Search

Transposon mutagenesis and marker rescue were used to isolate and identify an 8.5-kb contiguous region containing six open reading frames constituting the operon for the sorbitol P-enolpyruvate phosphotransferase transport system (PTS) of Streptococcus mutans LT11. The first gene, srlD, codes for sorbitol-6-phosphate dehydrogenase, followed downstream by srlR, coding for a transcriptional regulator; srlM, coding for a putative activator; and the srlA, srlE, and srlB genes, coding for the EIIC, EIIBC, and EIIA components of the sorbitol PTS, respectively. Among all sorbitol PTS operons characterized to date, the srlD gene is found after the genes coding for the EII components; thus, the location of the gene in S. mutans is unique. The SrlR protein is similar to several transcriptional regulators found in Bacillus spp. that contain PTS regulator domains (J. Stülke, M. Arnaud, G. Rapoport, and I. Martin-Verstraete, Mol. Microbiol. 28:865-874, 1998), and its gene overlaps the srlM gene by 1 bp. The arrangement of these two regulatory genes is unique, having not been reported for other bacteria.

Oxaloacetate decarboxylase was purified to 136-fold from the oral anaerobe Veillonella parvula. The purified enzyme was substantially free of contaminating enzymes or proteins. Maximum activity of the enzyme was exhibited at pH 7.0 for both carboxylation and decarboxylation. At this pH, the Km values for oxaloacetate and Mg2+ were at 0.06 and 0.17 mM, respectively, whereas the Km values for pyruvate, CO2, and Mg2+ were 3.3, 1.74, and 1.85 mM, respectively. Hyperbolic kinetics were observed with all of the aforementioned compounds. The Keq' was 2.13 X 10(-3) mM-1 favoring the decarboxylation of oxaloacetate. In the carboxylation step, avidin, acetyl coenzyme A, biotin, and coenzyme A were not required. ADP and NADH had no effect on either the carboxylation or decarboxylation step, but ATP inhibited the carboxylation step competitively and the decarboxylation step noncompetitively. These types of inhibition fitted well with the overall lactate metabolism of the non-carbohydrate-fermenting anaerobe.

A human oral strain of Actinomyces viscosus, GN431/75, was grown anaerobically in a defined medium in continuous culture with a glucose limitation at dilution rates (D) between 0.025 and 0.2 h-1 and with a nitrogen limitation at D = 0.005 and 0.1 h-1. With 5 mg of glucose per ml, the culture was limited for carbon at D = 0.025 and 0.05 h-1, but became nitrogen limited (asparagine) at D = 0.1 and 0.2 h-1. The molar growth yield (Yglucose) decreased from 50.0 to 40.9 g of cells per mol of glucose as the dilution rate was increased from 0.025 to 0.2 h-1, reflecting the limitation of asparagine. With high glucose and low amino acid concentrations (nitrogen limited), the cell yields at D = 0.05 and 0.1 h-1 were 37 to 33% lower than in the glucose-limited culture. The major products of metabolism were succinic and lactic acids with lesser amounts of acetic and formic acids and ethanol. The rate of glucose fermentation by resting cells removed from the glucose-limited culture and assayed in a pH stat increased with the dilution rate and was always higher than that for the fermentation of sucrose (60%) and fructose (40%). Activity for the glucose-P-enolpyruvate phosphotransferase system was observed in whole homogenates, with the highest activity evident at D = 0.1 h-1 with the glucose-limited culture. The observed activity was significantly lower than the rate of glucose metabolism at each dilution rate, suggesting that glucose-P-enolpyruvate phosphotransferase system was underestimated or that an additional transport system exists in the organism. The glucose-limited culture showed considerable ability to synthesize glycogen during the transition from carbon to nitrogen limitation, when 35% of the cell mass was present at this polymer. The organism was shown to possess the glycogen synthetic enzymes ADP glucose synthase and ADP glucose transferase, as well as the degradative enzyme maltodextrin phosphorylase. Washed cells of A. viscosus GN431/75 were shown to be relatively insensitive to the inhibiting actions of NaF in pH-fall and constant-pH experiments at all dilution rates. At pH 7.0, 25 mM NaF was required to completely inhibit glycolysis by glucose-limited cells at D = 0.05 h-1, whereas a concentration of only 11 mM NaF was required with cells of Streptococcus mutans grown and incubated under identical conditions. An interesting feature of the growth of A. viscosus GN431/75 in the chemostat was the shift from individual nonadherent cells at the low dilution rates to the appearance at D = 0.2 h-1 of large cell aggregates forming tenacious adherent films reminiscent of its characteristics in the oral cavity.

Streptococcus mutansDR0001 and a glucose-phosphotransferase (PTS)-defective mutant, DR0001/6, were grown anaerobically in a chemostat with a glucose limitation at dilution rates (D) of 0.04 to 0.6 h−1(mean generation time, 17 to 1.2 h). The mutant possessed only 15% of glucose-PTS activity of the wild type and gave cell yields (19%) less than those of the wild type. Glucose-PTS activity in strains DR0001 was maximum at D= 0.1 h−1and was adequate to account for transport in the chemostat at all dilution rates except D= 0.6 h−1, at which it was 80% of the actual glucose uptake activity. The mutant DR0001/6, on the other hand, possessed only sufficient glucose-PTS activity to sustain growth at below D= 0.1 h−1, indicating the presence of an alternate transport activity. This was confirmed in glycolytic rate experiments with washed cells, which demonstrated that the mutant showed rates 11- to 27-fold higher than that accountable via glucose-PTS activity alone. The wild-type organism contained both a high (Ks, 6.7 to 8.0 μM)- and a low (Ks, 57 to 125 μM)-affinity transport system, whereas the glucose-PTS-defective mutant contained only the low-affinity system (Ks, 62 to 133 μM). The glucose-PTS was shown to be the high-affinity system. Glucose uptake by the mutant was unaffected by 8 mM sodium arsenate, 10 mM azide, and 10 mM dinitrophenol but was completely inhibited by 0.05 mM sodium iodoacetate. Glycolysis in the organism was almost completely inhibited by 0.25 mM N′,N′ -dicyclohexylcarbodiimide (DCCD), indicating the involvement of an ATPase in glucose uptake. The ionophores carbonylcyanide-m-chlorophenylhydrazone and tetrachlorosali-cylanilide were inhibitory at concentrations of 10 μM, suggesting that a proton gradient was important in the transport process. Higher levels of DCCD and the ionophores were required to inhibit the wild-type organism to the same degree. A mechanism is proposed for the alternative transport system whereby proton motive force is created by the extrusion of protons by the DCCD-sensitive ATPase and glucose is transported down a proton gradient in a symport with protons.

Growth of Streptococcus mutans Ingbritt on limiting sucrose in a chemostat at dilution rates of 0.05 to 0.4 h-1 (mean generation time, 14 to 1.7 h) resulted in a heterofermentative pattern of metabolic end products. During fast growth, lactic acid was the major end product, whereas at slower growth rates, acetic and formic acids, as well as ethanol, increased to be major end products. The patterns obtained were similar to those seen with the same organism growing on glucose. The glycolytic rate by washed cells was maximum at the lowest dilution rates and decreased as the cells were made to grow faster. Transport of sucrose, glucose, and fructose via the phosphoenolpyruvate phosphotransferase system (PTS) was repressed during growth on sucrose after growth on glucose. Uptake rates suggested that sucrose was transported in the PTS as the intact disaccharide. Comparison of the rate of sugar uptake in the chemostat with the rate of PTS activity in the cells at each growth rate indicated that the PTS was capable of supporting growth only at a dilution rate of 0.05 h-1. Growth on sucrose at faster growth rates required the activity of a second transport system, supporting our earlier observations with glucose that S. mutans contains at least two sugar transport systems.

Longitudinal microbiological examinations have been made of dental plaque from a site approximal to the upper central incisors of 10 8-year-old children living in an area with water fluoridation. Differential counts of viable bacteria, made using a selective medium containing various levels of fluoride (0 to 100 μg/ml) at pH levels of 7.0 to 5.5, demonstrated an effect of both pH and fluoride on the numbers and types of bacteria isolated. Strains of Streptococcusand Neisseriagrew after only 16 h of incubation at pH levels as low as 6.0 with fluoride levels up to 50 μg/ml. The most commonly isolated streptococci were Streptococcus mitiorand S. salivarius. S. mutanswas isolated less frequently and was inhibited by 20 and 50 μg of fluoride per ml at pH 6.0 and 6.5, respectively. Veillonellastrains were the most resistant isolates, being isolated after 16 h of incubation on media at pH 6.0 with 100 μg of fluoride per ml. Despite their known fluoride resistance, Actinomycesspp. were often only detected on the selective media after 72 h of incubation. The pH of the medium had a definite selective effect, as the number of colonies growing on the fluoride-free basal media at pH 6.0 was only 30% of that at pH 7.0. Representative strains of S. mutans, S. mitior, S. sanguis, and S. milleriwere tested for their ability to utilize glucose at the pH and fluoride levels of the medium on which they were initially isolated. Fluoride reduced the initial glycolytic rate of the cells, but in 5 of the 13 strains tested the final amount of glucose used after 2 h of incubation was the same in the presence or absence of fluoride. The isolation of bacteria capable of growth in the presence of fluoride over a significant portion of the pH range that occurs in plaque in vivo could explain in part the finding that fluoride does not have a dramatic effect on the plaque community. Fluoride in plaque may reduce the ecological advantage afforded to aciduric S. mutansstrains by carbohydrate substances. In the in vivo situation this could mean that, even with high carbohydrate intake, fluoride may permit S. mitiorto compete with S. mutanswithin the plaque ecosystem.

A freshly isolated oral strain, Lactobacillus casei RB1014, was grown in continuous culture to compare the effects of pH and fluoride on growth and metabolism. The cells were grown at pH 7.0 to 3.2 in the absence of fluoride and from pH 7.0 to 5.4 with 20 mM NaF. Cell numbers varied from 3 X 10(9) to 30 X 10(9)/ml on blood agar during alterations in the growth pH from 7.0 to 4.27. Only when the culture was stressed by lowering the pH to 3.2 were cell numbers drastically reduced. Cells growing at pH 7.0 without fluoride were unable to grow when plated on fluoride agar (10.5 mM) at pH 5.5; however, when the growth pH was allowed to decrease to 4.94, cells grew on the fluoride plates in numbers equal to those growing on blood agar. This fluoride tolerance trait appeared rapidly once pH control was removed and was lost when the culture was returned to pH 7.0. The addition of 20 mM NaF to the culture medium did not adversely affect growth, provided that the pH was maintained at 6.0 or above; cells tolerant to 10.5 and 16 mM NaF appeared on pH 5.5 plates during this phase. In cells removed from the chemostat throughout the experiment and incubated at the pH of growth in a pH stat, glycolytic activity was optimum at pH 5.5 in the absence of NaF. Fluoride stimulated glycolytic activity by cells incubated at pH 7.0 and by cells growing with 20 mM NaF, provided that the pH of growth remained at or above 6.0. A more detailed examination of the adaptation to fluoride tolerance during shifts to acidic pH values revealed that cells capable of growth on acidic fluoride agar plates appeared within 2 h of the start of the fall in pH of the chemostat culture. Estimation of the intracellular pH during the period of the initial pH fall revealed that the intracellular pH was identical to the extracellular pH (i.e., no pH gradient [delta pH]), indicating that fluoride would not be transported into the cells to inhibit metabolism. However, once the pH of the medium was stabilized, delta pHs were generated, with the delta pH increasing as the pH declined. The inhibition of glycolysis by fluoride increased in proportion to the delta pH. Cells grown at pH 5.5 generated larger delta pHs than did cells grown at pH 7.0, although the values were normally small (approximately 0.9 U). The data suggest that the inherent fluoride tolerance of L. casei RB1014 was associated with relatively small delta pHs.

Streptococcus mutansIngbritt was grown anaerobically in a chemostat, at a rate (mean generation time, 13 h) similar to that in dental plaque, in a complex medium with excess glucose and at pH values of 6.5, 6.0, and 5.5. The yield of cells was constant at pH 6.5 and 6.0 (2.00 mg/ml) but fell to 1.25 at pH 5.5; Yglucosewas relatively constant under all conditions. Lactic acid was the major end product. Amino acid analysis of the culture supernatants indicated that growth was probably limited by the availability of cysteine. Cells were harvested and monitored for their capacity to produce acid from endogenous polysaccharide and exogenous sugars in the presence and absence of NaF, as well as for their glucose phosphoenolpyruvate (PEP)-phosphotransferase activity. Surprisingly, cells grown at pH 5.5 possessed two to three times more glycolytic activity, as measured by the rate of acid production, than cells grown at pH 6.5 and 6.0 when incubated in a washed suspension at constant pH with a sugar source. Furthermore, the cells grown at pH 5.5 were about twice as resistant to the effect of NaF in reducing the rate of acid production in this system. Fluoride inhibition could be reversed by increasing the pH of the system. Cells grown at all three pH values showed significant acid production from endogenous reserves, despite the fact that the glucoamylase-specific glycogen content of the cells dropped from 33% of the total carbohydrate during pH 6.5 growth to only 3% after growth at pH 6.0 and 6.5. Incubation of washed cells for 18 h in phosphate buffer resulted in the loss of 62% of the total carbohydrate, indicating that nonglycogen cellular polysaccharide was metabolized. A comparison of the fluoride effect on endogenous and exogenous metabolism under pH fall conditions showed that, with pH 6.5- and 6.0-grown cells, the inhibitor was more effective in the presence of an exogenous carbon source than in its absence. This effect was not seen with pH 5.5-grown cells. The decreased sensitivity of the pH 5.5-grown cells to fluoride was probably associated with the decreased glucose PEP-phosphotransferase activity (11%) in these cells compared with the activity of those grown at pH 6.5. This evidence supports the hypothesis that S. mutanspossesses at least two glucose transport systems, one of which is relatively fluoride insensitive.

A comparison was made of the properties of Streptococcus mutansIngbritt grown in continuous culture under conditions of excess glucose (nitrogen limitation) and limiting glucose at mean generation times of 1.7 to 14 h. Only low levels of glucoamylase-specific glycogen were formed in cells from either culture, and the total carbohydrate content of the cells under excess glucose was only at most 1.6-fold higher than in the glucose-limited culture. A negligible amount of cell-free polysaccharide was formed in either culture, although a significant level of glucosyltransferase activity was observed in both, with the highest activity at D= 0.2 and 0.4 h-1with a glucose limitation. Other differences were observed. (i) Lactate was the main end product of the glucose-excess culture, whereas acetate, formate, and ethanol were the main products of the glucose-limited culture except at a mean generation time of 1.5, when lactate represented 30% of the products. (ii) The yield (in grams per mole of glucose) of the latter culture was 2.6- to 4.0- fold higher than the yield of the glucose-excess culture. (iii) Washed cells from the glucose-limited culture were much more acidogenic (1.7- to 6.2-fold) than the glucose-excess cells when incubated with glucose, sucrose, and fructose. Endogenous glycolytic activity by the latter cells was significant, being 31 to 92% of the exogenous glucose rate at the four dilution rates. (iv) Cells from the glucose-excess culture were more insensitive to fluoride than cells from the glucose-limited culture. The NaF 50% inhibition dose values for the effect of fluoride on the metabolism of glucose, sucrose, and fructose were calculated for the four dilution rates at four pH values. This analysis indicated that rapidly metabolizing cells were more sensitive to fluoride than cells that metabolized the sugars more slowly.

Growth of Streptococcus mutans Ingbritt in continuous culture (pH 7.0, dilution rate of 0.1 h-1) at medium glucose concentrations above 2.6 mM resulted in repression of the sugar-specific membrane components, enzyme IIGlc (EIIGlc) and EIIMan, of the phosphoenolpyruvate:sugar phosphotransferase system (PTS). In one experiment, significant repression (27-fold) was observed with 73 mM glucose when the glycolytic capacity of the cells was reduced by only 2-fold and when the culture was still glucose limited. In a more comprehensive experiment in which cells were grown in continuous culture at eight glucose concentrations from 2.6 to 304 mM, in addition to repression of specific EII activities for glucose, mannose, 2-deoxyglucose, and fructose, synthesis of the general protein, EI, was repressed at all glucose levels above 2.6 mM to a maximum of 4-fold at 304 mM glucose when the culture was growing with excess glucose (i.e., nitrogen limited). The other PTS general protein, HPr, was less sensitive to the exogenous glucose level but was nevertheless repressed fourfold under glucose-excess conditions. The Km for glucose for EIIGlc increased from 0.22 mM during growth at 3.6 mM glucose (glucose limited) to 0.48 mM at 271 mM glucose (glucose excess). The shift from heterofermentation to homofermentation during growth with increasing glucose levels suggests the involvement of glycolytic intermediates, ATP, or another high-energy phosphate metabolite in regulation of the synthesis of the PTS components in S. mutans.

The membrane-bound, sugar-specific enzyme II (EII) component of the phosphoenolpyruvate:sugar phosphotransferase system (PTS) in Streptococcus mutans Ingbritt is repressed by growth on glucose under various conditions in continuous culture. Compared with optimal PTS conditions (i.e., glucose limitation, dilution rate [D] of 0.1 h-1, and pH 7.0), EII activity for glucose (EIIGlc) and mannose (EIIMan) in cells grown at a D of 0.4 h-1 and pH 5.5 with the same glucose concentration was reduced 24- to 27-fold. EII activity with methyl alpha-glucoside and 2-deoxyglucose was reduced 6- and 26-fold, respectively. Growth with excess glucose (i.e., nitrogen limitation) resulted in 26- to 88-fold repression of EII activity with these substrates. The above conditions of low pH, high dilution rate, and excess glucose also repressed EII activity for fructose (EIIFru), but to a lesser extent (two- to fivefold). Conversely, growth of S. mutans DR0001 at a D of 0.2 h-1 and pH 5.5 resulted in increased EIIGlc and EIIMan activity. Unlike the EII component, the HPr concentration in S. mutans Ingbritt varied only twofold (5.5 to 11.4 nmol/mg of protein) despite growth at pH 5.5 with limiting and excess glucose. The HPr concentrations in S. mutans DR0001 and the glucose-PTS-defective mutant DR0001/6 were similar. In a companion study, the soluble components of the PTS (i.e., HPr, EI, and EIIILac) in Streptococcus sobrinus grown on limiting lactose in a chemostat were not influenced significantly by growth at various pHs (7.0 and 5.0) and growth rates (D of 0.1, 0.54, and 0.8 h-1). However, growth on lactose resulted in repression of both EIIGlc and EIIFru, confirming earlier results with batch-grown cells. Thus, the glucose-PTS in some strains of S. mutans is regulated at the level of EII synthesis by certain environmental conditions.

The nonglycolytic, anaerobic organism Veillonella parvula M4 has been shown to contain an active pyruvate kinase. The enzyme was purified 126-fold and was shown by disc-gel electrophoresis to contain only two faint contaminating bands. The purified enzyme had a pH optimum of 7.0 in the forward direction and exhibited sigmoidal kinetics at varying concentrations o-f phosphoenol pyruvate (PEP), adenosine 5'-monophosphate (AMP), and Mg-2+ ions with S0.5 values of 1.5, 2.0, and 2.4 mM, respectively. Substrate inhibition was observed above 4 m PEP. Hill plots gave slope values (n) of 4.4 (PEP), 2.8 (adenosine 5'-diphosphate), and 2.0 (Mg-2+), indicating a high degree of cooperativity. The enzyme was inhibited non-competitively by adenosine 5'-triphosphate (Ki = 3.4 mM), and this inhibition was only slightly affected by increasing concentration of Mg-2+ ions to 30 mM. Competitive inhibition was observed with 3-phosphoglycerate, malate, and 2,3-diphosphoglycerate but only at higher inhibitor concentrations. The enzyme was activated by glucose-6-phosphate (P), fructose-6-P, fructose-1,6-diphosphate (P2), dihydroxyacetone-P, and AMP; the Hill coefficients were 2.2, 1.8, 1.5, 2.1, and 2.0, respectively. The presence of each these metabolites caused substrate velocity curves to change from sigmoidal to hyperbolic curves, and each was accompanied by an increase in the maximum activity, e.g., AMP greater than fructose-1,6-P2 greater than dihydroxyacetone-P greater than glucose-6-P greater than fructose-6-P. The activation constants for fructose-1,6-P2, AMP, and glucose-6-P were 0.3, 1.1, and 5.3 mM, respectively. The effect of 5 mM fructose-1,6-P2 was significantly different from the other compounds in that this metabolite was inhibitory between 1.2 and 3 mM PEP. Above this concentration, fructose-1,6-P2 activated the enzyme and abolished substrate inhibition by PEP. The enzyme was not affected by glucose, glyceraldehyde-3-P, 2-phosphoglycerate, lactate, malate, fumerate, succinate, and cyclic AMP. The results suggest that the pyruvate kinase from V. parvula M4 plays a central role in the control of gluconeogenesis in this organism by regulating the concentration of PEP.

The addition of lactose, galactose, or isopropyl-beta-D-thiogalactoside (IPTG) to glucose-grown cells of Streptococcus salivarius 25975 resulted in the co-induction of both the lactose-P-enolpyruvate phosphotransferase system (lactose-PTS) and beta-galactosidase, with the latter the predominant metabolic system. With various strains of Streptococcus mutans and Streptococcus sanguis 10556, on the other hand, the lactose-PTS was the major metabolic pathway with beta-galactosidase induced either to low or negligible levels. In all cases, induction of the lactose-PTS resulted in the concomitant induction of 6-P-beta-galactosidase. The induction by lactose of both the lactose-PTS and beta-galactosidase in all strains was repressed by glucose and other catabolites, notably, fructose. Induction of beta-galactosidase in S. salivarius 25975 by IPTG was, however, relatively resistant to glucose repression. Induction experiments with IPTG and lactose suggested that a cellular metabolite of lactose metabolism was a repressor of enzyme activity. Exogenous cAMP was shown to reverse the transient repression by glucose of beta-galactosidase induction in cells of S. salivarius 25975 receiving lactose, provided the cells were grown with small amounts of toluene to overcome the permeability barrier to this nucleotide, cAMP, was however, unable to overcome the permanent repression of beta-galactosidase activity to a significant extent under these conditions.

Streptococcus mutansIngbritt was grown anaerobically in a chemostat with a glucose limitation, as well as with an excess of glucose (amino acid limitation) at dilution rates (D) between 0.05 and 0.4 h−1(mean generation time = 12 to 1.5 h). The glucose-limited culture produced cells having 1.5- to 6.0-fold greater glycolytic activity than the cells from the glucose-excess culture. The preferred substrate for these cells was glucose, with the glycolytic rate for sucrose being only slightly lower; the rate for fructose was half that of glucose. The glycolytic rate of the glucose-limited cells was maximum at D= 0.1 h−1, with a decline in rate as the growth rate approached D= 0.4 h−1. A comparison of the activity of phosphoenolpyruvate phosphotransferase system (PTS) in the two types of cells showed that the glucose-limited cells had 1.7- to 5.6-fold greater PTS activity for the three sugars than the glucose-excess-grown cells. Whereas little difference was seen between the three sugars with the latter cells, the glucose-PTS had the greatest activity with glucose-limited cells, with the maximum in cells grown at D= 0.1 h−1. Comparison of the rate of sugar uptake in the chemostat with the rate of PTS transport activity in the cells at each growth rate demonstrated that only under conditions of slow growth with a glucose limitation was the PTS system capable of supporting growth on glucose. Furthermore, PTS activity in cells grown with an excess of glucose was insignificant when compared with glucose uptake during growth in the chemostat. This evidence supports the observation that S. mutanspossesses at least one other system, in addition to the PTS, for the transport of glucose into the cell. The organism was, however, devoid of glucose-proton symport transport activity.

Hamilton, I. R. (University of Wisconsin, Madison), R. H. Burris, P. W. Wilson, and C. H. Wang. Pyruvate metabolism and carbon dioxide assimilation by an Achromobacterspecies. J. Bacteriol. 89:647–653. 1965.—Carbon dioxide fixation by washed whole cells of AchromobacterN4-B has been observed during anaerobic pyruvate metabolism with both nitrogen- and NH4+-grown cells. Labeled sodium bicarbonate-C14was assimilated into cells by a mechanism requiring pyruvate under conditions of nitrogen fixation, nitrogenase induction, and assimilation of NH4+. Of the assimilated radioactivity, 89% appeared in six amino acids and two ninhydrin-positive unknown compounds, with the distribution of the label essentially independent of the nitrogen nutritional state of the organism. Aspartic and glutamic acids were the most highly labeled, with lesser amounts in glycine, alanine, ornithine, arginine, and the unknowns. All of the radioactivity extracted from these cells by ethanol-boiling water appeared in a protein fraction precipitated by 20% trichloroacetic acid. Radiorespirometric experiments with individually labeled pyruvate substrates demonstrated the preferential decarboxylation of the C-1 of pyruvate by this organism in a flowing helium gas phase. This decarboxylation was almost completely inhibited by using flowing nitrogen in place of helium; the addition of 0.5% CO2to the flowing nitrogen prevented inhibition and allowed 70% of the expected CO2evolution. These results, coupled with those from growth experiments, indicate a carbon dioxide requirement for anaerobic growth and pyruvate metabolism, which appears to be coupled to the formation of protein precursors.