APPLID AND NVIRONMNTAL MICROBIOLOGY, JUlY 199, p. 246-25 99-224/9/7246-5$2./ Copyright 199, American Society for Microbiology Vol. 56, No. 7 Fuel Alcohol Production: ffects of Free Amino Nitrogen on Fermentation of Very-High-Gravity Wheat Mashes KOLOTHUMANNIL C. THOMAS AND W. M. INGLDW* Department of Applied Microbiology and Food Science, University of Saskatchewan, Saskatoon, Saskatchewan, Canada 57N OWO Received 29 January 199/Accepted 2 April 199 Although wheat mashes contain only growth-limiting amounts of free amino nitrogen, fermentations by active dry yeast (Saccharomyces cerevisiae) were completed (all fermentable sugars consumed) in 8 days at 2 C even when the mash contained 35 g of dissolved solids per 1 ml. Supplementing wheat mashes with yeast extract, Casamino Acids, or a single amino acid such as glutamic acid stimulated growth of the yeast and reduced the fermentation time. With.9% yeast extract as the supplement, the fermentation time was reduced from 8 to 3 days, and a final ethanol yield of 17.1% (vol/vol) was achieved. Free amino nitrogen derived in situ through the hydrolysis of wheat proteins by a protease could substitute for the exogenous nitrogen source. Studies indicated, however, that exogenously added glycine (although readily taken up by the yeast) reduced the cell yield and prolonged the fermentation time. The results suggested that there are qualitative differences among amino acids with regard to their suitability to serve as nitrogen sources for the growth of yeast. The complete utilization of carbohydrates in wheat mashes containing very little free amino nitrogen presumably resulted because they had the "right" kind of amino acids. The production of fuel alcohol from renewable resources has received considerable interest in recent years, despite the fact that alcohol produced by fermentation may require subsidies in order to compete economically with petroleum fuels. The costs of raw materials, capital equipment, and processing all influence greatly the profitability of fuel alcohol manufacture. While cellulolytic and other agricultural waste products may eventually serve as the cheapest possible raw materials, the technology needed 'for their efficient and most economical conversion to fuel alcohol has not yet been fully realized. Starch and other carbohydrate-based raw materials can be converted to fuel alcohol with a high degree of efficiency. Yields of ethanol as high as 9 to 95% of the theoretical maximum have been realized under industrial conditions. Further improvement can be achieved only by reducing the fermentation time (increasing the rate of production) and increasing the concentration of alcohol by fermenting greater amounts of sugar by a process known as very-high-gravity fermentation (5). Fuel alcohol production would be greatly helped by the availability of ethanol-tolerant strains of Saccharomyces cerevisiae. This would allow the fermentation of high concentrations of carbohydrates. Usually, sugar concentrations in excess of 2% (wt/vol) are not used under industrial conditions because increasing concentrations of ethanol retard the growth of the yeast and fermentation eventually stops (2, 5, 15, 17). The critical concentration of ethanol at which a yeast ceases to grow is influenced by several factors. These have been comprehensively reviewed by Casey and Ingledew (5), Jones (14), and van Uden (21). thanol tolerance was thought to be independent of the nutritional conditions or the physiological state of the yeast and purely a reflection of its genetic makeup. It is now known that by altering nutritional conditions, it is possible to increase ethanol yield as well as the survival of yeast at high concentrations of ethanol (6, 7, 15). For example, during sake fermentation, concentrations of ethanol of up to 2% (vol/ * Corresponding author. 246 vol) can be achieved. This high yield results from the unique fermentation conditions used and not because of inherent genetic differences among the yeast strains (16). For the production of fuel alcohol, the starch in wheat and other grains must be first converted to fermentable sugars by a procedure known as mashing. The mashing procedure liberates nutrients needed for yeast growth. Our preliminary studies have shown that in wheat mash, only assimilable nitrogen is limiting. Although wheat is rich in proteins, the yeast cannot use these complex nitrogenous materials for growth unless they are first hydrolyzed to simple amino acids, dipeptides, or perhaps tripeptides (4). A small amount of usable nitrogen (collectively referred to as free amino nitrogen [FAN]) is liberated during the mashing, but this amount is not sufficient to support fermentation at the fastest rate, especially when high concentrations of sugars are to be fermented. The results presented in this report show the relationship between the availability of FAN and the rate of fermentation of sugars derived from wheat starch. Results also show that in spite of the low FAN levels in wheat mash, it is possible to ferment relatively large amounts of carbohydrate. Fermentation can be accelerated further by supplementing the mash with assimilable forms of nitrogen or by generating FAN from wheat proteins through hydrolysis with protease supplied exogenously. MATRIALS AND MTHODS Materials. Active dry yeast, high-temperature a-amylase, and Allcoholase II (glucoamylase) were supplied by Alltech Biotechnology Center, Nicholasville, Ky. The high-temperature a-amylase was found to have maximal activity at temperatures between 8 and 9 C and a specific activity of 1.14 g of starch (hydrolyzed) min-1 mg of protein-' at 8 C. The preparation supplied by the manufacturer contained 1 mg of protein per ml. Allcoholase II contained 98.5 mg of protein per ml and had a specific activity of 1 mg of glucose (produced) min-1 mg of protein-1 at 3 C. Neutrase (a protease active in the neutral ph range) was manufactured by Novo Industri A/S, Bagsvaerd, Denmark, and was used
VOL. 56, 199 at.2% (vol/vol). Wheat samples were purchased from a local supplier. All routinely used chemicals were of reagent grade and were purchased from local suppliers. Chemicals required for the measurement of ethanol were purchased from Sigma Chemical Co., St. Louis, Mo. Casamino Acids were obtained from Difco Laboratories, Detroit, Mich. The yeast extract used in this study was a dried water-soluble extract of autolyzed yeast obtained originally from Anheuser- Busch, Inc., St. Louis, Mo., but now marketed by Gist- Brocades, Charlotte, N.C. Grinding and mashing of wheat. Wheat was ground with a plate grinder (type KT-3; Falling Number AB, Stockholm, Sweden) at setting number 2. For mashing, 7 g of ground wheat was dispersed with constant agitation in 2,1 ml of hot (6 C) water containing 1,uM calcium chloride. To this, 3.5 ml of the high-temperature oa-amylase preparation was added. After 5 min, the temperature was raised to 95 to 97 C and held at this temperature for 6 min with continuous stirring. The volume lost because of evaporation was made up by the addition of sterile distilled water, and the temperature was lowered to 8 C. The gelatinized starch was liquefied further by the addition of another 3.5 ml of hightemperature a-amylase and incubated at 8 C for 3 min. This liquefaction step hydrolyzed all the starch into soluble sugars and dextrins. The mash thus prepared contained 19 to 2 g of dissolved solids per 1 ml (water-soluble portion). This mash was saccharified and fermented directly, or its dissolved solid content was further raised by adding freezedried wheat hydrolysate (preparation outlined below), and then the mash was used for fermentation. In most experiments, 5-g samples of the wheat mash were transferred to jacketed sterile Celstir fermentors (Wheaton Instruments, Millville, N.J.). If additives were used, they were sterilized in these fermentors by autoclaving at 121 C for 15 min. The fermentors containing the mashes were connected to a D3-G water bath circulator (Haake Inc., Saddle Brook, N.J.) maintained at 3 C. To each fermentor, 1 ml of Allcoholase II was added to saccharify the dextrins to fermentable sugars. After 3 min, the temperature of the fermentors was lowered and maintained at 2 C throughout the fermentation. Preparation of inocula. leven grams of active dry yeast were dispersed in 99 ml of prewarmed (38 C).1% sterile peptone water. After 2 to 25 min at 38 C, the required volume of inoculum was added to each fermentor. Preparation of wheat hydrolysate. Liquefied (but not saccharified) wheat mash was strained through a double layer of cheesecloth, and the clear liquid was centrifuged (1,3 x g for 15 min) to remove the remaining suspended particles. The resulting supernatant liquid was freeze-dried. The freeze-dried product, called wheat hydrolysate, was used to raise the dissolved solid contents of wheat mashes when very-high-gravity (in this case, 25 g or more of dissolved solids per 1 ml) fermentations were attempted. Measurement of total solids. Total dissolved solids were measured by estimating the specific gravity of the watersoluble portion of the mash obtained by centrifugation at 1,3 x g for 15 min. The specific gravity was determined at 2 C with a digital density meter (DMA-45; Anton Paar KG, Graz, Austria). With the aid of appropriate tables, the results were converted to grams of dissolved solids (expressed as grams of sucrose) per 1 ml. FAN. The FAN in supernatant liquid was determined by the ninhydrin method of the uropean Brewing Convention (9). VRY-HIGH-GRAVITY FRMNTATION OF WHAT 247 thanol. thanol was measured by alcohol dehydrogenase assay (Sigma Bulletin no. 331 U.V., Sigma Chemical Co., St. Louis, Mo.). This method overestimates ethanol by about 7%, as shown by assays of standard solutions of ethanol (unpublished results). All values reported here have been corrected for this overestimation. Cell number and percent viability. Total cell counts and viable-cell counts were determined by the direct microscopic method at a magnification of x4 with an improved Neubauer counting chamber. Samples were diluted with modified Ringer solution containing 25 mg of methylene blue per liter. Modified Ringer solution contained the following ingredients (per liter): 8.6 g of NaCl,.3 g of KCl,.33 g of CaCl2,.5 g of Na2S23 * 5H2, and 1. g of sucrose. Cells which stained blue were considered to be nonviable. Results are expressed as cells per gram of mash rather than per milliliter because it was not possible to pipette the viscous mash accurately. RSULTS An all-malt brewery wort of 16 P (where P corresponds to grams of dissolved solids expressed as sucrose per 1 g of wort) would normally contain about 25 to 3 mg of FAN per liter. Wheat mashes of similar sugar contents had only 54 to 58 mg of FAN per liter. In brewing, adequate rates of attenuation and complete utilization of fermentable sugars require 15 mg of FAN per liter in a 12 P wort (or 2 mg of FAN per liter in a 16 P wort). If the FAN content is less than this value, the wort is likely to be deficient in nitrogen, and incomplete or protracted (stuck or sluggish) fermentations could result (18). With such a criterion, wheat mash might be considered severely limited in utilizable nitrogen. Wheat mashes (16 g of dissolved solids per 1 ml) were inoculated with active dry yeast to give an initial cell number of 1.75 x 17/g of mash and fermented at 2 C with stirring. Virtually all of the fermentable sugars were exhausted by 12 h (Fig. la), a result unattainable in wort of similar gravity with such low levels of nitrogen. Nearly 8% of the total of 58 mg of FAN per liter was taken up by the yeast. Much of this uptake of nitrogen occurred during the early stages of fermentation. The fact that, even under severe nitrogen deficiency (compared with wort of equivalent gravity), complete attenuation was possible suggested that the FAN in wheat mash was qualitatively different from that of wort. The rate of fermentation of wheat mash could be stimulated either by providing the yeast with extra amounts of utilizable nitrogen through in situ hydrolysis of wheat protein, or by adding an external source of FAN such as yeast extract. Wheat protein was partially hydrolyzed by adding.2% (vol/vol) of a commercial protease (Neutrase) to the fermentor 3 min after inoculation. Addition of the protease reduced the fermentation time from 12 h to less than 72 h (Fig. la). The stimulatory effect of Neutrase apparently resulted from an increase in available FAN for yeast growth and not through addition of other growth factors such as vitamins and minerals in the enzyme preparation. There was no reduction in the fermentation time when Neutrase inactivated by boiling (1 min) was added to a control fermentor. A separate experiment was conducted to determine the amount of FAN liberated through the action of the Neutrase. An uninoculated fermentor was treated with Neutrase, and the increase in the FAN was determined at various times. A maximum of 165 mg of FAN per liter was liberated during the first 24 h, during which all of the FAN uptake in an inoculated fermentor would have been completed.
248 THOMAS AND INGLDW C (I) cn -o cn -Q 2 15 1 5 5 1 15 3 2 1 5 1 15 4 3 2 1 1 Time (hours) 2 FIG. 1. Changes in dissolved-solid contents and number of yeast cells during fermentation of wheat mashes by active dry yeast at 2 C. (a) Wheat mashes initially contained 16 g of dissolved solids per 1 ml. Symbols:, no additions (control);,.2% (vol/vol) Neutrase added 3 min after inoculation. (b) Unsupplemented mash initially contained 25.7 g of dissolved solids per 1 ml. Symbols:, no additions (control); other symbols, 3. g of Casamino Acids (-), glutamic acid (O), or yeast extract (A) per liter. (c) Wheat mash initially contained 35 g of dissolved solids per 1 ml. Symbols:, no additions (control);, 1.5% yeast extract;,.8% glycine. The observed stimulation of fermentation was apparently mediated through increased growth and cell multiplication. Without protein hydrolysis, a maximal yeast cell number of 1.5 x 18/g of mash was reached within the first 24 h, and TABL 1. Time (h) APPL. NVIRON. MICROBIOL. ffects of different sources of FAN on fermentation of very-high-gravity wheat mash Concn of FAN (mg/liter) with indicated supplement Yeast Casamino Glutamic Control extracta Acidsa acid' 69.9 216 25 321 24 19. 31. 31. 13.2 48 17.1 23. 21.8 18.5 72 18. 22.6 19.8 18.6 Total taken upb 52.8 193 186 33 % taken up in 24 h 96.4 95.7 94. 72. a Concentration, 3. g/liter. b These amounts were considered to be 1% for the calculation of percent taken up in the first 24 h. there was no appreciable change thereafter (Fig. la, inset). When extra FAN was made available through protein hydrolysis, the cell population reached 4.1 x 18/g of mash by 48 h. The unsupplemented wheat mash contained insufficient amounts of FAN to promote cell growth to the extent required for a fast rate of attenuation and complete conversion of fermentable sugars to ethanol. Addition of utilizable nitrogen in the form of yeast extract, Casamino Acids, or even a single amino acid such as glutamic acid was also found to stimulate the rate of attenuation (Fig. lb). For this study, very-high-gravity mashes (25.7 g of dissolved solids per 1 ml) which contained 74.4 mg of FAN per liter were used. Supplementations were made at the rate of 3. g of the selected nitrogen source per liter of mash. Without supplementation, the attenuation rate was slow and a small amount of sugar remained even after 144 h. Addition of yeast extract or glutamic acid resulted in complete utilization of the sugar within 96 h. Casamino Acids were not as stimulatory as yeast extract or glutamic acid, although fermentations with this additive were still considerably faster than that observed with the control. A single amino acid such as glutamic acid was as effective in stimulating growth as the mixture of amino acids contained in the yeast extract. However, it must be pointed out that more FAN is contained in a given weight of glutamic acid than in either yeast extract or Casamino Acids. No attempt was made to balance FAN because not all FAN in Casamino Acids or yeast extract preparations is usable. Most of the FAN uptake (94 to 96%) occurred during the first 24 h in the control as well as in mashes supplemented with yeast extract or Casamino Acids (Table 1). During the same period, only 72% uptake occurred when glutamic acid was the nitrogen supplement, although the total amount taken up was greater than that in other samples. The high rate of uptake and nearly complete removal of FAN (the residual amounts may be nonutilizable forms of FAN such as higher peptides [.4 amino acid residues] and soluble proteins) suggested that even when 3. g of supplement per liter was used, the wheat mash was still deficient in usable nitrogen. In conditions under which nitrogen is present in excess, one would expect a larger portion of FAN to remain unused (. S. C. O'Connor-Cox, Ph.D. thesis, University of Saskatchewan, Saskatoon, Saskatchewan, Canada, 1989). Supplementation of wheat mash with an external source of assimilable nitrogen not only resulted in an increased rate of fermentation but also yielded greater amounts of ethanol (data not shown).
VOL. 56, 199 TABL 2. ffects of various concentrations of yeast extract on the uptake by yeast of FAN from wheat masha Time (h) FAN uptake (mg/liter) with indicated amount of yeast extract added to wheat mash (g/liter) 3. 6. 9. 113 283 434 593 24 3.7 62.9 215 355 48 28.8 32.1 37.2 43.4 72 24.5 29.7 35. 48. 96 24. 31.2 37.7 58.7 144 22.5 3.2 NDb ND Total taken upc 9. 253 399 549 % taken up in 24 h 9.9 86.9 54.8 43.3 a Dissolved solid content of wheat mash, 31 g/1 ml. b ND, Not determined. c These amounts were considered to be 1% for the galculation of percent taken up in the first 24 h. The low level of nutrient supplement (135 to 24 mg of FAN per liter) was insufficient to promote fermentation at the highest rate. Separate experiments showed that when wheat mash was supplemented with increasing amounts of yeast extract, the fermentation rate increased. The fermentation was completed within 72 h even when the wheat mash contained 31 g of dissolved solids if the concentration of yeast extract supplement was raised from.3 to.9%. Nearly 8% of the available FAN was used by the yeast, and virtually all of the FAN uptake occurred during the first 24 h, when the initial FAN value was low. The uptake period extended to 48 h when the initial FAN values were higher (Table 2). In the unsupplemented wheat mash, the cell number increased from 1.5 x 17 to 2.18 x 18 cells per g of mash within the first 48 h. There was no significant change in cell number after this, although 55% of the sugar was fermented after cell proliferation had ceased. The maximum cell number in each case appeared to be a function of the initial FAN values. With.9% yeast extract, there was continuous growth throughout the fermentation period and a maximum cell number of 4. x 18 cells per g of mash was reached towards the end of fermentation. Wheat mash was not deficient in any minerals. Supplementing the wheat mash with phosphate, magnesium, or other trace elements (Zn, Mn, Fe, Co, Mo, Cu, B, and I) did not affect the growth rate or attenuation (data not shown). The nature of the amino acid which served as the source of nitrogen did, however, influence the rate of fermentation and growth. This was made clear by an experiment in which glycine was used as a source of nitrogen. The unsupplemented mash contained 97.7 mg of FAN per liter and 35 g of dissolved solids per 1 ml. Without any added nutrient, 9.6% of the dissolved-solid content was used up by the yeast within 8 days (Fig. lc). This indicated that if the sugar concentration was raised above 3% (wt/vol), the fermentation time would increase significantly. thanol production paralleled the attenuation rate (Fig. 2). An ethanol yield of 17.1% (vol/vol) was obtained within 3 days when the wheat mash was supplemented with yeast extract. In the absence of nutrient supplement, the fermentation rate was slow, but a final ethanol yield of 16.9% (vol/vol) was still achieved after 8 days. Supplementing the wheat mash with 1.5% (wt/wt) yeast extract (which, along with the FAN derived from the wheat, gave a total FAN value of 859 mg of N per liter of mash) resulted in the complete removal of the sugar VRY-HIGH-GRAVITY FRMNTATION OF WHAT 249 cn. ) 1 2 Time (hours) FIG. 2. Production of ethanol from very-high-gravity wheat mashes during fermentation. The unsupplemented mashes initially contained 35 g of dissolved solids per 1 ml and were fermented at 2 C with active dry yeast with and without various supplements. Symbols:, no additions (control);, 1.5% yeast extract; r1,.8% glycine. in less than 72 h. In fact, the sugar uptake may have been completed by 55 h if it is assumed that the initial rate of uptake observed during the first 48 h continued. A total of 629 mg of FAN per liter of mash was taken up by the yeast when yeast extract was used as the supplement. In the unsupplemented mash, 73 mg of FAN per liter (75% of the available FAN) was assimilated. In contrast, the amount of FAN taken up by the yeast from the mash supplemented with glycine (.8%, wt/wt) was considerable (943 mg of N per liter taken up in 12 h). However, the large uptake of FAN in the form of glycine did not result in a corresponding increase in the rate of attenuation or in increased cell proliferation (Fig. lc). In fact, the rates of attenuation and cell multiplication were lower in glycine-supplemented mashes than in the unsupplemented controls, in which the magnitude of the FAN uptake was only 7.7% of that observed with glycine. The results clearly indicated that glycine slowed down the fermentation and resulted in a lower yield of ethanol than was found in the unsupplemented control (Fig. 2). DISCUSSION The results presented here have shown clearly that wheat mash is deficient in assimilable nitrogen and that a fast rate of fermentation by yeast requires the supply of a source of nitrogen provided exogenously or produced in situ through the hydrolysis of wheat protein. Within the range of dissolved solids reported in this paper (16 to 35 g of dissolved solids per 1 ml), the amount of FAN present in the wheat mashes prepared as described can be calculated by using the following equation derived from the linear regression analy-
25 THOMAS AND INGLDW sis of the data (r =.95): milligrams of FAN per liter (supernatant) = -7.92 + 3.86 x (grams of dissolved solids per 1 ml). The mechanism of stimulation of fermentation by yeast extract or other "yeast foods" has recently been the subject of discussion. While many investigators have observed that yeast extract (6, 7), Casamino Acids (G. P. Casey, Ph.D. thesis, University of Saskatchewan, 1984), or yeast foods containing nitrogen sources and minerals (1, 12, 13, 17) stimulate fermentation under conditions of brewing, vinification, or fuel alcohol production, the roles of these additives have been thought to be purely nutritional. Nonnutritional factors which stimulate fermentation have also been reported (1, 11, 16, 22). These include mere physical effects brought about by substances of particulate nature acting as sites for nucleation of carbon dioxide bubbles (1). The stimulation of fermentation observed in our study does not appear to be such a physical effect. Glutamic acid, which stimulated the fermentation as well as did the yeast extract, is completely soluble and should not provide particulate nucleation sites any more than would soluble compounds such as sugars. In addition, the enzyme Neutrase stimulated the fermentation through liberation of free amino acids. Inactivated Neutrase had no effect, which suggests that it was not the particulate nature of the enzyme or nonenzymatic components of the preparation which caused the stimulation. It is clear that not all amino acids have the same effect in promoting the growth of yeast. For example, glycine is readily taken up by yeast but is inhibitory to growth and fermentation. This inhibition may have resulted from the inability of the yeast to dispose of the two-carbon skeleton (glyoxylate) derived from glycine. Induction of the glyoxylate cycle in yeast occurs only under aerobic and nonrepressing conditions (3, 4, 1), and an active glyoxylate cycle is essential for the assimilation of two carbon compounds. Thus, the effectiveness of an amino acid as a source of nitrogen may be linked to the ability of the yeast to use the carbon skeleton derived from that amino acid. The fact that very-high-gravity wheat mashes containing relatively low amounts of FAN could be fermented to near completion seems to suggest that these mashes contained the "right" kind of amino acids or contained relatively low amounts of those amino acids or other compounds which may be inhibitory. It is known that some basic amino acids such as lysine and arginine inhibit growth and cell division in yeast (8, 2), although they are readily taken up by the yeast and belong to group A (a classification based on the rate of absorption by yeast) (19). It is not surprising that wort which requires considerably more FAN than wheat mash needs to complete fermentation also contains large quantities of arginine and lysine (1). Adding to this rationale is the fact that mashing procedures result in the production of amino acids which are mostly assimilable while other fermentation media such as wort may contain much of their FAN in the form of peptides not usable by yeast. ACKNOWLDGMNTS We gratefully acknowledge research support provided by the Saskatchewan Agriculture-Agricultural Development Fund of the Province of Saskatchewan and the Natural Sciences and ngineering Research Council of Canada. APPL. NVIRON. MICROBIOL. LITRATUR CITD 1. Axcell, B., L. Kruger, and G. Allen. 1988. Some investigative studies with yeast foods, p. 21-29. In Proceedings of the 2th Convention of the Institute of Brewing (Australia and New Zealand Section). Institute of Brewing, Sydney, Australia. 2. Beaven, M. J., C. Carpentier, and A. H. Rose. 1982. Production and tolerance of ethanol in relation to phospholipid fatty-acyl composition in Saccharomyces cerevisiae NCYC 431. J. Gen. Microbiol. 128:1447-1455. 3. Beck, C., and H. K. von Meyenburg. 1968. nzyme pattern and aerobic growth of Saccharomyces cerevisiae under various degrees of glucose limitation. J. Bacteriol. 96:479-486. 4. Berry, D. R., and C. Brown. 1987. Physiology of yeast growth, p. 159-199. In D. R. Berry, I. Russell, and G. G. Stewart (ed.), Yeast biotechnology, Allen and Unwin, London. 5. Casey, G. P., and W. M. Ingledew. 1986. thanol tolerance in yeasts. Crit. Rev. Microbiol. 13:219-28. 6. Casey, G. P., C. A. Magnus, and W. M. Ingledew. 1983. High gravity brewing: nutrient enhanced production of high concentrations of ethanol by brewing yeasts. Biotechnol. Lett. 5: 429-434. 7. Casey, G. P., C. A. Magnus, and W. M. Ingledew. 1984. High-gravity brewing: effects of nutrition on yeast composition, fermentative ability, and alcohol production. Appl. nviron. Microbiol. 48:639-646. 8. Cooper, T. G., C. Britton, L. Brand, and R. Sumrada. 1979. Addition of basic amino acids prevents G-1 arrest of nitrogenstarved cultures of Saccharomyces cerevisiae. J. Bacteriol. 137:1447-1448. 9. uropean Brewery Convention. 1987. Free amino nitrogen, p. 141-142. Analytica-BC, 4th ed. Brauerei- und Getranke- Rundschau, Zurich. 1. Gonzfilez,. 1977. Two-carbon assimilative capacity and inhibition of isocitrate lyase in Saccharomyces cerevisiae. J. Bacteriol. 129:1343-1348. 11. Hayashida, S., K. Ohata, P. Q. Flor, N. Namri, and I. Miyahara. 1982. High concentration-ethanol fermentation of raw ground corn. Agric. Biol. Chem. 46:1947-195. 12. Ingledew, W. M., and R.. Kunkee. 1985. Factors influencing sluggish fermentations of grape juice. Am. J. nol. Vitic. 36:65-76. 13. Ingledew, W. M., F. W. Sosulski, and C. A. Magnus. 1986. An assessment of yeast foods and their utility in brewing and enology. J. Am. Soc. Brew. Chem. 44:166-17. 14. Jones, R. P. 1989. Biological principles for effects of ethanol. nzyme Microb. Technol. 11:13-153. 15. Kalmokoff, M. L., and W. M. Ingledew. 1985. valuation of ethanol tolerance in selected Saccharomyces strains. J. Am. Soc. Brew. Chem. 43:189-196. 16. Kawaharada, H., S. Hayashida, and M. Hongo. 197. The mechanism of formation of high concentration of alcohol in sake brewing. Part VI. Stimulation of yeast growth by koji mold. J. Ferment. Technol. 48:29-33. 17. Nagodawithana, T. W., and K. H. Steinkraus. 1976. Influence of the rate of ethanol production and accumulation on the viability of Saccharomyces cerevisiae in "rapid fermentation." Appl. nviron. Microbiol. 31:158-162. 18. O'Conner-Cox,. S. C., and W. M. Ingledew. 1989. Wort nitrogenous sources. Their use by brewing yeasts: a review. J. Am. Soc. Brew. Chem. 47:12-18. 19. Pierce, J. S. 1987. The role of nitrogen in brewing. J. Inst. Brew. 93:378-381. 2. Sumrada, R., and T. G. Cooper. 1978. Basic amino acid inhibition of cell division and macromolecular synthesis of Saccharomyces cerevisiae. J. Gen. Microbiol. 18:45-56. 21. van Uden, N. 1989. ffects of alcohols on the temperature relations of growth and death in yeasts, p. 77-88. In N. van Uden (ed.), Alcohol toxicity in yeasts and bacteria. CRC Press, Inc. Boca Raton, Fla. 22. Viegas, C. A., I. Sfi-Correia, and J. M. Novais. 1985. Nutrientenhanced production of remarkably high concentrations of ethanol by Saccharomyces bayanus through soy flour supplementation. Appl. nviron. Microbiol. 5:1333-1335.