymocin) K l plasmid and 1990) immunity integration Sensory Attributes -Acidity Malate permease/ mael/mleA S. pombe/O. ceni PGK1/PGK1 None Integration Industrial (Husnik et al., adjustment Malolactic enzyme 2006) Malate permease/ mael/mleS S. pombe/L. PGK1/PGK1 URA3 2u Laboratory (Volschenk et Malolactic enzyme lactis al., 1997b) Malate permease/Malic mael/mae2 S. pombe PGKl/PGKl SMR1-140 Integration Industrial (Volschenk et enzyme al.,2001) Acetaldehyde ALD6 S. cerevisiae (Deletion) kanMX4 Integration Laboratory (Remize et al., dehydrogenase of marker 2000) Lactate dehydrogenase LDH Lactobacillus ADH1/ADH1 Tn903 2n Industrial (Dequin et al., casei (G418r) 1999) -Increase Glycerol-3-phosphate GPDl S. cerevisiae ADH1/ADH1 Tn5We 2u Laboratory (Michnick et glycerol dehydrogenase and a l , 1997; production industrial Remize et a l , 1999) -Liberate grape Endoglucanase egU Trichoderma ACT CYH2 2u Industrial (Perez-terpenoids longibrachiatum Gonzalez et al., 1993) Arbinofuranosidase abjB Aspergillus niger ACT CYH2 2u Industrial (Sanchez-Torres et al., 1996) Target Protein Gene(s) Source Promoter/ Terminator Marker Plasmid/ Integration Host Strain* Reference Sensory Attributes -Liberate grape Endoxylanase xlnA Aspergillus ACT CYH2 2u Industrial (Ganga et al., terpenoids nidulans 1999) Rhamnosidase rhaA Aspergillus GPD/PGK CYH2 Industrial (Manzanares aculeatus et a l , 2003) -Acclerated A A A ATPase cscl-l S. cerevisiae TDH3 URA3 2u Laboratory (Cebollero et autolysis a l , 2005) -Volatile Phenolic acid pdc Lactobacillus PGKl/PGKl URA3 and 2u and Laboratory (Smit et al., phenol decarboxylase plantarum SMR1-140 integration and 2003) formation of plasmid industrial -Optimise ester Alcohol acetyltransferase ATF1 S. cerevisiae PGKl/PGKl LEU2 and 2u and Laboratory (Lilly et a l , production SMR1-140 integration and 2000) of plasmid industrial Alcohol acetyltransferase ATF2 S. cerevisiae PGKl/PGKl SMR1-140 Integration Industrial (Lilly et al., of plasmid 2006) Isoamyl acetate- IAH1 S. cerevisiae PGKl/PGKl SMR1-140 Integration Industrial (Lilly et a l , hydrolyzing esterase of plasmid 2006) Ethanol hexanoyl EHT1 S. cerevisiae PGKl/PGKl SMR1-140 Integration Industrial (Lilly et a l , transferase of plasmid 2006) -Decrease Sulphite reductase MET 10 S. cerevisiae MET3 LEU2 Laboratory (Sutherland et sulphur al., 2003) volatiles Wine Processing -Improve Endopolygalacturonase PGU1 S. cerevisiae PGKl/PGKl kanMX 2(x Industrial (Vilanova et clarificaiton a l , 2000) Pectate lyase pelA Fusarium solani ACT CYH 2fi Industrial (Gonzalez-Candelas et a l , 1995) Target Protein Gene(s) Source Promoter/ Marker Plasmid/ Host Terminator Integration Strain* Reference Microbial spoilage control -Production of Pediocin antimicrobials Chitinase Leucocin Health Aspects -Ethyl carbamate reduction -Resveratrol production Glucose oxidase Urea amidolyase ? -Glucosidase Resveratrol synthase Coenzyme-A ligase -Reduction of Glycerol-3-phosphate ethanol dehydrogenase Glycerol-3-phosphate dehydrogenase and Acetaldehyde dehydrogenase pedA CTS1-2 IcaB gox bglN 4CL216 Vstl Pediococcus acidilactici S. cerevisiae Leuconostoc carnosum ADH1/ADH1 PGKl/PGKl ADH1/ADH1 Aspergillus niger PGKl/PGKl DUR1,2 S. cerevisiae PGKl/PGKl Candida molischiana Hybrid poplar Vitis vinifera GPD2 S. cerevisiae GPD2/ALD6 S. cerevisiae ADH1 promoter and ALD6 deletion URA3 URA3 URA3 URA3 None ACT/ACT CYH2 ADH2/ADH2 URA3 EN02/ENQ2 LEU2 ADH1 SMRl-140 promoter LEU2 2u 2u 2|i 2u 2u 2jx 2(i Laboratory Laboratory Laboratory Integration Laboratory Integration Industrial Industrial Laboratory Laboratory Industrial Laboratory (Schoeman et a l , 1999) (Carstens et a l , 2003) (Du Toit. and Pretorius, 2000) (Malherbe et al., 2003) (Coulon et al. 2006) (Gonzalez-Candelas et al., 2000) (Becker et al., 2003) (Becker et al., 2003) (de Barros Lopes et al., 2000) (Eglinton et al., 2002) ON Target Protein Gene(s) Source Promoter/ Marker Plasmid/ Host Terminator Integration Strain* Reference Health Aspects -Reduction of ethanol Glycerol-3-phosphate dehydrogenase and Acetaldehyde dehydrogenase Hexose transporter 1 and 7 GPD1/ALD6 S. cerevisiae Hxtl/Hxtl S. cerevisiae ADHl/ADHl and ALD6 deletion kanMX Integration Industrial (Cambon et a l , 2006) HXT7pvom/ URA3 Integration Laboratory (Henricsson et HXTlterm a l , 2005) industrial strains are defined here as wine yeast of the genus Saccharomyces that are used by the wine industry. Host strains that are derived from industrial yeast but have undergone significant alterations (multiple auxotrophies) in order to facilitate genetic modification are defined here as laboratory strains. In general, G M wine yeast for commercial use should not contain any drug resistant markers such as CYH2, Tn5ble, kanMX(Tn903) or SMR1-140, conferring resistance to cycloheximide, phelomycin, geneticin (G418) and sulphometuron methyl ( S M M ) , respectively; nor should they contain vectors or vector sequences with bacterial resistance markers such as the E. coli ampicilin resistance (bid) gene. Genetic modifications relying on plasmids should also be re-constructed with the genes/elements of interest stably integrated into the genome of the host strains. Moreover, all genetic material should be derived from the host species ("self-cloning") or from G R A S (Generally Regarded A s Safe) organisms with a history of safe use in the food industry; the use of D N A sequences from pathogenic species or organisms known to produce allergens should be avoided. One of the main features of recombinant D N A technology versus classical genetics is the possibility to introduce a specific gene(s) into the host without loss of the phenotypic characteristics of the host strain. Apart from the introduced change, a commercially available G M wine yeast strain should produce wine that is not significantly different from the parent strain. Currently this last requirement may be very difficult for polygenic traits that are not well-characterised in wine yeast and only well-characterised traits requiring modifications to one or two genes have been successful (Coulon et al., 2006; Husnik et al., 2006). A G M wine yeast strain for commercial use requires approval from the regulatory authorities of each country where the yeast w i l l be sold. There are two major approaches to regulating G M products in the world. The first is exemplified by the United States and the other is the approach adopted by the European Union. The U S system essentially evaluates the final product and is based on the principle of substantial equivalence. 28 Substantial equivalence means that i f a food from a genetically modified organism ( G M O ) is "as safe as" the corresponding conventional food product then they should both be treated equally. A detailed knowledge of the genotype, phenotype and supporting transcriptomic, proteomic and metabolomic data are necessary for a successful evaluation. The U S system works within existing regulations for the commercialisation of foods and does not require labelling unless an introduced gene encodes for a novel product that has never been part of any other food. Other countries like Canada, Australia and Japan are also based on the U S system but with varying degrees of additional requirements and levels of stringency. The E U system is considerably more complex. The E U approach, in addition to evaluating the final product, also takes into consideration the technique used to obtain the product and it is primarily based on the precautionary principle. The recently amended European Novel Food Regulations define the procedures for authorisation, labelling and traceability. The regulations for authorisation require the applicant to demonstrate that the product is safe to human and animal health, that it does not differ nutritionally from the conventionally produced product, it does not mislead consumers, and it is safe for the environment (Schuller and Casal, 2005). The government o f the country introducing the proposed product must also inform all the commissions of the other members of the E U and the European Food Safety Authority in Brussels, Belgium (Ramon et al., 2005). In the E U , labelling is compulsory for G M food regardless of whether D N A or protein can be detected in the final product, although it is not required for foods with accidental or technologically unavoidable trace amounts of G M O s (<0.9%) (Schuller and Casal, 2005). The E U also requires that G M O s and products derived from G M O s must be traceable 29 through all o f the stages of production and distribution (Schuller and Casal, 2005). However, G M 'processing aids', which are products used in the food and feed production process, do not fall under the strict E U regulations outlined above. G M yeast can be viewed as a processing aid in the E U : i f less than 0.9% of the yeast is present in the final product, no labelling is required. In addition to the demanding technical requirements to construct a suitable G M wine yeast strain and the regulatory hurdles that must be traversed, the largest challenge to a commercial G M wine yeast strain is consumer and winemaker acceptance. Several authors have suggested that the first G M wine products should demonstrate a direct benefit to consumers (Dequin, 2001; Pretorius and Bauer, 2002; Pretorius and Hoj, 2005; Ramon et al., 2005). Ideally, this benefit should be related to health, such as an increased concentration of a nutritional ingredient or the reduction in a naturally present toxic compound (Tables 1 and 2). Although the increase in a health-related compound such as resveratrol or vitamins could be viewed as a direct benefit to the consumer, an alcohol-based beverage may not be the most appropriate delivery method for bringing a health-related ingredient to the public. However, the reduction of a naturally occurring toxic or harmful compound already present in wine, such as biogenic amines and/or ethyl carbamate (Tables 1 and 2) would provide a direct benefit to wine consumers and the removal of such compounds should not affect the organoleptic qualities of the product. This direct benefit, in addition to the fact that the G M wine yeast would be removed by filtration, could increase the consumer acceptance of the first G M wine product. A G M yeast strain that provided a direct benefit to the winemaker, in addition to the consumer, is more likely to be commercially acceptable. In conclusion, a successful G M wine yeast 30 strain should provide a direct benefit to the consumer and to the winemaker without affecting the organoleptic qualities of the product. 1.6 Proposed Research 1.6.1 Significance of research The research reported in this thesis addresses the need for a wine yeast strain capable of completing the M L F . M L O l , the malolactic yeast described herein, is the first genetically engineered wine yeast to be commercialised. The development of a commercially-acceptable malolactic yeast strain is of great benefit to the wine industry. The application of M L O l w i l l provide a direct advantage to the winemaker (by reducing processing time), improve wine sensory qualities (by reducing volatile acidity and preventing deterioration of colour by L A B ) , and possibly a direct health benefit to the wine consumer (by preventing the production of biogenic amines and ethyl carbamate by L A B ) . The preservation of wine immediately after alcoholic fermentation could also improve the organoleptic characteristics of the wine by preventing chemical oxidation and microbial spoilage often associated with a delayed bacterial M L F . 1.6.2 General hypotheses The hypotheses that guided this research were that the construction of a genetically engineered wine yeast strain capable of M L F is possible and that this malolactic yeast could produce high quality wine. 31 1.6.3 Main objectives The main objectives of this study are to: 1. Construct a genetically stable industrial wine yeast strain of S. cerevisiae that w i l l be capable of performing an efficient M L F . 2. Characterise the genome, transcriptome and proteome of the malolactic yeast. 3. Characterise the phenotype of the malolactic yeast and evaluate the wine produced by the genetically engineered wine yeast. 32 2 MATERIALS AND METHODS 2.1 Strains and plasmids employed in the genetic construction and characterisation of MLOl The different strains and plasmids used in this study are listed in Tables 3 and 4, respectively. Table 3. Different strains used in the metabolic engineering and characterisation o f the malolactic yeast M L O l . Strains Description Reference E. coli D H 5 a F end Al glnY 44 thi-l reck.1 relAl gyrA96 deoR nupG F 80d/acZ? M l 5 ? (lacZYA-argF)U\69 hsdR17(rK m K + ) , ?" (Hanahan, 1983) O. ceni Viniflora® Oenos, a freeze dried pure culture of 0. ami. Chr. Hansen Ltd. , Hoersholm, Denmark S. cerevisiae 35977 78-1 A hybrid industrial baker's yeast strain containing an integrated Tn5ble gene cassette at the SUC2 locus. Lesaffre Development, Marcq-en-Barceul, France S. cerevisiae M L O l Industrial "Prise de Mousse" strain S92 containing the malolactic cassette integrated into the URA3 locus. (This study) S. cerevisiae S92 Industrial "Prise de Mousse" strain originally isolated from the Champagne region in France. Bio Springer, Maisons-Alfort, France (a division of Lesaffre International, Marcq-en-Barceul, France) 33 Table 4. Different plasmids used in the metabolic engineering and characterisation o f malolactic clone M L 0 1 . Plasmids Description Reference pJH2 YEp352(? Kpnl) containing the mael and the mleA expression cassettes cloned between URA3 flanking sequences. pJH3 Y C p l a c 3 3 - K a n M X containing the malolactic cassette subcloned from pJH2. pUT332 E. co/z'/yeast episomal shuttle vector containing the Tn5ble dominant marker. YCplac33- E. co/7/yeast centromeric shuttle vector containing K a n M X the K a n M X dominant marker. (Husnik, 2001) (Husnik, 2001) (Gatignol et al., 1987) Lesaffre Development, Marcq-en-Baroeul, France 2.2 Cul ture conditions Escherichia coli D H 5 a was used for plasmid propagation and cultured according to standard methods (Ausubel et al., 1995). A l l S. cerevisiae strains were cultured in Y P D broth (Difco, Becton, Dickinson and Co. , Sparks, U S A ) according to standard methods (Ausubel et al., 1995). Active dry yeast ( A D Y ) were used either directly or re-hydrated at 40 °C in 7% synthetic must for 30 min ( D N A mieroarray and proteomics studies) or re-hydrated at 37 °C in sterile distilled water for 15 min (post-fermentation and effect of residual M L 0 1 concentrations on M L F studies). A D Y was mixed intermittently during re-hydration and prior to inoculation. The medium Y E G (1% yeast extract, 2% dextrose, 1.5% Pastagar B [Bio-Rad, Hercules, U S A ] ) supplemented with 100 ug/mL of phleomycin (Invivogen, San Diego, U S A ) was used for selecting co-transformed yeast colonies and testing for phleomycin sensitivity. Synthetic must was prepared by adding 1.0 m L / L of Tween 80 to synthetic 34 broth (Denayrolles et al., 1995). The p H of the synthetic must was adjusted to 3.5 for all studies except the D N A mieroarray, Real-Time R T - P C R and i T R A Q experiments. The tartaric acid concentration of the synthetic must for the growth kinetics, D N A mieroarray, Real-Time R T - P C R and i T R A Q experiments was adjusted to 4.5 g/L using L-tartaric acid. Pasteurized, commercially available Chardonnay grape must ("Vine Fresh", Wine Ki t z , Vancouver, Canada; 22.65 Br ix , p H 2.9, T A 4.39 g/L, 0.92 g/L malate, Y A N C 348.1 mg/L) was used to study growth kinetics and viability of M L 0 1 post-fermentation. Malate concentration was adjusted to 4.5 g/L and the must was filter-sterilized (0.22 pm). Utilization of malate as a sole carbon source by M L 0 1 was examined in modified Y P D medium containing 10 g/L Yeast Extract (Difco, Becton, Dickinson and Co. , Sparks, U S A ) , 20 g/L Peptone (Difco, Becton, Dickinson and Co., Sparks, U S A ) , 5 g/L Dextrose and 20 g/L L-malic acid. The p H was adjusted to 6.5 with K O H pellets and the medium was filter-sterilized (0.22 pm). 2.3 Genetic construction of malolactic wine yeast 2.3.1 Co-transformation of the malolactic cassette and pUT332 The malolactic cassette was isolated from plasmid pJH2 (Husnik, 2001) by digestion with Srfl and subsequent gel extraction and purification of an 8683 bp fragment containing the mael and mleA expression cassettes flanked by homologous ura3 sequences. S. cerevisiae S92 was co-transformed with the malolactic cassette and pUT332 (Gatignol et al., 1987) combined at a 10:1 (malolactic cassette:pUT332) molar ratio. After electroporation (Ausubel et al., 1995) 1 m L of ice-cold Y P D media was used to recover the yeast. Yeast cells were transferred to a 1.5 m L microfuge tube and gently 35 mixed at 30 °C for 4 h. Aliquots of the yeast suspension were then directly plated onto Y E G plates containing 100 (ig/mL phleomycin and incubated at 30 °C for 3 days. 2.3.2 Screening of transformants for the integrated malolactic cassette Co-transformed yeast colonies and the positive control strain, S92 transformed with pJH13 (YCplac33-KanMX with the malolactic cassette cloned into the Xbal site) (Husnik, 2001), were inoculated into sterile 96 round bottom micro-well plates containing 200 uL of synthetic must, wrapped in parafilm and incubated at 30 °C for 3-5 days. After incubation, 75 uL of the supernatant from each well was removed and placed into a new micro-well plate. A 25 uL volume of the L-lactic acid dehydrogenase/NAD/ phenazine methosulfate/nitro blue tetrazolium reaction mixture (pH 8.3) (Subden et al., 1982) was added to each well containing 75 uL of the test sample and plates were incubated at 37 °C in the dark for 30 min. The reaction mixture causes wells containing L-lactic acid to show a purple/blue colour and yeast cells from the corresponding plate would be recovered by inoculation into 5 m L of Y P D and incubated at 30 °C for 1-2 days. After recovery, malolactic clones were re-inoculated into 5 m L synthetic must and streaked onto Y E G plates and incubated at 30 °C for 3-5 days. The degradation of malate and production of lactate in synthetic must was confirmed by enzymatic analysis (R-Biopharm, Darmstadt, Germany). Confirmation o f integration was performed by P C R on genomic D N A (Promega, Madison, U S A ) from individual colonies from each clone identified in the screening. The P C R reaction was done according to manufacturer's recommendations (Invitrogen, Carlsbad, U S A ) using primers 5 ' - T T G T A A T G T G A C C A A T G A G - 3 ' (inside the cassette, PGK1 promoter) and 5 ' - C T C T T T A T A T T T A C A T G C T A 36 A A A A T G G -3' (outside the cassette, 3'-end URA3 flanking region). The 1095 bp P C R product was visualized by 0.8% agarose gel electrophoresis and ethidium bromide staining (Ausubel et al., 1995). 2.3.3 Loss of plasmid pUT322 Each positively identified clone was sub-cultured daily in non-selective Y P D for one week to ensure loss of the co-transformed plasmid pUT332. Cultures were grown at 30 °C on a shaking platform (180 rpm) and sub-cultured by inoculating yeast from a 24 h culture into 5 m L o f Y P D to a final O D of 0.05. Malolactic clones, S92 (parental strain and negative control), and S. cerevisiae 3597? 78-1 (positive control containing phleomycin resistance gene Tn5ble) were plated onto Y E G medium with and without 100 ug/mL o f phleomycin to demonstrate sensitivity. 2.4 Functionality of malolactic wine yeast 2.4.1 Malate decarboxylation and residual sugar concentrations of wine produced by malolactic clones The parental strain S92 and three individual colonies selected from each sub-cultured malolactic clone were grown in 250 m L Erlenmeyer flasks containing 50 m L of Y P D at 30 °C with agitation (180 rpm) for two days. S92 transformed with p J H l 3 and Y C p l a c 3 3 - K a n M X were cultured in the same way as the parental strain and the malolactic clones except for the addition 200 ug/mL of G418. Yeast cells were harvested by centrifugation for 5 min at 3000 x g and re-suspended in 1 m L of sterile distilled water. A l l yeast cultures were inoculated into 500 m L sterile Erlenmeyer flasks 37 containing 500 m L synthetic must to a final cell density o f 4 x 10 6 cells/mL. Synthetic must inoculated with S92 transformed with Y C p l a c 3 3 - K a n M X or pJH13 also contained 200 ug/mL o f G418. Each fermentation flask was fitted with an autoclaved glass vapour lock. Fermentations were incubated at 20 °C with agitation (65 rpm) for 14 days. Malate was monitored for 14 days and residual sugar on day 14. L - M a l i c acid, D-glucose and D -fructose concentrations were determined by enzymatic analysis (R-Biopharm, Darmstadt, Germany). 2.4.2 Functionality of active dry wine yeast MLOl Malolactic yeast clone 4 ( M L O l ) was produced as active dry yeast at the Lesaffre Development pilot plant facility (Lesaffre Development, Marcq-en-Baroeul, France) and tested for functionality in synthetic must. Active dry yeast strains M L O l and S92 were inoculated at a concentration of 0.2 g/L into 500 m L sterile Erlenmeyer flasks containing 500 m L synthetic must in triplicate. Each fermentation flask was fitted with an autoclaved glass vapour lock. Flasks were incubated at 20 °C with agitation (65 rpm) for 14 days and monitored for malate degradation, lactate production and residual sugar on day 14. L - M a l i c acid, L-lactic acid, D-glucose and D-fructose concentrations were determined by enzymatic analysis (R-Biopharm, Darmstadt, Germany). 2.5 Genetic characterisation of MLOl 2.5.1 Chromosome karyotyping of MLOl and S92 Yeast chromosomes for pulsed field gel electrophoresis were prepared in low melting point agarose plugs (Schwartz and Cantor, 1984). Cells were harvested by 38 centrifugation and washed with 5 m M E D T A , p H 7.5. The cell pellet was re-suspended in 10 uL of 1 M Sorbitol, 0.1 m M E D T A (pH 8), 14 m M p-mercaptoethanol and 10 mg/mL o f Driselase (Sigma-Aldrich, St. Louis, U S A ) . Cel l suspensions were mixed with low melting point agarose (Invitrogen, Carlsbad, U S A ) and placed into molds. Agarose plugs were incubated overnight at 37 °C in 0.5 M E D T A (pH 8), 10 m M Tris (pH 8) and 20 m M D T T . The plugs were subsequently incubated in 0.5 M E D T A (pH 8), 10 m M Tris (pH 8), 1% laurylsarcosine, 1 mg/mL proteinase K (Merck, Whitehouse Station, U S A ) for 6 h at 55 °C. After extensive washing in solutions of 50 m M E D T A (pH 7.5), T E , and I X T B E the plugs were inserted into a 1% agarose-TBE gel. Yeast chromosomes were separated by Gene Navigator P F G E apparatus (Pharmacia Biotech, Uppsala, Sweden) in T B E buffer at 12 °C. Electrophoresis was performed at 165 V for 20 h with 90 s pulses, followed by 10 h with 110 s pulses, followed by 10 h with 125 s pulses and 4 hours with 30 s pulses. The gel was stained with ethidium bromide and photographed. 2.5.2 Southern blot analyses Southern blotting, labelling of probes, hybridization, stringency washes and detection were completed as recommended by the E C L direct nucleic acid labelling and detection system (GE Healthcare, Buckinghamshire, U K ) . Genomic D N A was digested with EcoRV, Ncol, Nsil, or Pvull. After fractionation by electrophoresis in a 1% agarose gel, the D N A was blotted onto positively charged membranes (Roche, Basel, Switzerland) and fixed by heating at 80 °C for 1 h. Probes corresponding to the malolactic cassette were excised from pJH2 (Husnik, 2001) with the following restriction 39 enzymes: XballKpnl to produce the 5' end 938 bp URA3 probe, Sphl/BamHl to produce the 842 bp mael probe, NaeVPmel to produce the 721 bp mleA probe, and ClaVNoil to produce the 674 bp PGK1 promoter probe. Probes corresponding to pUT332 were either excised from the plasmid or P C R amplified (Appendix A ) . The 471 bp probe for the Tn5ble gene was prepared by P C R amplification using primers J20 ( 5 ' - A A T G A C C G A C C A A G C G A C G - 3 ' ) and J21 ( 5 ' - A T C C T G G G T G G T G A G C A G - 3 ' ) and Pwo polymerase (Roche, Basel, Switzerland). A probe for the bla gene and 970 bp o f non-Saccharomyces sequences o f pUT332 was prepared by digesting pUT332 with SspVClal resulting in a 1758 bp fragment (Appendix A ) . 2.5.3 Sequencing of the malolactic cassette integrated into the genome of MLOl Synthesis of primers (Table 5) and sequencing was completed at the Nucleic A c i d Protein Service Unit (University of British Columbia, Vancouver, Canada). Synthesis was performed on a Perkin Elmer - Applied Biosystems synthesizer using solid support phosphoramidite chemistry (Applied Biosystems, Foster City, U S A ) . Sequencing of the malolactic cassette from S. cerevisiae M L O l (both strands) was completed using an A B I P R I S M 377 D N A sequencer and ABI ' s Ampl iTaq FS DyeDeoxy Terminator Cycle Sequencing chemistry (Applied Biosystems, Foster City, U S A ) . A minimum of two sets of 13 unique templates spanning the entire malolactic cassette were obtained by P C R using PfuTurbo (Stratagene, L a Jolla, U S A ) and primers listed in Table 5. One set of templates was used for the sequencing o f one strand while the other set was used for the sequencing of the reverse strand. Primers used for sequencing are also listed in Table 5. 40 The two complete sequences were aligned and analyzed for differences using Discovery Studio Gene v l . l software (Accelrys, San Diego, U S A ) . If anomalies occurred, another round of P C R and sequencing of the region in doubt was run to determine the correct sequence. Genomic D N A for the P C R reactions was obtained using a Promega, Wizard® Genomic D N A Purification K i t (Promega, Madison, U S A ) . A l l P C R templates used for sequencing were unique to the malolactic cassette. A schematic representation of the sequencing strategy is presented in Appendix B . Table 5. Primers used for sequencing of the malolactic cassette integrated into S. cerevisiae M L 0 1 Primer name Primer sequence For l9g For20g F o r l 5 B For3p For4g For5g For6g For7p For l8g For21g For8g For9p ForlOg F o r l l p Fo r l2g For l3p For l4p F o r l 7 B R e v l 8 g Rev2p Rev3bp R e v l 9 g Rev20g Rev4bg Rev5g 5'' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' 5' A A C T A A T G A G A T G G A A T C G G - 3 ' G A A G G T T A A T G T G G C T G T G G - 3 ' A A G G A A C G T G C T G C T A C T C - 3 ' A G A T C T C A T A T G C A A G A C G C - 3 ' T G G A G G A T G G G C A T C T T C G - 3 ' T G A G A A A G C T G G T G G A C C G - 3 ' T G G C A A G C A T G T C G A T G A A C - 3 ' A G T T C A C C C A T G T C G A A T C G - 3 ' T C T T G A G T T G A A G T C A G G A A T C - 3 ' T G A T G C G T T C A T G C C T G A T C - 3 ' T G G T A C C G C G G C C G C A A G -3 ' A C T T C A A A T C G T C G A C G G C -3 ' • T G G T C T T T C A G T A T A A C C A G -3 ' • C C A G T T C C T T G A A T A T C A T C -3 ' • T G A T A T C G C G G C C A T T A G C -3 ' A C T T A C T G G A T C T G T C A T G -3 ' G C T T G C G G C C G C A C A A A G -3 ' A G G T A G A G G G T G A A C G T T A C -3 ' • T T G T T C C G T T T G A C T T G T C G C -3 ' • T C G C A A T G T C A A C A G T A C C C -3 ' A C T A T A G T A G A G A T A A C G T C -3 ' A C A A G C A A T C G A A G G T T C T G -3 ' T T G T T G A A C C G C A A G G T G C -3 ' T G C T A A T G G C C G C G A T A T C -3 ' A C C A G A T T A G A G T A C A A A C G C -3 ' 41 Primer name Primer sequence Rev6g Rev7p Rev8gb Rev9p RevlOg R e v l l g R e v l 2 g Rev l3p Rev l5p R e v l 7 B Rev21g 5' - T T G C C G G C G T T C T T G G A G -3 ' 5' - A A G C C T T G A T C G G T A C T G G -3 ' 5' - A C T A A T G A G A T C T C C T C G A G -3 ' 5' - A G C T T G C G G C C G C G G T A C -3 ' 5' - A T C T C T C G A T T C G A C A T G G G -3 ' 5' - T G G A T G G T G T G G G T C A T T C -3 ' 5' - A A C T C A T C C G A G T A T C T T G G -3 ' 5 ' - A C A G G T G T C C T T A A C C C T A C -3 ' 5' - C T T G G T A C C T A C T T C T T C C -3 ' 5' - A T G A G T A G C A G C A C G T T C C -3 ' 5 ' - T C C T T C T G C T C G G A G A T T A C -3 ' 2.5.4 Sequence analysis The assembled sequence of the malolactic cassette from M L 0 1 was aligned to an assembled sequence containing previously published sequences. The published sequences for URA3 and the PGK1 promoter and terminator sequences were obtained from the Saccharomyces Genome Database (SGD) (http://www.yeastgenome.org/), and for mael and mleA from the National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov/). The sequence of the malolactic cassette was also analysed for the presence of ORFs of more than 100 codons in length. A l l analyses were completed using Discovery Studio Gene v l . l (Accelrys, San Diego, U S A ) . 2.5.5 Genetic stability of the malolactic cassette in the genome of ML01 The M L 0 1 strain produced as A D Y (large-scale 'trade' fermentation) (Bio Springer, Maisons-Alfort, France) was inoculated into filter-sterilised Chardonnay grape must from fruit harvested in 2004 (23.75 Brix , p H 3.41, T A 8.78 g/L, malate 5.5 g/L, Y A N C 401.2 mg/L, 25 mg/L total S 0 2 ) by Calona Vineyards, Okanagan Valley, B C . 42 M L O l was inoculated at a concentration of 0.05 g/L into a 500 m L sterile Erlenmeyer flask containing 500 m L grape must. The fermentation flask was fitted with a disinfected vapour lock and was incubated at 20 °C for 6 days. The flask was stirred once daily and samples for monitoring O D were obtained by puncturing the #6.5 rubber stopper (Fermenthaus, Victoria, Canada) with a sterile syringe and 18G x 3" needle (Air-Tite, Virginia Beach, U S A ) . After absorbance readings, samples were centrifuged and the supernatant stored at -30 °C until chemical analyses were completed. L -Ma l i c acid, and L-lactic acid concentrations were determined by enzymatic analysis (Megazyme, Wicklow, Ireland). On day 6 yeast cells were plated onto Y P D medium and incubated at 30 °C for 4 days. Four hundred and four randomly selected colonies were inoculated into 200 u L of synthetic juice and incubated at 20 °C for 6 days. The presence/absence of L -lactic acid in each micro-vinification was determined by enzymatic analysis (Megazyme, Wicklow, Ireland). 2.5.6 G l o b a l gene expression analyses One m L of re-hydrated M L O l (25 mg/mL) and one m L o f S92 (25 mg/mL) were each inoculated in triplicate into 500 m L sterile flasks (with an additional 130 m L head space), containing 500 m L of synthetic must and a magnetic stir bar. Each fermentation flask was fitted with a disinfected vapour lock. Additional flasks were similarly prepared in order to monitor weight loss, measure optical density and take samples for chemical analyses. A l l flasks were stirred once daily and incubated at 20 °C. After stirring, samples for monitoring O D were obtained by puncturing the #6.5 rubber stopper (Fermenthaus, Victoria, Canada) with a sterile syringe and 18G x 3" needle (Air-Tite, 43 Virginia Beach, U S A ) . After absorbance readings, samples were centrifuged and the supernatant stored at -30 °C until chemical analyses were completed. A t 48 and 144 hours, five 30 m L volumes from each culture were centrifuged in 40 m L tubes for 3 min at 3500 x g. The supernatant was decanted and the pellet re-suspended (briefly vortexed) in 10 m L o f d H 2 0 . A second centrifugation was completed for 3 min at 3500 x g. The supernatant was decanted and the cell pellet was placed in liquid nitrogen for 30 s and then stored at -80 °C. Optical density measurements and chemical analyses were conducted on the remaining fermentation broth and the initially decanted supernatant. Total R N A , poly ( A ) + R N A purification, c D N A synthesis and biotin-labelled c R N A synthesis and fragmentation procedures have been previously described (Erasmus et al., 2003). The only modification was that the isolated total R N A was also passed through an RNeasy M i d i kit (Qiagen, Hilden, Germany). Twelve Y G S 9 8 oligonucleotide arrays (Affymetrix, Santa Clara, U S A ) were used as targets for hybridization. Hybridization, fluidics and scanning procedures have been described previously (Erasmus et al., 2003). Data were analyzed using Affymetrix Microarray Suite v 5.0. A l l detection and comparison tunable parameters were set to default values (Affymetrix, Santa Clara, U S A ) . Absolute analysis was completed on triplicate data for M L O l and S92 at 48 hrs and 144 hrs. Nine comparative analyses were generated for each time point. Only probe sets that had a change call o f T or ' M F and change p value of < 0.003 across all nine comparison were called 'increasers', probe sets with ' D ' or ' M D ' and a p value > 0.997, 'decreasers'. Average Signal Log (base 2) Ratio values were used to calculate the fold change. Probe sets were linked to their target descriptions and to their gene ontology (GO) annotations using the NetAffx analysis centre (Affymetrix, 44 Santa Clara, U S A ) (http://www.affymetrix.corn/analysis/index.affx) and S G D (Saccharomyces Genome Database; http://www.yeastgenome.org/). 2.5.7 Confirmat ion of D N A microarray results by Real-Time P C R Confirmation of the D N A microarray data was done by semi-quantitative reverse transcriptase Real-Time P C R . Total R N A isolated for D N A microarray assays was treated with DNase I as per manufacturer's instructions (RNeasy M i n i kit, Qiagen, Hilden, Germany). After three applications o f DNase I, 100 ng of total R N A was used as template in a P C R amplification (Fermentas, Burlington, Canada) to show the absence of genomic D N A using primers specific to ACT1 (Table 6). Control genomic D N A was isolated (Ausubel et al., 1995) and 50 ng was used for the P C R reaction (Fermentas, Burlington, Canada). The 100 bp P C R product for ACT7 was visualized by 4% NuSieve G T G Agarose (Cambrex Corp., East Rutherford, U S A ) gel electrophoresis and ethidium bromide staining (Ausubel et al., 1995). One microgram of clean total R N A was used to synthesize c D N A using Superscript and a random hexamer primer mix as per manufacturer's instructions (Invitrogen, Carlsbad, U S A ) . Semi-quantitative real-time P C R was conducted using S Y B R Green P C R Master mix and an A B I P R I S M 7500 instrument (Applied Biosystems, Foster City, U S A ) . P C R reactions were done in triplicate using 100 ng of c D N A as template. ACT1 was used as the control gene. The genes and primer sets used are shown in Table 6. 45 Table 6. Primers used in semi-quantitative reverse transcriptase real-time P C R to confirm D N A microarray data. Gene Primer Primer Sequence Symbol Direction ACT! Forward 5' - G T T T C C A T C C A A G C C G T T T T G - 3 ' ACT1 Reverse 5' - G C G T A A A T T G G A A C G A C G T G A G - 3 ' AQR1 Forward 5' - T C G A G C A A G A C A A A G C T A A C G G - 3 ' AQR1 Reverse 5' - G C T A C G A C G G C C A A G A A A T T T T - 3 ' CTT1 Forward 5' - G A G A A A G A G T T C C G G A G C G T G T - 3 ' CTT1 Reverse 5 - A T T C T G G T A T G G A G C G G C G T A - 3 ' DIP5 Forward 5' - T T T G T G G C T T G G C G T A C A T G - 3 ' DIP5 Reverse 5' - G G T G A T C C A A C T C A A G A T T C C G - 3 ' ENA2 Forward 5' -C A T T C G A C T C A A C T G T G A A G C G - 3 ' ENA2 Reverse 5' - G C A A C A A C T G A T G A T G C T T T C G - 3 ' PH084 Forward 5' - G C C A T T A T T G C A C A A A C C G C - 3 ' PH084 Reverse 5' - C G A A A A T T T C C A T G A C G T G A G G - 3 ' PRR2 Forward 5 - G A C T G C A G A A C A C G C C T A T T C C - 3 ' PRR2 Reverse 5 - T T A T T T A G C G C C T C A C C C G T C - 3 ' PUT4 Forward 5 -C A T C C A C G G C A G A C G T G T T T A - 3 ' PUT4 Reverse 5 - A G C C T T G C G G A A T C T C A G G T A C - 3 ' SUE1 Forward 5 - T T G T T T G G T G A A C G T G G C A C T - 3 ' SUE1 Reverse 5 - C C A C C A A T T G A A T G G C A A C A G - 3 ' YPC1 Forward 5 - A C T G C T T G A A C C A C A C G G A T G - 3 ' YPC1 Reverse 5 - T G A C G T T G A G C G T A A T G A C C C - 3 ' Y M L 0 8 9 C Forward 5 - C A A T G A A A T G C A A G A G C G C A - 3 ' Y M L 0 8 9 C Reverse 5 - G G A A T T G T A A G G C A C A C C G A G T - 3 ' 2.5.8 Transcription of URA3 and transgenes mael and mleA Transcription of URA3 and transgenes mael and mleA was analysed by reverse transcriptase P C R (One-Step R T - P C R , Qiagen, Hilden, Germany). R N A isolation, DNase treatment and clean-up were completed as previously described and 1 jig of clean total R N A was used as a template for each R T - P C R reaction. URA3 primers F o r l 5 B and Rev3bp, mael primers For4g and R e v l l g , and mleA primers For9p and Rev6g are shown in Table 5. Primers ACTFor269 ( 5 ' - A A G A G A G G T A T C T T G A C T T T A C G - 3 ' ) and A C T R e v 1 c (5' - A C A A T A C C A G T A G T T C T A C C G - 3 ' ) were used as internal controls. 46 Control genomic D N A was isolated (Ausubel et al., 1995) and 50 ng was used for each R T - P C R reaction (One-Step R T - P C R , Qiagen, Hilden, Germany). The 269 bp, 783 bp, 807 bp and 499 bp P C R products for ACT1, mael, mleA, and URA3 respectively, were visualized by 1.5% agarose gel electrophoresis and ethidium bromide staining (Ausubel etal., 1995). 2.5.9 Analysis of the proteome of M L 0 1 Cel l pellets obtained for mieroarray analysis at 48 h were also used for protein extraction. Each cell pellet was re-suspended in ice-cold 5 m L lysis buffer (50 m M Tris, pH 8.5, 0.1% SDS) and aliquoted into four 1.5 m L tubes also on ice containing sterile glass beads. Cells were lysed in a cold room using a mini-beadbeater 8 (Glen M i l l s , Clifton, U S A ) with intermittent vortexing. After lysis the samples were centrifuged at 18000 x g for 2 min at 2 °C. The supernatant was transferred to clean tubes and centrifuged at 18000 x g for 10 min at 2 °C. Trichloroacetic acid precipitation of the proteins required adding an equal volume of ice cold 40% trichloroacetic acid and incubation (on ice) for 60 min. Samples were centrifuged at 18000 x g for 30 min at 2 °C. Precipitated proteins were washed twice with 100% acetone (-20 °C) and centrifuged at 2 °C for 30 min and 20 min. Pellets were dried at room temperature until all visible liquid was evaporated and then stored at -80 °C. Proteins were examined by one dimensional S D S - P A G E (Ausubel et al., 1995) and total protein determination was completed using a commercial Bradford assay reagent (BioRad, Hercules, U S A ) . Protein pellets were shipped on dry ice to Genome B C Proteomics Centre for i T R A Q analysis (University of Victoria, Victoria, Canada). Denaturation, blocking of 47 cysteines, digestion with trypsin and labelling with i T R A Q tags were done according to the i T R A Q protocol (Applied Biosystems, Foster City, U S A ) . Protein samples were labelled with the i T R A Q tags as follows: M L 0 1 replicate 1, i T R A Q l 14; S92 replicate 1, i T R A Q l 15; M L 0 1 replicate 2, i T R A Q l 16; and S92 replicate 2, i T R A Q l 17. Strong cation exchange ( S C X ) chromatography, fractionation and L C M S / M S analyses have been previously described (DeSouza et al., 2005). The modifications to the S C X method were that the flow rate was set to 0.5 mL/min and the gradient applied was 0-35% buffer B in 30 min. Fractions were collected every min. Prior to L C M S / M S analysis fractions were brought up to 20 p L with 5% acetonitrile and 3% formic acid and transferred to autosampler vials (Dionex/LC Packings, Sunnyvale, U S A ) . The modifications to the L C M S / M S procedure were that the mobile phase (solvent A ) consisted of water/acetonitrile (98:2 [v/v]) with 0.5% formic acid for sample injection and equilibration on the guard column at a flow rate of 100 uL/min. A linear gradient was created upon switching the trapping column inline by mixing with solvent B that consisted of acetonitrile/water (98:2 [v/v]) with 0.5% formic acid and the flow rate reduced to 200 nL/min. A 10 p L of sample was injected in 95% solvent A and allowed to equilibrate on the trapping column for 10 min. Upon switching inline with the M S , a linear gradient from 95% to 40% solvent A developed for 40 min and in the following 5 min the composition of mobile phase was decreased to 20% A before increasing to 95% A for a 15 min equilibration before the next sample injection. The M S data acquisition method consisted of a 1 s T O F M S survey scan of mass range 400-1200 amu and two, 2.5 s product ion scans of mass range 100-1500 amu. The two most intense peaks over 20 counts, with charge state 2-5 were selected for fragmentation and a 6 amu window was 48 used to prevent the peaks from the same isotopic cluster from being fragmented again. Once an ion was selected for M S / M S fragmentation it was put on an exclude list for 180 s. Curtain gas was set at 23, nitrogen was used as the collision gas and the ionization tip voltage used was 2700 V . Data files were processed using the ProQuant software (v. 1.0) (Applied Biosystems, Foster City, U S A ) in Analyst using the following parameters: The M S and M S / M S tolerances were set to 0.15. A S. cerevisiae subset database of the Celera Discovery Systems database (Ver. 3.0, 01/12/2004) (Celera, Rockville, U S A ) was used for the protein searches. Methyl methanethiosulphonate ( M M T S ) modification of cysteines was used as a fixed modification. A l l results were written to a Microsoft Access database (Microsoft, Redmond, U S A ) . In order to reduce protein redundancy, experimental software, ProGroup viewer (Applied Biosystems, Foster city, U S A ) was used to assemble and report the data. Four comparative analyses were generated using S92 replicate 1 and S92 replicate 2 as denominators. Weighted average ratios were calculated for protein ratios with a p-value < 0.05 across all comparisons. 2.6 Phenotypic characterisation of ML01 2.6.1 Growth kinetics of ML01 A Bioscreen C™ Automated Microbiology Growth Curve Analysis System (Thermo Electron Co. , Waltham, U S A ) was used to examine the growth kinetics of the M L 0 1 or S92 yeasts in Y P D and commercially available Chardonnay must. A single colony from stock culture plates was inoculated into 5 m L of Y P D broth which was incubated at 30 °C until the culture reached stationary phase. One u L of the culture was 49 inoculated into one well o f a microtiter plate (Thermo Electron Co. , Waltham, U S A ) containing 99 u L of Y P D test media or Chardonnay must. A total of nine replicates and one blank control were inoculated per type of test media. The plate was incubated at 30 °C for 64 h with continuous shaking at high intensity. Optical densities for all wells were measured every 10 min using the wide band measurement filter (ODeoonm)- Maximum specific growth rate (p m a x ) was calculated by converting the O D readings to natural log values and the maximum slope during the exponential growth phase was determined for each replicate. Generation time was calculated as described by (Reed and Nagodawithana, 1991). Data was analyzed using Excel 2000 (Microsoft, Redmond, U S A ) . 2.6.2 Utilization of malate as sole carbon source by ML01 and S92 Single colonies from stock culture plates of the parental strain S92 or the M L 0 1 strain were transferred to 5 m L Y P D broth and were incubated at 30 °C until the cultures reached stationary phase. Four 1-L Erlenmeyer flasks were filled with 150 m L of modified Y P D containing 2% L-malic acid and were closed with cotton plugs. Two of the flasks were inoculated with M L 0 1 to achieve an initial O D 6 o o n m o f 0.01 and two additional flasks were similarly inoculated with S92. The flasks were incubated at 30 °C under constant agitation at 180 rpm. Samples were removed at each sampling time and optical densities were measured at abs 600 nm. The samples were then centrifuged in a micro-centrifuge at 18,000 x g for 10 min (Eppendorf, Hamburg, Germany), and supernatants were frozen at -30 °C for later analysis. The OD6oonm readings from the two M L 0 1 flasks and the two S92 flasks were averaged and growth curves were generated 50 using Microsoft Excel 2000 (Microsoft, Redmond, U S A ) . L - M a l i c concentrations were determined in triplicate for each sample at 350 h by enzymatic analysis (Roche, Basel, Switzerland). 2.6.3 Winemak ing wi th M L 0 1 Wines were made from Chardonnay and Cabernet Sauvignon grapes. Chardonnay grape must from fruit harvested in the year 2000 (22.5 Br ix , p H 3.18, T A 13.45 g/L, 9.2 g/L malate, Y A N C 285.7 mg/L, 60 mg/L total S 0 2 ) was obtained from Quails ' Gate Estate Winery, Okanagan Valley, B C , Canada. Two carboys (11.7 L capacity) and two flasks (3 L capacity) of Chardonnay grape must were directly inoculated with 0.2 g/L o f M L 0 1 and four carboys and four flasks were directly inoculated with 0.2 g/L o f S92. After the alcoholic and M L F fermentations were completed by the malolactic yeast M L 0 1 , the wines were racked, topped-up, provided with sulphite (40 mg/L total S 0 2 ) , and aged at 7 °C for nine months. The alcoholic fermentations were considered complete once the wines reached a specific gravity of 0.990 - 0.996. After alcoholic fermentation with the parental strain S92 was completed, two carboys were racked, topped-up, sulphited (40 mg/L total S 0 2 ) , and aged at 7 °C for nine months. Wines in the remaining two carboys fermented by S92 were racked, topped-up, and inoculated with a freeze dried preparation of O. ceni as per manufacturer's recommendations (Vinoflora Oenos, Chr. Hansen, Hoersholm, Denmark) and placed at 20 °C. The two carboys inoculated with O. ceni were re-inoculated one week later with O. ceni (recommended concentrations) and again two weeks later with a double inoculum of O. ceni and placed at 25 °C. Six months after the initial inoculation with O. ceni, the 51 M L F was stuck at 0.25 g malate/L in one carboy and at 2.98 g/L in the second carboy. The wine was racked, sulphited (40 mg/L total SO2) and aged at 7 °C for 3 months. After ageing, all wines were racked and bottled (total SO2 adjusted to 40 mg/L). Sensory analyses were performed after four months and again after four years of bottle ageing at 14 °C. Physical, chemical and sensory analyses were performed on at least three bottle replicates from one carboy. Analyses for wines inoculated with S92 and O. ceni were conducted on the carboy with a residual of 0.25 g malate/L. In order to obtain sufficient biological replicates to analyse the volatile compounds in wine by G C / M S , the fermentations with M L O l , S92 and S92 plus O. ceni were repeated in Chardonnay grape must from fruit harvested in 2004 (23.75 Brix , p H 3.41, T A 8.78 g/L, malate 5.5 g/L, Y A N C 401.2 mg/L, 25 mg/L total S 0 2 ) by Calona Vineyards, Okanagan Valley, B C , Canada. Eight 500-mL flasks, two carboys and two 2-L flasks were directly inoculated with 0.05 g/L of M L O l . Sixteen 500-mL flasks, four carboys and four, 2-L flasks were directly inoculated with 0.05 g/L o f S92. The must was incubated initially at 19 °C for 35 h, then 13 °C for 1 week and finally 19 °C until completion o f the fermentation. After the alcoholic and M L F fermentations were completed by the malolactic yeast M L O l , the wines were racked, topped-up, sulphite levels adjusted to 0.8 mg/L molecular SO2), and kept at 4 °C for 11 months. After alcoholic fermentation with the parental strain S92 was completed, half the wines were racked, topped-up, sulphite levels adjusted to 0.8 mg/L molecular SO2, and aged at 4 °C for 11 months. The remaining S92 produced wine was racked, topped-up, and inoculated with O. ceni as per manufacturer's recommendations (Vinoflora Oenos, Chr. Hansen, Hoersholm, Denmark) and placed at 20 °C. The S92 wine inoculated with O. ceni was re-52 inoculated 2 weeks later with a double inoculum of O. ceni. Nine days after the double inoculum the temperature was increased to 25 °C. One month after the temperature increase, 1.5 L of wine undergoing M L F was mixed with wine not showing active M L F . Two months later a double inoculum of O. ceni and 50 mg/L of Leucofood (Gusmer Enterprises, Mountainside, U S A ) were added. Nine months from the initial inoculation with O. ceni, the bacterial M L F was completed. The wine was racked, sulphite adjusted to 0.8 mg/L molecular SO2 and aged at 4 °C for two months. After ageing, all wines were racked and bottled. Flasks for chemical analyses (500 m L fermentations) were stirred once before daily sampling in order to obtain a homogenous sample. Sampling was done anaerobically as described for the transcriptome/proteome study (Section 2.5.6). Physical, chemical and G C / M S analyses were performed on three biological replicates. Cabernet Sauvignon must (22.9 Brix , p H 3.72, T A 7.41 g/L, malate, 6.2 g/L, Y A N C 325.9 mg/L, 30 mg/L total S 0 2 ) was obtained from Hawthorne Mountain Vineyards, Okanagan, B C in the 2000 vintage. The grapes were crushed and must was vinified without skin contact at the Pacific Agri-Food Research Centre ( P A R C ) using standard winemaking procedures. Fermentations were conducted using the same procedures as described for the Chardonnay 2000 must. The two carboys inoculated with O. ceni (Vinoflora Oenos, Chr. Hansen, Hoersholm, Denmark) after the completion o f the alcoholic fermentation were re-inoculated two weeks later with O. ceni (recommended concentrations). After completion of the M L F the wine was racked, sulphited (80 mg/L total SO2) and stored at 7 °C for eight months. After ageing, all wines were racked and 53 molecular S O 2 adjusted to 0.4 mg/L before bottling. Physical and chemical analyses were performed on at least three bottle replicates. 2.6.4 Analyses of must and wine Microbiological populations in grape musts were determined in samples diluted with 0.1% peptone that were spread onto Modified Rogosa Agar (Pilone and Kunkee, 1976) and Dichloran Rose Bengal Agar (Difco, Becton, Dickinson and Co. , Sparks, U S A ) . Plates were incubated for two and seven days at 25 °C to enumerate yeast and lactic acid bacteria, respectively. Plating was done in duplicate. Soluble solids (Brix) in wines were determined by specific gravity and with a Reichert A B B E Mark II Refractometer (Reichert Analytical Instruments, Depew, U S A ) . Titratable acidity was determined according to A O A C method number 926.12 and by using a Metrohm 686 Titroprocessor (Metrohm, Herisau, Switzerland) (tartaric acid as reference). The p H was determined using a Corning 455 pH/ion analyzer (Corning, Corning, U S A ) and a Metrohm 686 Titroprocessor (Metrohm, Herisau, Switzerland). The viscosity o f the wine was determined with a Brookfield viscometer Model DV- I I (Brookfield Engineering Labs, Stoughton, U S A ) equipped with a L V spindle. The viscometer was set at 60 rpm at 25 °C. Colour according to C I E L A B tristimulus scales was measured in Chardonnay 2000 wines and Cabernet Sauvignon after four months of bottle ageing at 14 °C in the dark. The wines were analyzed with a Beckman DU640B scanning spectrophotometer (Beckman Coulter, Fullerton, C A ) and /4420nm+520nm with an Ultrospec 3000 (GE Healthcare, Chalfont St. Giles, U K ) . A l l colour analyses were completed without dilutions. Malate, lactate, glucose, fructose, acetate and ethanol determinations 54 were done by enzymatic analysis (Megazyme, Wicklow, Ireland). Physical and chemical analyses were performed on at least three biological replicates or at least three bottle replicates. 2.6.5 Sensory analysis Chardonnay wines produced in 2000 were evaluated in duplicate after four years of bottle ageing at 14 °C for colour, aroma, flavour-by-mouth and overall quality, by thirteen trained judges. The methodology of the sensory study was approved by the Clinical Research Ethics Board at the University of British Columbia (Appendix C) . A l l tasters provided written informed consent to participate in the study (Appendix D). Prior to the sensory evaluation, five experienced wine judges bench tested the samples to select attributes that would characterise the wines. The bench testers were also instructed to screen out any defect wines. The attributes selected for aroma were fruity and buttery; the attributes selected for flavour-by-mouth/taste were fruity, sweet, buttery, acidity and body (Table 7). Thirteen judges (six females and seven males) evaluated the wines in duplicate using a completely randomized design. A l l judges were experienced wine tasters. Judges participated in a training session to become familiarized with the tasting protocol and unstructured line scale. The tasting/rinsing protocol required judges to swirl and sniff the glass for the aroma assessment, and sip and swirl the wine in their mouth for the flavour/ quality assessment. Judges were instructed to rinse well with water between assessments. A l l aroma, flavour-by-mouth and quality assessments were conducted in individual tasting booths. A 30 m L wine sample was presented at room temperature in 250 m L 55 I A N O - ISO glasses. Glasses were labelled with a three-digit random number and covered with plastic Petri dishes. The colour assessment was conducted on 20 m L samples in 25 m L plastic Petri dishes. Evaluations were conducted against a white background and were evaluated under natural light. Different random codes were used for the colour, aroma, flavour-by-mouth and quality evaluations to prevent bias among assessments. Evaluations took place on two successive afternoons. Judges scored each attribute on a 10 cm unstructured line scale, anchored at 1 cm and 9 cm with low and high (or light and dark for colour). Data were quantified by measuring the distance of the judge's mark from the origin. Table 7. Definition of sensory attributes for visual, olfactory and gustatory evaluations Attributes Definitions Visual Colour Olfactory Fruity Aroma Buttery Aroma Gustatory Fruity Sweet Buttery Taste Acidi ty Body Overall Quality Relative degree of colour intensity from light yellow to dark yellow Intensity of fruity aromas (generic) from low to high Intensity of diacetyl and/or lactic qualities from low to high Intensity of fruity tastes (generic) from low to high Intensity of sweetness from low to high Intensity of diacetyl and/or lactic qualities from low to high Intensity of sour taste from low to high Tactile sensation (mouth coating) differentiating low-ethanol (thin) from high-ethanol (full-bodied) wines within the context of these wines A composite response o f all sensations (Visual, aroma, flavor and aftertaste) from low to high quality 56 2.6.6 Analysis of volatile compounds in wine by gas chromatography/mass spectrometry Volatile compounds in Chardonnay wines were analyzed by G C / M S headspace analysis as described by Danzer et al., (1999) except that no S P M E was used. A 10 m L sample of wine was placed into a 20 m L headspace vial containing 3 g of N a C l and then positioned into the headspace auto sampler. Sample equilibration was done at 85 °C for 10 min with agitation set on high. 3-octanol (100 u L of 0.565 mg/L) was used as the internal standard. Volatile compounds were analyzed and quantified using an Agilent 6890N G C (Agilent Technologies, Palo Al to , U S A ) interfaced to a 5973N Mass Selective Detector. A 60 m x 0.25 mm ID, 0.25 urn thickness D B W A X fused silica open tubular column ( J & W Scientific, Folstom, U S A ) was used. The carrier gas was ultra high purity helium at a constant flow of 1.3 mL/min . The headspace sample valve and transfer line temperatures were set at 100 °C and 110 °C, respectively. The G C oven temperature was initially set at 40°C for 5 min, then raised to 100 °C at 5 °C/min, then raised to 200 °C at 3 °C/min, held for 1 min, and then raised to 240 °C at 20 °C/min. The injection volume was 1 m L and the injection mode was split with a ratio of 10:1. The M S was operated in scan mode (35-400). The analysis was completed in triplicate, data were analyzed using Enhanced Chemstation software ( M S D Chemstation Build75, Agilent Technologies, Palo Al to , U S A ) , and compounds found were matched with the Wiley275 library (Wiley and Sons, Hoboken, U S A ) . 57 2.6.7 Quantification of ethyl carbamate in wine produced by ML01 Determination of ethyl carbamate concentration in wine was done by solid phase microextraction and G C / M S as described by Coulon et al., (2006). Chardonnay wines were heated at 70 °C for 48 h to maximize ethyl carbamate production. 2.6.8 Effect of residual ML01 populations on M L F in wine fermented primarily with the parental strain S92 The effect of different concentrations of M L 0 1 on the decarboxylation of malic acid to lactic acid was examined in 500 m L fermentation flasks containing 500 m L of synthetic must inoculated with S92 and M L 0 1 . S92 and M L 0 1 were inoculated in the following ratios: 50:0.005, 50:0.05, 50:0.5, 50:5 and 50:50 mg/L. Control fermentations were inoculated to a final concentration of 50 mg/L, 55 mg/L, and 100 mg/L with S92 or M L 0 1 strains alone. The flasks were incubated at 20 °C without shaking and ODgoonm was measured in two samples from each flask at each sampling time. After measurement of O D , individual samples were centrifuged (18000 x g, 10 min) and the supernatants were stored at -30 °C. L - M a l i c and L-lactic acid concentrations were determined by enzymatic analysis as described earlier. The OD 6oonm readings and the L-malic or L-lactic acid concentrations from each replicate were averaged and plots were generated using Microsoft Excel 2000 (Microsoft, Redmond, U S A ) . Fermentations were conducted in duplicate. 58 2.6.9 Post-fermentation viability of M L O l Fermentation flasks (250 m L capacity) containing 200 m L Chardonnay grape must were inoculated with 100 mg/L of re-hydrated M L O l or S92 yeast. Each fermentation flask was fitted with an ethanol-disinfected vapour lock.. The flasks were incubated at 20 °C without agitation for 269 days. Samples were removed from stirred flasks after 0, 6, 20, 50, 81, 115, 170, and 269 days. Samples were vigorously vortexed, serially diluted in 0.1% peptone water (with vortexing between serial dilutions) and cultured on Y P D agar at 30 °C to estimate viable cell populations. A l l fermentations were conducted in duplicate. Results were expressed as mean colony forming units (CFU) /mL. 2.7 Statistical analyses A three-way analysis of variance ( A N O V A ) was used to examine the main effects of judge, wine and replication for each of the sensory attributes. A l l two-way interactions were calculated (judge x wine, judge x replicate, and wine x replicate), in order to track the panel consistency, judge reproducibility and sample-to-sample variation, respectively. Mean scores and least significant differences (Fisher L S D p < 0.05) were calculated. The mean scores with significant differences among wines were plotted using a cobweb diagram (Figure 17). A correlation principal component analysis (PCA) was performed on the mean sensory scores (n=13) from each of the replications (n=2). This allowed the location o f both replicates to be on the diagram (Figure 18). A N O V A s were used to pre-select the most relevant attributes prior to P C A analysis. The term 'overall quality' was 59 not included in the P C A in order to clearly delineate the objective flavour profile analysis (attribute intensities) from the more subjective quality assessment. One-way A N O V A s were used to evaluate the variation in physico-chemical properties and volatile compounds among the wines. Duncan's post-hoc tests were performed to determine which treatment means were statistically different for each o f the measurements (p < 0.05). Differences in generation times between M L 0 1 and S92 as well as differences in sugar concentrations of wines produced by M L 0 1 and S92 were evaluated using a two-tailed independent samples t tests (p < 0.05). A l l statistical calculations were performed using Minitab 14 (Minitab, State College, U S A ) , SPSS v. 11.5 (SPSS, Chicago, U S A ) and Excel 2002 (Microsoft, Redmond, U S A ) . 60 3 R E S U L T S 3.1 Integration of the malolactic cassette into the genome of S92 After screening approximately 2000 yeast transformants for integration of the malolactic cassette (Figure 1), five clones were found that produced lactic acid from malic acid using the screening method developed for the production of L-lactic acid (Figure 2). After recovery of the malolactic clones in Y P D and subsequent inoculation into synthetic must, clone 2 was removed since it no longer continued to produce lactic acid. The PCR with primers specific to PGKlp and downstream of the 3'-end integration site were used to show that the four remaining clones had the malolactic cassette integrated into the URA3 locus (Appendix E). ura3 PG •4— mael PGK1p PGKU ' •*—mleA PGK1p ura3 Figure 1. Schematic representation of the malolactic cassette integrated into the URA3 locus of S. cerevisiae S92. The linear cassette was co-transformed with pUT332 that contains the Tn5ble gene that encodes resistance to phleomycin. 3.2 Functionality of malolactic wine yeast Three individual colonies from each malolactic clone were inoculated into synthetic juice containing 4.5 g/L of malate. Each of the three selected colonies from clones 1, 3 and 4, and two colonies from clone 5 was able to degrade the 4.5 g/L of malate in the synthetic must to = 0.05 g/L by day six (Table 15, Appendix F). A l l three of the selected colonies from clone 4 were also capable of completing the alcoholic fermentation within 14 days (Tables 16 and 17, Appendix F). Clones 1, 3 and 5 were 61 A B Clone 4 , M L 0 1 Figure 2. Test for L-lactic acid production in S92 cells transformed with the malolactic cassette and pUT332. (A) S92 industrial wine yeast transformants growing in a micro-well plate containing 200 pL of synthetic must per well (4 days at 30 °C). (B) Micro-well plate containing 75 pL of the L-LDH/NAD/phenazine methosulfate/nitro blue tetrazolium reaction mixture per well and 25 pL of supernatant from micro-wells in panel A. Wells containing L-lactic acid were a purple/blue colour. The positive controls were S92 yeast transformed with pJH13 (a CEN-based plasmid containing the malolactic cassette and KanMX marker) (Husnik, 2001). either unable to complete the alcoholic fermentation by day 14 or did not ferment as rapidly as clone 4 by day 7 (Appendix F). The residual sugar (glucose and fructose) in the synthetic must after 14 days was = 1.3 g/L for all of the colonies selected from clone 4 (Appendix F). Based on these results, malolactic clone 4 (ML01) was selected for pilot-scale A D Y production. ML01 produced as A D Y was capable of decarboxylating 4.46 ± 0.02 g/L of malate (99.1%) within the first three days and producing equimolar amounts of lactic acid (3.01 ± 0.07 g/L) by day 14. In contrast the parental strain S92 consumed 0.89 ± 0.1 g/L of malate and only 0.02 g/L of lactate was detected in the media after 14 days of fermentation. Residual sugar (glucose and fructose) was 1.71 ± 0.99 g/L and 3.79 ± 0.95 g/L for ML01 and S92, respectively. The final pH of the fermentations was 3.31 ± 0.01 and 3.15 ± 0.01 for ML01 and S92, respectively. 62 3.3 Genetic characteristics of ML01 3.3.1 Conf i rmat ion of the identity of the parental strain Chromosome separation by pulsed field gel electrophoresis (Figure 3) and PCR amplification patterns of genomic D N A regions in between d elements of the Ty l retrotransposon (Appendix G) confirmed that the parent strain of ML01 was S92. ML01 S 9 2 Figure 3. ML01 and S92 chromosomes as separated by pulsed field gel electrophoresis. Chromosomal patterns for ML01 and the parent S92 were identical. In addition, PCR profiles of amplified D N A between Ty l retrotransposon d sequences were also identical for both strains (Appendix G). 3.3.2 Correct integration of the malolactic cassette into the genome of ML01 A Southern blot was performed using a 5,ura3 probe on /Vs/I-digested genomic DNA of ML01 and S92. Two signals were detected for ML01 corresponding to 1.7 kbp and 2.8 kbp D N A fragments, and one signal was detected for S92 corresponding to a 1.7 kbp DNA fragment (Figure 4). The 1.7 kbp fragment matches the expected fragment size for a non-disrupted URA3 locus and the 2.8 kbp fragment is in accordance with the 63 presence of a malolactic cassette integrated into the URA3 locus. To clearly characterize the integration event in the MLOl strain, Southern blot analyses were performed using probes corresponding to the mael and mleA genes (Figures 5 and 6), and the PGK1 promoter (Appendix H). Blots with the mael, mleA and PGK1 promoter probes confirmed the presence of these genes in the MLOl strain and correct integration into only the URA3 locus. S92 ML01 U2URA3 M2URA3 < 2.8 Kbp < 1.7 Kbp 2.8 Kbp URA3, Malolactic cassette 1.7 Kbp Figure 4. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot analysis using a URA3 probe. The schematic representation illustrates the Nsil restriction sites with vertical lines and the URA3 probe (area with hatched boxes in panel on the right). 64 S92 ML01 URA3 No hybridization M2URA3 M2URA3 < 4.9 Kbp Malolactic casi j i 4.9 Kbp Figure 5. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot using a mleA probe. The schematic representation illustrates the Pvull restriction sites with vertical lines and the mleA probe hybridization site is depicted as a hatched box in the panel on the right. S92 ML01 URA3 No hybridization M2URA3 M2URA3 < 2.8 Kbp 2.8 Kbp Malolactic cassette Figure 6. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot using a mael probe. The schematic representation illustrates the Nsil restriction sites with vertical lines and the mael probe hybridization site is depicted as a hatched box in the panel on the right. 65 Both S92 and ML01 form ascospores (Appendix I). Analysis of the ML01 spores revealed 2:2 segregation; cells from two spores were auxotrophic for uracil but positive for M L F while cells from the remaining two ascopores had the opposite phenotype. 3.3.3 Genetic stability of the malolactic cassette in the genome of ML01 After large-scale A D Y production and completion of M L F in Chardonnay grape must, 401/404 randomly chosen ML01 colonies tested positive for the M L F phenotype (99.3 ±1 .0 %, p < 0.05). The integration of the malolactic cassette in the URA3 locus of S92 strain is therefore sufficiently stable for A D Y production and the subsequent M L F during the winemaking process. 3.3.4 ML01 does not contain bla and TnSble antibiotic markers After transformation, ML01 was cultured successively for seven days on a non-selective medium to eliminate pUT332, whose only purpose was to serve in the early steps of screening for transformants. A single colony was plated on non-selective media and on phleomycin containing media (Figure 7). A Southern blot using a probe specific to the Tn5ble genes of pUT332 revealed that this gene was absent in ML01 (Figure 7). Southern blot analyses also verified that ML01 does not contain the bla gene (nor an additional 970 bp of bacterial derived pUT322 sequences) (Figure 8). 66 Figure 7. The phleomycin resistance gene is absent in the genome o f M L 0 1 . (A) Growth of M L 0 1 , S92, S92 transformed with pUT332 and S. cerevisiae 3597 ? 78-1 (integrated 7n5ble gene) yeast strains on Y E G media with or without 100 pg/ml phleomycin. (B) Agarose gel showing fractionated genomic D N A of S92, M L 0 1 and S. cerevisiae 3597 ? 78-1 digested with Ncol restriction enzyme. (C) Southern blot analysis of the genomes of S92, M L 0 1 and S. cerevisiae 3597 ? 78-1 probing for the presence o f the Tn5ble gene. A B C D E F G Figure 8. The ampicillin resistance gene and 970 bp of pUT332 non-Saccharomyces vector are absent in the genome of M L 0 1 . Lane A represents a D N A Ladder, lane B contains 3 pg of EcoRY digested S92 genomic D N A , lanes C and D each contain 3 pg of M L 0 1 genomic D N A , lanes E to G contain increasing concentrations o f EcoRV digested D N A from S92 transformed with pUT332 (0.3 pg, 0.75pg and 1.5 pg). A description of the probe used is given in Appendix A . 3.3.5 Sequence of the malolactic cassette integrated into the genome of ML01 The sequence of the malolactic cassette integrated at the URA3 locus as well as 35 bp upstream and 184 bp downstream of the genomic flanking sequences were verified by sequencing (Figure 9). A detailed description of the sequence is given in Appendix J. 68 Alignment of DNA sequences: ML01 and Native URA3 locus Upper line: ML01, from 1 to 8901 Lower line: Native URA3 locus, from 1 to 2081 1 C A G C A A T T A A T A C T T G A T A A G A A G A G T A T T G A G A A G G G C A A C G G T T C A T C A T C T C A T G G A I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 1 C A G C A A T T A A T A C T T G A T A A G A A G A G T A T T G A G A A G G G C A A C G G T T C A T C A T C T C A T G G A 61 T C T G C A C A T G A A C A A A C A C C A G A G T C A A A C G A C G T T G A A A T T G A G G C T A C T G C G C C A A T T I'l I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 61 T C T G C A C A T G A A C A A A C A C C A G A G T C A A A C G A C G T T G A A A T T G A G G C T A C T G C G C C A A T T continuation of malolactic cassette 8 641 C G C T G C C T T G G G A C A A G G C T T G G G C C G A T A A G G T G T A C T G G C G T A T A T A T A T C T A A T T A T I I I I I I I Ii I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I II I I I I I I I 1821 C G C T G C C T T G G G A C A A G G C T T G G G C C G A T A A G G T G T A C T G G C G T A T A T A T A T C T A A T T A T 8701 G T A T C T C T G G T G T A G C C C A T T T T T A G C A T G T A A A T A T A A A G A G A A A C C A T A T C T A A T C T A I I I II II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 1881 G T A T C T C T G G T G T A G C C C A T T T T T A G C A T G T A A A T A T A A A G A G A A A C C A T A T C T A A T C T A 87 61 A C C A A A T C C A A A C A A A A T T C A A T A G T T A C T A T C G C T T T T T T C T T T C T G T A T C G C A A A T A A I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I II I I I I I I I I I I I I I I I I I I I I I I I I I I 1941 A C C A A A T C C A A A C A A A A T T C A A T A G T T A C T A T C G C T T T T T T C T T T C T G T A T C G C A A A T A A 8821 GTGAAAATTAAAAAAGAAAGATTAAATTGGAAGTTGGATATGGGCTGGAACAGCAGCAGT I II I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 2001 GTGAAAATTAAAAAAGAAAGATTAAATTGGAAGTTGGATATGGGCTGGAACAGCAGCAGT 8881 A A T C G G T A T C G G G T T C G C C A C I I I I I I I I I I I I I I I I I I I I I 2061 A A T C G G T A T C G G G T T C G C C A C Figure 9. The upstream and downstream sequences flanking the malolactic cassette in S. cerevisiae MLOl are 100% identical. The sequence obtained from M L O l is shown in the upper line and the native sequences are shown in the lower line. Only a partial sequence of the malolactic cassette is displayed. The Srfl half sites (integration sites) are indicated in bold and underlined. Comparison to previously published sequences obtained from the Saccharomyces Genome Database for URA3, PGK1 promoter and terminator, and the National Center for Biotechnology Information for mael and mleA showed four nucleotide differences in the URA3 flanking sequences and two nucleotide differences in one of the PGK1 promoter sequences (Appendix K). These anomalies could be due to genetic polymorphisms or 69 amplification errors during the construction of the cassette. Two changes were found in the coding sequence of the mleA gene (Appendix K ) . The first difference resulted in a change from aspartic acid to glutamic acid (aa position 538); both amino acids are acidic and did not affect the functionality o f the malolactic enzyme. The second change (C-T, mleA nt position 996) resulted in a silent mutation (aspartic acid, aa position 332). The mael sequence was identical to the published sequence. In silico analysis of the integrated cassette revealed that four new ORFs were created during construction of the malolactic cassette; these ORFs were primarily composed of Saccharomyces sequences (Figure 10). In addition to the S. cerevisiae sequences, three ORFs contained one or two restriction endonuclease sites that were required for construction of the cassette. Novel O R F 1 (435 bp) contains one Kpnl (6 bp) restriction endonuclease site; novel O R F 2 (663 bp) contains a Kpnl and a Noil (8 bp) site; novel O R F 3 (636 bp) contains a Noil site; novel O R F 4 (360 bp) is entirely composed of S. cerevisiae sequences. 5' URA3 flanking PGK1I mael PGK1p PGKH mleA PGK1p 3' URA3 flanking Novel O R F 1 5' ura3 truncated Novel ORF 2 5' adpl truncated Novel O R F 3 , 5' adpl truncated Novel ORF 4 3' ura3 truncated Figure 10. A schematic representation of new ORFs o f more than 100 codons generated during construction of the malolactic cassette; four new ORFs primarily composed of S. cerevisiae sequences, were created. 70 3.3.6 Effect of the integrated malolactic cassette on the transcriptome of ML01 Global gene expression patterns in M L 0 1 and S92 were investigated using the Affymetrix GeneChip® Yeast Genome S98 Array. The transcriptome o f M L 0 1 and S92 was analyzed at 48 hours and 144 hours, corresponding to the middle of the M L F and completion of the M L F (Figure 11). The 48 h and 144 h time points also corresponded to log phase and stationary phase, respectively (Figure 12). A t 48 h the M L 0 1 strain had consumed 1.73 ± 0.04 g/L of malate and produced 0.92 ± 0.04 g/L of lactate. In contrast, the parental strain S92 had consumed 0.28 ± 0.16 g/L of malate and only 0.05 ± 0.01 g/L of lactate was detected in the media. Equal amounts of glucose/fructose remained; 169.9 ± 4.0 g/L and 166.7 ± 7.3 g/L for M L 0 1 and S92, respectively (difference not significant atp < 0.05). Transcription of 19 genes was affected = two-fold in M L 0 1 after 48 hours; 11 genes were expressed at higher levels and eight genes were expressed at lower levels (Table 8). A t 144 h the transcription of six genes was affected = two-fold; three genes were expressed at higher levels and three genes were expressed at lower levels (Table 9). After 144 hours of fermenting, the M L 0 1 yeast had consumed all of the malate (4.47 ± 0.002 g/L) and produced 3.05 ± 0.14 g/L of lactate, whereas the control strain consumed 0.51 ± 0.05 g/L of malate and negligible amounts of lactate (0.06 ± g/L) was detected in the medium (Figure 11). The fermentation rate (monitored as ethanol production and CO2 loss) was similar to that of the parental yeast S92 (Figure 13) and the sugar concentrations (glucose/fructose) at this time point were 37.09 ± 2.79 g/L and 36.5 ± 1.74 g/L for M L 0 1 and S92, respectively (difference not significant atp < 0.05). 71 I ' i 48 h 144 h 48 h 144 h Figure 11. Malate degradation and lactate production by M L 0 1 and S92 yeast strains in synthetic must (n=3). (A) Malate degradation by yeast strains M L 0 1 and S92; at 144 h M L 0 1 and S92 degraded 4.47 g/L and 0.51 g/L of malate, respectively. (B) Lactate production by M L 0 1 and S92; at 144 h M L 0 1 had produced 3.05 g/L of lactate (negligible amounts of lactate could be detected in S92 fermentations). Yeast were harvested at 48 h (during M L F by ML01) and 144 h (completion of M L F by ML01) for D N A mieroarray analysis (represented schematically as vertical lines). 0 1 2 3 4 5 6 7 8 .9 10 11 12 13 14 15 pj ,i Days 48 h 144h Figure 12. Growth o f M L 0 1 and S92 yeast strains in synthetic must (n=3). Dai ly absorbance readings showed no difference in growth. Yeast were harvested at 48 h (log phase) and 144 h (stationary phase) for D N A mieroarray analysis (represented schematically as vertical lines). The introduction of the malolactic cassette into S. cerevisiae S92 thus had a minimal effect on the transcription of the 5773 ORFs (Saccharomyces Genome Database; June 1, 2005) in the M L 0 1 strain. Moreover, no metabolic pathway was affected by the presence of the malolactic expression cassette integrated into the genome o f M L 0 1 . Only one gene, AQR1, was found to be expressed differently at the two time points; -3.23 and 1.77 fold change at 144 hours and 48 hours, respectively. 73 Table 8. Effect of the integrated the malolactic cassette in the genome of S92 on global gene expression patterns in S. cerevisiae M L O l at 48 hours (> 2 fold change). Genes Expressed at Higher Levels in M L O l Gene Symbol Fold Change Biological Process DIP5 2.81 Amino acid transport YLR073C 2.79 Unknown PCL1 2.69 Cell cycle SUL1 2.47 Sulfate transport OPT2 2.32 Oligopeptide transport RPL7B 2.19 Protein biosynthesis PH084 2.18 Manganese ion transport and phosphate transport RLP24 2.08 Ribosomal large subunit biogenesis YOR315W 2.08 Unknown MAK16 2.02 Ribosomal large subunit biogenesis and host-pathogen interaction HAS1 2.01 rRNA processing Genes Expressed at Lower Levels in M L O l SUE1 -5.13 Protein catabolism PRR2 -3.44 M A P K K K cascade CTT1 -3.29 Response to stress PUT4 -3.13 Neutral amino acid transport YGR243W -2.41 Unknown YR02 -2.31 Unknown JID1 -2.23 Unknown YPC1 -2.18 Ceramide metabolism Table 9. Effect of the integrated malolactic cassette in the genome of S92 on global gene expression patterns in S. cerevisiae M L O l at 144 hours (> 2 fold change). Genes Expressed at Higher Levels in M L O l Gene Symbol Fold Change Biological Process ENA2 5.27 Sodium ion transport RDH54 2.18 Double-strand break repair via break-induced replication, meiotic recombination, and heteroduplex formation YOL048C 2.10 Unknown Genes Expressed at Lower Levels in M L O l AQR1 -3.23 Monocarboxylic acid transport and drug transport YML089C . -3.14 Unknown YIL152W -2.46 Unknown 74 A B Figure 13. Ethanol production and CO2 loss of M L O l and S92 yeast strains in synthetic must (n=3). (A) Ethanol production by M L O l and S92 in synthetic must; no difference was detected. (B) CO2 loss in fermentations by M L O l and S92 in synthetic must; no difference could be detected. Yeast cells were harvested at 48 h and 144 h (represented schematically as vertical lines) for D N A microarray analysis. -a The D N A mieroarray data was verified by semi-quantitative reverse transcriptase P C R of 10 genes shown to have a = 2 fold change in the mieroarray experiments (Table 6); similar levels of expression for nine genes were obtained (one gene, ENA2 was below the threshold for the real-time P C R experiment) (Appendix L) . Transcripts of the mael, mleA and URA3 were also detected by reverse-transcriptase P C R at 48 and 144 hours (Appendix M ) . 3.3.7 Effect of the integrated malolactic cassette on the proteome of ML01 The proteomes of M L 0 1 and S92 were analyzed after 48 hours using i T R A Q (DeSouza et al., 2005) and multidimensional liquid chromatography and tandem mass spectrometry. Proteins were extracted from the same samples used for mieroarray analysis. i T R A Q analysis identified 559 proteins (confidence level >94%) using a subset of the S. cerevisiae Celera Discovery System database (Ver 3.0, 01/12/2004). Only one protein, lanosterol 14-demethylase cytochrome P450 (Ergl lp) was shown to be different at a p-value < 0.05 (default parameters) and across duplicate experiments. Lanosterol 14-demythylase cytochrome P450 had a weighted average ratio of 0.799 (using the S92 data as the denominator); E r g l l p is involved in ergosterol biosynthesis. Furthermore, 199 of the 559 proteins detected are involved in major metabolic pathways including carbohydrate and amino acid metabolism, and pyrimidine, purine, fatty acid, ergosterol and formate biosynthesis. Other than Erg l l p , no difference in protein ratios for M L 0 1 and S92 were detected for any other identified protein. In order to search for unique M L 0 1 proteins, a custom database was constructed which included the sequences for the mleAp, maelp, and the three new ORFs (5' ura3 76 truncated O R F , and the two 5' adpl truncated ORFs). The fourth new O R F could not be tested using this method since it contained only homologous URA3 sequences. The mleAp was present in M L O l (confidence level >99%) but absent in S92. The membrane bound maelp and the three putative proteins that might be encoded by the novel O R F ' s created by cloning were not identified at a confidence level as low as 50%. 3.4 Phenotypic properties of MLOl 3.4.1 Growth kinetics In Y P D , there was a slight difference in |a.m a x between M L O l (0.54 ± 0.01 h"1) and S92 (0.55 ± 0.005 h"1); the corresponding generation times were 1.28 ± 0.02 h and 1.26 ± 0.01 h for M L O l and S92, respectively (p < 0.05, n=9). In Chardonnay must no statistical difference was observed between the pmax for M L O l (0.37 + 0.03 h"1) and the parental S92 (0.37 ± 0.02 h"1); the corresponding generation times were 1.88 + 0.13 h and 1.86 ± 0.08 h for M L O l and S92, respectively (n=9). Growth during the commercial production of M L O l A D Y was not affected by introduction of the malolactic cassette into the URA3 locus of S92 (Didier Colavizza, Lesaffre Development, rue Gabrie Peri 137, F-59700 Marcq-en-Baroeul, France, personal communication). 3.4.2 Utilization of malate as sole carbon source by MLOl and S92 The M L O l and parental strains were unable to consume malate as a sole carbon source. When grown aerobically in modified Y P D media containing only 5 g/L of glucose (to trigger the PGK1 promoter) and 20 g/L of malate, the strains had similar growth kinetics (Figure 14) and no malate was consumed by M L O l or S92. After 350 77 hours, malate concentrations in the media inoculated with M L O l and S92 were 20.4 ± 0.99 g/L and 20.2 ± 1.03 g/L of malate, respectively. 12 n 0 4- > > r 1 1 1 • 1 1 0 50 100 150 200 250 300 350 400 Time (hours) Figure 14. M L O l and S92 cannot consume L-malate as a sole carbon source. M L O l (?) and S92 (? ) strains were inoculated (0.01 ODgoonm final concentration) into modified Y P D medium containing 20 g/L of malate and 5 g/L of glucose and grown aerobically for 350 hours (n=2). Malate analyses after 350 h showed no reduction in malate levels in the medium; 20.4 ± 0.99 g/L and 20.2 ± 1.03 g/L of malate remained in media inoculated by M L O l and S92, respectively. 3.4.3 Malolactic fermentation in Chardonnay and Cabernet Sauvignon musts by MLOl The sulphited and cold stabilized Chardonnay must (from fruit harvested in 2000) contained 105 C F U / m L of yeast and 40 C F U / m L of lactic acid bacteria prior to inoculation with M L O l or S92 A D Y . The M L O l and S92 strain both attained a final specific gravity of 0.996 in the high-acid Chardonnay wine at the end of the alcoholic 78 fermentation. The M L 0 1 strain completed the alcoholic fermentation in 22 days and the S92 strain in 32 days (Figure 15). The M L 0 1 strain consumed 9.04 ± 0.03 g/L of malate (n=2) within the first five days of the alcoholic fermentation (Figure 16a) and produced an approximately equimolar amount of 6.0 ± 0 . 1 g/L of lactate in the must by day seven (Figure 16b). In contrast the S92 strain consumed only 0.93 ± 0.26 g/L of malate (n=4) and no lactate was produced in the Chardonnay wine at the end of the alcoholic fermentation. O. ceni required 171 days after alcoholic fermentation to consume 5.29 g/L and 8.02 g/L of malate (n=l) (Figure 16a) and produce 3.96 g/L and 5.42 g/L of lactate (n=l) (Figure 16b), respectively in wine fermented with S92. N o further decarboxylation o f malic acid was observed (Figure 16b). Analysis of titratable acidity, acetate, p H , viscosity and colour properties of the Chardonnay wines produced by M L 0 1 , S92 and S92 plus O. ceni are shown in Table 10. 79 Figure 15. Ethanol production by M L 0 1 (? ) and S92 (?) in high-acid Chardonnay grape must fermented at 20 °C was positively affected by introduction of the malolactic cassette into a URA3 locus in the industrial wine yeast S92. M L 0 1 completed the alcoholic fermentation to a specific gravity o f 0.996 in 22 days and the parental strain in 32 days. Duplicate biological replications were analysed in triplicate. 80 A 12 -, 0 20 40 60 80 100 120 140 160 180 200 220 Days B 7 i 0 20 40 60 80 100 120 140 160 180 200 220 Days Figure 16. M L F by M L 0 1 is completed in the first five days of the alcoholic, fermentation in high-acid Chardonnay grape must (9.2 g/L). (A) Malate degradation and (B) lactate production by M L 0 1 (? /?) and S92 (? /? ) and S92 + O. ceni (? I\ ). The M L 0 1 strain fully and efficiently degraded 9.08 g/L of malate and produced equimolar amounts of lactate (6.07 g/L of lactate). Wine inoculated with O. ceni required 171 days post-alcoholic fermentation to complete the M L F in carboy replicate 2 (0.25 g/L residual malate). A stuck M L F was observed for S92 + O. ceni in carboy replicate 1; 2.98 g/L of malate remained despite four inoculums of O. ceni. The parental strain consumed approximately 10% of the malate during the alcoholic fermentation. 81 Table 10. Physicochemical and colour measurements of high-acid Chardonnay wines (2000 harvest) produced by M L 0 1 , S92 and S92 plus O. ceni1. ML01 S92 S92 + O. ceni Titratable acidity (g/L) 7.7 a b 10.9 b 7.4 c *** Acetate (g/L) 0.452 a 0.399 b 0.5 c *** p H 3.22 a 3.09 b 3.24 c *** Viscosity (mPa.s) 1.64 a 1.62 ab 1.60 b * Colour measurements L (degree o f lightness) 99.07 a 98.72 b 99.06 a *#* a (greenness) -0.44 a -0.53 b -0.82 c *** b (yellowness) 4.77 a 5.50 b 5.84 c *** ^420nm + ^520nm 0.151 0.158 0.174 ns The mean values for bottle replicates are given for all quantities (n=3) a*, n s ; significant at/? < 0.05, 0.01, 0.001, or not significant bMeans separated atp < 0.05 by Duncan's post-hoc test In Chardonnay grape must from fruit harvested in 2004 the M L 0 1 and S92 strains both attained a final specific gravity of 0.990 at the end of the alcoholic fermentation. This specific gravity corresponded to a residual glucose/fructose concentration of 1.23 ± 0.05 g/L and 0.47 ± 0.02g/L for the M L 0 1 and S92 produced wines, respectively. The M L 0 1 and S92 strains completed the alcoholic fermentation in 21 days (Appendix N) . The M L 0 1 strain consumed 5.43 ± 0.002 g/L of malate (98.7% complete) and produced approximately equimolar amount of 3.69 ± 0.03 g/L o f lactate by the end of the alcoholic fermentation; 97.7% of the malate was consumed within the first nine days predominantly at a temperature of 13 °C (Appendix N) . The S92 strain consumed only 0.50 ± 0 . 1 g/L of malate and no lactate was produced in the Chardonnay wine at the end of the alcoholic fermentation (Appendix N) . O. ceni required 9 months at 20 and 25 °C after the alcoholic fermentation to complete the M L F . Analysis of titratable acidity, 82 acetate, p H and colour properties of the Chardonnay wines produced by M L O l , S92 and S92 plus O. ceni are shown in (Appendix O). The sulphited Cabernet Sauvignon must (without skin contact) contained 7.2 x 10 3 C F U / m L o f yeast and no lactic acid bacteria prior to inoculation with M L O l or S92 A D Y . The M L O l and S92 strains completed the alcoholic fermentation in 16 days. The M L O l and S92 strains both attained a final specific gravity of 0.990 in Cabernet Sauvignon wines at the end of the fermentation. The M L O l strain consumed 6.13 ± 0.02 g/L of malate (n=2) within the first four days of the alcoholic fermentation. In contrast, the S92 strain consumed 1.87 ± 0.07 g/L of malate (n=2) by the end of the alcoholic fermentation. O. ceni required 42 days after alcoholic fermentation to consume 6.13 ± 0.03 g/L of malate (n=2) in wine fermented with S92. Analysis of titratable acidity, acetate, p H , and colour properties of the Cabernet Sauvignon wines produced by M L O l , S92 and S92 plus O. ceni are shown in Table 11. Table 11. Physicochemical and colour measurements of Cabernet Sauvignon wines produced by M L O l , S92 and S92 plus O. ceni1. M L O l S92 S92 + O. ceni Titratable acidity (g/L) 4.39 a b 6.38 b 4.27 a *** Acetate (g/L) 0.324 a 0.237 b 0.355 c *** p H 3.98 a 3.80 b 4.05 c *** Colour measurements L (degree o f lightness) 84.60 a 84.56 a 88.05 b *** a (redness) 12.67 a 14.32 b 8.01 c *** b (yellowness) 24.34 a 22.05 b 24.78 a *** ^420nm +^520nm 1.00 a 1.14b 0.84 c *** The mean values of bottle replicates are given for all quantities (n=3) in the title a relates only to pH a*. **^ n s : significant atp < 0.05, 0.01, 0.001, or not significant bMeans separated atp < 0.05 by Duncan's post-hoc test 83 3.4.4 Sensory profile of Chardonnay wines produced by MLOl The A N O V A results of the sensory attribute ratings for wines produced by M L O l , S92 and S92 with M L F are summarized in Table 12. Significant differences among wines were observed for seven sensory attributes (colour, fruity aroma, fruity taste, sweetness, acidity, body and overall quality) of the nine which were measured. Judge variation was significant for all attributes, except for acidity (Table 12). This was expected due to individual physiological and scoring differences. Panel inconsistencies as indicated by significant judge x wine interactions (Table 12) were present for two o f the seven significantly different attributes observed for wines (fruity aroma and overall quality). Therefore F-values were re-calculated for these two attributes using a more conservative random effects model (MS t i e atment /MSj* w ) (Goniak and Noble, 1987). Whi le the new F-value for fruity aroma (2.29) was not significant, the new F-value for overall quality (7.99) was statistically significant. This indicates that the panel inconsistencies were relatively minor compared to the magnitude of the overall quality effects. Judge x replication and wine x replication interactions are shown in Table 12. The significantly different mean sensory attributes were plotted on a cobweb diagram (Figure 17) and a P C A plot (Figure 18). Wine produced by M L O l was significantly highest in overall quality, body, and perception of sweetness and lowest in acidity when compared to wines produced by S92 with and without a M L F . Wine produced by M L O l was also significantly highest for fruity taste when compared to wines produced by S92 with a bacterial M L F . The main characteristics of wine produced by S92 without a M L F were its darker colour and high acidity. Wines produced by S92 with 84 bacterial M L F were judged to be more acidic, less sweet, have less body, less fruity taste and lower in quality than wines produced with M L 0 1 . Table 12. F-values from analysis of variance of Chardonnay wines for sensory attributes (three wines, 13 judges, two replications) Descriptor Wine Judge Rep Judge X Judge X WineX Wine Rep Rep Yel low colour 16.02*** a 9.81 *** 0.19 1.48 1.10 4.85* Fruity aroma 6.36* 1 ] 7^*** 0.01 2.78* 1.44 1.44 Buttery aroma 1.14 4.73** 0.00 1.74 0.86 1.27 Fruity taste 16.71*** 7.68*** 1.72 1.27 2.01 0.74 Buttery taste 1.55 7 j g*** 1.39 4.50*** 2.57* 0.95 Sweet 49.46*** 8.46*** 0.92 1.40 2.14 0.92 Acidi ty 23.05*** 1.81 4.01 1.24 1.03 1.12 Body 12.01*** 3.37* 0.09 1.45 1.85 0.67 Overall quality 16.51*** 5 72*** 0.29 2.02* 1.16 1.29 a*, **, ***. s i g n i f l c a n t a t p < Q.05, 0.01, 0.001 85 Overall Quality*** + O. oeni Body*' Figure 17. Cobweb diagram showing the significantly different mean sensory attributes of Chardonnay wines produced by M L O l , S92, and S92 plus O. oeni (n=26). ***p< .001. 86 Yellow Color P C 1 (82.4%) Acidity Fruity Taste Sweetness B o d y # ML01 rep 2 • S 9 2 + O. oeni rep 1 o ML01 rep 1 S 9 2 + O. oeni rep 2 Figure 18. A principal component analysis plot showing the significantly different mean sensory attributes for Chardonnay wines produced by M L O l , S92, and S92 plus O. oeni. P C A factors 1 and 2 explain 82.4% and 15.2% o f the variability, respectively (n=13). The first two factors o f the P C A mean sensory scores accounted for 97.7% of the variance among the wines (Figure 18). The attributes fruity taste, body, sweetness and acidity were most heavily loaded on P C I accounting for 82.4% of the variability. Colour was most heavily loaded on PC2 accounting for 15.2% of the variability. The P C A plot showed that body was negatively correlated with acidity, as indicated by the 180 degree angles between the vectors. Body was positively correlated with sweetness and fruity taste, as indicated by the small angles. Overall quality (not shown in the P C A plot of objective aroma/flavour assessments) was also strongly correlated with body (Appendix 87 P). In contrast, yellow colour and fruity taste were uncorrected, as shown by their 90 degree orientation. The wines from M L 0 1 , located on the right, were characterized by their full body, sweet taste and fruity flavour. In contrast, S92 and S92 + O. ceni wines, located to the left, were higher in acidity and lower in body, sweetness and fruitiness. S92 wines located slightly higher in the plot were distinguished by their darker yellow colour. These traits are consistent with characteristics identified from the cobweb diagram (Figure 17). 3.4.5 Volat i le compounds in wine produced by ML01 G C / M S headspace analysis (Table 13) revealed that no additional compounds were detected in wine produced by M L 0 1 when compated to wine produced with the parental strain S92 or wine produced with S92 and malolactic bacteria. One compound, ethyl 2-methylbutanoate, was detected in wine produced with S92 plus O. ceni that was not present in wines produced with M L 0 1 and S92 without a M L F . Wines produced with M L 0 1 and S92 without a M L F contained several compounds, acetal, 2,4,5-trimethyl-l,3-dioxolane, 1,1-diethoxyisopentane, n-hexanal, and benzaldehyde that were not detected in wines produced with S92 plus O. ceni. 88 Table 13. Concentration of volatile compounds in Chardonnay wines produced with M L 0 1 , S92, S92 plus O. ceni. Wines were analyzed by G C / M S headspace assay1. Compounds M L 0 1 S92 S92 + O. ceni Pa (mg/L) (mg/L) (mg/L) acetaldehyde 83.71 78.33 43.52 ns dimethylsulfide 0.04 a b 0.05 a 0.43 b ** ethyl formate 0.2 0.28 0.28 ns methyl acetate 0.32 0.33 0.35 ns ethyl acetate 180.21 a 173.18 a 277.08 b * isobutyl acetate 0.008 0.008 0.007 ns ethyl butanoate 8.16 8.04 12.24 ns propanol 71.93 64.02 91.54 ns ethyl isovalerate 0.003 a 0.005 a 0.014 b * isobutyl alcohol 229.09 a - 262.89 a 393.11 b * isoamyl acetate 1.24 1.24 2.33 ns n-butanol 1.29 0.91 1.40 ns 2-methyl-1 -butanol 8.21 8.79 12.64 ns 3 -methyl-1 -butanol 94.66 102.34 150.7 ns ethyl hexanoate 0.94 0.89 1.63 ns 1-hexyl acetate 0.16 0.16 0.15 ns acetoin 5.7 ab 1.56 a 9.96 b * 3 -methyl-1 -pentanol 0.06 0.05 0.04 ns ethyl lactate 177.9 a 5.63 b 295.88 c *** 1 -hexanol 21.69 20.27 30.92 ns 3 -ethoxy-1 -propanol 20.6 16.27 20.99 ns 3-octanol (IS) 0.21 0.21 0.21 -ethyl octanoate 0.96 a 0.95 a 2.9 b * acetic acid 6.35 a 5.33 a 14.27 b * ethyl decanoate 0.46 a 0.46 a 1.3 b ** diethyl succinate 0.21 a 0.35 a 1.22 b *** phenylethyl acetate 0.33 0.22 0.2 ns hexanoic acid 1.31 1.76 1.06 ns phenylethyl alcohol 0.95 0.98 1.12 ns octanoic acid 2.62 3.13 2.85 ns The mean values for biological replicates are given for all compounds (n=3) *' **, ***, ns: significant atp < 0.05, 0.01, 0.001, or not significant cMeans separated at p < 0.05 by Duncan's post-hoc test N o significant differences were found for acetaldehyde, ethyl formate, methyl acetate, isobutyl acetate, ethyl butanoate, propanol, isoamyl acetate, n-butanol, 2-methyl-1-butanol, 3-methyl-l-butanol, ethyl hexanoate, 1-hexyl acetate, 3-methyl-1-pentanol, 1-89 hexanol, 3-ethoxy-l-propanol, phenylethyl acetate, hexanoic acid, phenylethyl alcohol and octanoic acid concentrations when the three wines were compared. The only significant difference between wines produced with M L O l and S92 without a M L F was that ethyl lactate was detected at a higher concentration in wine fermented with M L O l (177.9 mg/L) than S92 without a M L F (5.63 mg/L). Ethyl lactate concentration in wine produced with S92 plus O. ami was 295.88 mg/L. Wines produced with S92 plus O. ami also contained significantly higher concentrations of dimethyl sulfide, ethyl acetate, ethyl isovalerate, isobutyl alcohol, ethyl octanoate, acetic acid, ethyl decanoate, and diethyl succinate than wine produced with M L O l or S92 without a M L F . Acetoin was also significantly higher in wines produced with S92 plus O. ami than wines produced with S92 without a M L F . Physicochemical characteristics of Chardonnay wine produced for G C / M S analysis is given in Appendix O. 3.4.6 Ethyl carbamate in wine produced by MLOl Ethyl carbamate (EC) produced by the M L O l yeast and the S92 yeast with a bacterial M L F in Chardonnay grape musts is shown in Table 13. The S92 yeast and O. ami bacterium produced Chardonnay wine (2000 harvest) that had a maximum potential E C concentration of 71.32 pg/L and wine produced by M L O l had 50.89 pg/L, a reduction of 28.6%. In the Chardonnay (2004 harvest) wine, S92 and O. oeni produced wine with 99.14 ug/L of E C and the M L O l yeast produced wine with 89.58 pg/L, a reduction of 9.6%. N o significant difference in maximum potential E C concentration was determined between wines produced by M L O l and the parent strain S92 without a bacterial M L F . 90 Table 14. The production of ethyl carbamate in Chardonnay wines produced with M L 0 1 , S92 and S92 with a bacterial M L F . Wines ML01 S92 S92 + O. P* (Hg/L) ceni (Hg/L) Chardonnay (2000 harvest)b 50.89 a c 44.92 a 71.32 b *** Chardonnay (2004 harvest)d 89.58 a 84.24 a 99.14b *** a***, significant atp < 0.001 bThe mean values for bottle replicates (n=3 ) cMeans separated atp < 0.05 by Duncan's post-hoc test dThe mean values of biological replicates (n=3 ) 3.4.7 Effect of increasing populations of ML01 on M L F in wine conducted with the parental yeast To test the effect of different levels of M L 0 1 cell populations on fermentations conducted with S92, fermentations in synthetic must were performed using mixed cultures of M L 0 1 and S92 in different ratios. The M L F did not occur when the M L 0 1 strain was present at < 1 % of the total yeast cell population at the beginning of the alcoholic fermentation (Figures 19 a, b). Fermentations containing a 10% M L 0 1 inoculum resulted in a partial (33.3 %) M L F (Figures 19 c, d). If 50% o f the yeast population comprised M L 0 1 , an almost complete (95.3%) M L F was observed. 91 - ML01 S92 * S92 + 1%MI01 S92 + 0.1%ML01 - S92 + 0 01%ML01 8 10 12 14 Time (days) 22 -ML01 •S92 -S92 + 10% ML01 —i— 12 — i — 14 —i— 16 — i — 20 22 Time (days) B 3H m c •a 3 1 4 i 3H -ML01 •S92 -S92 + 1% ML01 • S92 + 0,1% MLOl -S92 + 0.01%ML01 10 12 14 Time (days) 18 20 22 D MLOl S92 S92 + 10% ML01 8 10 12 14 16 18 20 22 Time (days) Figure 19. M L F was not detected in wines containing an inoculum < 1% of M L O l yeast. (A) Malate degradation and (B) lactate production by M L O l (?) and S92 (?), and S92 + 1% M L O l (?), S92 + 0.1% M L O l (? ) and S92 + 0.01% M L O l (?) co-cultures in synthetic must containing 4.5 g/L of malate. (C) Malate degradation and (D) lactate production by a co-culture of S92 + 10% M L O l (] ); only 33% of the malate was consumed when M L O l was present at 10% of the total inoculum. The control M L O l culture completely decarboxylated malate within 5 days (n=2). 3.4.8 Post-fermentation viability of MLOl The viability of M L O l and S92 yeast cells post-fermentation was determined by plate counts on Y P D for 269 days (Figure 20). The viability of M L O l and S92 cells post-fermentation declined at similar rates for M L O l and S92. Days Figure 20. Post-fermentation viability of M L O l (? ) is similar to that of S92 (? ) in Chardonnay wine. Both strains were inoculated (100 mg/L) in duplicate into filter-sterilized Chardonnay must and viability o f cells was determined by plate counts on Y P D for 269 days (n=2). 93 4 DISCUSSION The objective of this study was to genetically engineer and characterise a commercially acceptable S. cerevisiae wine yeast strain capable of decarboxylating extracellular L-malate (one of the major organic acids found in grape must) to L-lactate. The correct integration of the S. pombe malate transporter (mael) and the O. ceni malolactic enzyme (mleA) under the control of the S. cerevisiae PGK1 promoter and terminator signals into the genome of the industrial wine yeast S92, yielded the malolactic yeast M L O l . The genetically stable M L O l yeast strain does not contain any antibiotic marker sequences and can complete the M L F during the alcoholic fermentation ( in a variety o f musts; M L O l cannot appreciably decarboxylate L-malic acid to L-lactic acid when present at levels below 1% of the total inoculum. Analysis o f the transcriptome and the proteome showed that no metabolic pathway was affected by the introduction of the malolactic cassette. G C / M S analysis of the volatile compounds showed that wine produced by M L O l did not contain any compounds that were not detected in wine produced with the parental strain S92 or with S92 and O. ami. Wine produced by M L O l also revealed improved colour properties and lower volatile acidity than wines produced with a bacterial M L F . Moreover, processing time after alcoholic fermentation is reduced and wine produced by M L O l is judged highest in overall quality by trained tasters when compared to control wines. 94 4.1 Integration of the malolactic cassette into the genome of S92 yielded the functional malolactic yeast M L 0 1 S. cerevisiae cannot effectively metabolise extracellular malate due to its lack of an active malate transporter (Grobler et al., 1995; Volschenk et al., 1997b) and the low substrate affinity of its NAD-dependent malic enzyme (Km = 50 m M ) (Fuck et al., 1973) that is also subject to catabolite repression (Redzepovic et al., 2003). The successful integration of the malolactic cassette (Figure 1) into the URA3 locus of the wine yeast S92 yielded the first malolactic wine yeast M L 0 1 . In addition to the ura3 flanking sequences required for homologous recombination, the malolactic cassette contains the S. pombe malate transporter (mael) gene (Grobler et al., 1995) and the O. ceni malolactic enzyme (mleA) (Husnik, 2001); both genes are under control of the S. cerevisiae PGK1 promoter and terminator sequences. It is important to note that the transgenes, mael and mleA were acquired from wine microorganisms (Barnett et al., 1990; Garvie, 1967; Lodder, 1970). The S. pombe strain used in this study can trace its roots back to a strain isolated from sulphited grape must (Osterwalder, 1924). Moreover, S. pombe has recently been commercialised (Proenol, Industria Biotecnologica Lda., Portugal) as an alternative method for the deacidification of wines (Silva et al., 2003). O. ceni, formerly designated Leuconostoc cenos (Dicks et al., 1995), has only been isolated from wines and related habitats such as wineries and vineyards (Williams et al., 1989). The ura3 flanking sequences and the PGK1 promoter and terminator sequences were acquired from S. cerevisiae strains GC210 (Cunningham and Cooper, 1991) and AB972 (Olson et al., 1986), respectively. The parental strain o f M L 0 1 is S92 and this was confirmed by electrophoretic karyotyping (Figure 3) and P C R amplification using d sequences of the 95 T y l retrotransposon (Appendix G). S. cerevisiae S92 is an isolate from the Champagne region in France and belongs to a family of very close or identical commercial strains designated as "Prise de Mousse" (PDM) strains. P D M strains are some of the most popular commercial strains and are found in every wine region in the world. The S. cerevisiae M L O l strain is the first genetically engineered wine yeast strain to be constructed without the integration of an antibiotic marker or E. coli vector sequences (Schuller and Casal, 2005). Since the malolactic cassette did not have a selectable marker to detect transformants, a phenotypic screen was developed based on the previous work of Subden et al. (1982). This colorimetric method relies on the specific reaction between L-lactate present in the media and L-lactate dehydrogenase. The screening method was specific, economical, undemanding and effective (Figure 2). Individual colonies from each of the identified malolactic clones were tested for functionality in synthetic must. A l l o f the randomly chosen colonies of clone 4 were able to completely decarboxylate malate to = 0.05 g/L within the first five days of the alcoholic fermentation (Appendix F). The M L F is generally considered complete when the malate concentrations are - 0.05 to 0.2 g/L (Avedovech et al., 1992; du Plessis et al., 2002; Henick-Kling and Park, 1994). Individual colonies from clone 4 were also capable o f consuming the glucose and fructose in the synthetic must to less than 2.0 g/L of residual sugar by day 14 of the alcoholic fermentation (Appendix F). The alcoholic fermentation is considered complete when the residual sugar concentration (glucose and fructose) is < 2.0 g/L (Bisson, 1999). 96 Based on these results malolactic clone 4 ( M L O l ) was selected for pilot-scale A D Y production. Originally, wine yeasts were commercialised as compressed cakes and as liquid cultures. In 1963, the first active dry wine yeast were produced and marketed to the wine industry (Reed and Chen, 1978). Since that time wine A D Y have become widely accepted and are the preferred form of commercial yeast at wineries. M L O l was successfully produced as A D Y and after direct inoculation into synthetic must (without rehydration) it was capable of completely decarboxylating the malate in the medium to equimolar amounts of lactic acid. M L O l was also capable of completing the alcoholic fermentation within 14 days (1.71 ± 0.99 g/L residual sugar), whereas the residual sugar for the parental strain under the same conditions was 3.79 ± 0.95 g/L. A s expected, the final p H of the fermented synthetic must was 0.16 p H units higher than the fermentation with S92. The p H of wine that has undergone a M L F is generally 0.1 - 0.2 p H units higher than wine that has not gone through a M L F due to the decarboxylation of malate (Bartowsky, 2005; Beelman and Gallander, 1979; Boulton, 1996). The successful construction and subsequent production of a functional malolactic yeast as A D Y (pilot plant and eventual large scale manufacturing) was a major achievement towards the commercialisation of the first genetically modified wine yeast. 4.2 MLOl completes the M L F during alcoholic fermentation in Chardonnay and Cabernet Sauvignon musts Most red and some white wine styles (such as Chardonnay) are subjected to the M L F . The M L F is especially favoured in cooler wine regions where the grapes at harvest tend to have naturally higher acidity. High-acid Chardonnay must is one of the most 97 difficult musts winemakers are confronted with to produce a well-balanced quality wine. After inoculation of the M L 0 1 strain into a high-acid Chardonnay must ( T A 13.45 g/L, p H 3.18), 98.3% (9.06 g/L) o f the malate was consumed in the first five days of the alcoholic fermentation and an equimolar amount o f lactate (6.0 g/L) was produced by day seven (Figure 16). The parental strain S92 consumed only 10.3% (0.95 g/L) of the malate in the media and no lactate was produced. The two carboys inoculated with O. ceni after the alcoholic fermentation required 171 days, three additional inoculums and an increase in temperature (25°C) to consume 64.0% (5.29 g/L) and 97.0% (8.02 g/L) of malate in S92 produced wine (Figure 16). These vinification trials, conducted under ideal conditions at a research facility, confirm that the bacterial M L F relies on long and demanding protocols, which are not always in accordance with good winemaking practices. Indeed, wines had to be kept at a relatively high temperature that is conducive to the growth of O. ceni for a long time; this prolonged fermentation at a higher temperature could alter wine aromatic volatile compounds and increase chances of spoilage by unwanted microorganisms and lead to oxidation of wines in wineries. In contrast, M L F by M L 0 1 occurred rapidly, which w i l l allow for early stabilization of wine in the cellar. The rapid M L F in high-acid Chardonnay conducted by M L 0 1 was also demonstrated in the Chardonnay must from fruit harvested in 2004 (Titratable acidity, 8.78 g/L, p H 3.41) and the Cabernet Sauvignon must (Titratable acidity, 7.41 g/L, p H 3.72). In Chardonnay must from fruit harvested in 2004, the M L 0 1 strain efficiently decarboxylated 5.5 g/L of malate and produced equimolar amounts of lactate by day nine (Appendix N) . In contrast, O. ceni required nine months after alcoholic fermentation to 98 complete the M L F despite four additional inoculums, an increase in temperature (25°C) and an addition of L A B nutrients (Leucofood). In Cabernet Sauvignon must, the M L 0 1 strain completed the M L F within the first four days of the alcoholic fermentation; O. ceni required 42 days after alcoholic fermentation to complete the M L F . The parental strain S92 required 10 additional days to complete the alcoholic fermentation in the high-acid chardonnay must compared to the M L 0 1 strain (Figure 15). The low p H of Chardonnay must (pH 3.18) may have contributed to the slower and longer fermentation times for S92, whereas the M L 0 1 was able to complete the fermentation more easily due to the slight p H increase resulting from malate degradation over the first five days (Table 10). In higher p H wines, the alcoholic fermentations by the two yeast strains were completed at the same time: 21 days in the Chardonnay from fruit harvested in 2004 (pH 3.41) and 16 days for the Cabernet Sauvignon must (pH 3.72) . These data indicate that the presence of the malolactic cassette in M L 0 1 does not negatively affect ethanol production when compared to the parental strain S92. 4.3 Wines produced by ML01 have improved physicochemical and organoleptic properties The main effect of the M L F is a decrease in titratable acidity due to the decarboxylation of L-malic acid to L-lactic acid. Titratable acidity of Chardonnay and Cabernet Sauvignon wines produced by M L 0 1 , S92 and S92 with a bacterial M L F are shown in Tables 10 and 11 and Appendix O. A s expected, titratable acidity was considerably reduced in wines that had undergone a M L F . This deacidification is crucial for the production of less sour and balanced wines, particularly for higher-acid wines 99 from cool-climate regions. The decrease in titratable acidity after a M L F is also accompanied by an increase in p H which could affect microbiological stability in higher p H wines (Tables 10 and 11 and Appendix O). However, wines produced by the M L 0 1 strain had lower p H than those produced with the parental strain and O. ceni (Tables 10 and 11 and Appendix O). Therefore, winemakers conducting a M L F using L A B or the malolactic yeast in higher p H wines can adjust sulphur dioxide to appropriate concentrations (and acidity, i f desired) to provide microbiological and chemical stability. The slightly lower p H o f wines fermented by M L 0 1 than S92 with a bacterial M L F (Tables 10 and 11, Appendix O) are probably due to the amount of L-lactic acid present in the wines. M L 0 1 decarboxylates all o f the malate present initially in the must to lactate, whereas O. ceni starts the M L F with less malate in the wine since S92 has already consumed approximately 10 - 30% o f the malate, probably via the malo-ethanolic pathway. A n increase in p H also plays a role in loss o f colour in red wines. The substantial loss of colour after a bacterial M L F is generally attributed to the p H effects on anthocyanins (Boulton, 1996). In an acid medium, anthocyanins are red and as the p H increases their colour changes to blue and eventually fade to yellow (Ribereau-Gayon et al., 2000b). Interestingly, Cabernet Sauvignon wines produced by M L 0 1 and S92 without a M L F , had a darker colour than wine produced with S92 with a bacterial M L F (Table 11, degree of lightness). A s expected, the wine with the lowest p H produced by S92 without a M L F (pH 3.80) had the highest value for redness and colour intensity (^ 420nm + ^520nm) and the lowest value for yellowness; wine produced by S92 with a bacterial M L F had the highest p H (pH 4.05) and was significantly lighter, the least red, 1 and the least intense (Table 11). Wine produced by M L 0 1 (pH 3.98) had a similar degree of yellowness to wine produced by S92 with a M L F , but redness and intensity were significantly higher (Table 11). These data clearly indicate that the metabolic activity of O. ceni impacts negatively on anthocyanins in red wine and the loss o f colour cannot be attributed simply to an increase in p H . Volatile acidity is another important component o f wine acidity. Acetic acid, the main component of volatile acidity in wine, was found in significantly lower concentrations in wines produced with M L 0 1 and S92 without a M L F than in wines produced with S92 and a bacterial M L F (Tables 10 and 11 and Appendix O). High-acid Chardonnay wine produced with M L 0 1 had lower concentrations o f acetate (0.452 g/L) than the wine produced with a bacterial M L F (0.5 g/L) but higher levels of acetate than wine fermented with S92 alone (0.399 g/L). The Chardonnay wine produced by M L 0 1 from fruit harvested in 2004 had lower levels of acetate (0.25 g/L) than the wine produced with and without a bacterial M L F (0.424 g/L and 0.328 g/L, respectively). Acetate concentrations in Cabernet Sauvignon wines produced by M L 0 1 was 0.324 g/L, by S92 with a bacterial M L F 0.355 g/L, and by S92 alone 0.237 g/L. The higher concentration of acetic acid in wines produced with O. ceni can be attributed to the metabolism of remaining sugars and citric acid in the wine (Liu , 2002; Ribereau-Gayon et al., 2000a; van Vuuren and Dicks, 1993). In addition to the acidity properties o f wine, numerous other metabolites influence the organoleptic characteristics of wine. G C / M S headspace analysis o f volatile compounds was conducted on 2004 Chardonnay wine since no biological replicates were available for the 2000 Chardonnay wines (only one carboy completed the bacterial M L F ) . 101 G C / M S analysis showed that no additional compounds were detected in wine produced with M L O l when compared to wine produced with the parental strain with or without a M L F . However, concentrations of certain compounds such as ethyl lactate, an aroma compound (buttery), that also gives the wine a 'broader', 'fuller' taste (Henick-Kling, 2002), varied greatly. Ethyl lactate concentrations were high in wines produced by M L O l (177.9 mg/L) and S92 with a bacterial M L F (295.88 mg/L) but were low in wine produced by the parental strain without a M L F (5.63 mg/L). The different ethyl lactate concentrations are most l ikely related to the lactate concentrations found in the wines. Concentrations of the following compounds were higher in wines produced with S92 and O. ceni than wines produced by M L O l or S92 alone: dimethyl sulfide (quince, truffle odour), ethyl acetate (fruity at lower concentrations, acescence and unpleasant at higher concentrations), ethyl isovalerate (fruity, vinous odour), isobutyl alcohol (fruity), acetoin (milky and fatty odour), ethyl octanoate (soapy, candlewax odour), acetic acid (vinegar), ethyl decanoate (oily, fruity, floral), and diethyl succinate (faint, pleasant odour) (Table 13) (Clarke and Bakker, 2004; Ribereau-Gayon et al., 2000b). Although these volatile compounds were found in higher concentrations in wine produced with the S92 and O. ceni, it is difficult to ascertain i f the higher concentrations of several different chemicals in a complex matrix wi l l detract from the quality o f wine or improve it. Sensory analysis of Chardonnay wines aged for four years showed that wine produced by M L O l was judged highest for overall quality (Figure 17). Overall quality was also strongly correlated with body and negatively correlated with acidity (Appendix P). Chardonnay wine produced by S92 without a M L F was judged to be significantly darker than wines produced by M L O l or S92 with M L F and this correlates well to the 102 degree o f lightness measured by C I E L A B colour measurements (Table 10). C I E L A B colour measurements also indicated that Chardonnay wine produced with M L 0 1 had the lowest amount o f 'greenness' and 'yellowness' and S92 with a M L F had the highest amount of'greenness' and 'yellowness' (Table 10). These colour subtleties were not perceived by the panellists. Body and perceived sweetness was highest for wines produced with M L 0 1 and lowest for acidity. The complete degradation of malate by M L 0 1 may have contributed to the perceived sweetness of wines produced by M L 0 1 . The main descriptive attributes that are associated with wine produced by M L 0 1 are highest overall quality, fruity taste, sweetness (perceived because o f a lack o f acidity when compared to other wines) and body, whereas, high acidity is an attribute of wine produced with S92 with and without a M L F , and dark yellow colour is an attribute of wine produced with S92 without a M L F (Figures 17 and 18). In addition to compounds that affect organoleptic properties of wine, lactic acid bacteria produce harmful compounds such as biogenic amines and ethyl carbamate (Lonvaud-Funel, 1999). Biogenic amines are produced from their respective precursor amino acids by specific amino acid decarboxylases. Histamine, the best studied biogenic amine, can cause headaches, hypotension, sneezing, flush, skin itch, shortness of breath and digestive problems (Soufleros et al., 1998; Wantle et al., 1994). The increase in biogenic amines after M L F is well documented (Marcobal et al., 2006) and could be dramatically reduced i f wine were to be sulphited and processed immediately after alcoholic fermentation. Arginine metabolism by certain strains of O. ceni and other heterofermentative L A B leads to the formation of ethyl carbamate (urethane) precursors (Liu and Pilone, 103 1998). Ethyl carbamate is a potential human and known animal carcinogen found in wine (Ough, 1976; Ough, 1993). The main ethyl carbamate precursors in wine are urea (produced by yeast), and citrulline (produced by L A B ) . The utilisation of M L O l to conduct the M L F during alcoholic fermentation prevents the formation ethyl carbamate from citrulline. The concentration of ethyl carbamate in Chardonnay wines produced by M L O l and S92 without a bacterial M L F were significantly lower than wines produced with a bacterial M L F (Table 14). Complete reduction o f ethyl carbamate is not possible with the M L O l yeast since other precursors such as urea and other carbamyl compounds are still available for E C production 4.4 Integration of the malolactic cassette in S92 does not confer a growth advantage to MLOl nor does it not affect the production of ADY or wine making The M L F by M L O l coincides with the log growth phase of the yeast. In laboratory media ( Y P D ) the maximum specific growth rate was slightly lower for the M L O l strain (0.54 ± 0.01 h"1) than the parental strain S92 (0.55 ± 0.005 h"1). However, the slightly longer doubling time observed in Y P D media did not affect the growth of M L O l on molasses during the production of A D Y (Didier Colavizza, Lesaffre Development, rue Gabrie Peri 137, F-59700 Marcq-en-Baroeul, France, personal communication). Moreover, the M L O l strain and the parental strain, S92, have the same maximum specific growth rate in Chardonnay must (0.37 h"1) and die at a similar rate (Figure 20) indicating that the introduction of the malolactic cassette should not affect wine making. 104 The M L 0 1 strain, like the parental strain S92, is unable to consume malate as a sole carbon source (Figure 14) showing that the malolactic cassette does not confer any competitive advantage to the M L 0 1 strain in this respect. Furthermore, although the M L 0 1 yeast can efficiently decarboxylate malate to lactate in a variety of musts, it cannot appreciably decarboxylate malic acid to lactic acid when present at levels below 1% o f the total inoculum (Figures 19 a,b). Even at inoculum levels of 10% M L 0 1 , the decarboxylation of malate to lactate was limited (33.3%) and the M L F did not continue after the first few days (Figures 19 c,d). These data indicate that the M L F mainly coincided with the growth phase of the M L 0 1 strain and that the M L F was not active during the later stages of the fermentation. Hence, cross-contamination of M L 0 1 yeast in must not destined for deacidification in wineries should not be a concern under commercial conditions. 4.5 The malolactic cassette without any antibiotic resistance markers is integrated correctly and stably into the genome of M L 0 1 Screening for the integration of the malolactic cassette in the genome of S92 was completed with the assistance of a co-transforming plasmid pUT332. The purpose o f pUT332 was to reduce the number of transformants to be screened by the colorimetric method. The linear malolactic cassette was combined in a 10:1 molar ratio with pUT332 in order to select yeast cells that were permeable to D N A . After successive subculturing in non-selective media the plasmid was lost (Figure 7). Southern blots probing for Tn5ble (phleomycin resistance) (Figure 7) and the bla gene (ampicillin resistance) (Figure 8) confirmed absence of these antibiotic genes. The probe for the bla gene also 105 contained 970 bp of pUT332 non-Saccharomyces sequences comprising 67.1% of all the pUC19 sequences distributed throughout the sequence of pUT322 (Appendix A ) . Southern analyses confirmed that pUT332 or components o f the plasmid (antibiotic markers and E. coli vector sequences) are not present in M L O l or S92. M L O l is therefore the first genetically engineered wine yeast strain to be constructed without the integration o f an antibiotic marker or E. coli vector sequences (Schuller and Casal, 2005). The screening method and the colony P C R results (Appendix E) demonstrated that at least one copy o f the malolactic cassette was integrated into one copy of the URA3 locus. Southern blots were performed in order to confirm the absence of concatamers and illegitimate integrations elsewhere in the genome (Figures 4-6 and Appendix H). Uti l is ing different restriction enzyme digests (EcoRV, Nsil and PvulT) and four different probes corresponding to the mael, mleA, and URA3 genes, and the PGK1 promoter, it was confirmed that a single copy of the URA3 gene was disrupted by the malolactic cassette without the formation of concatamers; no additional copies of the malolactic cassette were intergrated elsewhere in the genome. Furthermore, analysis of M L O l ascospores confirmed that one copy of the URA3 gene was disrupted by the malolactic cassette since yeast cells from two spores were auxotrophic for uracil but positive for M L F and cells from the remaining two ascospores were prototrophic for uracil and negative for M L F . The targeted integration of the malolactic cassette to the URA3 locus ( Y E L 0 2 1 W ) was completed using flanking sequences that are 925 bp (5' end) and 933 bp (3' end) long. Although it has been previously shown that flanking sequences of 30 bp are sufficient for homologous recombination in lab strains of Saccharomyces sp. (Baudin et 106 al., 1993; Manivasaskam et al., 1995), the efficiency largely depends upon perfect homology at both ends o f the transforming D N A to the target locus (Wach, 1996). Sequence polymorphism in different strain backgrounds reduces the efficiency of homologous recombination when short-flanking sequences are used. Transformation efficiencies can be increased approximately 30- to 50-fold using flanking sequences of several hundred base pairs in length (Wach, 1996). Both strands of the integrated malolactic cassette were sequenced (Appendix B) . Eight unexpected nucleotide differences were found after comparing the sequence of the integrated malolactic cassette to previously published sequences (Appendix K ) . Six differences were found in either non-coding regions (PGK1 promoter) or in the non-functional disrupted URA3 O R F (Appendix K ) . Two changes were found in the mleA sequence, one corresponding to a silent mutation, the other involving the exchange of one amino acid for another amino acid o f the same family (Appendix K ) . These anomalies could be due to genetic polymorphisms or amplification errors during the construction o f the cassette. Alternatively, the differences may be attributed to errors in the original published sequences. The two differences involving one of the PGK1 promoter sequences have l ikely originated from polymerase errors during the amplification steps involved in the cloning o f the malate expression cassette because the other PGK1 promoter did not contain these two nucleotide differences. The two differences found in the coding sequence of the mleA gene could have originated from a mistake during the amplification steps or could be attributed to errors in the published sequence or a mutation within the strain isolate used for the cloning of mleA. Regardless, one change resulted in a silent mutation and the other change resulted in a switch between two acidic 107 amino acids, aspartic acid to glutamic acid and did not affect the functionality of the enzyme. N o differences were found in the mael sequence when compared to the published sequence which comes from the same S. pombe strain. These results indicate that the sequences composing the malolactic cassette that is integrated in M L O l are not significantly different than the original sequences isolated from the donor strains. Other than the eight nucleotide differences, the malolactic cassette contained only synthetic polylinkers and D N A from the two donor microorganisms (Appendices J and K ) . The upstream and downstream sequences of the malolactic cassette were also sequenced and show the native half-Sr/I sites and perfect homology to native sequences surrounding the integration sites at each end of the malolactic cassette (Figure 9). The blunt-end restriction enzyme Srfi was specifically used so that after digestion from pJH2 the remaining half sites o f the linear malolactic cassette would be completely homologous to native sequences (Figure 9). The integrated malolactic cassette must be stable for A D Y yeast production and subsequent winemaking procedures. In general, yeast production starts with pure culture slants and then proceeds through several pure culture laboratory scale fermentations (0.2 - 25 kg of yeast), followed by several larger batch and fed-batch fermentations (25 -15,000 kg of yeast), and then it concludes with the final fed-batch "trade" fermentation (15,000 - 100,000 kg of yeast) (Reed and Nagodawithana, 1991). Throughout yeast production and growth in grape must, S. cerevisiae cells divide mitotically by forming a bud that eventually leads to a daughter cell. Theoretically, i f one diploid yeast cell (80 x 10"1 2 g) (Sherman, 1997) and its progeny continuously doubled (with each mother cell forming no more than 20 - 30 daughter cells) it would take 60 generations before 108 approximately 92,230 kg of yeast is produced. During commercial winemaking, yeast generally undergo a maximum of seven generations since large initial inoculums of 1.5 -6 x 10 6 cells/mL (5 - 20 g/hL of A D Y ) are used and the cells multiply to a maximum of 5 -15 x 10 7 cells/mL (Reed and Chen, 1978; Reed and Nagodawithana, 1991). Hence a relatively low number of generations (< 70) are required to produce wine using A D Y . Over this length of time the majority of M L 0 1 cells are expected to retain a functional malolactic cassette since the frequencies of molecular mechanisms that could affect stability of the malolactic cassette, and therefore the M L F phenotype, are relatively low. Spontaneous mutations (10~9 to 10"8 per generation in S. cerevisiae) (Magni and von Borstel, 1962), loss of the malolactic cassette via mitotic gene conversion or crossing over (10~6 to 10"5 per generation) (Petes et al., 1991; Puig et al., 2000) or loss of a portion of the malolactic cassette (either the mael or mleA expression cassettes) due to recombination at the direct repeats (the two PGK1 promoter or terminator sequences) (10 4 to 10"3 per generation) (Estruch and Prieto, 2003; Wach et al., 1994) are infrequent and w i l l have a minimal effect on the M L F performed by the majority of functional M L 0 1 yeast cells. To date, several wine fermentations with M L 0 1 A D Y (produced at a pilot plant and large-scale fermentation facility) have been done and all have achieved a complete M L F . Even wine fermentations conducted with a 1:1 ratio of M L 0 1 and S92 cells are able to accomplish an almost complete M L F (95.3% of malate degraded, data not shown). Moreover, after growth and completion of M L F in Chardonnay must by M L 0 1 A D Y (produced at a large-scale A D Y production facility, Lesaffre, Turkey), 99.3 ± 1.0 % of M L 0 1 colonies tested positive for a M L F phenotype. 109 4.6 The malolactic cassette has a minimal effect on the transcriptome and proteome of MLOl Global gene expression patterns of M L O l and S92 were analysed at 48 hours and 144 hours. These two time points were selected because at 48 hours the M L O l yeast cells perform the M L F and at 144 hours the M L F is complete (Figure 11). The two time points also correspond to log and stationary growth phase o f the yeast (Figure 12). A t 48 hours the transcription of 19 genes were affected = two-fold, and at 144 hours, the transcription of six genes were affected = two-fold (Table 9). With only 25 genes having a = two-fold change at both time points it is clear that the introduction of the malolactic cassette into S. cerevisiae S92 had a minimal effect on the transcription of the 5773 ORFs (4286 verified amd 1487 uncharacterised, Saccharomyces Genome Database, June 1, 2005) in the yeast cell. Moreover, the data suggests that no metabolic pathway was affected by the presence of the malolactic expression cassette integrated into the genome of M L O l . The difference in the number of genes being affected at 48 hours (19 genes) and 144 hours (6 genes) is likely due to the M L F that is proceeding within M L O l cells at 48 hours. Only one gene, AQR1, was found to be expressed differently at both time points. At 48 hours, AQR1 is expressed 1.77 fold higher in M L O l than S92 and at 144 hours AQR1 is expressed -3.23 fold lower. AQR1 is a gene involved in the excretion of short-chain (C2-C6) monocarboxylic acids (such as lactate), quinidine, and excess amino acids (Tenreiro et al., 2002; Velasco et al., 2004). A t 48 hours, lactic acid is present within M L O l yeast cells and it is likely that A q r l p is involved in the transport of lactic acid out of the yeast. Interestingly, an A q r l p - G F P fusion protein was shown to be localized to 11 multiple internal membrane structures and appears to cycle between these components and the cell surface (Velasco et al., 2004). According to this model, it is possible that A q r l p catalyzes transport of lactic acid into vesicles that subsequently release the lactic acid into the external medium by exocytosis. Excretion of lactic acid via exocytosis may also explain the observation in the lag of lactic acid accumulation in the external medium during M L F . A t 144 hrs there was a -3.23 fold decrease in expression of AQR1 in the M L 0 1 strain (Table 9). A t 144 hrs, the malolactic fermentation was complete with the production of equimolar amounts o f lactic acid from malic acid and no further increases in lactic acid were observed. This indicates that the yeast has transported all o f the lactic acid into the surrounding media and no longer requires higher levels of A q r l p to remove the intracellular lactic acid. AQR1 is also involved in the secretion of excess amino acids (Velasco et al., 2004). The amino acids reported to be present in the highest concentrations in the cytosol of cells growing in glucose-ammonium medium (such as the synthetic must) are glutamate, aspartate and alanine, followed by asparagine, glutamine, serine and glycine (Messenguy et al., 1980). Other amino acids present at high levels are predominately found in the vacuole. The higher levels of AQR1 during M L F could also result in the excretion of the main cytosolic amino acids, in addition to exporting lactic acid out of the cell. This depletion of amino acids may explain the increase in expression of DIP5 (2.81 fold, Table 8) at 48 hours. DIP5 is a dicarboxylic amino acid permease that mediates high affinity and high-capacity transport of glutamate and aspartate (Regenberg et al., 1998). DIP5p also transports alanine, glutamine, asparagine, serine and glycine; the same 111 amino acids potentially excreted by the over expression of AQRI (Regenberg et al., 1999). Another gene of interest, PH084, is a high-affinity inorganic phosphate transporter and a low-affinity manganese transporter involved in manganese homeostasis (Jensen et al., 2003). It is l ikely expressed at higher levels at 48 hours since the malolactic enzyme requires manganese as a cofactor to complete the decarboxylation of malate to lactate during the M L F . PH084 (2.18 fold, Table 8) may be required at higher levels in order to fulfill the manganese demand on the cell due to the presence of the transgenic mleAp. Global gene expression patterns of M L O l indicate that no metabolic pathway was affected by the presence of the malolactic cassette integrated into its genome. The analysis also suggests that the genes AQR1, DIP5 and PH084 are affected in order to remove intracellular lactic acid and maintain homeostasis. The proteomes o f M L O l and S92 were analysed after 48 hours of fermentation in synthetic must. The time point was selected because at 48 hours the M L O l yeast cells are performing the M L F (Figure 11) and the analysis of the transcriptome indicated that more genes are affected at this time point (Table 8). i T R A Q analysis identified only one protein, lanosterol 14-demethylase cytochrome P450 that was shown to have a weighted average ratio of 0.799 (±0.031) at a p-value < 0.05 across duplicate experiments. The protein corresponds to the gene ERG11 and is one of the 19 enzymes involved in ergosterol biosynthesis. Since i T R A Q analysis identified 12 of 19 enzymes involved in the ergosterol pathway and only 1 protein showed a lower ratio, it is unlikely that the ergosterol pathway is affected by the malolactic cassette. 11 The analysis of the proteome of M L 0 1 showed that 199 of the 559 detected proteins were involved in many of the major metabolic pathways including carbohydrate and amino acid metabolism, and pyrimidine, purine, fatty acid, ergosterol and formate biosynthesis. Other than E R G 1 l p , protein ratios for M L 0 1 and S92 were identical. The malolactic cassette encodes unique proteins such as the mleAp, maelp, and possibly the three putative proteins created by cloning (Figure 10). The mleAp was identified in both M L 0 1 duplicate samples (confidence >99%); the maelp and the three putative proteins were not identified at the default parameters (confidence >94%, or at a confidence >50%). The difficulty in detecting the maelp could be due to its membrane bound nature and possibly lower levels of protein due to degradation. The maelp contains a P E S T region (aa 421-434) at the C-terminal end that targets the protein for degradation (Grobler et al., 1995). Many proteins with short intracellular half-lives contain a P E S T region that consists of proline, glutamic acid, serine, threonine and to a lesser extent aspartic acid (Rogers et al., 1986). Since the mael transcript was detected by R T - P C R (Appendix M ) and the M L 0 1 yeast performs the malolactic fermentation efficiently; it is reasonable to suggest that the maelp was present, even though the protein was not detected. The three putative proteins composed entirely o f Saccharomyces sequences and one or two common restriction endonuclease sites (Kpnl and/or Notl), were not detected at a confidence level of >50%. The introduction of the malolactic cassette into S. cerevisiae S92 affected the concentration of one protein out of 559 identified proteins; no metabolic pathway in the yeast cell was found to be affected. Therefore, the introduction of the malolactic cassette had minimal effect on global gene expression and protein levels. 11 4.7 Ethical considerations concerning use of the genetically modified yeast MLOl The issues surrounding the use of the malolactic yeast M L O l are the same as those facing every G M O designed for use in the food industry. Broadly speaking, these issues include possible effects on the health of consumers, potential environmental impact and social considerations. S. cerevsiae is an organism which has an extensive history of safe use. It has been used for millennia in fermentation processes such as bread leavening, and wine or beer production. The introduction of non-harmful, limited and well-characterised D N A from two other wine microorganisms should not adversely affect the safety of S. cerevisiae M L O l . The mleA gene was isolated from O. ceni which is not only found in wine but attains very high populations during M L F (10 to 10 cfu/mL) (Wibowo et al., 1985). Despite the two nucleotide differences discovered in the mleA gene integrated into M L O l , the primary sequence of the malolactic enzyme is similar to the native malolactic enzyme since one difference corresponds to a silent mutation and the other involves the change of an acidic amino acid for another acidic amino acid. Moreover, the M L O l strain displays an efficient L-malate decarboxylating activity indicating that the secondary and tertiary structure of the enzyme is conserved. Sequencing of the inserted mael gene of M L O l showed that no sequence discrepancy could be found between the mael gene in M L O l and the native mael gene from S. pombe. Therefore the primary structure of the malate permeases are the same and the fact that the malate permease is functional in M L O l indicates that the secondary and tertiary structures of the proteins are similar as well . Although S. pombe is a wine related microorganism and it may occasionally participate in spontaneous grape must fermentations (and was recently commercialised for use in the wine industry), this 114 organism is not as predominant in wine fermentations as S. cerevisiae and O. ceni. Therefore, a literature search was carried out by Lesaffre's library department using Medline and Biosis databases to evaluate the allergenicity of S. pombe. Not a single paper implicating S. pombe in allergenic reactions was found (Didier Colavizza, Lesaffre Development, rue Gabrie Peri 137, F-59700 Marcq-en-Baroeul, France, personal communication). Moreover, it is l ikely that at the end of the fermentation very little malate permease wi l l remain in M L 0 1 yeast cells due to rapid intracellular degradation o f the protein as a consequence of the presence of a P E S T region in its C-terminus (Grobler et al., 1995). Storing the wine on lees (sediment composed primarily of yeast), i f desired, would also further enhance this degradation of the malate permease. During storage on lees, cell proteins and nucleic material first undergo intracellular enzymatic degradation due to the liberation of intracellular proteases, aminopetidases, nucleases and phosphatases (Charpentier and Feuillat, 2002; Charpentier et al., 2005; Moreno-Arribas and Polo, 2005; Perrot et al., 2002). The yeast cell gradually loses its hydrolyzed contents as integrity of the cell wall is compromised. In the wine, further degradation can occur by proteases present in the extracellular media (Charpentier and Feuillat, 2002). Therefore the protein content of wines stored on M L 0 1 lees or S92 lees should result in very similar hydrolysis products, namely small peptides and amino acids. Given the origin of the malate permease and the safe history of the presence o f S. pombe strains in fermented beverages, as well as the absence of data concerning the allergenicity of this yeast, it can reasonably be concluded that i f any small peptides hydrolyzed from the maelp are present in wine they wi l l not constitute a health safety issue. 11 Standard winemaking procedures also consist of various forms o f clarification, wine stabilisation and filtration procedures that drastically reduce yeast cell numbers after alcoholic fermentation. It is a common practice to employ filtration prior to bottling to improve clarity and stabilisation of the wine. Various types of cellulose and membrane filters can reduce the viable yeast cell count from 50 cells/100 m L to <1 cell/100 m L , depending on the porosity of the individual filters (Ribereau-Gayon et al., 2000b). Although most wines (especially white wines) are filtered prior to bottling, some winemakers may rely only on clarification o f their red wines with gelatins or egg white albumin and bottle without filtration. The use of M L 0 1 is compatible with such a procedure although the final concentration of yeast in the bottle wi l l be greater than filtered wine. The thorough genetic characterisation of M L 0 1 has confirmed that the integration site contains no D N A sequences other than those described in the malolactic cassette. . The discovery that four new ORFs were created during construction of the malolactic cassette also does not affect the safety status of M L 0 1 . The three of the four putative proteins that could be checked by i T R A Q , composed entirely of Saccharomyces sequences and one or two common restriction endonuclease sites (Kpnl and/or Noil), were not detected. The fourth new O R F is entirely composed of S. cerevisiae sequences (truncated URA3, 3' end) and the i T R A Q method is incapable of differentiating it from the native U R A 3 p . None of the putative proteins of these novel ORFs were detected by liquid chromatography and tandem mass spectrometry ( i T R A Q ) even at the 50% confidence limit. 11 The environmental impact of the M L O l strain should be no greater than the environmental impact of the industrial S92 strain. Commercial yeasts such as S92 are annually released in large quantities into the environment surrounding wineries. Recently, a large-scale three-year study of six different vineyards revealed that dissemination of commercial yeast in the vineyard is limited to short distances over short periods of time and is largely favoured by the presence of water run-off (Valero et al., 2005). Despite the annually intensive dissemination of commercial yeast into the local environment, 94% of the commercial yeast strains were found between 10 to 200m from the winery. This underscores the limited range of possible environmental impact beyond the winery. Moreover, analysis of population variations from year to year indicated that commercial strains do not settle in the vineyard or predominate over the indigenous flora (Valero et al., 2005). It has also been shown that colonisation of damaged grapes, where the modified ecology favours fermenting yeast species, by a selected S. cerevisiae wine strain is no different from colonisation of undamaged grapes (Comitini and Ciani, 2006). In both cases the inoculated wine strain could not out-compete the indigenous microflora resident on the grapes (Comitini and Ciani, 2006). Survival of G M yeast strains in a confined wine cellar and greenhouse vineyard has been studied (Bauer et al. 2003 - see Schuller and Casal, 2005). In this study, four G M yeast containing resistance markers (KanMXox SMR1-140) and expressing the transgenes (with strong yeast promoters) for a-amylase, endo-6-l,4-glucanase, xylanase or pectate lyase were sprayed onto vines in a confined greenhouse vineyard. Results showed that despite high initial cell counts, few S. cerevisiae cells were isolated from grapes, leaves, stems and soil during weekly monitoring. Furthermore, no significant 11 difference between the occurrences of the modified strains compared to the parental strain was detected, including G M strains secreting glucanases and pectinases (modifications thought to provide a selective advantage). The total yeast population of treated vines was also very similar to the untreated control vines and spontaneous micro-vinifications resulted in no significant differences in the fermentations performances amongst the trials (Schuller and Casal, 2005). Survival of G M and "self-cloned" baker's yeast in a simulated natural environment (water and soil) also showed that G M and self-cloned yeast decreased at an equal or faster rate than the wild-type control (Ando et al., 2005). In this case, the G M and Self-Cloned yeast were both modified for freeze tolerance by disruption o f the acid trehalase gene (ATH1). The possibility o f horizontal gene transfer to other organisms cannot be ignored. Horizontal transfer o f D N A can occur via interspecies mating among yeast belonging to the Saccharomyces sensu stricto complex and across species barriers via transfer o f plasmid D N A (Marinoni et al., 1999; Mentel et al., 2006; Nevoigt et al., 2000). A s no antibiotic resistance markers are present in the M L 0 1 strain, much of the concern associated with horizontal gene transfer is not applicable. The transfer of the entire malolactic cassette or part of it to another organism would also not constitute a threat to the environment. The transfer of both transgenes to another organism in the environment (most likely another Saccharomyces sensu stricto strain) may give that particular cell the ability to degrade malate, like M L 0 1 ; an ability that is already present in numerous lactic acid bacteria and malate degrading yeast such as S. pombe present in the ecology in and around the winery. Individually the mleA and mael genes are also predominant in the environment and it is presumable that i f they conferred a selective advantage to an 11 organism this transfer of genetic material would have already occurred. The combination of S. cerevisiae promoter sequences with transgenes is also not a major concern since recognition of these regulatory sequences would only be effective in S. cerevisiae strains (and possibly other phylogenetically close species) and less so in distantly related organisms. It is also conceivable that due to the massive world-wide production and utilisation of baker's yeast, brewer's yeast and wine yeast (as well as other industries utilising Saccharomyces sp.) D N A fragments containing PGK1 regulatory sequences have been dispersed into the environment over the centuries. The addition of the malolactic cassette to this mileu is of minimal concern. The well-characterised malolactic cassette is safe, does not confer any advantage to M L O l and should be considered as very low risk to the environment. Specific social consequences of using the M L O l strain in the wine industry also exist. Wine is a traditional product that is perceived to be more than just an alcoholic beverage. Wine also has strong geographical ties to certain regions/countries and in many cases the wine industry in these regions is unreceptive to the use of G M O s . Wi th this background in tradition it is very difficult to incorporate new technologies and i f they are to be successful, they have to offer advantages to the consumer. In this regard the M L O l yeast can succeed i f the consumer is educated about the toxic effects of naturally occurring bioamines and how improvements in winemaking (early sulphiting) and the use of M L O l can reduce or eliminate these compounds. In this respect the M L O l yeast is unique as a G M O since it provides a direct health advantage to consumers and a direct benefit to the producer (specifically a complete M L F during alcoholic fermentation). 119 Although there w i l l always be wine makers that prefer inoculation of malolactic bacteria or even the use of the natural microflora of the wine cellar to complete the M L F , the malolactic yeast is an important additional tool for winemakers to produce wholesome, well-balanced high quality wines. 120 5 CONCLUSIONS The production of high quality wines that are enjoyable, healthful and produced by environmentally sustainable production methods, has become important in a globalized world where there has been a paradigm shift from a production-driven to a market-driven wine industry (Pretorius and Hoj, 2005). Consumers have become sensitized and concerned about food safety issues (e.g. mad cow disease and dioxin in chocolates) and are increasingly demanding safe food and beverages. Moderate consumption of wine protects consumers against cardiovascular disease, dietary cancers, ischaemic stroke, peripheral vascular disease, diabetes, hypertension, peptic ulcers, kidney stones and macular degeneration (Bisson et al., 2002). These potential health benefits may have contributed to the increased popularity o f wine in recent years. However, wine also contains compounds such as biogenic amines that exert negative effects on the central nervous and vascular systems of consumers. Naturally occurring lactic acid bacteria present in wine as well as O. oeni, the bacterium used in commercial starter cultures, produce these bioamines by decarboxylating naturally occurring amino acids in grape must to their corresponding bioamines. After consumption, biogenic amines are usually metabolized by amine oxidases such as monoamine and diamine oxidase and histamine N-methyltransferase to physiologically less active products (Santos, 1996). However, in consumers without sufficient bioamine detoxifying enzymes, these amines can be absorbed into the bloodstream causing negative health affects. Genetic engineering o f wine yeast can be used to prevent the formation of bioamines in wine. It has also been used to minimize the formation of E C , a well-known carcinogen in wines (Coulon et al., 2006). Moreover, powerful genomic, transcriptomic, 121 proteomic and metabolomic techniques are now available to demonstrate that recombinant yeasts are substantially equivalent to the parental strains. The genetically stable commercial wine yeast strain, M L 0 1 , was constructed by expressing the S. pombe malate transporter gene (mael) and the O. ceni malolactic enzyme gene (mleA) under control of the S. cerevisiae PGK1 promoter and terminator signals in a popular industrial wine yeast strain of S. cerevisiae. The malolactic cassette integrated into the URA3 locus o f M L 0 1 contains no vector sequences or antibiotic resistance marker genes. D N A sequencing confirmed that the integration site contains no D N A sequences other than those present in the isolated malolactic cassette. Results obtained from the hybridization of genomic D N A with various probes indicate that the malolactic cassette is correctly integrated into a URA3 locus. D N A mieroarray and i T R A Q analyses of the transcriptome and proteome of the yeast indicated that the introduction of the malolactic cassette had little effect on global gene expression patterns and protein levels. Phenotypic results show that the novel malolactic yeast M L 0 1 is capable o f efficiently decarboxylating malate to lactate within the first five days o f the alcoholic fermentation at 20°C; at lower temperatures (13°C) M L F by M L 0 1 can take up to nine days. Wines produced by the M L 0 1 yeast had lower volatile acidity than wine produced with the parental strain S92 and a bacterial M L F . M L 0 1 also produced Chardonnay wines lighter in color than wine produced by the parental strain and Cabernet Sauvignon wines darker in color than wines produced with S92 and a bacterial M L F . G C / M S analysis of volatile compounds and sensory analyses of wine produced by M L 0 1 , the parental yeast S92 and S92 plus O. ceni indicated that M L 0 1 is ideal for the 122 production of wine on a commercial scale as MLOl-produced wines were judged to be superior in overall quality by trained tasters. The bacterial M L F is unpredictable and often results in stuck M L F , and the production of off-flavors and biogenic amines. S. cerevisiae M L O l conducts an efficient malolactic fermentation that w i l l solve the important issue of cellar capacity and prevent oxidation and microbial spoilage o f wines that result in financial losses to wineries. Early sulfiting of wine produced with M L O l w i l l prevent the growth of undesirable lactic acid bacteria that produce biogenic amines. It is therefore conceivable that wines produced with M L O l should be free of toxic bioamines, providing great relief to wine lovers who were previously unable to consume many wines that underwent the bacterial malolactic fermentation. This thesis represents an integrated approach, conducting analyses at the phenotypic, D N A , R N A , and protein levels to determine i f recombinant yeast M L O l is substantially equivalent to the parental strain S92. Based on these results it was concluded that the M L O l strain is substantially equivalent to the parental strain S92. M L O l has been approved for use in Canada (Appendix Q) and has G R A S status with the U S F D A (Appendix R). It is the first metabolically engineered yeast to be commercialised by the wine industry and is currently available in Canada, the U S A and Moldova. Notifications are currently being submitted to all o f the major wine producing countries in the world. 123 5.1 Future Directions Several lines of future investigations with the novel M L 0 1 yeast and its impact on the wine industry should be pursued. First, analysis of the bioamine content of wines produced by M L 0 1 in several different commercial wineries should be conducted. It is expected that wines produced by the M L 0 1 yeast would contain significantly lower levels of bioamines and these data could be used to educate wine consumers on how yeast biotechnology can reduce the toxic properties of naturally occurring biogenic amines in wine. Secondly, the ageing o f commercially prepared wines using M L 0 1 and appropriate control wines should be studied. This experiment would provide important data to the wine industry which places high regard on wines that age well . Thirdly, the construction of other strains containing the malolactic cassette would be of particular importance to winemakers. Although the strain S92 can be used to produce red and white wines, it is generally preferred for white wine production. Creating other malolactic strains well suited to red wine production could increase the options available to winemakers seeking to complete the M L F during the alcoholic fermentation of their red wines. 124 R E F E R E N C E S Akada, R. 2002. Genetically modified industrial yeast ready for application. J. Biosci . Bioeng. 94:536-544. Alexandre, H . , P.J. Costello, F. Remize, J. Guzzo, and M . Guilloux-Benatier. 2004. Saccharomyces cerevisiae - Oenococcus ceni interactions in wine: Current knowledge and perspectives. Int. J. Food Microbiol . 93:141-154. Ando, A . , C . Suzuki, and J. Shima. 2005. Survival of genetically modified and self-cloned strains of commercial baker's yeast in simulated natural environments: Environmental risk assessment. App l . Environ. Microb. 71:7075-7082. Ansanay, V . , S. Dequin, B . Blondin, and P. Barre. 1993. Cloning, sequence and expression of the gene encoding the malolactic enzyme from Lactococcus lactis. F E B S Lett. 332:74-80. Arena, M . E . , M . C . M . de Nadra, and R. Munoz. 2002. The arginine deiminase pathway in the wine lactic acid bacterium Lactobacillus hilgardii X I B : structural and functional study of the arcABC genes. Gene 301:61-66. Ausubel, F . M . , R. Brent, R . E . Kingston, D . D . Moore, J .G. Seidman, J A . Smith, and K . Struhl. 1995. Short Protocols in Molecular Biology. 3rd ed. Wi l ey & Sons, New York . Avedovech, R . M . , M . R . Mcdaniel, B .T . Watson, and W . E . Sandine. 1992. A n evaluation of combinations of wine yeast and Leuconostoc oenos strains in malolactic fermentation of Chardonnay wine. A m . J. Enol . Viticult. 43:253-260. Baranowski, K . , and F. Radler. 1984. The glucose-dependent transport of L-malate in Zygosaccharomyces bailii. Anton. Leeuw. Int. J. G . 50:329-340. 125 Barnett, J .A. , and H . L . Kornberg. 1960. The utilization by yeasts of acids of the tricarboxylic acid cycle. J. Gen. Microbiol . 23:65-82. Barnett, J .A. , R . W . Payne, and D . Yarrow. 1990a. Yeasts: Characteristics and Identification. 2nd ed. Cambridge University Press, New York. Bartowsky, E.J . 2005. Oenococcus oeni and malolactic fermentation - moving into the molecular arena. Aust. J. Grape Wine R. 11:174-187. Battaglia, R., H . B . S . Conacher, and B . D . Page. 1990. Ethyl carbamate (urethane) in alcoholic beverages and foods: A review. Food Add . Cont. 7:477-496. Battermann, G . , and F. Radler. 1990. A comparative study of malolactic enzyme and malic enzyme o f different lactic acid bacteria. Can. J. Microbiol . 37:211-217. Baudin, A . , O. Ozier-Kalogeropoulos, A . Denouel, F. Lacroute, and C. Cul l in . 1993. A simple and efficient method for direct gene deletion in Saccharomyces cerevisiae. Nucleic Acids Res. 21:3329-3330. Becker, J .V .W. , G.O. Armstrong, M . J . van der Merwe, M . G . Lambrechts, M . A . Vivier , and I.S. Pretorius. 2003. Metabolic engineering of Saccharomyces cerevisiae for the synthesis of the wine-related antioxidant resveratrol. F E M S Yeast Res. 4:79-85. Beelman, R . B . , and J.F. Gallander. 1979. Wine deacidification. Adv . Food Res. 25:1-53. Bisson, L . F . 1999. Stuck and sluggish fermentations. A m . J. Enol . Viticult. 50:107-119. Bisson, L . F . 2004. Biotechnology of wine yeast. Food Biotechnol. 18:63-96. Bisson, L . F . , A . L . Waterhouse, S.E. Ebeler, M . A . Walker, and J.T. Lapsley. 2002. The present and future of the international wine industry. Nature 418:696-699. 126 Boone, C , A . M . Sdicu, J. Wagner, R. Degre, C. Sanchez, and H . Bussey. 1990. Integration of the yeast K l killer toxin gene into the genome of marked wine yeasts and its effect on vinification. A m . J. Enol . Viticult. 41:37-42. Boulton, R . B . 1996. Principles and Practices ofWinemaking. Chapman & Hal l , New York . Bourdineaud, J.P., B . Nehme, S. Tesse, and A . Lonvaud-Funel. 2003. The ftsHgene of the wine bacterium Oenococcus ceni is involved in protection against environmental stress. App l . Environ. Microbiol . 69:2512-2520. Bourdineaud, J.P., B . Nehme, S. Tesse, and A . Lonvaud-Funel. 2004. A bacterial gene homologous to A B C transporters protect Oenococcus ceni from ethanol and other stress factors in wine. Int. J. Food Microbiol . 92:1-14. Bousbouras, G .E . , and R . E . Kunkee. 1971. Effect of p H on malolactic fermentation in wine. A m . J. Enol. Viticult. 22:121-126. Cambon, B . , V . Monteil , F. Remize, C . Camarasa, and S. Dequin. 2006. Effects o f GPD1 overexpression in Saccharomyces cerevisiae commercial wine yeast strains lacking ALD6 genes. App l . Environ. Microbiol . 72:4688-4694. Canas, B.J . , F . L . Joe, Jr., G . W . Diachenko, and G . Burns. 1994. Determination o f ethyl carbamate in alcoholic beverages and soy sauce by gas chromatography with mass selective detection: Collaborative study. J. A O A C Int. 77:1530-6. Carstens, M . , M . A . Vivier , P. Van Rensburg, and I.S. Pretorius. 2003. Overexpression, secretion and antifungal activity of the Saccharomyces cerevisiae chitinase. Ann . Microbiol . 53:15-28. 127 Caspritz, G . , and F. Radler. 1983. Malolactic enzyme oi Lactobacillus plantarum: Purification, properties, and distribution among bacteria. J. B i o l . Chem. 258:4907-4910. Cavalieri, D . , P .E . McGovern, D . L . Hartl, R. Mortimer, and M . Polsinelli. 2003. Evidence for S. cerevisiae fermentation in ancient wine. J. M o l . Evol . 57:S226-S232. Cebollero, E . , A . Martinez-Rodriguez, A . V . Carrascosa, and R. Gonzalez. 2005. Overexpression of cscl-1. A plausible strategy to obtain wine yeast strains undergoing accelerated autolysis. F E M S Microbiol Lett. 246:1-9. Charpentier, C , and M . Feuillat. 2002. Yeast autolysis, p. 225-242, In G . H . Fleet, ed. Wine Microbiology and Biotechnology. Taylor and Francis, New York. Charpentier, C , J. Aussenac, M . Charpentier, J .C. Prome, B . Duteurtre, and M . Feuillat. 2005. Release of nucleotides and nucleosides during yeast autolysis: Kinetics and potential impact on flavor. J. Agr i . Food Chem. 53:3000-3007. Clarke, R.J . , and J. Bakker. 2004. Wine flavour chemistry. Blackwell Pub., Oxford. Codon, A . C . , T. Benitez, and M . Korhola. 1998. Chromosomal polymorphism and adaptation to specific industrial environments of Saccharomyces strains. A p p l . Microbio l . Biot. 49:154-163. Cogan, T . M . 1987. Co-metabolism of citrate and glucose by Leuconostoc spp: Effects on growth, substrates and products. J. App l . Bacter id. 63:551-558. Comitini, F. , and M . Ciani. 2006. Survival of inoculated Saccharomyces cerevisiae strain on wine grapes during two vintages. Lett. App l . Microbiol . 42:248-253. 1 Conacher, H .B .S . , B . D . Page, B . P . Y . Lau, J.F. Lawrence, R. Bailey, P. Calway, J.P. Hanchay, and B . M o r i . 1987. Capillary column gas-chromatographic determination of ethyl carbamate in alcoholic beverages with confirmation by gas-chromatography mass-spectrometry. J. A O A C 70:749-751. Costello, P.J., and P . A . Henschke. 2002. Mousy off-flavor of wine: Precursors and biosynthesis of the causative N-heterocycles 2-ethyltetrahydropyridine, 2-acetyltetrahydropyridine, and 2-acetyl-l-pyrroline by Lactobacillus hilgardii D S M 20176. J. Agr. Food. Chem. 50:7079-7087. Costello, P.J., G.J . Morrison, T . H . Lee, and G . H . Fleet. 1983. Numbers and species of lactic acid bacteria in wines during vinification. Food Technol. Austr. 35:14-18. Coton, E . , G .C . Rollan, and A . Lonvaud-Funel. 1998a. Histidine carboxylase of Leuconostoc oenos 9204: Purification, kinetic properties, cloning and nucleotide sequence o f the hdc gene. J. App l . Microbiol . 84:143-151. Coton, E . , G . Rollan, A . Bertrand, and A . Lonvaud-Funel. 1998b. Histamine-producing lactic acid bacteria in wines: Early detection, frequency, and distribution. A m . J. Enol . Viticult . 49:199-204. Coulon, J., J.I. Husnik, D . L . Inglis, G . K . van der Merwe, A . Lonvaud, D.J . Erasmus, and H.J.J, van Vuuren. 2006. Metabolic engineering of Saccharomyces cerevisiae to minimize the production of ethyl carbamate in wine. A m . J. Enol. Viticult. 57:113-124. Cox, D.J . , and T. Henick-Kling. 1989. Chemiosmotic energy from malolactic fermentation. J. Bacteriol. 171:5750-5752. 129 Cox, D.J . , and T. Henick-Kling. 1995. Protonmotive force and A T P generation during malolactic fermentation. A m . J. Enol . Viticult . 46:319-323. Cunningham, T.S., and T . G . Cooper. 1991. Expression of the DAL80 gene, whose product is homologous to the G A T A factors and is a negative regulator of multiple nitrogen catabolic genes in Saccharomyces cerevisiae, is sensitive to nitrogen catabolite repression. M o l . Ce l l B i o l . 11:6205-6215. Danzer, K . , D . D . Garcia, G . Thiel, and M . Reichenbacher. 1999. Classification of wine samples according to origin and grape varieties on the basis of inorganic and organic trace analyses. A m . Lab. 31:26-34. Davis, C , D . Wibowo, G . H . Fleet, and T . H . Lee. 1988. Properties of wine lactic acid bacteria: Their potential enological significance. A m . J. Enol . Viticult. 39:137-142. Davis, C.R. , D . Wibowo, R. Eschenbruch, T . H . Lee, and G . H . Fleet. 1985. Practical implications o f malolactic fermentation: A review. A m . J. Enol. Viticult. 36:290-301. de Barros Lopes, M . , A . ur-Rehman, H . Gockowiak, A . J. Heinrich, P. Langridge, and P . A . Henschke. 2000. Fermentation properties of a wine yeast over-expressing the Saccharomyces cerevisiae glycerol 3-phosphate dehydrogenase gene (GPD2). Aus. J. Grape Wine Res. 6:208-215. de las Rivas, B . , A . Marcobal, and R. Munoz. 2004. Al le l i c diversity and population structure in Oenococcus ozni as determined from sequence analysis of housekeeping genes. App l . Environ. Microbiol . 70:7210-7219. 130 Denayrolles, M , M . Aig le , and A . Lonvaud-Funel. 1994. Cloning and sequence analysis of the gene encoding Lactococcus lactis malolactic enzyme: Relationships with malic enzymes. F E M S Microbiol . Lett. 116:79-86. Denayrolles, M . , M . Aig le , and A . Lonvaud-Funel. 1995. Functional expression in Saccharomyces cerevisiae of the Lactococcus lactis mleS gene encoding the malolactic enzyme. F E M S Microbiol . Lett. 125:37-43. Dequin, S. 2001. The potential of genetic engineering for improving brewing, wine-making and baking yeasts. App l . Microb. Biotech. 56:577-588. Dequin, S., E . Baptista, and P. Barre. 1999. Acidification of grape musts by Saccharomyces cerevisiae wine yeast strains genetically engineered to produce lactic acid. A m . J. Enol . Viticult. 50:45-50. DeSouza, L . , G . Diehl, M . J . Rodrigues, J. Guo, A . D . Romaschin, T.J . Colgan, and K . W . Siu. 2005. Search for cancer markers from endometrial tissues using differentially labeled tags i T R A Q and c I C A T with multidimensional l iquid chromatography and tandem mass spectrometry. J. Proteome Res. 4:377-86. Dicks, L . M . , F. Dellaglio, and M . D . Collins. 1995. Proposal to reclassify Leuconostoc oenos as Oenococcus ceni [corrig.] gen. nov., comb. nov. Int. J. Syst. Bacteriol. 45:395-397. D ivo l , B . , T. Tonon, S. Morichon, E . Gindreau, and A . Lonvaud-Funel. 2003. Molecular characterization o f Oenococcus ceni genes encoding proteins involved in arginine transport. J. A p p l . Microbiol . 94:738-746. 13 du Plessis, H . W . , C . L . C . Steger, M . du Toit, and M . G . Lambrechts. 2002. The occurrence of malolactic fermentation in brandy base wine and its influence on brandy quality. J. App l . Microbio l . 92:1005-1013. D u Toit., M . , and I.S. Pretorius. 2000. Microbial spoilage and preservation o f wine: using weapons from nature's own arsenal. S.A. J. o f Enol . and Vi t ic . 21:74-96. Eglinton, J . M . , A . J . Heinrich, A . P . Pollnitz, P. Langridge, P . A . Henschke, and M . D . Lopes. 2002. Decreasing acetic acid accumulation by a glycerol overproducing strain of Saccharomyces cerevisiae by deleting the ALD6 aldehyde dehydrogenase gene. Yeast 19:295-301. Erasmus, D.J . , G . K . van der Merwe, and H.J.J, van Vuuren. 2003. Genome-wide expression analyses: Metabolic adaptation of Saccharomyces cerevisiae to high sugar stress. F E M S Yeast Res. 3:375-399. Estruch, F., and J .A. Prieto. 2003. Construction of a Trp- commercial baker's yeast strain by using food-safe-grade dominant drug resistance cassettes. F E M S Yeast Res. 4:329-338. Fleet, G . H . 1998. The microbiology of alcoholic beverages, p. 217-262, In B . J. B . Wood, ed. Microbiology of Fermented Foods, V o l . 1. Blackie Academic and professional, Glasgow. Fleet, G . H . , and G . M . Heard. 1993. Yeasts - Growth during fermentation, In G . H . Fleet, ed. Wine Microbiology and Biotechnolgy. Harwood Academic Pub, Switzerland. Fortier, L . C . , R. Tourdot-Marechal, C. Divies, B . H . Lee, and J. Guzzo. 2003. Induction of Oenococcus oeni H + -ATPase activity and m R N A transcription under acidic conditions. F E M S Microbiol . Lett. 222:165-169. 132 Fuck, E . , G . Stark, and F. Radler. 1973. Apfelsaurestoff-wechsel bei Saccharomyces II. Anreicherung und eigenschaften eines malatenzymes. Arch. Mikrobio l . 89:223-231. Gallander, J.F. 1977. Deacidification of eastern table wines with Schizosaccharomyces pombe. A m . J. Enol. Viticult . 28:65-68. Ganga, M . A . , F. Pinaga, S. Valles, D . Ramon, and A . Querol. 1999. Aroma improving in microvinification processes by the use of a recombinant wine yeast strain expressing the Aspergillus nidulans xlnA gene. Int. J. Food Microbiol . 47:171-178. Gao, C , and G . H . Fleet. 1995. Degradation of malic and tartaric acids by high density cell suspensions of wine yeasts. Food Microbiol . 12:65-71. Garcia, M . J . , M . Zuniga, and H . Kobayashi. 1992. Energy production from L-malic acid degradation and protection against acidic external p H in Lactobacillus plantarum C E C T 220. J. Gen. Microbiol . 138:2519-2524. Garmyn, D . , C. Monnet, B . Martineau, J. Guzzo, J.F. Cavin, and C. Divies. 1996. Cloning and sequencing of the gene encoding alpha-acetolactate decarboxylase from Leuconostoc oenos. Ferns Microbiol . Lett. 145:445-450. Garvie, E.I. 1967. Leuconostoc oenos sp.nov. J. Gen. Microbiol . 48:431-438. Gatignol, A . , M . Baron, and G . Tiraby. 1987. Phleomycin resistance encoded by the ble gene from transposon T n 5 as a dominant selectable marker in Saccharomyces cerevisiae. M o l . Gen. Genet. 207:342-348. 133 Gindreau, E . , E . Wall ing, and A . Lonvaud-Funel. 2001. Direct polymerase chain reaction detection of ropy Pediococcus damnosus strains in wine. J. App l . Microbiol . 90:535-542. Giudici , P., L . Solieri, A . M . Pulvirenti, and S. Cassanelli. 2005. Strategies and perspectives for genetic improvement of wine yeasts. App l . Microbiol . Biotechnol. 66:622-628. Goffeau, A . , B . G . Barrell, H . Bussey, R . W . Davis, B . Dujon, H . Feldmann, F. Galibert, J.D. Hoheisel, C. Jacq, M . Johnston, E.J . Louis, H . W . Mewes, Y . Murakami, P. Philippsen, H . Tettelin, and S.G. Oliver. 1996. Life with 6000 genes. Science 274:546, 563-567. Goniak, O.J., and A . C . Noble. 1987. Sensory study o f selected volatile sulfur-compounds in white wine. A m . J. Enol. Viticult. 38:223-227. Gonzalez-Candelas, L . , A . Cortell, and D . Ramon. 1995. Construction of a recombinant wine yeast strain expressing a fungal pectate lyase gene. F E M S Microbiol . Lett. 126:263-269. Gonzalez-Candelas, L . , J .V. G i l , R . M . Lamuela-Raventos, and D . Ramon. 2000. The use of transgenic yeasts expressing a gene encoding a glycosyl hydrolase as a tool to increase resveratrol content in wine. Int. J. Food. Microbiol . 59:179-183. Grobler, J., F. Bauer, R . E . Subden, and H.J.J, van Vuuren. 1995. The mael gene of Schizosaccharomyces pombe encodes a permease for malate and other C-4 dicarboxylic acids. Yeast 11:1485-1491. Hanahan, D . 1983. Studies on transformation of Escherichia coli with plasmids. J. M o l . B i o l . 166:557-80. 134 Henick-Kling, T. 1993. Malolactic fermentation, p. 289-326, In G . H . Fleet, ed. Wine Microbiology and Biotechnology. Harwood Academic Publishers, Switzerland. Henick-Kling, T. 1995. Control of malolactic fermentation in wine: Energetics, flavor modification and methods of starter culture preparation. J. App l . Bacteriol. 79:S29-S37. Henick-Kling, T., and Y . H . Park. 1994. Considerations for the use of yeast and bacterial starter cultures: SO2 and timing o f inoculation. A m . J. Enol Viticult. 45:464-469. Henricsson, C , M . C . D . J . Ferreira, K . Hedfalk, K . Elbing, C. Larsson, R . M . B i , J. Norbeck, S. Hohmann, and L . Gustafsson. 2005. Engineering of a novel Saccharomyces cerevisiae wine strain with a respiratory phenotype at high external glucose concentrations. App l . Environ. Microbiol . 71:6185-6192. Herskowitz, I. 1988. Life cycle of the budding yeast Saccharomyces cerevisiae. Microbiol . Rev. 52:536-553. Husnik, J.I. 2001. Genetic construction o f malolactic wine yeasts. M . S c , University of Guelph, Guelph. Husnik, J.L, H . Volschenk, J. Bauer, D . Colavizza, Z . Luo, and H.J.J, van Vuuren. 2006. Metabolic engineering of malolactic wine yeast. Metab. Eng. 8:315-23. Jackson, D . , and D . Schuster. 1987. The Production of Grapes and Wine in Cool Climates Nelson, Melbourne. Jackson, R .S . 1994. Wine Science: Principles and Applications. Academic Press, San Diego. 135 Jensen, L .T . , M . Ajua-Alemanji, and V . C . Culotta. 2003. The Saccharomyces cerevisiae high affinity phosphate transporter encoded by PH084 also functions in manganese homeostasis. J. B i o l . Chem. 278:42036-42040. Jobin, M . P . , D . Garmyn, C . Divies, and J. Guzzo. 1999a. The Oenococcus ceni clpX homologue is a heat shock gene preferentially expressed in exponential growth phase. J. Bacteriol. 181:6634-6641. Jobin, M . P . , D . Garmyn, C. Divies, and J. Guzzo. 1999b. Expression of the Oenococcus ceni trxA gene is induced by hydrogen peroxide and heat shock. Microbiology 145:1245-1251. Jobin, M . P . , F. Delmas, D . Garmyn, C. Divies, and J. Guzzo. 1997. Molecular characterization of the gene encoding an 18-kilodalton small heat shock protein associated with the membrane of Leuconostoc oenos. App l . Environ. Microbiol . 63:609-614. Klaenhammer, T., E . Altermann, F. Arigoni , A . Bolotin, F. Breidt, J. Broadbent, R. Cano, S. Chaillou, J. Deutscher, M . Gasson, M . van de Guchte, J. Guzzo, A . Hartke, T. Hawkins, P. Hols, R. Hutkins, M . Kleerebezem, J. Kok , O. Kuipers, M . Lubbers, E . Maguin, L . M c K a y , D . M i l l s , A . Nauta, R. Overbeek, H . Pel, D . Pridmore, M . Saier, D . van Sinderen, A . Sorokin, J. Steele, D . O'Sullivan, W . de Vos, B . Weimer, M . Zagorec, and R. Siezen. 2002. Discovering lactic acid bacteria by genomics. Anton. Leeuw. Int. J. G . 82:29-58. Kourkoutas, Y . , A . Bekatorou, I . M . Banat, R. Marchant, and A . A . Koutinas. 2004. Immobilization technologies and support materials suitable in alcohol beverages production: A review. Food Microbio l . 21:377-397. 136 Kuczynski , J.T., and F. Radler. 1982. The anaerobic metabolism of malate o f Saccharomyces bailii and the partial purification and characterization of malic enzyme. Arch. Microbiol . 131:266-270. Kunkee, R . E . 1967. Malo-lactic fermentation. A d v . App l . Microbiol . 9:235-279. Kunkee, R . E . 1991. Some roles o f malic-acid in the malolactic fermentation in wine-making. F E M S Microbiol . Rev. 88:55-72. Labarre, C , C. Divies, and J. Guzzo. 1996a. Genetic organization of the mle locus and identification of a mleR-like gene from Leuconostoc oenos. App l . Environ. Microbiol . 62:4493-4498. Labarre, C , J. Guzzo, J.F. Cavin, and C. Divies. 1996b. Cloning and characterization of the genes encoding the malolactic enzyme and the malate permease of Leuconostoc oenos. A p p l . Environ. Microbiol . 62:1274-1282. Lafon-Lafourcade, S., A . Lonvaud-Funel, and E . Carre. 1983. Lactic acid bacteria o f wines: Stimulation of growth and malolactic fermentation. Anton. Leeuwen. Int. J. 49:349-352. Landry, C.R. , J.P. Townsend, D . L . Hartl, and D . Cavalieri. 2006. Ecological and evolutionary genomics of Saccharomyces cerevisiae. M o l . Ecol . 15:575-591. Lautensach, A . , and R . E . Subden. 1984. Cloning of malic-acid assimilating activity from Leuconostoc oenos in Escherichia coli. Microbios 39:29-39. Lehtonen, P. 1996. Determination of amines and amino acids in wine: A review. A m . J. Enol . Viticult. 47:127-133. 137 L i l l y , M . , M . G . Lambrechts, and I.S. Pretorius. 2000. Effect of increased yeast alcohol acetyltransferase activity on flavor profiles of wine and distillates. App l . Environ. Microbiol . 66:744-753. L i l l y , M . , F .F . Bauer, M . G . Lambrechts, J .H. Swiegers, D . Cozzolino, and I.S. Pretorius. 2006. The effect of increased yeast alcohol acetyltransferase and esterase activity on the flavour profiles of wine and distillates. Yeast 23:641-659. L i u , S.Q. 2002. Malolactic fermentation in wine: Beyond deacidification. J. App l . Microbiol . 92:589-601. L i u , S.Q., and G.J . Pilone. 1998. A review: Arginine metabolism in wine lactic acid bacteria and its practical significance. J. A p p l . Microbiol . 84:315-327. L i u , S.Q., and G.J . Pilone. 2000. A n overview o f formation and roles of acetaldehyde in winemaking with emphasis on microbiological implications. Int. J. Food Sci . Tech. 35:49-61. L i u , S.Q., C R . Davis, and J.D. Brooks. 1995a. Growth and metabolism of selected lactic-acid bacteria in synthetic wine. A m . J. Enol . Viticult. 46:166-174. L i u , S.Q., G . G . Pritchard, M . J . Hardman, and G.J . Pilone. 1995b. Occurrence of arginine deiminase pathway enzymes in arginine catabolism by wine lactic acid bacteria. A p p l . Environ. Microbiol . 61:310-316. Lodder, J. 1970. The yeasts: A taxonomic study. 2nd ed. North Holland Publishing, Amsterdam. London, J. 1976. The ecology and taxonomic status o f the LactobacUli. Ann . Rev. Microbiol . 30:279-301. 138 Lonvaud, A . 1975. Recherches sur renzyme des bacteries lactiques assurant la transformations du malate en lactate, University of Bordeaux, Bordeaux. Lonvaud-Funel, A . 1995. Microbiology of the malolactic fermentation: Molecular aspects. F E M S Microbiol . Lett. 126:209-214. Lonvaud-Funel, A . 1999. Lactic acid bacteria in the quality improvement and depreciation of wine. Anton. Leeuw. Int. J. G . 76:317-331. Lonvaud-Funel, A . 2001. Biogenic amines in wines: Role of lactic acid bacteria. F E M S Microbiol . Lett. 199:9-13. Lonvaud-Funel, A . , and A . M . S . Desaad. 1982. Purification and properties of a malolactic enzyme from a strain of Leuconostoc-mesenteroides isolated from grapes. A p p l . Environ. Microb. 43:357-361. Lucas, P., J. Landete, M . Coton, E . Coton, and A . Lonvaud-Funel. 2003. The tyrosine decarboxylase operon of Lactobacillus brevis I O E B 9809: Characterization and conservation in tyramine-producing bacteria. F E M S Microbiol . Lett. 229:65-71. Magni, G .E . , and R . C . von Borstel. 1962. Different rates of spontaneous mutation during mitosis and meiosis in yeast. Genetics 46:1097-1108. Magyar, I., and I. Panyik. 1989. Biological deacidfication o f wine with Schizosaccharomycespombe entrapped in Ca-alginate gel. A m . J. Enol . Vi t i c . 40. Maicas, S. 2001. The use of alternative technologies to develop malolactic fermentation in wine. App l . Microbiol . Biotechnol. 56:35-39. Maicas, S., P. Gonzalez-Cabo, S. Ferrer, and I. Pardo. 1999. Production of Oenococcus ceni biomass to induce malolactic fermentation in wine by control of p H and substrate addition. Biotechnol Lett. 21:349-353. 1 Malgoire, J .Y. , S. Bertout, F. Renaud, J . M . Bastide, and A . Mal l ie . 2005. Typing of Saccharomyces cerevisiae clinical strains by using microsatellite sequence polymorphism. J. C l in . Microbiol . 43:1133-1137. Malherbe, D.F . , M . du Toit, R . R . C . Otero, P. van Rensburg, and I.S. Pretorius. 2003. Expression of the Aspergillus niger glucose oxidase gene in Saccharomyces cerevisiae and its potential applications in wine production. App l . Microbiol . Biot. 61:502-511. Manca de Nadra, M . , and A . M . Strasser de Saad. 1995. Polysaccharide production by Pediococcuspentosaceus from wine. Int. J. Food Microbiol . 27:101-106. Manivasaskam, P., S.C. Weber, J. McElver , and R . H . Schiestl. 1995. Micro-homology mediated PCR-targeting in Saccharomyces cerevisiae. Nuc l . Acids Res. 23:2799-2800. Manzanares, P., M . Orejas, J .V. G i l , L . H . de Graaff, J. Visser, and D . Ramon. 2003. Construction of a genetically modified wine yeast strain expressing the Aspergillus aculeatus rhaA gene, encoding an alpha-L-rhamnosidase of enological interest. App l . Environ. Microbiol . 69:7558-7562. Marcobal, A . , B . de las Rivas, M . V . Moreno-Arribas, and R. Munoz. 2004. Identification of the ornithine decarboxylase gene in the putrescine-producer Oenococcus ceni BIFI-83. F E M S Microbiol . Lett. 239:213-220. Marcobal, A . , P J . Martin-Alvarez, M . C . Polo, R. Munoz, and M . V . Moreno-Arribas. 2006. Formation of biogenic amines throughout the industrial manufacture of red wine. J. Food Protect. 69:397-404. 140 Marinoni, G . , M . Manuel, R .F . Petersen, J. Hvidtfeldt, P. Sulo, and J. Piskur. 1999. Horizontal transfer of genetic material among Saccharomyces yeasts. J. Bacteriol. 181:6488-96. Martineau, B . , and T. Henick-Kling. 1995. Formation and degradation of diacetyl in wine during alcoholic fermentation with the Saccharomyces cerevisiae strain EC1118 and malolactic fermentation with Leuconostoc oenos strain M C W . A m . J. Enol. Viticult. 46:442-448. Martineau, B . , T. Henickkling, and T. Acree. 1995a. Reassessment of the influence of malolactic fermentation on the concentration of diacetyl in wines. A m . J. Enol . Viticult. 46:385-388. Martineau, B . , T .E . Acree, and T. Henickkling. 1995b. Effect o f wine type on the detection threshold for diacetyl. Food Res. Int. 28:139-143. Martini , A . 1993. The origin and domestication of the wine yeast Saccharomyces cerevisiae. J. Wine Res. 4:165-176. Mathews, C . K . , and K . E . van Holde. 1990. Biochemistry. Benjamin/Cummings Publ. Co, California. Mayer, K . , and A . Temperli. 1963. The metabolism of L-malate and other compounds by Schizosaccharomycespombe. Arch . Mikrobiol . 46:321-328. Mentel, M . , M . Spirek, D . Jorck-Ramberg, and J. Piskur. 2006. Transfer of genetic material between pathogenic and food-borne yeasts. A p p l . Environ. Microbiol . 72:5122-5125. 141 Messenguy, F. , D . Col in , and J.P. ten Have. 1980. Regulation of compartmentation of amino acid pools in Saccharomyces cerevisiae and its effects on metabolic control. Eur. J. Biochem. 108:439-447. Michnick, S., J .L. Roustan, F . Remize, P. Barre, and S. Dequin. 1997. Modulation of glycerol and ethanol yields during alcoholic fermentation in Saccharomyces cerevisiae strains overexpressed or disrupted for GPD1 encoding glycerol 3-phosphate dehydrogenase. Yeast 13:783-793. M i l l s , D . A . , H . Rawsthorne, C. Parker, D . Tarnir, and K . Makarova. 2005. Genomic analysis of Oenococcus ozni PSU-1 and its relevance to winemaking. F E M S Microbiol . Rev. 29:465-475. Moreno-Arribas, M . V . , and M . C . Polo. 2005. Winemaking biochemistry and microbiology: Current knowledge and future trends. Crit. Rev. Food Sci . 45:265-286. Moreno-Arribas, V . , and A . Lonvaud-Funel. 1999. Tyrosine decarboxylase activity of Lactobacillus brevis IOEB 9809 isolated from wine and L. brevis A T C C 367. F E M S Microbiol . Lett. 180:55-60. Moreno-Arribas, V . , and A . Lonvaud-Funel. 2001. Purification and characterization of tyrosine decarboxylase of Lactobacillus brevis IOEB 9809 isolated from wine. F E M S Microbiol . Lett. 195:103-107. Mortimer, R., and M . Polsinelli . 1999. On the origins o f wine yeast. Res. Microbiol . 150:199-204. Mortimer, R . K . , and J.R. Johnston. 1986. Genealogy of principal strains of the yeast genetic stock center. Genetics 113:35-43. 142 Naouri, P., P. Chagnaud, A . Arnaud, and P. Galzy. 1990. Purification and properties o f a malolactic enzyme from Leuconostoc oenos A T C C 23278. J. Basic Microbiol . 30:577-585. Nathan, H . A . 1961. Induction of malic enzyme and oxaloacetate decarboxylase in three lactic acid bacteria. J. Gen. Micro . 25:415-420. Nevoigt, E . , A . Fassbender, and U . Stahl. 2000. Cells o f the yeast Saccharomyces cerevisiae are transformable by D N A under non-artificial conditions. Yeast 16:1107-1110. Nielsen, J .C., C. Prahl, and A . Lonvaud-Funel. 1996. Malolactic fermentation in wine by ' direct inoculation with freeze-dried Leuconostoc oenos cultures. A m . J. Enol . Viticult . 47:42-48. Olsen, E . B . , J .B. Russell, and T. Henick-Kling. 1991. Electrogenic L-malate transport by Lactobacillusplantarum: A basis for energy derivation from malolactic fermentation. J. Bacteriol. 173:6199-6206. Olson, M . V . 1991. Genome structure and organization in Saccharomyces cerevisiae, p. 1-39, In J. R. Broach, et al., eds. The Molecular and Cellular Biology of the Yeast Saccharomyces, V o l . 1. Co ld Spring Harbor Laboratory Press, N e w York. Olson, M . V . , J .E. Dutchik, M . Y . Graham, G . M . Brodeur, C. Helms, M . Frank, M . Maccol l in , R. Scheinman, and T. Frank. 1986. Random-clone strategy for genomic restriction mapping in yeast. P N A S 83:7826-7830. Osothsilp, C . 1987. Genetic and biochemical studies o f malic acid metabolism in Schizosaccharomyces pombe. Ph.D., University o f Guelph, Guelph. 143 Osothsilp, C , and R . E . Subden. 1986a. Malate transport in Schizosaccharomyces pombe. J.Bacteriol. 168:1439-1443. Osothsilp, C , and R . E . Subden. 1986b. Isolation and characterization of Schizosaccharomyces pombe mutants with defective NAD-dependent malic enzyme. Can. J. Microbiol . 32:481-486. Osterwalder, A . 1924. Schizosaccharomyces liquefaciens n.sp., eine gegen freie schweflige Saure widerstandsfahige Garhefe. Mitt . Geb. Lebensmitt. Hyg . 15:5-28. Ough, C S . 1976. Ethyl carbamte in fermented beverages and foods. J. Agric . Food Chem. 24:323-328. Ough, C S . 1992. Winemaking Basics. Food Products Press, New York. Ough, C S . 1993. Ethyl carbamate in foods and wine. B u l l . Soc. Med. Friends Wine 25:7-8. Paul, H . W . 1996. Science, Vine, and Wine in Modern France. Cambridge University Press, New York. Perez-Gonzalez, J .A. , R. Gonzalez, A . Querol, J. Sendra, and D . Ramon. 1993. Construction of a recombinant wine yeast strain expressing beta-(l,4)-endoglucanase and its use in microvinification processes. App l . Environ. Microb. 59:2801-2806. Perez-Torrado, R., J .V. Gimeno-Alcaniz, and E . Matallana. 2002. Wine yeast strains engineered for glycogen overproduction display enhanced viability under glucose deprivation conditions. App l . Environ. Microbiol . 68:3339-3344. 144 Perrot, L . , M . Charpentier, C. Charpentier, M . Feuillat, and D . Chassagne. 2002. Yeast adapted to wine: Nitrogen compounds released during induced autolysis in a model wine. J. Ind. Microbiol . Biotech. 29:134-139. Petes, T .D. , R . E . Malone, and L . S . Symington. 1991. Recombination in yeast, p. 407-521, In J. R. Broach, et al., eds. The Molecular and Cellular Biology o f the Yeast Saccharomyces, V o l . 1. C S H L Press, Cold Spring Harbor, N Y . Phaff, H.J . , and W . T . Starmer. 1987. Yeasts associated with plants insects and soils, p. 123-179, In A. H . Rose and J. S. Harrison, eds. The Yeasts, V o l . 1. Academic Press, London. Pilone, G.J. , and R . E . Kunkee. 1976. Stimulatory effect of malolactic fermentation on growth-rate of Leuconostoc oenos. Appl . Environ. Microb. 32:405-408. Pilone, G.J. , M . G . Clayton, and R.J . Vanduivenboden. 1991. Characterization of wine lactic acid bacteria: Single broth culture for tests of heterofermentation, mannitol from fructose and ammonia from arginine. A m . J. Enol . Viticult . 42:153-157. Pretorius, I.S. 2000. Tailoring wine yeast for the new millennium: Novel approaches to the ancient art of winemaking. Yeast 16:675-729. Pretorius, I.S., and F.F. Bauer. 2002. Meeting the consumer challenge through genetically customized wine-yeast strains. Trends Biotechnol. 20:426-432. Pretorius, I.S., and P .B . Hoj. 2005. Grape and wine biotechnology: Challenges, opportunities and potential benefits. Aust. J. Grape Wine Res. 11:83-108. Puig, S., A . Querol, E . Barrio, and J.E. Perez-Ortin. 2000. Mitotic recombination and genetic changes in Saccharomyces cerevisiae during wine fermentation. App l . Environ. Microbiol . 66:2057-61. 145 Querol, A . , M . T . Fernandez-Espinar, M . del Olmo, and E . Barrio. 2003. Adaptive evolution of wine yeast. Int. J. Food Microbiol . 86:3-10. Radler, F. 1993. Yeasts-metabolism of organic acids, p. 165-182, In G . H . Fleet, ed. Wine Microbiology and Biotechnology. Harwood Academic Publishers, Switzerland. Ramon, D. , S. Genoves, J .V . G i l , O. Herrero, A . MacCabe, P. Manzanares, E . Matallana, M . Orejas, G . Uber, and S. Valles. 2005. Milestones in wine biotechnology. Minerva Biotecnol .17:33-45. Ramos, A . , J.S. Lolkema, W . N . Konings, and H . Santos. 1995. Enzyme basis for p H regulation of citrate and pyruvate metabolism by Leuconostoc oenos. A p p l . Environ. Microbiol . 61:1303-1310. Rankine, B . C . 1977. Developments in malolactic fermentation of Australian red table wines. A m . J. Enol . Viticult. 28:27-33. Redzepovic, S., S. Orlic, A . Majdak, B . Kozina, H . Volschenk, and M . Vil joen-Bloom. 2003. Differential malic acid degradation by selected strains of Saccharomyces during alcoholic fermentation. Int. J. Food Microbiol . 83:49-61. Reed, G . , and S.L. Chen. 1978. Evaluating commercial active dry wine yeasts by fermentation activity. A m . J. Enol . Viticult. 29:165-168. Reed, G . , and T . W . Nagodawithana. 1988. Technology o f yeast usage in winemaking. A m . J. Enol . Viticult. 39:83-90. Reed, G . , and T .W. Nagodawithana. 1991. Yeast technology. 2nd ed. Van Nostrand Reinhold, New York. 146 Regenberg, B . , S. Holmberg, L . D . Olsen, and M . C . Kielland-Brandt. 1998. Dip5p mediates high-affinity and high-capacity transport of L-glutamate and L-aspartate in Saccharomyces cerevisiae. Curr. Genet. 33:171-7. Regenberg, B . , L . During-Olsen, M . C . Kielland-Brandt, and S. Holmberg. 1999. Substrate specificity and gene expression of the amino-acid permeases in Saccharomyces cerevisiae. Curr Genet 36:317-28. Remize, F., E . Andrieu, and S. Dequin. 2000. Engineering of the pyruvate dehydrogenase bypass in Saccharomyces cerevisiae: Role of the cytosolic M g 2 + and mitochondrial K + acetaldehyde dehydrogenases Ald6p and Ald4p in acetate formation during alcoholic fermentation. App l . Environ. Microbiol . 66:3151-3159. 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. App l . Environ. Microbiol . 65:143-149. Renault, P., C. Gaillardin, and H . Heslot. 1988. Role of malolactic fermentation in lactic acid bacteria. Biochimie 70:375-379. Renault, P., C . Gaillardin, and H . Heslot. 1989. Product of the Lactococcus lactis gene required for malolactic fermentation is homologous to a family of positive regulators. J. Bacteriol. 171:3108-3114. Ribereau-Gayon, P., D . Dubourdieu, B . Doneche, and A . Lonvaud. 2000a. Handbook of Enology: The Microbiology of Wine and Vinifications John Wi ley & Sons, Chichester. 147 Ribereau-Gayon, P., Y . Glories, A . Maujean, and D . Dubourdieu. 2000b. Handbook of Enology: The Chemistry of Wine Stabilization and Treatments John Wiley and Sons, Ltd, Chichester. Robinson, J. 1994. The Oxford Companion to Wine. Oxford University Press, New York . Rodriguez, S.B., and R.J . Thornton. 1990. Factors influencing the utilization of L-malate by yeasts. F E M S Microbiol . Lett. 72:17-22. Rodriguez, S.B., E . Amberg, R .J . Thornton, and M . R . Mclel lan. 1990. Malolactic fermentation in Chardonnay: Growth and sensory effects o f commercial strains o f Leuconostoc oenos. J. App l . Microbiol . 68:139-144. Rodriquez, S.B., and R.J . Thornton. 1989. A malic acid dependent mutant of Schizosaccharomyces malidevorans. Arch . Microbiol . 152:564-566. Rogers, S., R. Wells, and M . Rechsteiner. 1986. Amino acid sequence common to rapidly degraded proteins: The P E S T hypothhesis. Science 234:364-368. Salou, P., P. Loubiere, and A . Pareilleux. 1994. Growth and energetics of Leuconostoc oenos during cometabolism of glucose with citrate or fructose. A p p l . Environ. Microbiol . 60:1459-1466. Sanchez-Torres, P., L . Gonzalez-Candelas, and D . Ramon. 1996. Expression in a wine yeast strain of the Aspergillus niger abfB gene. F E M S Microbiol . Lett. 145:189-194. Santos, M . H . S . 1996. Biogenic amines: Their importance in foods. Int. J. Food Microbiol . 29:213-231. 148 Schoeman, H . , M . A . Vivier , M . D u Toit, L . M . T . Dicks, and I.S. Pretorius. 1999. The development of bactericidal yeast strains by expressing the Pediococcus acidilactici pediocin gene (pedA) in Saccharomyces cerevisiae. Yeast 15:647-656. Schuller, D . , and M . Casal. 2005. The use of genetically modified Saccharomyces cerevisiae strains in the wine industry. A p p l . Microbiol . Biotechnol. 68:292-304. Schutz, M . , and F. Radler. 1973. [The "malic enzyme" from Lactobacillusplantarum and Leuconostoc mesenteroides]. Arch. Mikrobiol . 91:183-202. Schwartz, D . C , and C R . Cantor. 1984. Separation of yeast chromosome-sized D N A s by pulsed field gradient gel-electrophoresis. Ce l l 37:67-75. Sethi, N . , and W . Mandell . 1988. Saccharomyces fungemia in a patient with A I D S . New York State J. Med. 88:278-279. Shalaby, A . R . 1996. Significance of biogenic amines to food safety and human health. Food Res. Int. 29:675-690. Sherman, F. 1997. Yeast genetics, p. 302-325, In R. A . Meyers, ed. The Encyclopedia of Molecular Biology and Molecular Medicine, V o l . 6. V C H Publishing, Weinheim, Germany. Silva, S., F. Ramon-Portugal, P. Andrade, S. Abreu, M . D . Texeira, and P. Strehaiano. 2003. Mal i c acid consumption by dry immobilized cells of Schizosaccharomyces pombe. A m . J. Enol. Vi t i c . 54:50-55. Slavikova, E . , and R. Vadkertiova. 1997. Seasonal occurrence o f yeasts and yeast-like organisms in the river Danube. Anton. Leeuwen. Int. J. G . 72:77-80. 149 Smit, A . , R . R . C . Otero, M . G . Lambrechts, I.S. Pretorius, and P. Van Rensburg. 2003. Enhancing volatile phenol concentrations in wine by expressing various phenolic acid decarboxylase genes in Saccharomyces cerevisiae. J. Agr. Food Chem. 51:4909-4915. Sniegowski, P .D. , P .G . Dombrowski, and E . Fingerman. 2002. Saccharomyces cerevisiae and Saccharomyces paradoxus coexist in a natural woodland site in North America and display different levels of reproductive isolation from European conspecifics. F E M S Yeast Res. 1:299-306. Snow, R. 1983. Genetic improvement of wine yeast, p. 439-459, In J. F. T. Spencer, et al., eds. Yeast Genetics: Fundamental and Applied Aspects. Springer-Verlag, New York . Soufleros, E . , M . L . Barrios, and A . Bertrand. 1998. Correlation between the content of biogenic amines and other wine compounds. A m . J. Enol . Viticult. 49:266-278. Sousa, M . J . , M . Mota, and C. Leao. 1992. Transport o f malic acid in the yeast Schizosaccharomyces pombe: evidence for a proton-dicarboxylate symport. Yeast 8:1025-31. Sousa, M . J . , M . Mota, and C. Leao. 1995. Effects of ethanol and acetic acid on the transport of malic acid and glucose in the yeast Schizosaccharomyces pombe: Implications in wine deacidification. F E M S Microbiol . Lett. 126:197-202. Spettoli, P., M . P . Nuti , and A . Zamorani. 1984. Properties of malolactic activity purified from Leuconostoc oenos M L 3 4 by affinity chromatography. App l . Environ. Microb. 48:900-901. 150 Sponholz, W . R . 1993. Wine spoilage by microorganisms, p. 395-420, In G . H . Fleet, ed. Wine Microbiology and Biotechnology. Harwood Academic Publishers, New York . Strasser de Saad, A . M . , A . A . Pesce de Ruiz Holgado, and G . Oliver. 1984. Purification and properties of malolactic enzyme from Lactobacillus murinus C N R Z 313. J. A p p l . Biochem. 6:374-383. Straub, B . W . , M . Kicherer, S . M . Schilcher, and W.P . Hammes. 1995. The formation of biogenic amines by fermentation organisms. Z . Lebensm. Unters. Forsch. 201:79-82. Subden, R . E . , W . L . L i n , A . Lautensach, and A . G . Meiering. 1982. A n L-lactic acid dehydrogenase based method for detecting microbial colonies performing a malo-lactic fermentation. Can. J. Microbiol . 28:883-886. Subden, R . E . , A . Krizus, C. Osothsilp, M . Viljoen, and H.J.J, van Vuuren. 1998. Mutational analysis of malate pathways in Schizosaccharomyces pombe. Food Res. Int. 31:37-42. Sutherland, C M . , P . A . Henschke, P. Langridge, and M . D . Lopes. 2003. Subunit and cofactor binding of Saccharomyces cerevisiae sulfite reductase: Towards developing wine yeast with lowered ability to produce hydrogen sulfide. Aus. J. Grape Wine Res. 9:186-193. Taillandier, P., and P. Strehaiano. 1991. The role of malic acid in the metabolism of Schizosaccharomyces pombe: Substrate consumption and cell growth. A p p l . Microbio l . Biotechnol. 35:541-543. 15 Taillandier, P., J.P. Riba, and P. Strehaiano. 1988. Malate utilization by Schizosaccharomyces pombe. Biotechnol. Lett. 10:469-472. ten Brink, B . , C . Damink, H . M . L . J . Joosten, and J.H.J.H.I. Tveld. 1990. Occurrence and formation of biologically-active amines in foods. Int. J. Food Microbiol . 11:73-84. Tenreiro, S., P . A . Nunes, C A . Viegas, M . S . Neves, M . C . Teixeira, M . G . Cabral, and I. Sa-Correia. 2002. AQR1 gene (ORF Y N L 0 6 5 w ) encodes a plasma membrane transporter o f the major facilitator superfamily that confers resistance to short-chain monocarboxylic acids and quinidine in Saccharomyces cerevisiae. Biochem. Biophys. Res. Commun. 292:741-748. Tonon, T., J.P. Bourdineaud, and A . Lonvaud-Funel. 2001. The a r cABC gene cluster encoding the arginine deiminase pathway of Oenococcus ceni, and arginine induction o f a CRP- l ike gene. Res. Microbiol . 152:653-661. Torok, T., R . K . Mortimer, P. Romano, G . Suzzi, and M . Polsinelli . 1996. Quest for wine yeasts: A n old story revisited. J. Ind. Microbiol . 17:303-313. Tracey, R.P. , and T.J. Britz. 1989. The effect o f amino-acids on malolactic fermentation by Leuconostoc-oenos. J. App l . Bact. 67:589-595. Valero, E. , D . Schuller, B . Cambon, M . Casal, and S. Dequin. 2005. Dissemination and survival o f commercial wine yeast in the vineyard: A large-scale, three-years study. F E M S Yeast Research 5:959-969. van Vuuren, H.J.J. , and L . M . T . Dicks. 1993. Leuconostoc oenos: A review. A m . J. Enol . Viticult. 44:99-112. 152 Vaughan-Martini, A . , and A . Martini. 1995. Facts, myths and legends on the prime industrial microorganism. J. Ind. Microbio l . 14:514-522. Velasco, I., S. Tenreiro, I .L. Calderon, and B . Andre. 2004. Saccharomyces cerevisiae AQR1 is an internal-membrane transporter involved in excretion of amino acids. Eukaryot Ce l l 3:1492-1503. Vilanova, M . , P. Blanco, S. Cortes, M . Castro, T . G . V i l l a , and C. Sieiro. 2000. Use of a PGU1 recombinant Saccharomyces cerevisiae strain in oenological fermentations. J. App l . Microbio l . 89:876-883. Volschenk, H . , H.J.J , van Vuuren, and M . Vil joen-Bloom. 2003. Malo-ethanolic fermentation in Saccharomyces and Schizosaccharomyces. Curr. Genet. 43:379-391. Volschenk, H . , M . Vil joen-Bloom, R . E . Subden, and H.J.J, van Vuuren. 2001. Malo-ethanolic fermentation in grape must by recombinant strains of Saccharomyces cerevisiae. Yeast 18:963-970. Volschenk, H . , M . Vil joen, J. Grobler, F. Bauer, A . Lonvaud-Funel, M . Denayrolles, R . E . Subden, and H.J.J . Van Vuuren. 1997a. Malolactic fermentation in grape musts by a genetically engineered strain of Saccharomyces cerevisiae. A m . J. Enol . Viticult. 48:193-197. Volschenk, H . , M . Vil joen, J. Grobler, B . Petzold, F. Bauer, R . E . Subden, R A . Young, A . Lonvaud, M . Denayrolles, and H.J.J, van Vuuren. 1997b. Engineering pathways for malate degradation in Saccharomyces cerevisiae. Nat. Biotechnol. 15:253-257. 153 Wach, A . 1996. PCR-synthesis of marker cassettes with long flanking homology regions for gene disruptions in S. cerevisiae. Yeast 12:259-265. Wach, A . , A . Brachat, R. Pohlmann, and P. Philippsen. 1994. New heterologous modules for classical or PCR-based gene disruptions in Saccharomyces cerevisiae. Yeast 10:1793-1808. Wall ing, E . , E . Gindreau, and A . Lonvaud-Funel. 2005. A putative glucan synthase gene dps detected in exopolysaccharide-producing Pediococcus damnosus and Oenococcus ceni strains isolated from wine and cider. Int. J. Food Microbiol . 98:53-62. Wantle, F. , M . Gotz, and R. Jarisch. 1994. The red wine provocation test: Intolerance to histamine as a model for food intolerance. Al lergy Proc. 15:27-32. Wibowo, D . , R. Eschenbruch, C R . Davis, G . H . Fleet, and T . H . Lee. 1985. Occurrence and growth of lactic acid bacteria in wine: A review. A m . J. Enol . Viticult . 36:302-313. Will iams, S., R . A . Hodges, T . L . Strike, R. Snow, and R . E . Kunkee. 1984. Cloning the gene for the malolactic fermentation of wine from Lactobacillus delbruekii in Escherichia coli and yeasts. App l . Microbiol . Biotechnol. 47:288-293. Will iams, S.T., M . E . Sharpe, and J . C Holt. 1989. Bergey's Manual of Systematic Bacteriology. Wil l iams & Wilkins, Baltimore. Zee, J .A. , R . E . Simard, L . Lheureux, and J. Tremblay. 1983. Biogenic-amines in wines. A m . J. Enol. Viticult . 34:6-9. Zhang, D.S. , and R . W . Lovitt. 2005. Studies on growth and metabolism of Oenococcus ceni on sugars and sugar mixtures. J. App l . Microbiol . 99:565-572. 154 Zhang, D.S. , and R . W . Lovitt. 2006. Strategies for enhanced malolactic fermentation wine and cider maturation. J. Chem. Technol. Biot. 81:1130-1140. APPENDIX A Schematic Representation of pUT332 and Probes Used to Confirm the Absence of Antibiotic Markers in ML01 Clal Figure 21. Plasmid map of pUT332 and schematic representation of the probes used in Southern blot experiments to confirm the absence of antibiotic markers in S. cerevisiae M L 0 1 . The bla gene (ampicillin resistance) probe was retrieved as a Clal-Sspl fragment after restriction digest of the pUT332 plasmid. This probe includes an additional 970 bp of plasmid pUT332 which trace back directly to plasmid pUC19 (Gatignol et al., 1990). The 1758 bp plasmid probe comprises 67.1% o f all the pUC19 sequences distributed throughout the sequence of pUT332. The Tn5ble probe was retrieved by P C R amplification of a region of the pUT332 plasmid using primers J20 and J21. 1 A P P E N D I X B Strategy for Sequencing the Malolactic Cassette from the Genome of M L O l Insert For19g For20g For15B For3p For4g For5g For6g For7p t I ) \ I I I 5' U R A 3 f lanking P G K 1 tsi 1K 2K r 3K For8g ForQp FonOg For11p For12g For13p I I I I I I For18g PGK1 promoter -I 1 - l 1 1 1 1 I 6 K Rev2 1g Rev15p Rev13p i { i Rev17B Rev11g Rev1Og Rev20g Rev19g Rev9p RevSgb Rev7p Rev6g Rev4bg Rev20g Rev19 Rev1 2g I Rev5g 11 12 6 • 13 For14p For17B 3' URA3 flanking Rev3fap Rev2p Rev18g 1 { { 10 Figure 22. Strategy used to sequence the malolactic cassette from the genome of S. cerevisiae M L O l . A minimum of two sets of 13 unique PCR templates (lines), spanning the entire malolactic cassette, were used in the sequencing. Primers for sequencing of each strand are indicated as arrows. APPENDIX D Participant Consent Form Title of Study: Sensory evaluation of wine produced by the malolactic yeast MLOl Principal Investigator: Professor Hennie J.J. van Vuuren (Ph.D.) Wine Research Centre UBC (604) 822-0418 Sponsors: South African Wine Industry 1994-1996; Bio-Springer 1997-2002 1. I N T R O D U C T I O N Y o u are being invited to take part in this research study because of your expertise and extensive experience in the sensory evaluation of wine. Y o u w i l l not receive any remuneration to participate in this tasting. 2. Y O U R P A R T I C I P A T I O N IS V O L U N T A R Y Your participation is entirely voluntary, so it is up to you to decide whether or not to take part in this study. If you wish to participate, you w i l l be asked to sign this form. If you do decide to take part in this study, you are still free to withdraw at any time and without giving any reasons for your decision. Please take time to read the following information carefully. 3. W H O IS C O N D U C T I N G T H E S T U D Y ? This study was initiated in the laboratory of Dr. Hennie van Vuuren at the University of Stellenbosch in August 1994. Financial support was received from the South African Wine Industry and B io Springer. John Husnik, Ph.D. student under supervision of Dr. Hennie J.J. van Vuuren, continued with the research in the Wine Research Centre at U B C . B i o Springer sponsored the research at U B C from January I, 2001 - June 2002. 4. B A C K G R O U N D Most red wines and some white wines undergo the bacterial malolactic fermentation that catalyses the bio-conversion of L-malate to L-lactate. Some naturally occurring lactic acid bacterial strains in wine produce undesirable compounds, such as biogenic amines, from amino acids present in grape musts. The presence of bio-amines can be of great concern for consumers since these molecules, particularly histamine, have been shown to be the causative agent of head aches and other allergenic symptoms such as, diarrhea, Page 1 o f 4 160 palpitations, rashes and vomiting. Moreover, strong scientific evidence suggests that other biogenic amines such as cadaverine, putrescine, spermine, and tyramine can potentiate the toxic effect of histamine. During the last two decades new technologies such as metabolic engineering, protein engineering, and novel enzyme and fermentation technologies have been developed that have greatly enhanced our capabilities to produce safer wines of a higher quality. The malolactic yeast M L 0 1 is such an example. Over the last 10 years we have succeeded to genetically enhance an industrial wine yeast strain to perform the alcoholic and malolactic fermentations simultaneously. We have conducted rigorous scientific examination o f the M L O l malolactic yeast and this yeast has now received Generally Regarded as Safe ( G R A S ) status from the F D A . A l l food grade microorganisms are required to have this status before they can be applied in the food or wine industry. The application of the M L O l malolactic yeast w i l l minimize or prevent growth o f lactic acid bacteria capable o f producing allergens and at least ensure a reduction, or elimination, of these allergens from wine. Consumers sensitive to bioamines are unlikely to get headaches when drinking wines produced with the M L O l yeast. 5. W H A T IS T H E P U R P O S E OF T H E S T U D Y ? The purpose of this study is to evaluate and compare the colour, aroma, flavour and overall quality of the wine produced by the malolactic wine yeast, wine produced with the parental yeast, and wine produced with the parental yeast and malolactic bacteria. The yeast is being used for the production of commercial wines in Moldavia and large-scale fermentation trials have been conducted in South Africa and are currently being conducted in the U S A . To the best of our knowledge, wine produced with the malolactic yeast M L O l is safe to consume. 6. W H O C A N P A R T I C I P A T E IN T H E S T U D Y ? A n y experienced wine taster. 7. W H A T D O E S T H E S T U D Y I N V O L V E ? Sensory evaluation of the wines at P A R C in Summerland. Wines w i l l be tasted blind and each panel member w i l l be given a questionnaire to complete. The time required for tasting is approximately 2 hours. 8. A F T E R T H E S T U D Y IS F I N I S H E D M r . Husnik wi l l use the data as part of his Ph.D. thesis and in publications resulting from this study. Every member of the panel w i l l receive a copy of the paper. 9. W I L L M Y T A K I N G P A R T IN THIS S T U D Y B E K E P T C O N F I D E N T I A L ? Your confidentiality w i l l be respected. N o information that discloses your identity w i l l be released or published without your specific consent to the disclosure. Page 2 o f4 161 10. W H O D O I C O N T A C T IF I H A V E Q U E S T I O N S A B O U T T H E S T U D Y D U R I N G M Y P A R T I C I P A T I O N ? If you have any questions or desire further information about this study before or during participation, you can contact Dr. Hennie J.J. van Vuuren at (604) 822-0418. I f you have any concerns about your rights as a research subject and/or your experiences while participating in this study, contact the 'Research Subject Information Line in the University o f British Columbia Office o f Research Services' at (604) 822-8598. Thanks for your willingness to participate in this study. Sincerely yours Hennie J.J. van Vuuren (Ph.D.) Professor and Eagles Chair Director September 22, 2003 Page 3 of4 162 CONSENT I understand that my participation in this study is entirely voluntary and that I may refuse to participate or withdraw from this study at any time. To the best of our knowledge, wine produced with the malolactic yeast MLOl is safe to consume and there are no foreseeable risks. There are no direct benefits to subjects from participating in this tasting. I have received a copy of this consent form for my own records. I consent to participate in this study. Subject signature Printed name Date Signature o f a witness Printed name Date Signature o f PI Printed name Date Version 2 November 18, 2003 Page 4 of 4 163 APPENDIX E PCR Confirmation of the Screening Method CD CM CD CO CO LO 0 CD CD c c c C o o o o O O O O 4072 bp 3054 bp 2036 bp 1636 bp 1018 bp 506, 517 bp 396 bp Figure 23 . PCR confirmation of the screening method used to detect integration of the malolactic cassette into the URA3 locus of S92. The 1095 bp PCR product was visualized by 0.8% agarose gel electrophoresis and ethidium bromide staining. PCR primers specific to PGKJp and outside the malolactic cassette (3'-end) were used. 164 APPENDIX F Degradation of Malate and Consumption of Glucose and Fructose by Malolactic Wine Yeast Clones Table 15. Degradation of malate (g/L) by individual colonies o f malolactic clones inoculated into synthetic must containing 4.5 g/L of malate. Malolactic clone Malolactic clone Malolactic clone Malolactic clone S92 S92 + S92 + Day 1 3 4 5 YCplac33- pJH13 1 2 3 1 2 3 1 2 3 1 2 3 K a n M X 0 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 4.38 1 2.38 2.27 3.26 3.08 3.29 3.30 3.94 4.08 4.05 4.02 3.67 4.34 4.29 4.35 3.94 2 0.28 0.25 0.30 0.91 0.36 0.33 0.88 1.20 1.11 1.88 2.50 4.08 4.31 4.19 1.39 3 0.07 0.07 0.07 0.11 0.07 0.07 0.08 0.08 0.17 0.16 0.50 nd 3.84 4.04 0.15 4 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.05 0.08 0.06 0.07 nd 3.89 3.88 0.05 5 0.06 0.06 0.06 0.06 0.05 0.05 0.04 0.04 0.05 0.05 0.06 nd 3.90 3.95 0.04 6 0.05 0.04 0.05 0.05 0.04 0.04 0.03 0.02 0.04 0.04 0.05 nd 3.91 3.61 0.03 7 0.04 0.04 0.04 0.04 0.05 0.04 0.02 0.02 0.03 0.04 0.04 nd 3.31 3.33 0.03 14 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 0.02 0.02 0.03 3.48 2.98 3.21 0.01 Table 16. Consumption o f glucose (g/L) by individual colonies o f malolactic clones inoculated into synthetic must containing 100 g/L of glucose (and 100 g/L of fructose). Day Malolactic clone 1 Malolactic clone 3 Malolactic clone 4 Malolactic clone 5 S92 S92 + YCplac33-K a n M X S92 + pJH13 1 2 3 1 2 3 1 2 3 1 2 3 0 107 107 107 107 107 107 107 107 107 107 107 107 107 107 107 7 10.0 5.18 9.18 20.5 11.2 8.04 1.19 .704 1.27 5.46 10.2 19.6 3.22 6.31 1.46 14 0.13 0.08 0.10 0.21 0.10 0.10 0.05 0.06 0.06 0.07 0.17 0.46 0.05 0.05 0.04 Table 17. Consumption of fructose (g/L) by individual colonies of malolactic clones inoculated into synthetic must containing 100 g/L o f fructose (and 100 g/L of glucose). Day Malolactic clone 1 Malolactic clone 3 Malolactic clone 4 Malolactic clone 5 S92 S92 + YCplac33-K a n M X S92 + pJH13 1 2 3 1 2 3 1 2 3 1 2 3 0 105 105 105 105 105 105 105 105 105 105 105 105 105 105 105 7 23.9 30.6 29.5 27.5 21.8 20.5 12.7 10.3 12.1 20.2 33.0 43.4 22.9 26.8 12.3 14 3.07 1.52 2.60 4.34 2.22 2.10 0.57 1.24 0.93 1.82 3.41 16.5 1.58 1.76 0.71 165 A P P E N D I X G Confi rmat ion of the Identity of the Parental Strain by P C R Ampli f ica t ion of the d Elements of the T y l Retrotransposon CD s_ CD E O CN > —I O) > ^ CO ^ Figure 24. Genetic patterns of the M L O l and the S92 yeast strains based on amplification of genomic D N A regions in between d elements of the T y l retrotransposon. A detailed method of the P C R amplification procedure is given by Ness et al., 1993. 166 A P P E N D I X H Southern Blo t Conf i rming Integration of the Malolact ic Cassette into the URA3 Locus of S92 (PGK1 promoter probe) S92 ML01 6.3Kbp 6.0Kbp < 5.1 Kbp PGK1 6.3Kbp Malolac 5.1 Kbp cassette 6.0Kbp URA3 No hybridization Figure 25. Integration of the malolactic cassette into the URA3 locus of S92 was confirmed by Southern blot analysis using a PGK1 promoter probe. The schematic representation illustrates the EcoRV restriction sites with vertical lines and the PGK1 promoter probe hybridization site is depicted as hatched boxes in the lower panel. 167 APPENDIX I Ascospore formation by the M L O l and S92 strains Sporulation was done according to standard protocols (Ausubel et al., 1995). The only modification was that cells were plated on S A A medium (6.5 g/ sodium acetate and Pastagar B 15 g/L) and incubated for 72 h at ambient temperature. Table 18. Comparison of-the M L O l and S92 strains capability to sporulate. Following spore formation on S A A medium, the percentage of ascospores (containing 3 or 4 spores) was determined by microscopic evaluation. M L O l S92 Total counted 782 612 Percentage of triads and tetrads formed 11% 8% per total number of cells Analysis of the M L O l spores revealed that cells from two spores were auxotrophic for uracil but positive for M L F and cells from the remaining two ascospores were prototrophic for uracil and negative for M L F . 168 APPENDIX J Detailed Description of the DNA Sequences that Comprise the Malolactic Cassette Table 19. The source and description of the malolactic cassette DNA sequences. Nucleotide position Designation Reference for cloning details Source 1-4 Srfl half cloning site This study Saccharomyces cerevisiae GC210 1-928 URA3 sequence This study Saccharomyces cerevisiae GC210 1-508 5' non coding sequence 509-928 Part of open reading frame 929-934 Kpn\ cloning site This study Synthetic 935-1198 PGK1 terminator Crousefa/., 1995 Saccharomyces cerevisiae AB972 935-940 Hindlll site 941-1194 Rest oiPGKl terminator 1195-1198 Remainder of Clai site 1199-1218 Part of linker used in cloning strategy Crous et al., 1995 Synthetic 1199 C residue from linker 1200-1204 Remainder of BgUl cloning site 1205 C residue 1206-1211 Xhol cloning site 1212 G residue 1213-1218 BgUl cloning site 1219-2818 mael gene Grobler et al., 1995 Volschenk et al., 1997a and 1997b Schizosaccharomyces pombe 972 IT (leul-32) 1219-1224 Ndel site 1225-1456 3' non coding region 1457-1459 STOP codon 1460-2770 Coding region 2771-2773 START codon 169 Nucleotide Designation position 2774-2812 5' non coding region 2813-2818 Ball site Reference for cloning details Source 2819-2829 Part of linker used in cloning strategy 2819-2824 EcoRl cloning site 2825 C residue 2826-2829 remainder of BgUl cloning site Crous et al., 1995 Synthetic 2830-4316 PGK1 promoter Crous et al., 1995 Saccharomyces cerevisiae AB972 2830-4310 Part of PGKlp 4311 -4316 Hindlll cloning site 4317-4322 Kpnl cloning site This study Synthetic 4323-4330 NotI cloning site This study Synthetic 4331-4594 PGK1 terminator Crous et al., 1995 Saccharomyces cerevisiae AB972 4331-4336 Hindlll site 4337-4590 Res to fPGKl t 4591-4594 Remainder of Clal site 4595-4614 Part of linker used in cloning strategy 4595 C residue 4596-4600 Remainder of Bglll cloning site 4601 C residue 4602-4607 Xhol cloning site 4608 G residue 4609-4614 BgUl cloning site Crous et al., 1995 Synthetic 4615-4616 CA residues left from oligonucleotide used for mleA amplification Volschenk, unpublished Synthetic 4617-6242 mleA gene 4617-4619 STOP codon Volschenk, unpublished (Enococcus ceni Lo 8413 170 Nucleotide position 4620-6239 Designation mleA open reading frame Reference for cloning details Source 6240-6242 START codon 6243-6253 Part of linker used in cloning strategy Crous et al., 1995 Synthetic 6243-6248 EcoRl cloning site 6249 C residue 6250-6253 Remainder of BgUl cloning site 6254-7740 PGK1 promoter Crous et al., 1995 Saccharomyces cerevisiae AB972 6254-7734 Part of PGKlp 7735-7740 Hindlll cloning site 7741-7748 NotI cloning site This study Synthetic 7749-8683 URA3 sequence This study Saccharomyces cerevisiae GC210 7749-8132 Part of open reading frame 8133-8683 3' non coding region 8680-8683 SrfL half cloning site This study Saccharomyces cerevisiae GC210 171 APPENDIX K Discrepancies between the Integrated Malolactic Cassette and Published Sequences Table 20. Discrepancies found when comparing the sequence of the integrated malolactic cassette and previously published sequences. Nucleotide position Description 821 Difference in the 5' region of the URA3 open reading frame 929-934 Additional sequence resulting from cloning strategy 1199-1218 Additional sequence resulting from cloning strategy 2819-2829 Additional sequence resulting from cloning strategy 2896 Difference in the PGKlp sequence 4298 Difference in the PGKlp sequence 4317-4330 Additional sequence resulting from cloning strategy 4595-4614 Additional sequence resulting from cloning strategy 4629 Difference in the mleA open reading frame. This difference corresponds to a change of amino acids from aspartic acid (in the published sequence) to glutamic acid (in the 5247 Difference in the mleA open reading frame. This difference corresponds to no change of the amino acid sequence 6243-6253 Additional sequence resulting from cloning strategy 7741-7748 Additional sequence resulting from cloning strategy 7751 Difference in the 3' region of the URA3 open reading frame 8234 Difference in the 3' region of the URA3 non coding sequence 8543 Difference in the 3' region of the URA3 non coding sequence 172 APPENDIX L Confirmation of DNA Mieroarray Data by Real-Time Reverse Transcription PCR Table 21. Comparison of fold change data for ten genes as determined by DNA mieroarray and by real-time reverse transcription P C R . Gene Symbol Fold change by semi-quantitative R T - P C R Fold change by DNA mieroarray 48 h 48 h DIPS 4.92 2.81 PH084 3.60 2.18 SUE1 -3.15 -5.13 PRR2 -1.11 -3.44 CTT1 -2.99 -3.29 PUT4 -6.33 -3.13 YPC1 -1.74 -2.18 144 h 144 h ENA2 Below threshold 5.27 AQR1 -2.74 -3.23 Y M L 0 8 9 C -3.14 -3.14 173 APPENDIX M Transcripts of mael, mleA and URA3 in the M L O l yeast mael mleA URA3 CN cn » 0 0 — -I f — -783 bp mael -269 bp ACT! 4 8 h 1 4 4 h 4 8 h 1 4 4 h CM 0 0 CM O oo 2 807 bp m/&4 269 bp ACT! E -499 bp 3 -269bp^C7-7 4 8 h 1 4 4 h Figure 26. The presence of the mael, mleA and transcripts in the MLOl yeast during fermentation. Reverse transcriptase PCR was conducted on total RNA extracted from cells harvested from fermentations at 48 and 144 hours. The mael and mleA transcripts are absent in S92 at both time points. 174 APPENDIX N Vinification Trials with MLOl and S92 in Chardonnay Must (2004 harvest) A 6 i Time (days) Figure 27. Malate degradation and lactate and ethanol production by M L O l and S92 in Chardonnay must (from fruit harvested in 2004). (A) Efficient conversion of L-malate (?) to L-lactate ( A ) during alcoholic fermentation by M L O l ; no significant degradation of L-malate (? ) or production of L-lactate (? ) was observed for S92. (B) Production of ethanol by M L O l (?) and S92 (| ); introduction of the malolactic cassette did not affect ethanol formation. 175 APPENDIX O Physicochemical Characteristics of Chardonnay Wines (2004 Harvest) Produced by MLOl, S92 and S92 plus O. oeni Table 22. Physicochemical and colour measurements3 o f Chardonnay wines (2004 harvest) produced by M L O l , S92 and S92 plus O. ceni. MLOl S92 S92 + O. ceni Pb Titratable acidity (g/L) 6.1 a c 7.8 b 6.3a ** Acetate (g/L) 0.25 a 0.328 b 0.424 c *** p H 3.68 a 3.52 b 3.78 c *** Colour measurements L (degree of lightness) 97.86 a 97.96 b 96.55 a *** a (greenness) -1.85 a -1.73 b -2.39 c *** b (yellowness) 9.76 a 9.47 b 13.95 c *** ^420nm + ^520nm 0.198 a 0.179 a 0.255 b ** "The mean values for biological replicates are given for all quantities (n=3) b*'**, ***, ns: significant atp< 0.05, 0.01, 0.001, or not significant cMeans separated atp < 0.05 by Duncan's post-hoc test 176 A P P E N D I X P Correla t ion M a t r i x Table 23. Correlation matrix obtained from the sensory analysis of Chardonnay wines. Overall Quality Yellow Colour Fruity Taste Body Sweetness Acidity Overall Quality 1.00000 Yellow Colour -0.69810 1.00000 Fruity Taste 0.93026 -0.38672 1.00000 Body 0.99958 -0.71850 0.91927 1.00000 Sweetness 0.97510 -0.52192 0.98847 0.96828 1.00000 Acidity -0.99459 0.76872 -0.88710 -0.99718 -0.94677 1.00000 177 APPENDIX Q Health Canada Approval to use MLOl for the Commercial Production of Wine Canada ••• '• ¥'«niiey's Pasture - • -'•Po^lLw&for G7OTA*-• •' OTTAWA, Onteio, ' • -''••KiA-'Oli'' " ' • « L iftW;''" ' ' Dr. Ilcnric JJ . van Voiarcn ' FactrtiV of Afric«il«Ktt Sciences' Ifni versity -of Bf i:t»h. Columbia Suite 23i)f 2:205 E - M a i l " ' ; • " ; V6T1Z4 •. ..Dear DT . van Vuuren: • , ' .' • , Thisletier Defers toilve Nm'el;FaQd'S«bmissiei» wnccrriing WmsY^mt' Mt-i)l;'SB43 its use .its wiaemaMng in' Canada.- Officers of the MMrth'iybdacls-iu*d-:Foe^ l. 8ea«eh;htorr reviewed the iiifbniiaiiopyou pro^ id>stlfi>f ascssnKtit of the acwpfebility of tire WSf Yeast "•ML-OI and ofti»"wfn«s derived frond it far tasmanfood use fii CsBa&ti According to the subrojtt&i information,. Ms: yeast .'swain catties two Movel §erKS which peranits the yeast to coasiuet malolactic fenpehtelioti. Based m ©a«*atoaio« of;!%sulwiiiled:'data.; we haw no oi^cctiori to ftsfbod use of the Wine Yeast ML-01 i i Ciisaija,: . / • Ii should 6ie noted that iMs-opinion is solely with respect to the suitability for sifts as' food of prodnefe <teriw»*.#«BS Winci'mi ML-Cil. /2 V 3 f " tl)t ! and F i± B atl APPENDIX R US Food and Drug Administration GRAS Notice ( i M I M H i H H ». .1 I ] V A N P \ • ! I I I l> N l M i l U I I N FDA Home Page I C F S A N Home I Search/Subject Index [ Q & A I Help CFSAN/Office of Food Additive Safety June 30, 2003 Agency Response Letter GRAS Notice No. GRN 000120 Robert Biwersi Lesaffre Yeast Corporation 433 East Michigan Street Milwaukee, W I 53202 Re: G R A S Notice No. G R N 000120 Dear M r . Biwersi : The Food and Drug Administration (FDA) is responding to the notice, dated January 2, 2003, that you submitted in accordance with the agency's proposed regulation, proposed 21 C F R 170.36 (62 F R 18938; Apr i l 17, 1997; Substances Generally Recognized as Safe ( G R A S ) ; the G R A S proposal). F D A received the notice on January 7, 2003, filed it on January 10, 2003, and designated it as G R A S Notice No. G R N 000120. The subject of the notice is Saccharomyces cerevisiae strain M L O l (S. cerevisiae strain M L O l ) carrying a gene encoding malolactic enzyme from Oenococcus ceni and a gene encoding malate permease from Schizosaccharomyces pombe. The notice informs F D A of the view of Lesaffre Yeast Corporation (Lesaffre) that S. cerevisiae strain M L O l is G R A S , through scientific procedures, for use in winemaking as a yeast starter culture for grape must fermentation. Lesaffre recommends using between 0.1 to 0.2 grams of active dry yeast per liter of wine. 180 Lesaffre describes generally available information about traditional manufacturing processes for the production of wine from grapes. These processes include the harvesting, de-stemming and crushing of grapes (resulting in must), the separation of the juice from the skins and seeds, one or more distinct types o f microbial fermentation, clarification, stabilization, and bottling. Winemakers may vary the sequence of operational steps or modify procedures, depending upon the desired characteristics and nature of the wine. Lesaffre describes generally available information about two distinct fermentation processes (i.e., alcoholic fermentation and malolactic fermentation) that occur either through the action of microorganisms that already are present on the grapes or through the action of microorganisms that are specifically added by the winemaker. Alcohol ic fermentation (i.e., a process whereby the sugars glucose and fructose are converted to ethanol) is mediated by metabolic pathways associated with yeast (usually S. cerevisiae or closely related species). Malolactic fermentation (i.e., a process whereby the dicarboxylic acid malic acid is decarboxylated to the monocarboxylic acid lactic acid) is mediated by lactic acid bacteria through the combined action of a protein (called malate permease) that transports malic acid from the wine into the bacteria and an enzyme (called malolactic enzyme) that converts the malic acid to lactic acid. Because malolactic fermentation reduces the number o f carboxylic acid groups on organic acids present in the wine, it reduces the acidity of the must. Although alcoholic fermentation is an inherent process associated with all winemaking, malolactic fermentation is a secondary process that may or may not be induced by the winemaker, depending on the desired characteristics and nature of the wine. Lesaffre describes generally available information about clarification of wine, which can occur either at the end of the alcoholic fermentation or after the wine has been kept on the lees (the sediment formed by spent yeast cells and grape particulate matter). Wine clarification encompasses the removal of solid particles in the wine via gravity or centrifugation and subsequent elimination of the sediment or pellet. When clarification occurs at the end o f fermentation, the clarification process removes most yeast cells. When the wine is kept on the lees before clarification, the yeast cells undergo autolysis, which releases cellular material that ultimately is degraded through the action of enzymes such as proteases. Lesaffre describes generally available information about stabilization processes, which differ depending on whether the wine is a white wine or a red wine. For white wine, stabilization involves removing proteins via filtration with bentonite. For red wine, stabilization involves adding gelatin or egg white albumin to precipitate colloidal structures that include tannin-protein complexes. Prior to bottling, most wines undergo filtration (e.g., with diatomaceous earth, cellulose filters, or membrane filters) that eliminates any remaining yeast cells. Lesaffre describes published articles about bioengineered strains of S. cerevisiae, including strains of S. cerevisiae that have been modified to conduct malolactic fermentation. Lesaffre notes that the use of bioengineered strains that can conduct both 181 alcoholic and malolactic fermentation eliminates the need for separate additions of two distinct microorganisms (i.e., yeast and lactic acid bacteria). Lesaffre describes the development o f its own bioengineered strain of S. cerevisiae. The host strain, S. cerevisiae strain S92, was isolated from the Champagne region in France and is closely related or identical to commercial strains commonly used in winemaking. The microbial source of malate permease (i.e., Schizosaccharomyces pombe), is a yeast^ that was first isolated from African beer and has frequently been found in sugar-containing products in tropical and sub-tropical regions and in grape must and cider in moderate climates. The microbial source of malolactic enzyme (i.e., Oenococcus oeni) is a lactic acid bacterium that has been isolated from wines and related habitats such as wineries and vineyards. It is the preferred, and most commonly used, lactic acid bacterium for malolactic fermentation o f wines. The malate permease is a 49 k D a protein with a hydrophobicity profile typical o f membrane transport proteins. It contains a peptide sequence (composed of proline, glutamic acid, serine and threonine) that characterizes proteins with a rapid turnover. The malolactic enzyme is a dimer with a total molecular weight of approximately 130 kDa. Lesaffre describes the construction of an integration cassette that contains genes encoding malate permease from S. pombe and the malolactic enzyme from O. ceni, regulatory sequences associated with the expression of these genes, and sequences used for integration into an appropriate chromosomal site in S. cerevisiae strain S92. Lesaffre also describes the transformation strategy that it used to reduce the numbers of potentially transformed yeasts that needed to be screened for the successful integration of the integration cassette. This strategy involved co-transformation of S. cerevisiae strain S92 with a plasmid (pUT322) that carries a selectable marker conferring resistance to the antibiotic phleomycin and was based on the hypothesis that cells transformed with ** plasmid pUT322 are more likely to also have been transformed with the integration cassette. Using this strategy, Lesaffre first screened transformed yeast for resistance to phleomycin and then screened the selected phleomycin-resistant yeast for the ability to produce lactic acid. Lesaffre obtained a phleomycin-sensitive isolate and confirmed that it is free of plasmid pUT332 sequences. Lesaffre designated this strain as M L O l . Based on D N A analysis, Lesaffre concluded that the chromosomal patterns o f S. cerevisiae strains S92 and M L O l are the same except for the presence of the integration cassette. Lesaffre found that the integration cassette remained stably incorporated after 100 generations. Lesaffre also found that S. cerevisiae strain M L O l functions as intended in that it efficiently degrades malic acid. Based on studies that evaluated yeast physiology under different culture conditions, Lesaffre concluded that S. cerevisiae strain M L O l has the same growth kinetics, fermentation rate, and ethanol yield as S. cerevisiae strain S92 under winemaking conditions and that uptake and utilization of malic acid did not confer a growth advantage to S. cerevisiae strain M L O l . Lesaffre describes the method for routine production of S. cerevisiae strain M L O l and notes that this method is based on well-established procedures for the production of active dry yeast. The yeast is grown primarily under aerobic conditions to promote yeast 182 propagation rather than alcohol production. The yeast is harvested via centrifugation and is subsequently dewatered with a rotary vacuum filter, processed through an extruder, and dried, resulting in active dry yeast. The yeast is packaged in vacuum foil pouches prior to shipping. Lesaffre discusses potential dietary intake o f S. cerevisiae strain M L O l and of the proteins that Lesaffre has introduced into that strain. Lesaffre considers that exposure to the yeast itself or to the newly introduced proteins would be negligible because the processing procedures used in winemaking remove intact yeast cells, debris associated with autolyzed yeast cells, and proteins released during autolysis of yeast cells. Based on the information provided by Lesaffre, as well as other information available to F D A , the agency has no questions at this time regarding Lesaffre's conclusion that Saccharomyces cerevisiae strain M L O l is G R A S under the intended conditions of use. The agency has not, however, made its own determination regarding the G R A S status of the subject use of S. cerevisiae strain M L O l . A s always, it is the continuing responsibility of Lesaffre to ensure that food ingredients that the firm markets are safe, and are otherwise in compliance with all applicable legal and regulatory requirements. In accordance with proposed 21 C F R 170.36(f), a copy of the text of this letter, as well as a copy of the information in the notice that conforms to the information in proposed 21 C F R 170.36(c)(1), is available for public review and copying on the homepage of the Office of Food Additive Safety (on the Internet at http://www.cfsan.fda.gov/~lrd/foodadd.html). Sincerely, /s/ Laura M . Tarantino, Ph.D. Act ing Director Office o f Food Additive Safety Center for Food Safety and Applied Nutrition ( 1 'Al though malolactic fermentation is usually mediated by lactic acid bacteria, Lesaffre chose a yeast (rather than a lactic acid bacterium) as a source of the permease, because the permease must function in the membrane of the yeast S. cerevisiae. Food Ingredients and Packaging I Summary of all G R A S Notices C F S A N Home | C F S A N Search/Subject Index | C F S A N Disclaimers & Privacy Policy | C F S A N Accessibility/Help FDA Home Page | Search FDA Site | FDA A-Z Index I Contact FDA FDA/Center for Food Safety & Applied Nutrition Hypertext updated by rxm/pmg August 17, 2004 183 

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.831.1-0100406/manifest

Comment

Related Items