Nutrient Requirements for Growth of the Sherry Flor Yeast, Saccharomyces beticus

Nutrient Requirements for of the Sherry Flor Yeast, Saccharomyces beticus JOHN G. B. CASTOR AND T. E. ARCHER Department of Viticulture and Enology, University of California, Davis, California Several species of the genus Saccharomyces form films on nutrient media. One species, Saccharomyces beticus of Marcilla et al. (1936), occurs on certain types of Spanish wine (flor sherry) during aging of the wine under an air space in casks. The species apparently occurs in the natural flora of the grapes and participates in the alcoholic fermentation of the must pressed from them. Subsequent to fermentation this species, with several others, forms a film (flor) on the wine. When the species S. beticus predominates, the biochemical activities of the flor are considered to contribute to development of the characteristic flavor of flor sherry (Cruess, 1948; Fornachon, 1953). Strains of this species have been isolated and studied by Marcilla et al. (1936), Schanderl (1936, 1950), Niehaus (1937), Cruess (1948) and Fornachon (1953). A strain of S. beticus (probably variety -y of Marcilla et al., 1936) has been designated as S. fermentati by Lodder and Kreger-Van Rij (1952). Inoculated into grape must, this yeast produces a vigorous alcoholic fermentation, typical of industrial strains of Saccharomyces. Provided with sufficient sugar it produces 16 to 18 per cent by volume ethanol. Flor formation follows decrease of fermentable sugar to a low level. The outstanding biochemical activities of the flor on wine are indicated in table 1 (Fornachon, 1953). Occasionally the flor fails to grow well or at all on wines, even when favorable physio-chemical conditions are carefully maintained (Fornachon, 1953; Crowther and Truscott, 1955). Such conditions are air space over TABLE 1. Changes in chemical composition of wine caused by for development (Abridged from Fornachon, 1953) Received for publication September 10, 1956 Original Wine 12 Months Under Flor Specific gravity, 20 C... 0.9895 0.9880 Ethanol, per cent by volume... 14.5 14.1 Glycerol, g/l.......1 4.6 Titratable acid, g/l (as tartaric).6.2 5.75 Volatile acid, g/l (as acetic).0.48 0.24 Malic acid, g/l.0.98 0.94 Lactic acid, g/l.0.78 0.47 ph.3.23 3.21 Fermentable sugar g/l (as glucose)....79 0 0.10 Total aldehyde, mg/l... 37 444 Acetoin.. + 56 the wvine, suitable ethanol concentration, traces of fermentable sugar, optimum ph and temperature relationships, freedom from agitation and protection against competing film-forming yeasts and bacteria. Failure of flor growth under such conditions might be considered to be caused by an unfavorable nutritional situation. The nutrition of the flor with respect to nitrogen has been studied in some detail by Cruess (1948) and Fornachon (1953). The quality but not necessarily the quantity of nitrogen is a factor. A wvine with a low total nitrogen may support good flor growth, while a wine with high total nitrogen may not. Addition of ammonia nitrogen may or may not favorably affect flor growth, depending on ph and temperature relationships. Addition of yeast extract definitely favors flor development on both wine and artificial media. The effect of yeast extract introduces the question of accessory growth factors, in addition to the subject of nitrogen nutrition. Freiberg and Cruess (1955) have reported the effect of a variety of compounds, including organic nitrogen, amino acids, B-complex vitamins, and ascorbic acid, some of which were found stimulatory on the growth of flor on wine. Meager growth of a film yeast, isolated from Spanish sherry, was obtained by Neilsen (1952) on the artificial medium of Olson and Johnson (1949) without mineral supplement. The medium contained biotin, calcium pantothenate, inositol, thiamin, and pyridoxine. The present work was undertaken in view of the favorable effect of yeast extract on flor growth and the failure of earlier studies to critically define growth factors essential for flor development. It was considered desirable to work with a chemically defined medium made up to simulate the chemical composition of wine. This paper presents results obtained with such a defined, not strictly synthetic, medium. The flor development observed on various modifications of the medium indicates that pantothenate is an essential factor for initiation of growth, and that,b-alanine can replace pantothenate. Definition of the exact nitrogen requirements of the flor yeast, however, awaits further study. MATERIALS AND METHODS Strains of yeast used in earlier flor studies have been carefully and adequately described by Marcilla et al.

1957] 1NUTRIENT REQUIREMENTS FOR S. BETICUS GROWTH (1936) and Fornachon (1953). The strain of yeast used in this study (Enology no. 519) was isolated from Spanish wine by Schanderl (1936). It was selected for this work, on the basis of its fermentative, and film-forming activity, from a collection of film-forming fermentative yeasts. The strain conforms to the description of the true flor yeast (Fornachon, 1953), except that it has not formed spores on vegetable wedges, Gorodkowa agar, or plaster blocks. It is, however, differentiated from strains of the genera Torulopsis, Hansenula, and Pichia already studied (Castor and Archer, unpublished data) by its high ethanol production (14 g ethanol per 100 ml in 30 per cent sugar), film formation on media containing 15 per cent by volume ethanol, the strong aldehyde production of the film stage, low ethyl acetate production, and maintenance of vigorous growth for many years in stock culture. The strain has been carried as a stock culture on malt agar slants since 1939, and on the vegetable juice agar of Wickerham et al. (1946) since 1946, without notable change in ethanol production and film formation. The flor stage was obtained for experiments by permitting the yeast to completely ferment juice of the Mission grape (a variety of Vitis vinifera). For the first stage the juice, clarified by storage at about 0 C, was diluted 1: 1 with distilled water, nutritionally fortified with 0.1 per cent yeast extract (Difco) and autoclaved. Within a week after cessation of fermentation the flor appeared and was transferred to wine. During the experiments described, the flor was cultivated serially on wine cultures, which were used as a source of inoculum. The wine (from the Furmint grape, a variety of Vitis vinifera) at its natural 11 to 12 per cent by volume ethanol or fortified to 15 per cent as required, was sterilized by filtration through a microporous porcelain candle. Serial inoculum cultures were renewed at weekly intervals. Preparation of the flor from wine for use as inoculum was accomplished by washing on sterile M/15 KH2PO4 solution containing 10 to 15 per cent by volume ethanol. The flor was transferred from wine to washing solution on a 1-cm loop with aseptic precautions. It was left to float on the wash solution in a flat, thin film for 6 hr, in order to allow nutrients carried over from the wine to diffuse away from the cells. Phosphate solution containing ethanol was found necessary for washing. When distilled water alone was used, or ethanol was omitted, the flor sometimes failed to develop after the washed inoculum was placed again on wine. The composition of the artificial medium used fqr flor growth experiments is shown in table 2. The organic components used in the basal medium are known to be present in wine (Amerine, 1954). Inorganic components were those usually found in media used for investigation of yeast growth requirements (Williams et al., 1941). The basal medium was made up at twice the final concentration desired, sterilized by filtration and distributed as required in small, sterile Erlenmeyer flasks, capped with aluminum foil. Amino acids and B-complex vitamins were employed on the basis of recent studies on wine. The occurrence in musts and wines of 24 amino acids has been reported by Castor (1953a), Castor and Archer (1956), Dimotakis (1956), Luthi (1953), Procopio and CalR (1954), Sisakian and Bezinger (1953), Tarantola (1954), and Valaize and Dupont (1951). A quantitative study of the fate of amino acids during fermentation was made by Castor (1953a) and Castor and Archer (1956). The amounts of 10 B-complex vitamins or growth factors in musts and wines were investigated by Morgan et al. (1939), Hall et al. (1956), and Castor (1953b). The latter reported changes in eight growth factors during fermentation of grape musts. Amino acids and B-complex vitamins were added to the basal medium in flasks, as required, from concentrated sterile solutions. Amino acids, shown in table 2, were of CP grade, and chiefly in the i-form. When dl compounds were added, the amount given in table 2 was doubled. The volume of medium in each experimental flask was adjusted according to the amount of basal medium used, after addition of amino acids and vitamins. At inoculation a 1-cm loop of washed flor was floated on the medium in each experimental flask. Carry-over of nutrients was avoided by heating the loop to incandescence between each transfer. Inoculated flasks TABLE 2. Composition of defined medium simulating wine for study of flor growth Basal, 2 X Strength Additions Compound Amounts in Compound Amounts in 1-Arabinose 0.4 g Arginine HCl 10 mg Glucose, reagent 0.2 g Aspartic acid 0.5 mg Ethanol, 95% to Glutamic acid 1.0 mg conc. desired Glycine 1.5 mg Glycerol 1.5 g Histidine HCI 1.5 mg Acetaldehyde 6 mg H20 Ethyl acetate 5 mg Isoleucine 1.0 mg Leucine 1.5 mg Acetic acid 40 mg Methionine 0.5 mg Lactic acid 400 mg Phenylalanine 0.5 mg Tartaric acid 1 g Proline 200 mg Serine 40 mg KH2PO4 20 mg Tryptophan 0.5 mg K2HPO4 20 mg Tyrosine 0.5 mg MgSO4-7H20-8 mg Valine 1.0 mg MnSO4*4H20 0.4 mg FeSO4*7H20 0.4 mg Thiamin 25 pag NaCl 0.4 mg Riboflavin 20 pg Pyridoxine 30 pag KOH solution to Ca pantothenate 50 pg ph desired Nicotinic acid 50 pug Biotin 0. 1,ug p-aminobenzoic 5,ug acid Inositol 20 mg 57

58 J. G. B. CASTOR AND T. E. ARCHER [VOL. 5 TABLE 3. Effect on flor growth of omission of single vitamins from complete medium, with 10-volume per cent ethanol, ph 3.5, at 26 C All vitamins present.. 3+ All vitamins omitted. Thiamin omitted.. 3+ Riboflavin omitted.. 2+ Pyridoxine omitted.. 3+ Pantothenate omitted... i Nicotinic acid omitted.. 3+ Biotin omitted.... 2+ p-aminobenzoic acid omitted.. 3+ Inositol omitted.... 3+ Wine controls.... 4+ TABLE 4. Effect on for growth of presence of single vitamins in basal medium plus amino acids, with 10 per cent by volume ethanol, ph 3.5, at 26 C All vitamins present.. 3+ All vitamins omitted. Thiamin present. Riboflavin present. Pyridoxine present. Pantothenate present... + Nicotinic acid present. Biotin present. p-aminobenzoic acid present Inositol present. Wine controls.... 4+ were incubated at 20 3 C, or at 26 i 0.5 C, as desired, for 2 weeks. Flasks containing wine were used as controls. Experiments were set up in triplicate. Three of the basal medium components, arabinose, tartaric acid, and glucose, were obviously obtained from natural sources. Preliminary experiments were carried out with arabinose and tartaric acid, alone and in combination, to determine whether or not they would support flor growth. Hall et al. (1933) have shown that purified table sugars contained enough nutrient factors to support some yeast growth. The glucose employed in this study, however, was of reagent grade and was considered to be essentially free of accessory growth factors. Estimation of flor development on the defined medium offered little difficulty. The original inoculum could usually be seen, after 2 weeks, floating on the medium in flasks which gave negative results, and was used as an aid in estimating growth. Estimates of growth were recorded as follows: -, no increase in inoculum; z±, possible but not definite increase; +, definite increase, flor incomplete, smooth, thin; 2+, definite increase, flor complete, smooth, thin; 3+, definite increase, flor complete, smooth, thick; 4+, very heavy flor, wrinkled, thick. RESULTS AND DISCUSSION The results of trials of basal medium components, without addition of vitamins or amino acids, showed no growth of the flor with 10 and 15 per cent by volume ethanol at the temperatures employed. Arabinose and tartaric acid, alone and in combination, did not support flor growth within 3 weeks. Wine controls supported 4+ growth within 2 weeks. of the flor on the complete medium at 10 per cent ethanol during incubation at 26 C with glucose was 3+. Without glucose, growth was slower and did not go beyond the + stage. Similar results were obtained at 15 per cent ethanol. However, with the higher amount of ethanol, appreciable growth was obtained only upon incubation at 20 i 3 C. This result is in accordance with the observations of Fornachon (1953) that the alcohol tolerance of the flor is subject to temperature limitations. He found that 20 C was optimum for the growth of flor at 15 to 16 per cent by volume ethanol. Results of experiments with the complete medium, in which single vitamins were omitted, are shown in table 3. Flor growth was not initiated when all vitamins were omitted. Addition of all vitamins listed gave 3+ growth. Omission of thiamin, pyridoxine, nicotinic acid, p-aminobenzoic acid and inositol singly, resulted in 3+ growth. Omission of riboflavin and biotin resulted in 2+ growth. The medium without pantothenate failed to support definite growth. It is of interest that flor growth in the presence of pantothenate, when other vitamins were omitted singly, developed at least to the stage of complete coverage of this medium with a thin film (2+). Flor development on wine controls was 4+. This experiment was run 3 times, with essentially similar results at 10 and 15 per cent ethanol. Results of experiments with the basal medium plus amino acids, to which the vitamins listed were added singly, are shown in table 4. Flor growth was not initiated when all vitamins were omitted. Addition of all vitamins listed gave 3+ growth. Of the vitamins added singly, pantothenate alone supported detectable growth in 2 weeks. Flor development on wine controls was 4+. This experiment was repeated once with the same results. Flor growth in several sets of experiments was examined microscopically at various stages of development. The appearance of the cells was found to duplicate the observations on cell shape and size noted by Fornachon (1953) in his thorough study. The pantothenate requirement for development of the flor suggested that other compounds which are related to pantothenate might substitute for it in supporting growth of the flor. As f,-alanine forms part of the pantothenate molecule, the ability of this and the a form of alanine to support growth of the flor was tested. The results of these experiments are shown in table 5. In

1957] 1-7NUTRIENT REQU-IREMIENTS FOR S. BETICU-S GROWTH 59 TABLE 5. Effect on flor growth of a- and,3-alanine, as suibstitute for pantothenate, in the complete mtiediulm, with 10 per cent ethanol, ph 3.5, at 26 C Complete (table 2)... Complete, less pantothenate... Complete, less pantothenate, plus a-alanine.. Complete, less pantothenate, plus $-alanine.+... + Wine controls........ 4+ this case, the growth of the flor in 2 weeks was relatively meager on the complete medium. However, flor development was definite on the -medium containing pantotheniate, and on the medium in which f-alanine was substituted for pantothenate. No growth occurred on the media lacking pantothenate or containing a-alanine. Both forms of alanine were added at different levels, 1, 5, and 10 mg per 100 ml. Flor growth exhibited Ino varying effect of the different amounts of 3-alanine. Grow-th on the wine controls was 4+. These results also seem significant in the light of the kinown metabolic attack on acetic acid by the flor. The acetate metabolism of the flor and the flor growth requirement for pantothenate or the 3-alanine moiety of paintothenate, suggest that this stage of the organism uses the pantothenate in synthesis of coenzyme A. The acetate metabolism of this organism needs further investigation from this poinlt of view. Experiments with folic acid, found in wines by Hall et al. (1956), and pyridoxal as a more utilizable form of vitamin B6, were carried out. Neither folic acid nor pyridoxal appeared to be essential or notably stimulatory to the growth of the flor. Flor growth ratings on wiine controls and with pyridoxal and folic acid added singly or in combination were 3+. Flor growth response to addition or omission of single amino acids in the defined medium gave somewhat variable results. On the basal medium at 10 per cent ethanol without vitamins, wnith amino acids added singly, small but definite (+) flor development generally occurred only in the presenee of glycine or histidine. No growth occurred with addition of other amino acids singly. OIn the complete medium at 10 per cent ethanol, with omission of single amino acids, flor growth was small but definite (+), except for failure of flor growth with omission of histidine in some cases. The results suggested that there might be insufficient nitrogen present for growth when single amino acids were added in the amounts in w-hich they were present in a young wine. Amino acids waere added singly, in amounts of 10 mg per 100 ml, to the basal medium at 15 per cent ethanol plus vitaminis. In these experiments Ino flor growth occurred with 9 amino acids (cystine, glycine, histidine, leucine, lysine, methionine, phenylalanine, tyrosine, and valine). Six amino acids supported small but definite (+) growth (aspartic acid, glutamic acid, isoleucine, serine, threonine, and tryptophan). Two amino acids supported 2+ growth (arginine and proline). The complete medium supported 3+, and wine 4+, growth. These results are of interest in that earlier work (Castor 1953a; Castor and Archer, 1956) has indicated that aspartic acid, proline, serine, and threonine remain as the most abundant amino acids in young wine shortly after completion of alcoholic fermentation. The experimental evidence indicated that the complete medium as given in table 2 was sufficient to support good growth of the flor. However, the number of cases in which the complete medium supported 3+ or less growvth, while wine supported 4+ growth, suggested that the medium was not nutritionally equivalent to wine. The defined medium contained 17 amino acids. Wine has been reported by authors already cited to contain 24 amino acids. The defined medium was supplemen ted with small amounts of the amino acids additionally reported for wine by other workers. Amino acids in addition to the original 17 of the defined medium (table 2) were tested singly at 10 mg per 100 ml in basal medium at 15 per cent ethanol plus vitamins. Small but definite growth (+) was supported by a-alanine and -y-aminobutyric acid. -No growth appeared with f-alanine, a-aminobutyric acid, cysteine, hydroxyproline, and norvaline. When all 24 amino acids reported for wine were used in the basal medium plus vitamins, flor growth was apparently equal to that on wine (4+). Further work on the possible amino acid requirements of this and other strains of flor, and on the influence of accessory growth factors on the acetate and aldehyde metabolism of the flor, is in progress. SUMMARY The B-complex v-itamin requirements for growth of a Saccharomyces strain of sherry flor yeast were investigated. A chemically defined medium simulating wine was employed. The flor development observed on various modifications of the medium indicated that pantothenate is an essential factor for growth, and that f-alanine can replace caleium pantothenate. Although good growth of the flor yeast was obtained on the defined medium, the results suggested that the medium was not in general nutritionally equivalent to wine. Experiments on the amino acid requirements for flor growth were inconclusive. REFERENCES AMIERINE, MI. A. 1954 Composition of wines. I. Organic constituents. Advances in Food Research, 5, 353-510. CASTOR, J. G. B. 1953a The free amino acids of musts and w-ines. I. MNicrobiological estimation of fourteen amino

60 acids in California grape musts. II. The fate of amino acids of must during alcoholic fermentation. Food Research, 18, 139-145; 146-151. CASTOR, J. G. B. 1953b B-complex vitamins of musts and wines as microbial growth factors. Appl. Microbiol., 1, 97-102. CASTOR, J. G. B. AND ARCHER, T. E. 1956 Amino acids in must and wines, proline, serine, and threonine. Am. J. Enology, 7(1), 19-25. CROWTHER, R. F. AND TRUSCOTT, J. H. L. 1955 Note on the growth of flor yeasts. Can. J. Agr. Sci., 35(2), 211-212. CRUESS, W. V. 1948 Investigations oi the flor sherry process. Bulletin 710, College of Agr., University of California, Berkeley. DIMOTAKAS, P. 1956 Acides amines dans les vins grecs. Bull. off. int. vin, 29, 81. FORNACHON, J. C. M. 1953 Studies on the sherry flor. Australian Wine Board, Adelaide. FREIBERG, K. J. AND CRUESS, W. V. 1955 A study of certain factors affecting the growth of flor yeast. Appl. Microbiol., 3, 208-212. HALL, ALICE P., BRINNER, LISA, AMERINE, M. A., AND MOR- GAN, AGNES FAY. 1956 The B-vitamin content of grapes, musts and wines. Food Research, 21, 362-371. HALL, H. H., JAMES, L. H., AND STUART, L. S. 1933 Yeast growth stimulants in white sugars. Ind. Eng. Chem., 27, 1052-1054. LODDER, J., AND KREGER-VAN RIJ, N. J. W. 1952 The yeasts, a taxonomic study. North Holland Publishing Co., Amsterdam. LUTHI, H. 1953 Papier chromatographische Trennung und Bestimmung von Aminosauren in Traubenmost und Wein. Deutsche Weinbau, Wissensch. Beiheffte 7(2), 33-54. MARCILLA ARRAZOLA, J., ALAS, G., AND FEDUCHY, MARINO E. 1936 Contribucion al estudio de la levaduras que forman J. G. B. CASTOR AND T. E. ARCHER [VOL. 5 velo sobre ciertos vinos de elevado grado alcoholico. Anales centro invest. vinicolas, 1, 1-230. MORGAN, A. F., NOBLES, H. L., WIENS, A., MARSH, G. L., AND WINKLER, A. J. 1939 The B-vitamins of California grape juices and wines. Food Research, 4, 217-219. NEILSEN, NORA E. 1952 The effect of oxygen on growth and carbon metabolism of flor and related yeasts. Ph.D. Thesis, University of California, Berkeley. NIEHAUS, C. J. G. 1937 South African Sherries. Farming in S. Africa, 12, 82-85. OLSON, B. H. AND JOHNSON, M. J. 1949 Factors producing high yeast yields in synthetic media. J. Bacteriol., 57, 235-245. PROCOPIO, M. AND CALE, T. 1954 Amino acidi liberi nei vini. Riv. viticol. e enol. (Conegliano), 7(1), 12-16. SCHANDERL, H. 1936 Untersuchungen uber sogenannte Jerez-Hefen. Wein u. Rebe, 18, 16. SCHANDERL, H. 1950 Die Mikrobiologie des Weines. Handbuch der Kellerwirtschaft. II. Stuttgart. SISAKIAN, N. M. AND BEZINGER, E. N. 1953 Modification of the amino acid composition of wine during the first stages of its preparation (in Russian). Biokhimiya, 18, 412-422. TARANTOLA, C. 1954 Separazione e identificazione cromatografica degli amino-acidi nei vini. Atti accad. ital. vite e del vino, 6. VALAIZE, H. AND DUPONT, G. 1951 Les acides amin6 et le bouquet des vins. Inds. agret. aliment., 68, 245-250. WICKERHAM, L. J., FLICKINGER, MAY H., AND BURTON, K. A. 1946 A modification of Henrici's vegetable-juice sporulation medium for yeasts. J. Bacteriol., 52, 611-612. WILLIAMS, R. J., et al. 1941 Studies on the vitamin content of tissues. I. Univ. Texas Publ., No. 4137. October 1, 1941.