would tend to drop. The energy demand for export of protons and maintenance of intracellular ph may stimulate

Transcription

1 APPLID AND NVIRONMNTAL MICROBIOLOGY, Dec. 1991, p /91/ $2./ Copyright 1991, American Society for Microbiology Vol. 57, No. 12 Mechanism of Action of Benzoic Acid on Zygosaccharomyces bailii: ffects on Glycolytic Metabolite Levels, nergy Production, and Intracellular ph ALAN D. WARTHt Division of Food Processing, Commonwealth Scientific and Industrial Research Organization, P.O. Box 52, North Ryde, New South Wales 2113, Australia Received 7 March 1989/Accepted 8 September 1991 The effects of benzoic acid in the preservative-resistant yeast Zygosaccharomyces bailii were studied. At concentrations of benzoic acid up to 4 mm, fermentation was stimulated and only low levels of benzoate were accumulated. Near the MIC (1 mm), fermentation was inhibited, ATP levels declined, and benzoate was accumulated to relatively higher levels. Intracellular ph was reduced but not greatly. Changes in the levels of metabolites at different external benzoic acid levels showed that glycolysis was limited at pyruvate kinase and glyceraldehyde dehydrogenase-phosphoglycerate kinase steps. Inhibition of phosphofructokinase and several other glycolytic enzymes was not responsible for the inhibition of fermentation. Instead, the results suggest that the primary action of benzoic acid in Z. bailii is to cause a general energy loss, i.e., ATP depletion. Although benzoic acid and other weak-acid-type preservatives have been commonly used to preserve acidic foods, beverages, and pharmaceuticals for many years, there is uncertainty about which biochemical events are important in inhibiting cell growth under industrially relevant conditions. Inhibition of a variety of enzymes in vitro has been reported for sorbic acid (24), but this has not been clearly related to growth inhibition or cell death. There are few reports of significant inhibition of enzymes by benzoic acid or the aliphatic acids. In fact, glycolytic metabolism is stimulated by concentrations of preservative that inhibit cell growth (6, 26). Preservatives inhibit transport in a variety of cells (12). This may account for their effectiveness in preventing the growth of many sensitive bacteria and yeasts under some conditions, but it is not a likely cause of growth inhibition under conditions such as anaerobic fermentation of nutrientrich foods (29). On the other hand, weak acids are well known to reduce cytoplasmic ph, particularly under the acidic conditions in which the weak-acid-type preservatives are used and are most effective. Krebs et al. (14) suggested that benzoic acid inhibited the growth of Saccharomyces cerevisiae by lowering the intracellular ph, which inhibited glycolysis, specifically by inhibition of phosphofructokinase. Zygosaccharomyces bailii is a yeast which tolerates high levels of preservative and causes spoilage of beverages and other acidic foods (25, 29). When grown in the presence of benzoic acid, Z. bailii in particular, as well as several other yeasts, tolerate higher levels of the acid. These cells accumulate less benzoate than expected from their intracellular ph, and benzoic acid at sublethal concentrations stimulates fermentation rates and reduces growth yields (26, 3). It was suggested that cells grown in the presence of benzoic acid became permeable to or transported benzoate anion, thus reducing the concentration of benzoate in the cell. Benzoic acid would then cycle in and out of the cell in undissociated and dissociated forms, respectively, and the intracellular ph t Deceased 21 June 199. All revisions were completed by Dr. Kenneth W. Nickerson, University of Nebraska. Address reprint requests to Dr. Michael yles, Commonwealth Scientific and Industrial Research Organization. would tend to drop. The energy demand for export of protons and maintenance of intracellular ph may stimulate fermentation rates and lower the growth yields. In the present study, the effects of benzoic acid on fermentation rate, ATP level, intracellular ph, and benzoate accumulation level were determined over a range of concentrations up to that causing full inhibition of growth to elucidate the events that may be critical in producing growth inhibition in Z. bailii. Metabolite levels were measured to determine which reaction steps in glycolysis were affected. MATRIALS AND MTHODS Growth. Z. bailii FRR 1292 was grown in yeast extract medium (ph 3.5) (27) containing 5% glucose in place of the fructose. Benzoic acid (2 mm) was added (26). Cultures were stirred at 25 C in plugged flasks containing 3% headspace and harvested at the late exponential phase (.3 to 1 mg [dry weight] per ml). Cells were washed twice in.1 M potassium citrate buffer (ph 3.5) and kept at 1 C for 1 to 3 h prior to incubation. MICs were determined by using microwell plates with inocula grown in the presence of benzoic acid (28). Growth rates were measured at 25 C by using shaken screw-cap tubes. Incubation. Cells were incubated at 25 C in screw-cap tubes fitted with a septum and flushed with N2. Glucose (2%) was added after 1 min, and 2 mm potassium benzoate and.8 equivalent of 1 M HCI were added 2 min later. Final concentrations were as follows: cells, 13 g (dry weight) per liter; glucose, 5 g/liter; potassium citrate (ph 3.5),.1 M; benzoic acid, to 18 mm. Three independent experiments were done, and the same trends were found in each; results are plotted for one experiment only. Analyses. To determine the fermentation rate, samples (.5 ml) were withdrawn 3 and 13 min after the addition of benzoate, diluted with cold 1% 2-propanol, frozen with dry ice, and stored at -18 C. For analysis, the samples were rapidly thawed and filtered. thanol was determined by gas chromatography (27), and glycerol was determined enzymatically (Boehringer test combination; Boehringer Mannheim Australia Ltd.). Samples for metabolite determinations were Downloaded from on April 21, 218 by guest 341

2 VOL. 57, 1991 BNZOIC ACID AND ZYGOSACCHAROMYCS BAILII 3411 withdrawn at 14 min. Adenine nucleotides were extracted by injecting the sample rapidly into an equal volume of ethanol at 78 C. After 5 min, the samples were cooled, diluted fourfold, and stored at -18 C. ATP was measured by bioluminescence by using a model M17 luminometer and Lumit reagent (Lumac B.V., The Netherlands). AMP + ADP + ATP and ADP + ATP were determined as ATP by using pyruvate kinase and adenylate kinase (1). For other metabolites, the sample was mixed with an equal volume of 2 M HCl4 at C for 3 min, ph was adjusted to 6.2 to 6.7 with 2 M K2CO3, and the mixture was frozen with dry ice and thawed twice and centrifuged. Phosphoenolpyruvate and glycerate 2-phosphate were determined within 4 h, and other metabolites were determined after storage at -18 C. The following glycolytic metabolites were measured enzymatically: glucose 6-phosphate and fructose 6-phosphate (19), fructose 1,6-bisphosphate and triosephosphates (2), glycerate 3-phosphate (11), glycerate 2-phosphate and phosphoenolpyruvate (15), pyruvate (16), acetaldehyde (2), and NAD (13). Pi was estimated by the method of Lowry and Lopez (17). Intracellular concentrations of metabolites were calculated by using a value of 1.8 pil of cell water per mg (dry weight) (26). The amount of benzoic acid in cells and in the medium was determined by high-pressure liquid chromatography (27) after centrifugation for 2 min at 2, x g. A correction was applied for interstitial fluid. Intracellular ph determination. Since determination of intracellular ph by distribution of weak acid requires the assumption of impermeability of the anion, intracellular ph was estimated by the freeze-thaw method (4, 26, 3). Cells (25 mg [wet weight]) were rapidly filtered on Nuclepore membranes and washed with.5 ml of cold water, and the pellet was scraped off into a polythene cap (9 mm [inner diameter]) and frozen in liquid N2. The caps were then thawed in air or in hexane and immediately refrozen. After six cycles, the temperature was quickly raised to 25 C and the ph was determined by using a flat, combination electrode. No addition of fluid was necessary, as the pellet became a viscous fluid. Calculation of mass action substrate/product ratios. Levels of several metabolites were below the limit of detection by the methods used. Glyceraldehyde 3-phosphate and glycerate 2,3-bisphosphate concentrations were calculated by assuming equilibration of triosephosphate isomerase and phosphoglycerate mutase, respectively, as these steps are not commonly limiting (22). The lack of data for glyceric 1,3- bisphosphate means that glyceraldehyde-3-phosphate dehydrogenase and phosphoglycerate kinase reactions cannot be distinguished. Glyceraldehyde-3-phosphate dehydrogenase has been taken as being in equilibrium, but the results may apply to either step. The enolase reaction equilibrium required that the phosphoenolpyruvate concentration in the sample with the lowest benzoic acid concentration be less than.39 mm, which is less than the limit of detection (.1 mm). This value was used at each benzoate concentration. Modeling of higher values allowed at other benzoic acid concentrations did not significantly affect any conclusions. NAD+/NADH ratios were calculated from ethanol and acetaldehyde concentrations, assuming equilibration by alcohol dehydrogenase. RSULTS Growth inhibition. The MIC of benzoic acid for Z. bailii, determined by using microwell plates, was 1 mm (28). Benzoic acid progressively reduced the growth rate, causing I1 15 I 1 - o' S 2 n FIG. 1. ffect of benzoic acid on the production rate of ethanol (O) and glycerol (A) and on ATP levels with () and without () glucose. 3 2 i.-i I- 1 < full inhibition at 1 mm. In the absence of benzoic acid, Z. bailii grew faster under aerobic conditions; however, in the presence of 1 nmm benzoic acid, there was no difference between the aerobic and anaerobic growth rates. The following experiments were done under anaerobic conditions. Fermentation rate and ATP level. The rate of ethanol production was stimulated at benzoic acid concentrations up to 4 mm and then was progressively inhibited (Fig. 1). Glycolysis was not diverted to glycerol production, because glycerol production was reduced by benzoate (Fig. 1). ATP levels declined with increasing benzoic acid concentrations, and both ethanol production and the ATP level were very low near the MIC (Fig. 1). The energy charge (Y2ADP + ATP)/(AMP + ADP + ATP) declined from.97 at low benzoic acid concentrations to.49 at 7 mm benzoic acid. Benzoate accumulation and intracellular ph. Benzoate was accumulated in cells to approximately 1 times the concentration of benzoic acid in the medium (Fig. 2). A steeper rise between 6 and 8 mm benzoic acid corresponds to the range of inhibition of glycolysis. The maximum concentration of benzoate in the cytoplasm was 127 mm. The intracellular ph was reduced, but not greatly, and was 6.3 at fully inhibitory benzoic acid concentrations (Fig. 2). The cytoplasmic ph depended upon the length of time the washed cells were stored before the experiment. In two other experiments, phs I , ~~~ 4 6.~~~~~~~~~~~~ 2 m 5.5 j FIG. 2. ffect of benzoic acid on intracellular ph with () and without () glucose and on benzoate levels with (O) and without (-) glucose in Z. bailii. Downloaded from on April 21, 218 by guest

3 3412 WARTii APPL. NVIRON. MICROBIOL. 12 -z 1 L 8 C FIG. 3. ffect of benzoic acid on glycolytic metabolite levels in Z. bailii. Symbols:, glucose 6-phosphate;, fructose 6-phosphate;, fructose-1,6-bisphosphate. were 6.6 and 6.7 in the absence of benzoic acid and 6. and 6.1 at high concentrations of benzoic acid. The concentration of benzoate in the cells was much lower than that predicted from the phs of the medium and the cytoplasm, on the assumption of an equilibrium in which the cell membrane is permeable to benzoic acid but impermeable to benzoate. At a cytoplasmic ph of 6.77 and a concentration of benzoic acid in the medium of 5.8 mm, a cytoplasmic concentration of benzoate of 1.5 M would have been expected, whereas 45 mm was found. In the absence of glucose, the ph dropped to 5.9 but benzoate was accumulated to a much higher ratio (Fig. 2). Glycolytic metabolite levels. The levels of glycolytic metabolites are shown in Fig. 3 to 5. Glucose 6-phosphate, fructose 6-phosphate, and fructose 1,6-bisphosphate levels remained unchanged up to 5 mm benzoic and then increased (Fig. 3). Triosephosphate concentrations increased over the full range of benzoic acid concentrations, and glycerate 3-phosphate concentrations increased up to 5 mm benzoic acid (Fig. 4). Glycerate 2-phosphate and phosphoenolpyruvate levels were below the limit of detection (.5 mm) at all benzoic acid concentrations, and glyceric 1,3-bisphosphate was not determined. The pyruvate concentration (Fig. 4) decreased, and the acetaldehyde concentration (Fig. 5) showed no clear trend. NAD levels remained constant at all n.4 - o.2 3 c. 2 a FIG. 5. ffect of benzoic acid on glycolytic metabolite levels in Z. bailii. Symbols: A, acetaldehyde;, NAD;, Pi. benzoic acid concentrations, whereas P1 levels increased over the inhibitory range (Fig. 5). Limitation of glycolysis at specific steps. Comparison of the mass action ratios with published equilibrium values (7) showed that the major deviations from equilibrium were at the transport/hexokinase, pyruvate decarboxylase, and phosphofructokinase steps, whereas aldolase and phosphoglycerate kinase and pyruvate kinase steps were limiting to a lesser degree (Table 1). Phosphotriose isomerase and phosphoglycerate mutase reactions are not commonly rate limiting (22) and were taken as being in equilibrium, whereas equilibration of the alcohol dehydrogenase step was assumed in order to calculate NAD+/NADH ratios. With increasing benzoate concentrations, the phosphoglycerate and pyruvate kinase steps became progressively more limiting (Table 1) and appear a priori to be the sites of inhibition of glycolysis. Transport/hexokinase and the phosphofructokinase steps became progressively less limiting, as may result from relief of metabolic inhibitory controls in response to the energy demand or in compensation for inhibition at other stages of glycolysis. However, these results were caused largely by variations in the intracellular ATP levels. Appropriate elevation and depletion of the levels of the other substrates and products were not clearly evident. Modeling the reactions with a constant ATP/ADP ratio (Table 1) almost eliminated the apparent inhibitions of pyruvate and phosphoglycerate kinases and removed the apparent stimulation of hexokinase and phosphofructokinase. The model did not suggest any significant inhibition of aldolase or pyruvate decarboxylase. 2.5,> DISCUSSION 2. Z. bailii exhibits a remarkable resistance to benzoic acid. At low preservative concentrations, Z. bailii adopts a strato 1.5 egy of keeping the intracellular concentration of preservative anions low. Maintenance of the benzoate content below the chemical equilibrium value requires energy. It has been shown (18, 31) that Z. bailii is permeated rapidly by undissociated benzoic acid. Therefore, a low benzoate content.5 can be achieved only by removal of anion at a rate comparable to the rate of inflow of benzoic acid. It is probable that. the anion can leave without direct energy usage by flowing down the electrochemical gradient. However, energy is still required to maintain the intracellular ph, and the observed FIG. 4. ffect of benzoic acid on glycolytic metabolite levels in stimulation of fermentation is probably a response to this Z. bailii. Symbols:, triosephosphate;, glycerate 3-phosphate; energy demand. The accompanying reduced growth yields, pyruvate. (3) and lower growth rates (3) are due to diversion of C- Downloaded from on April 21, 218 by guest

4 VOL. 57, 1991 BNZOIC ACID AND ZYGOSACCHAROMYCS BAILII 3413 TABL 1. nzyme ffect of benzoic acid on the mass action substrate/product ratios for glycolytic reaction steps in Z. bailii SIPIKa Relative SIPIK valueb at benzoic acid concn of: Transport/hexokinase 1.8 x Phosphohexose isomerase Phosphofructokinase 1.1 x Aldolase Phosphoglycerate kinase nolase Pyruvate kinase Pyruvate decarboxylase 6.6 x Constant ATP/ADPC Transport/hexokinase Phosphofructokinase Phosphoglycerate kinase Pyruvate kinase a The mass action substrate/product ratios (SIP), taken from the lowest benzoic acid concentration (.2 mm) in Fig. 3 to 5, were compared with published (7) enzyme equilibrium values (K). Phosphotriose isomerase, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerate mutase, and alcohol dehydrogenase steps were assumed to be in equilibrium. The phosphoenolpyruvate concentration was assumed to be constant at.39 mm. Large numbers indicate rate-limiting deviations from equilibrium. b SIPIK values from Fig. 3 to 5 at increasing concentrations of benzoic acid relative to those at.2 mm benzoic acid. Large numbers (>1.) indicate a more limiting reaction, i.e., apparent inhibition, while small numbers (s1.) indicate a less limiting reaction, i.e., apparent stimulation. c Modeling calculated by using [ATP]/[ADP] = 5. mm at each benzoic acid concentration. energy and not, under these low-benzoate conditions, to inhibition of glycolysis. At benzoic acid concentrations between 5 and 1 mm, fermentation and growth rates were progressively inhibited. xamination of the glycolysis intermediates at each reaction step (Table 1) showed that as the benzoic acid concentration increased, glycolysis became more limiting at the phosphoglycerate kinase and pyruvate kinase reaction steps, whereas phosphofructokinase and the transport/hexokinase step became less restrictive. Both of these effects resulted from the change in ATP levels, since there were not significant changes in the concentrations of the other substrates and products. Apart from the effects on the kinases, there were no indications of specific inhibitions of glycolytic steps. In particular, there were no elevated levels of glycerate 3-phosphate and phosphoenolpyruvate as shown by a pyruvate kinase mutant or of triosephosphate as shown by a phosphoglycerate kinase mutant of S. cerevisiae (9). There was no evidence for inhibition of phosphofructokinase. The changes in hexose phosphate concentrations were quite different from those observed in benzoate-treated S. cerevisiae (14) and in a phosphofructokinase mutant of S. cerevisiae (9). Inhibition by benzoate of glucose transport in yeasts has been reported (23), but does not appear significant in Z. bailii since glucose 6-phosphate levels were not depressed. Furthermore, the similar response of yeasts to a number of weak-acid preservatives makes it unlikely that any very specific effects would be evident. Similarly, inhibition of glycolysis was not caused by a deficiency in NAD+ since its concentration remained constant over most of the benzoate range and decreased only at the highest benzoate level. Likewise, Pi was not limiting. Indeed, its threefold increase (Fig. 5) suggested mobilization of polyphosphate energy reserves. Production of glycerol could also cause a reduction in energy-yielding fermentation. This was found not to be the case since the rate of glycerol production was low compared with the rate of ethanol production and was more inhibited than ethanolic fermentation by low benzoate levels. Glycerol was nevertheless a major metabolite and reached levels equivalent to a 28 mm intracellular concentration during the 1-min incubation. The intracellular ph was near neutrality at low benzoic acid concentrations and then declined to 6.1 to 6.2 at benzoic acid concentrations fully inhibitory to fermentation and growth. There was no indication of inhibition of the phsensitive enzyme phosphofructokinase. Also, in growing cultures of Z. bailii (3), benzoic acid at concentrations up to 4 mm did not significantly reduce the intracellular ph. These results indicate that reduction of the intracellular ph is not a direct cause of inhibition of glycolysis or growth in Z. bailii. In contrast, Cole and Keenan (1) studied the effects of benzoic and sorbic acids on the intracellular ph of Z. bailii and did not find evidence that these weak acids were actively extruded. They measured intracellular phs by monitoring the distribution of benzoic and propanoic acids and by a fluorescence technique. They concluded that resistance can be explained in part by active ejection of acids produced during metabolism, and that Z. bailii has the ability to tolerate chronic reductions in intracellular ph. However, active ejection of metabolic acids is probably an insufficient explanation because of the high rate of benzoic acid permeation and the stoichiometry of 1 to 2 mol of ATP per mol of benzoic acid entering the cell (31). The cellular ph values found in the present study are much higher than those reported by Krebs et al. (14) and Cole and Keenan (1). Although the organisms and conditions were not identical, it appears likely that the difference is due to the methods used. The several methods available for the determination of intracellular ph in yeasts do not give consistent results (3). The most commonly used method, equilibration of a lipophilic weak acid, depends on the cell's being permeable to the undissociated acid and relatively impermeable to the anion. This condition has rarely been demonstrated. With preservative-adapted Z. bailii, it is not applicable (26), and another method is required. The intracellular ph was estimated in the present and previous (26) studies by freeze-thawing with liquid N2. Of the potential errors associated with this procedure, nearly all would be expected to give underestimates of the actual cytoplasmic Downloaded from on April 21, 218 by guest

5 3414 WARTH ph. Also, for yeast cells at external phs near 3.5 with an adequate energy supply, several studies have found nearneutral internal ph values. For S. cerevisiae, nuclear magnetic resonance studies gave a value of 7.1 (21), compared with 6.7 from freeze-thawing (5), and 6.4 from the weak-acid method (8). Lower values have frequently been obtained, but these can be attributed to an inadequate energy supply. Above the MIC for growth, cells contained high concentrations of benzoate. This is an appropriate response to conserve energy for long-term survival. Cells can retain viability under these conditions for many days while the high benzoate anion concentration could well cause direct inhibition of glycolysis and other cell functions. However, the intracellular ph values did not drop to levels which would require no energy to maintain, at least not within the time scale of this experiment. The residual fermentation rates, although low, may be necessary for survival. In summary, the action of benzoic acid on Z. bailii depends upon its concentration. At low concentrations, growth rates are reduced and fermentation rates are stimulated because energy is used to lower the internal concentration of benzoate. At higher concentrations, fermentation itself becomes inhibited, leading to a collapse of the system. The reasons for this inhibition of fermentation are not fully clear. Internal benzoate concentrations increased in parallel with the inhibition of fermentation, but they may not be the major direct cause of inhibition. ph reduction also does not appear to be a prime cause of inhibition in Z. bailii, since the internal ph was not severely depressed. Instead, the pattern of glycolytic intermediate changes suggests that ATP limitation was important. This ATP limitation may be direct or may result from regulatory mechanisms designed to conserve the remaining energy. RFRNCS 1. Bali, W. J., and D.. Atkinson Adenylate energy charge in Saccharomyces cerevisiae during starvation. J. Bacteriol. 121: Beutler, H.-O Acetaldehyde, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 3. Borst-Pauwels, G. W. F. H Ion transport in yeast. Biochim. Biophys. Acta 65: Borst-Pauwels, G. W. F. H., and J. Dobbelmann Determination of the yeast cell ph. Acta Bot. Neerl. 21: Borst-Pauwels, G. W. F. H., and P. H. J. Peters ffect of the medium ph and the cell ph upon the kinetical parameters of phosphate uptake by yeast. Biochim. Biophys. Acta 466: Bosund, I The action of benzoic and salicylic acids on the metabolism of microorganisms. Adv. Food Res. 11: Burton, K Free energy data of biological interest, p In H. A. Krebs and H. L. Kornberg (ed.), nergy transformations in living matter: a survey. Springer Verlag KG, Berlin. 8. Cartwright, C. P., J.-G. Juroszek, M. J. Beavan, F. M. S. Ruby, S. M. F. De Morais, and A. H. Rose thanol dissipates the proton-motive force across the plasma membrane of Saccharomyces cerevisiae. J. Gen. Microbiol. 132: Ceriacy, M., and I. Breitenbach Physiological effects of seven different blocks in glycolysis in Saccharomyces cerevisiae. J. Bacteriol. 139: Cole, M. B., and M. H. J. Keenan ffects of weak acids APPL. NVIRON. MICROBIOL. and external ph on the intracellular ph of Zygosaccharomyces bailii, and its implication in weak-acid resistance. Yeast 3: Czok, G D-Glycerate 3-phosphate, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 12. Freese,., and C. Levin Action mechanisms of preservatives and antiseptics. Dev. Ind. Microbiol. 19: Klingenberg, M Nicotinamide-adenine dinucleotides and dinucleotide phosphates, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 14. Krebs, H. A., D. Wiggins, M. Stubbs, A. Sols, and F. Bedoya Studies on the antifungal action of benzoate. Biochem. J. 214: Lamprecht, W., and F. Heinz Pyruvate, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 16. Lamprecht, W., and F. Heinz D-Glycerate 2-phosphate and phosphoenolpyruvate, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 17. Lowry,. H., and J. A. Lopez The determination of inorganic phosphorous in the presence of labile phosphate esters. J. Biol. Chem. 162: Macris, B. J Mechanism of benzoic acid uptake by Saccharomyces cerevisiae. Appl. Microbiol. 3: Michal, G D-Glucose 6-phosphate and D-fructose 6-phosphate, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 2. Michal, G D-Fructose 1,6-bisphosphate, dihydroxyacetone phosphate and D-glyceraldehyde 3-phosphate, p In H. U. Bergmeyer (ed.), Methods of enzymatic analysis, vol. VI. Academic Press, Inc., New York. 21. Navon, G., R. G. Shulman, Y. Yamane, T. R. ccleshall, K. B. Lam, J. J. Baronofsky, and J. Marmur Phosphorus-31 nuclear magnetic resonance studies of wild-type and glycolytic pathway mutants of Saccharomyces cerevisiae. Biochemistry 18: Rose, I. A., and Z. B. Rose Glycolysis: regulation and mechanisms of the enzymes, p In M. Florkin and. H. Stotz (ed.), Comprehensive biochemistry, vol. 17. lsevier Biomedical Press, Amsterdam. 23. Scharff, T. G., M. Z. Badr, and R. J. Doyle The nature of salicylate inhibition of sugar transport in yeast. Fundam. Appl. Toxicol. 2: Sofos, J. N., and F. F. Busta Antimicrobial activity of sorbate. J. Food Prot. 44: Thomas, D. S., and R. R. Davenport Zygosaccharomyces bailii-a profile of characteristics and spoilage activities. Food Microbiol. 2: Warth, A. D Mechanism of resistance of Saccharomyces bailii to benzoic, sorbic and other weak acids used as food preservatives. J. Appl. Bacteriol. 43: Warth, A. D Resistance of yeast species to benzoic and sorbic acids and to sulfur dioxide. J. Food Prot. 48: Warth, A. D ffect of nutrients and ph on the resistance of Zygosaccharomyces baiii to benzoic acid. Int. J. Food Microbiol. 3: Warth, A. D Preservative resistance of Zygosaccharomyces bailii and other yeasts. CSIRO Food Res. Q. 46: Warth, A. D ffect of benzoic acid on growth yield of yeasts differing in their resistance to preservatives. Appl. nviron. Microbiol. 54: Warth, A. D Transport of benzoic and propanoic acids by Zygosaccharomyces bailii. J. Gen. Microbiol. 135: Downloaded from on April 21, 218 by guest