PYRUVATE FERMENTATION BY STREPTOCOCCUS FAECALIS1

Transcription

1 JOURNAL OF BACTERIOLOGY Vol. 88, No. 1, p July, 1964 Copyright 1964 American Society for Microbiology Printed in U.S.A. PYRUVATE FERMENTATION BY STREPTOCOCCUS FAECALIS1 R. H. DEIBEL2 AND C. F. NIVEN, JR.3 Division of Bacteriology, American Meat Institute Foundation, and Department of Microbiology, The University of Chicago, Chicago, Illinois Received for publication 29 January 1964 ABSTRACT DEIBEL, R. H. (American Meat Institute Foundation, Chicago, Ill.), AND C. F. NIVEN, JR. Pyruvate fermentation by Streptococcus faecalis. J. Bacteriol. 88: Streptococcus faecalis, as opposed to S. faecium, utilizes pyruvate as an energy source for growth. The fermentation is adaptive, as demonstrated by growth experiments in a casein-hydrolysate medium and the fermentation of pyruvate by cell suspensions. The principal products of pyruvate catabolism were acetoin, CO2, and lactic, acetic, and formic acids, although carbon recoveries were low due to the formation of slime. End-product analyses suggested that both the phosphoroclastic and dismutation systems were active in pyruvate breakdown. Studies with cell-free extracts indicated a thiamine diphosphate requirement for active pyruvate catabolism. The involvement of lipoic acid in the phosphoroclastic system was investigated, and, although inconclusive results were obtained, no association of this cofactor with phosphoroclastic activity could be made. A separation of Sherman's (1937) enterococcus division into two distinct physiological types represented by Streptococcus faecalis and S. faecium has been the subject of previous communications (Deibel, Lake, and Niven, 1963; Deibel, 1964). In the study reported herein, it was observed that S. faecalis fermented pyruvate and utilized this substrate as an energy source, in contrast to S. faecium. Some characteristics of pyruvate fermentation, and the additional nutritive requirements imposed upon the microorganism when pyruvate is the sole energy source, are presented in this report. I Journal Paper No. 268, American Meat Institute Foundation. 2 Present Address: Division of Bacteriology, Cornell University, Ithaca, N.Y. 3 Present address: California Packing Corp., San Francisco, Calif. 4 MIATERIALS AND METHODS Strains and media. The source of strains and the methods used to maintain them were noted previously (Deibel et al., 1963). The complex basal medium and methods of test for the utilization of pyruvate as an energy source were identical to those described for the utilization of arginine as an energy source (Deibel, 1964). Sodium pyruvate (Calbiochem) was tested at the 1.0% level and added to the basal medium prior to autoclaving. Utilization studies indicated no necessity for filter sterilization or separate autoclaving of the pyruvate in routine growth studies. However, filter sterilization of this substrate was employed in fermentation balance studies. The semisynthetic, casein-hydrolysate medium employed in these studies (Table 1) was fortified with methionine, since Brodovsky, Utley, and Pearson (1958) observed an insufficiency of this amino acid in commercial preparations of acid-hydrolyzed casein commonly employed in microbiological assays with S. faecalis. The semisynthetic media were inoculated with one drop of a 24-hr culture that was washed twice in sterile distilled water and diluted tenfold. The lipoic acid used in this study was obtained from two sources: a natural preparation isolated from yeast extract was obtained through the courtesy of I. C. Gunsalus; a synthetic preparation was purchased from Sigma Chemical Co., St. Louis, Mo. Except where indicated specifically, the synthetic, DL-lipoic acid was used. It has been demonstrated that optically inactive synthetic preparations are one-half as active biologically as the optically active natural preparations (Hornberger et al., 1953). The methods employed for aerobic and anaerobic conditions of incubation were described previously (Deibel, 1964). Growth was estimated in a Bausch & Lomb Spectronic-20 colorimeter at a wavelength of 600 m,u. All cultures were incubated at 37 C.

2 l'ol. 88, 1964 PYRUVATE FERMENTATION BY S. FAECALIS 5 Chemical nethods. The culture media were clarified by the method of Neish (1950) prior to analyses for end products. Volatile neutrals and volatile acids were determined by steam distillation (Neish, 1950) with a 25-ml sample of the clarified culture medium. Volatile acids were titrated with 0.1 N sodium hydroxide, and formate was estimated by the method of Grant (1948). Acetic acid was calculated by difference. Carbon dioxide was displaced from the culture medium by acidification, flushing with nitrogen, and precipitation in 0.1 N barium hydroxide traps. The washed precipitate was dried to constant weight at 110 C. Pyruvic acid was determined by a modification of the method of Friedman and Haugen as described by Umbreit, Burris, and StaufTer (1957). A distilled, pyruvic acid standard was prepared by the method of Price and Levintow (1952), and it was used to standardize a sodium pyruvate preparation which was maintained under desiccation. For subsequent pyruvate analyses, a weighed amount of the sodium TABLE 1. Composition of the semisynthetic medium employed to culture Streptococcus faecalis Component Amt per 100 ml of medium Acid-hydrolyzed casein Enzyme-hydrolyzed casein Adenine sulfate... 1 Guanine hydrochloride... 1 Uracil... 1 L-Cystine... L-Tryptophan... 1 L-Methionine... L-Asparagine K2HPO NaCl Sorbitan monooleate FeSO4*7H MnCl2-4H20... MgSO4c7H20... mg Nicotinic acid Riboflavine Biotin Folic acid... Thiamine hydrochloride Pyridoxal hydrochloride Calcium pantothenate Ag TABLE 2. Typical growth responses of enterococci with glucose and pyruvate as energy sources* Addition to basal mediumt Strain ~ Condition of Strain incubation 1.0% 1.0% None glu- pyrucose vate Streptococcus faecalis R26... Anaerobic S. faecium F24. Anaerobic S. faecalis R26... Aerobic S. faecium F24... Aerobic * Results are expressed as optical density times 100 ṫ The complex basal medium was employed, and the cultures were incubated for 20 hr. pyruvate was employed. This procedure obviated the necessity of the distillation procedure each time a pyruvate analysis was performed. Acetoin, 2,3-butylene glycol, and diacetyl determinations were carried out as described by Neish (1950). Lactic acid was determined by the method of Barker and Summerson (1941). Standard methods of ascending chromatography were employed where indicated. The chromatographs were developed in a constant-temperature room maintained at 25 C. RESULTS Pyruvate utilization in complex media. Among a collection of enterococci tested for their ability to utilize pyruvate as an energy source, it was observed that 31 S. faecalis strains utilized pyruvate, in contrast to the inability of 38 strains of S. faecium and its variety durans. Pyruvate utilization occurred under both aerobic and anaerobic conditions of incubation, and the growth response equalled or surpassed that obtained with glucose (Table 2). The a-keto analogues of glutaric, butyric, and valeric acids were also tested for their ability to serve as energy substrates. The solutions were neutralized, filter sterilized (sintered glass), and added to the complex basal medium in concentrations ranging from 0.3 to 1.0%. None of these compounds supported the growth of ten representative S. faecalis strains or S. faecium F24. A collection of other lactic acid bacteria were tested to determine the incidence of pyruvate fermentation in the Lactobacillaceae. None of 12

3 6 DEIBEL AND NIVEN J. BACTERIOL. Lactobacillus, 6 Leuconostoc, 18 Pediococcus, and 19 Streptococcus strains (representing serological groups other than group D) fermented this compound. It appears that this characteristic is somewhat limited among the lactic acid bacteria and, indeed, it may be peculiar to S. faecalis. Pyruvate utilization in semisynthetic media. Initial attempts to cultivate representative strains of S. faecalis in the semisynthetic, caseinhydrolysate medium described in Materials and Methods met with failure. However, control tubes in which glucose was added as the energy source evidenced maximal growth, and thus indicated the nutritional adequacy of the medium. The inability to utilize pyruvate in this medium was traced to the absence of the growth factor, lipoic acid. When the semisynthetic growth medium was supplemented with adequate amounts of lipoate, pyruvate was utilized readily (Table 3). A lipoate requirement was not demonstrable when hexoses, pentoses, hexitols, glycerol, or gluconic acid were employed as energy sources. Thus, the lipoate requirement could be associated specifically with pyruvate catabolism. Adaptive nature of pyruvate fermentation. In growth experiments involving the semisynthetic medium with pyruvate as the energy source, erratic growth responses were obtained when a glucose-grown culture was employed as the inoculum and growth was tested as a function of the lipoate concentration. Quite often, growth initiation was delayed beyond 24 hr when the pyruvate-lipoate medium was inoculated with a glucose-grown inoculum. These erratic growth responses could be circumvented if the inoculum culture was grown with pyruvate instead of glucose as the energy source in the complex medium. By use of a pyruvate-grown inoculum, a prompt and reproducible growth response could TABLE 3. Typical growth response of Streptococcus faecalis FB82 to lipoic acid with glucose or pyruvate as the energy source* Lipoic acidt Optical density X 100 (misg/10 ml medium) Pyruvate Glucose * Semisynthetic medium; incubation period, 21 hr. t Natural lipoic acid. mpg Lipoate/lOml Medium Adapted FIG. 1. Growth responses of Streptococcus faecalis FB82 in a casein-hydrolysate medium with pyruvate as energy soulr ce and inoculated with cultures grown in conmplex miiedia containing either glucose or pyruvate. be effected in the pyruvate-lipoate, semisynthetic medium (Fig. 1). The results with growth experiments indicated that pyruvate fermentation was adaptive. An attempt was made to obtain more definitive evidence regarding this observation by determining the rate of pyruvate disappearance in pyruvate-grown and glucose-grown cell suspensions of S. faecalis FB82. Cells were harvested from 250 ml of the semisynthetic medium in which either glucose or pyruvate was the energy source. After washing in 0.03 m phosphate buffer (ph 7.0), the concentration of cells in the two suspensions was adjusted to equal density by comparing the optical density of the suspension with a previously prepared standard curve relating optical density to dry weight. The respective cell suspensions were divided equally into three portions, and the following test protocol was employed: 120,moles of glucose were added to one flask, 240,umoles of pyruvate were dispensed to another, and the third served as an endogenous control. Prior to the introduction of substrate, the flasks were gassed with nitrogen. Periodically, 1.0-ml samples were withdrawn and added to 2.0 ml of 0.5 N sulfuric acid to stop the reaction. The amount of residual pyruvate or glucose was determined in the respective samples. It was observed that the glucose-grown cells did not ferment appreciable quantities of pyruvate, in contrast to the rapid fermentation effected by the pyruvate-grown cells (Fig. 2).

4 VOL. 88, 1964 PYRUVATE FERMENTATION BY S. FAECALIS 7 Although not shown in Fig. 2, both cell suspensions fermented glucose at approximately the same rate. Consequently, these results indicate that the fermentation of pyruvate by S. faecalis is adaptive. Fermentation balances. Preliminary experiments were concerned chiefly with the qualitative detection of the end products of pyruvate fermentation. These tests were conducted with a protein-precipitated fermentation broth (Neish, 1]950) which was extracted continuously with ether at ph 2 for 48 hr in a Kutschner-Steudel apparatus. After extraction, 10 ml of water were added to the extract and, after neutralization, the ether was evaporated. Various concentrations of the extract were examined by the ascending chromatographic method with control samples of pyruvic, lactic, formic, acetic, propionic, and butyric acids. A variety of solvent systems were employed, and lactic, formic, and acetic acids were detected. Neither propionic nor butyric acid was detected. Furthermore, no volatile acids other than acetic and formic were detected when a steam-distilled, volatile fatty acid fraction Glucose-Grown Cells. Pyruvate-Grown Cells Time (Minutes) FIG. 2. Curves demonstrating adaptive nature of pyruvate fermentation with cell suspensions of Streptococcus faecalis FB82 grown in the caseinhydrolysate medium containing either glucose or pyruvate. Both flasks contained 11.4 mg (dry weight) of cells and 240 1Amoles of pyruvate. Total volume, 20 ml in ph 7.0 phosphate buffer (0.03M). Flasks were flushed with nitrogen prior to substrate addition; 1.0- ml samples were periodically withdrawn and added to 2.0 ml of 0.5 N H2S04 to stop the reaction. Residual pyruvate was quantitated as described in Materials and Methods. TABLE 4. Fermentation balance with pyruvate as the energy source* Amt Oxidation Cl/C2 Substrate and (mmoles/ Carbon balance balance pusroductsd prduts 100 recovery moles (mimoles) ofcs OfC3) Oxi- Redized duced Ci C2 Pyruvate Lactate Acetoin C Formate Acetate Totalst * Streptococcus faecalis FB82. Incubation period, 21 hr. Procedure of test given in Materials and Methods. t Carbon recovery = 93.3%; O/R = 1.03; C,/ C2 = was examined chromatographically. The possible presence of significant concentrations of dicarboxylic and a-keto acids other than pyruvate was eliminated by use of standard chromatographic techniques, as described by Block, Durran, and Zweig (1958). Qualitative tests for acetoin were strongly positive in the pyruvate-grown cultures. However, diacetyl could not be detected in significant amounts. Identical values were obtained when culture filtrates or their volatile, neutral, steamdistilled fractions were quantitated for acetoin (procedure quantitates both acetoin and diacetyl) and 2, 3-butvlene glycol (procedure quantitates both acetoin and 2,3-butylene glycol). Consequently, the concentrations of 2,3-butylene glycol and diacetyl were considered to be negligible, and the value obtained with the alkaline a-naphthol procedure was calculated as acetoin. Characteristically, low carbon recoveries were obtained in fermentation balances (Table 4). In a series of experiments, only 87 to 93 % of the carbon was recovered. An uncontrollable factor which may account for the low carbon recoveries was the formation of a viscous slime. Synthesis of this ribonucleic acid nucleoprotein-containing material is favored by slightly acidic or neutral ph values in the growth medium (Deibel, 1963). Thus, the fermentation of pyruvate (in which the final ph approaches 5.0 as contrasted to a final ph of 4.0 to 4.2 with glucose as the energy

5 8 DEIBEL AND NIVEN J. BACTERIOL. source) results in the production of this complex material. Slime also is produced when glucose is employed as the energy source, if neutrality is maintained throughout fermentation by intermittent addition of alkali. Different strains of S. faecalis form varying quantities of slime when grown with pyruvate as the energy source, and some strains form a sufficient quantity to produce a ropy mass of viscous material. Even though the strain chosen for balance studies did not form large quantities of this material, a detectable quantity was apparent in pyruvate-grown cultures. The formation of slime was noted in another study when S. faecalis was cultured at alkaline ph values (Gunsalus and Niven, 1942). These investigators associated a low carbon recovery under the experimental conditions employed with the formation of slime. Although definitive evidence associating low carbon recoveries with slime production is lacking, there appears to be a direct relationship. Further studies are needed to determine the amount of energy source, if any, that is incorporated in slime. Even though an acceptable carbon recovery was not obtained, the relatively high activity of the phosphoroclastic reaction (as indicated by formate accumulation) was unexpected. At first, it was assumed that only the pyruvate dismutation system was lipoate-linked. However, if it is assumed that both the phosphoroclastic and the dismutation systems yield energy as demonstrated by Lipmann (1941) and Gunsalus (1954), then it could be surmised that an absolute lipoate requirement for growth in the semisynthetic medium may reflect lipoate involvement in the phosphoroclastic system. These considerations formed the basis of subsequent experiments with cell suspensions and cell-free extracts in an attempt to demonstrate the involvement of lipoate in the phosphoroclastic system. Studies with the phosphoroclastic system. The cell crops employed in this phase of the study were grown in the semisynthetic medium with pyruvate as the energy source and a sufficient amount of lipoate to yield one-half maximal growth (0.12 m,ug/ml of medium). After 20 to 22 hr of incubation, the cell crop was harvested, washed once, and resuspended in 0.03 M phosphate buffer (ph 7.5). When cell-free extracts were desired, the sedimented, washed cells were ruptured in a Hughes press (-20 C), the preparation was suspended in cold phosphate buffer, and, after centrifugation (2,000 X g for 10 min at 7 C), the extract was decanted and used in the experiments. In preliminary experiments, the cell-free preparations were inactive consistently or evidenced feeble phosphoroclastic activity. With cell suspensions rather than cell-free extracts, it was observed that within a 3-hr period 51 j,moles of formate were formed from 178 j.moles of pyruvate. This conversion took place only in the absence of added lipoate, and insignificant quantities of formate were formed in the presence of this cofactor. In subsequent experiments, it was observed that, although the absolute value of formate produced was subject to variation, a significant decrease in the amount of formate produced consistently occurred when lipoate was added to the system. Later experiments revealed a significant increase in pyruvate catabolism when thiamine diphosphate was added to the cell-free system (Table 5), and maximal phosphoroclastic activity could be attained in a 30- to 40-min incubation period. Unlike the results obtained with cell suspensions, no decreased phosphoroclastic activity was noted when lipoate was added to the cell-free extracts. In a typical experiment, thiamine diphosphate was added with the cofactor mixture, lipoate was employed as the variable, and the unique end product for each reaction of pyruvate (lactate for the dismutation system and formate for the phosphoroclastic system) was quantitated. As noted in Table 6, some dismutation activity was observed in the absence of lipoate, as judged by the production of lactate. The inclusion of lipoate stimulated carbon dioxide and acetoin production, and almost doubled the dismutation activity. In this experiment, the quantity of carbon dioxide produced is not consistent with the quantities of acetoin and lactate formed. No explanation is apparent to account for this discrepancy. The phosphoroclastic activity was only slightly, if at all significantly, stimulated by lipoate. Thus, it would appear that the preparation was depleted of lipoate, although some of the cofactor must have been present to afford the dismutation activity observed without lipoate fortification. Whether this concentration was sufficient for maximal phosphoroclastic activity cannot be

6 VOL. 88, 1964 PYRUVATE FERMENTATION BY S. FAECALIS 9 determined from the data. However, it appears that a much greater stimulation of phosphoroclastic activity might be manifested if this system, like the dismutation system, were lipoatedependent. These results, coupled with the decreased phosphoroclastic activity due to lipoate addition in cell suspensions, cast doubt upon the involvement of lipoate in the phosphoroclastic svstem. DISCUSSION Studies by previous investigators indicated that the enterococci comprise two distinct metabolic types as represented by S. faecalis and S. faecium (Gunsalus, 1947; Skadhauge, 1950; Shattock, 1955; Barnes, 1956). The observation that S. faecalis possesses an energygaining mechanism involving the fermentation of pyruvate, and the lack of a similar mechanism in S. faecium, reinforces this basic separation of enterococcal species. In consideration of the role of lipoate in the oxidation of pyruvate by cell suspensions of S. faecalis (Gunsalus, 1954), the requirement for this cofactor in growing cultures with pyruvate as the energy source is not too surprising. This observation forms the basis for a rapid and reliable test-tube microbiological assay for this cofactor, as will be presented in another communication. As previously reported, a lipoate requirement exists for the fermentation of citrate (Deibel, Evans, and Niven, 1958) and serine TABLE 5. Effect of thiamine diphosphate on pyruvate metabolism in cell-free extracts of Streptococcus faecalis FB82 Amt (;&moles)* End product Without With measured thiamine thiamine diphosphate diphosphatet CO Formate * Pyruvate (200,gmoles per reaction vessel) and 0.1 mmag of D, L-lipoate. Reaction vessel was flushed with nitrogen prior to substrate addition. All values were corrected for endogenous metabolism, which was negligible. Reaction time, 4.5 hr at 37 C. Each flask contained cell-free preparation from 0.5 liter of medium and a total reaction volume of 5.0 ml. t Thiamine diphosphate, 100 Ag per reaction vessel. TABLE 6. Effect of lipoate on pyruvate metabolism by cell-free extracts of Streptococcus faecalis FB82 End product Amt (umoles)* measured Without lipoate With lipoatet CO Acetoin Lactate Formate * Thiamine diphosphate (100,g) and 200,Amoles of pyruvate per reaction flask, which was flushed with nitrogen prior to substrate addition. Each flask contained cell-free extract from growth in 0.5 liter of medium. Total reaction volume, 5 ml. Reaction time, 1 hr at 37 C. t D,L-Lipoate, 0.1 ug per flask. (Deibel and Niven, 1960) by S. faecalis. These observations lead to the conclusion that at least one major energy-yielding step in the fermentation of these compounds involves pyruvate. However, the fermentation of malate by S. faecalis (Deibel and Niven, 1960) is not lipoate-dependent and may involve a heretofore unsuspected energy-generating mechanism. Thus, the utilization of nutritional data in this manner may serve as an index regarding the involvement of pyruvate in unknown metabolic pathways of S. faecalis. The relatively high activity of the phosphoroclastic system in pyruvate catabolism by S. faecalis, coupled with the inability to demonstrate an absolute lipoate requirement for this activity, is not congruent with the necessity of this cofactor for growth. This apparent anomaly has not been resolved, and no explanation is consistent with the results obtained. Undoubtedly, additional study is necessary to clarify these confusing observations. The inability to demonstrate a lipoate requirement for phosphoroclastic activity with S. faecalis parallels the reports of Wolfe and O'Kane (1953) and Oster and Wood (1964), who studied the clostridial system and the enterococcal formate-pyruvate exchange systems, respectively. ACKNOWLEDGMENT This investigation was supported by Public Health Service grant E LITERATURE CITED BARKER, S. B., AND W. H. SUMMERSON The colorimetric determination of lactic acid

7 10 DEIBEL AND NIVEN J. BACTERIOL. in biological material. J. Biol. Chem. 138: BARNES, E. M Tetrazolium reduction as a means of differentiating S. faecalis from S. faecium. J. Gen. Microbiol. 14: BLOCK, R. J., E. L. DURRAM, AND G. ZWEIG A manual of paper chromatography and paper electrophoresis, 2nd ed. Academic Press, Inc., New York. BRODOVSKY, E. R., M. H. UTLEY, AND W. N. PEARSON Methionine inadequacy of casein hydrolysate as a source of difficulty in vitamin assays. Science 128: DEIBEL, R. H RNA nucleoprotein slime produced by Streptococcus faecalis. Bacteriol. Proc., p. 49. DEIBEL, R. H Utilization of arginine as an energy source for the growth of Streptococcus faecalis. J. Bacteriol. 87: DEIBEL, R. H., J. B. EVANS, AND C. F. NIVEN, JR Lipoic acid requirement for anaerobic utilization of pyruvate as an energy source by Streptococcus faecalis. Bacteriol. Proc., p DEIBEL, R. H., D. E. LAKE, AND C. F. NIVEN, JR Physiology of the enterococci as related to their taxonomy. J. Bacteriol. 86: DEIBEL, R. H., AND C. F. NIVEN, JR Lipoic acid and the fermentation of serine and malate by Streptococcus faecalis. Bacteriol. Proc., p GRANT, W. M Colorimetric microdetermination of formic acid based on reduction to formaldehyde. Anal. Chem. 20: GUNSALUS, I. C Products of anaerobic glycerol fermentation by Streptococci [sic] faecalis. J. Bacteriol. 54: GUNSALUS, I. C Oxidative and transfer reactions of lipoic acid. Federation Proc. 13: GUNSALUS, I. C., AND C. F. NIVEN, JR The effect of ph on the lactic acid fermentation. J. Biol. Chem. 145: HORNBERGER, C. S., R. F. HEITMILLER, I. C. GUNSALUS, G. H. F. SCHNAKENBERG, AND L. J. REED Synthesis of D, L-alphalipoic acid. J. Am. Chem. Soc. 75: LIPMANN, F Metabolic generation and utilization of phosphate bond energy. Advan. Enzymol. 1: NEISH, A. C Analytical methods for bacterial fermentations. Rept. No (revised) Natl. Res. Council Canada. OSTER, M. O., AND N. P. WOOD Formatepyruvate exchange reaction in Streptococcus faecalis. II. Reaction conditions for cell extracts. J. Bacteriol. 87: PRICE, V. E., AND L. LEVINTOW Sodium pyruvate, p In E. G. Ball [ed.], Biochemical preparations, vol. 2. John Wiley & Sons, Inc., New York. SHATTOCK, P. M. F The identification and classification of Streptococcus faecalis and some associated streptococci. Ann. Inst. Pasteur Lille 7: SHERMAN, J. M The streptococci. Bacteriol. Rev. 1:3-97. SKADHAUGE, K Studies on enterococci with special reference to the serological properties. Einar Munksgaards Forlag, Copenhagen, Denmark. UMBREIT, W. W., R. H. BURRIS, AND J. F. STAUF- FER Manometric techniques. 3rd ed., p Burgess Publishing Co., Minneapolis. WOLFE, R. S., AND D. J. O'KANE Cofactors of the phosphoroclastic reaction of Clostridiuwn butyricum. J. Biol. Chem. 205: