Metabolism of Carbohydrates by Pasteurella Pseudotuberculosis' R

Home , Gluconokinase, Glucose oxidase

JOURNAL OF BACrERiOLOGY, May 1968, p. 1698-1705 Vol. 95, No. 5 Copyright 0) 1968 American Society for Microbiology Printed In U.S.A. Metabolism of Carbohydrates by Pasteurella pseudotuberculosis' R. R. BRUBAKER2 Department ofMicrobiology, The University ofChicago, Chicago, Illinois 60637 Received for publication 26 February 1968

Cell-free extracts of Pasteurella pseudotuberculosis and P. pestis catalyzed a rapid and reversible exchange of electrons between pyridine nucleotides. Although the extent of this exchange approximated that promoted by the soluble nicotinamide adenine dinucleotide (phosphate) transhydrogenase of Pseudomonas fluorescens, the reaction in the pasteurellae was associated with a particulate fraction and was not influenced by adenosine-2'-monophosphate. The ability of P. pseudotuberculosis to utilize this system for the maintenance of a large pool of nicotinamide adenine dinucleotide phosphate could not be correlated with significant participation of the Entner-Doudoroff path or catabolic use of the hexose-monophosphate path during metabolism of glucose. As judged by the distribution of radioactivity in metabolic pyruvate, glucose and gluconate were fermented via the Embden-Meyerhof and Entner-Doudoroff paths, respectively. With the exception of hexosediphosphatase, all enzymes of the three paths were detected, although little or no gluconokinase or phosphogluconate dehydrase was present unless the organisms were cultivated with gluconate. The significance of these findings is discussed with respect to the regulation of carbohydrate metabolism in the pasteurellae, related enteric bacteria, and P. fluorescens.

Glucose is catabolized by Pasteurella pestis, fate of glucose-6-phosphate in P. pestis. The the causative agent of bubonic plague, via the possibility exists, however, that excess NADP Embden-Meyerhof path (29), whereas gluconate in P. pseudotuberculosis might increase the is metabolized by a combination of the Entner- turnover of glucose-6-phosphate dehydrogenase Doudoroff and hexose-monophosphate paths and thus permit hexose catabolism to occur via (24). Due to a deficiency of glucose-6-phosphate the nonglycolytic paths (8). The purpose of this dehydrogenase (2, 24, 25), P. pestis cannot utilize study was to define the central paths of carbo- the nonglycolytic paths during growth on hexose. hydrate metabolism in P. pseudotuberculosis In contrast, this key enzyme is present in P. and to determine the effect, if any, of the trans- pseudotuberculosis (25), an organism very closely hydrogenase activity on the participation of related to P. pestis (27). Few studies have been these paths. reported on the catabolism of carbohydrates by P. pseudotuberculosis; however, the biosynthesis MATERIALS AND METHODS of uncommon sugar compounds by this organism Bacteria. P. pseudotuberculosis strain PBI/+ was has received intensive study (23, 31). isolated by Burrows and Bacon (6) during a spon- Cell-free extracts of P. pestis and P. pseudo- taneous epidemic in laboratory guinea pigs. The tuberculosis catalyze a rapid oxidation of reduced properties of virulent (V+) and avirulent (V-) cells of this strain have already been described (5, 6). nicotinamide adenine dinucleotide phosphate P. pestis strain EV76, Pseudomonas fluorescens strain (NADPH2) at the expense of nicotinamide ade- B-10, and Escherichia coli strain B were used in com- nine dinucleotide (NAD). The resulting large parative experiments. pool of nicotinamide adenine dinucleotide phos- Media and cultivation. A modification of the salt phate (NADP) cannot, because of the metabolic component of Higuchi and Carlin (15), consisting of block already noted, influence the metabolic 0.025 M K2HPO4, 0.01 M citric acid, 0.0025 M MgC92, and 0.0001 M FeC12, was used in all culture media. I Presented in part at the 64th Annual Meeting Flasks (1-liter) containing 100 ml of this base plus of the American Society for Microbiology, Wash- 4 g of Casitone (Difco) were autoclaved and neu- ington, D.C., 3-7 May 1964. tralized with NaOH. Then the medium received 1 2 Present address: Department of Microbiology ml of stock solutions of 0.25 M CaCl2, 0.25 M Na2S2O3, and Public Health, Michigan State University, East and 1.0 M carbohydrate (sterilized by filtration). Lansing, Mich. 48823. After inoculation with 107 cells per rnl, the cultures 1698 VOL. 95,1968 CARBOHYDRATE METABOLISM IN P. PSEUDOTUBERCULOSIS 1699 were vigorously aerated at 37 C on a wrist-action Results are expressed in terms of specific activity or shaker, and the organisms were harvested during the units of enzyme per mg of protein. early stationary phase by centrifugation at 27,000 X g The rate of formation of glucose-6-phosphate was for 10 min in the cold. measured in 33.3 mM potassium phosphate buffer Isotopic experiments. Methods for the isolation and (pH 7.2), 3.3 mM MgCl2, 3.3 mM L-cysteine (pH 7.2), degradation of metabolic pyruvate were similar to 0.33 mm NADP, and excess yeast glucose-6-phosphate those used by Fraenkel and Horecker (12) and dehydrogenase. Phosphoglucomutase, glucosephos- Eisenberg and Dobrogosz (10). Bacteria were culti- phate isomerase, and glucokinase were determined in vated with glucose or gluconate, washed once with this system by the addition of 3.3 mm each of glucose- ;0.033 M potassium phosphate buffer (pH 7.2), and 1-phosphate, fructose-6-phosphate, and glucose, then suspended to yield a final concentration of 2 mg plus adenosine triphosphate (ATP), respectively. (dry weight) per ml in a reaction mixture containing The basal incubation mixture, without added glucose- 0.05 M potassium phosphate buffer (pH 7.0) and 6-phosphate dehydrogenase, was used to determine '0.002 M sodium arsenite. After aeration for 10 min endogenous levels of this enzyme plus phospho- -at 37 C, the volume was brought to 5 ml by addition gluconate dehydrogenase by the method of Glock of 0.005 M radioactive glucose or gluconate. Utili- and McLean (14); gluconokinase was also estimated zation of substrate and accumulation of pyruvate in the presence of 3.3 mm potassium gluconate, 3.3 -were complete after further incubation for 1 hr. The mM ATP, and the excess phosphogluconate dehy- incubation mixture was then acidified with 1 ml of drogenase present in 1.5 mg of protein obtained from 5 N HC104 and centrifuged at 27,000 X g for 10 min; glucose-adapted cells of E. coli. the clear supernatant fluid was neutralized by drop- Glyceraldehyde phosphate was estimated in a wise addition of 10 N KOH. After filtration to remove reaction mixture consisting of 73 mm Tris buffer KC104, pyruvate was assayed in a portion of the (pH 8.0), 10 mm sodium arsenate, 3.3 mM L-cysteine sample with lactic dehydrogenase, and other radio- (pH 7.8), 1 mM MgCl2, and 0.33 mM NADP. This -active products were determined by cochromatog- system was used to determine endogenous glyceral- raphy and radioautography. dehydephosphate dehydrogenase after addition of Essentially all radioactivitity was associated with 3 mm glyceraldehydephosphate. An excess of this pyruvate, although insignificant amounts of radio- enzyme in the presence of 3.3 mm fructose diphosphate active mannitol were sometimes noted. Therefore, the or dihydroxyacetone phosphate was used to measure radioactivity remaining in the incubation mixture was aldolase and triosephosphate isomerase, respectively. estimated directly (as pyruvate), and samples were The production of pyruvate was followed in an degraded with ceric sulfate to decarboxylate a-keto incubation mixture consisting of 33 mm potassium acids (12, 20). Evolved CO2 was collected in 0.2 ml phosphate buffer (pH 7.2), 3.3 mM reduced gluta- of 1 N KOH, transferred to a closed vessel, acidified thione, 3.3 mM MgCl2, and 0.17 mm reduced nicotin- with excess H2SO4, and collected in 1.0 ml of 1 N amide adenine dinucleotide (NADH2). Pyruvate Hyamine hydroxide in methanol. Radioactivity was kinase was determined with this reaction system in measured with a Packard Tri-Carb liquid scintilla- the presence of excess lactic dehydrogenase, plus tion counter (Packard Instrument Co., Inc., Downers 3.3 mm phosphoenolypyruvate, and 3.3 mm adenosine Grove, Ill.) in Bray's scintillation liquid (3). No more diphosphate; endogenous lactic dehydrogenase was than 0.1 ml of aqueous sample (in 2 ml of methanol) assayed directly with 3.3 mm potassium pyruvate. was mixed with scintillation fluid. The same proce- Phosphofructokinase was determined by estima- dures were used to determine '4CO2 released during tion of triose phosphate (28); pyruvate dehydrogenase, metabolism of glucose-1-'4C and glucose-6-14C. phosphate acetyltransferase, and acetate kinase were Cell-free extracts. Washed whole cells were sus- measured by minor modifications of the procedures pended to a concentration of approximately 101" outlined by Colowick and Kaplan (7). Methods for per ml in cold 0.1 M tris(hydroxymethyl)amino- determination of enzymes of the Entner-Doudoroff methane (Tris buffer, pH 7.8). After treatment for path and for qualitative assay of phosphoglycerate 1 min with an ultrasonic probe (Instrumentation kinase, phosphoglyceromutase, and phosphopyruvate Associates, New York, N.Y.), cellular debris was hydratase are given in Results. The exchange of elec- removed by centrifugation at 27,000 X g for 10 min, trons between pyridine nucleotides was measured and the resulting clear extract was dialyzed overnight by a procedure similar to that used by Eagon (8, 9). against cold 0.01 M potassium phosphate buffer Chromatography. Reactions involving the destruc- plus 0.001 M mercaptoethanol. tion of ribose-5-phosphate were stopped by addition Enzyme determinations. Most enzymes were deter- of 1 volume of 1 M HC104 to the reaction mixture. mined by methods similar to those used by Mortlock After removal of protein by centrifugation, the sam- (24) and Vandermark and Wood (32). Unless stated ples were neutralized with 5 N KOH, filtered, and then otherwise, a volume of 3 ml was used for spectro- adjusted to pH 5.5 by dropwise addition of 1 N photometric assays. Reactions linked to pyridine acetic acid. To each 1 ml of sample was added 1 mg nucleotides were followed at 340 m/M with a Beckman of acid phosphatase and, after incubation for 1 hr DU spectrophotometer (Beckman Instruments, Inc., at 25 C, the solution was brought to dryness on a Fullerton, Calif.) and were corrected, when necessary, steam bath. The free sugars were then extracted in for nonspecific oxidase or reductase activity. In all 5 ml of pyridine at 100 C for 10 min, the resulting quantitative determinations, one unit of activity is solutions were filtered, and then pyridine was removed defined as that amount of enzyme necessary to con- by distillation (22). The residue was dissolved in 10% vert 1 pmole of substrate to product per min at 25 C. isopropanol, and samples were chromatographed 1700 BRUBAKER J. BA=rRioL. on Whatman no. 1 paper in ascending systems com- NADPH2 was slowly oxidized under comparable posed of n-butyl alcohol-ethyl alcohol-water (13:8:4, conditions (Table 1). Although equine cyto- v/v), ethyl acetate-acetic acid-water (3:1:3, v/v), chrome c was reduced by NADH2 (specific ac- ethyl acetate-pyridine-water (2:1:2, v/v), and water- saturated benzyl alcohol-acetic acid-water (3:1:3, tivity = 0.137), NADPH2 did not promote v/v). After chromatography, the sugars were located significant reduction of this compound. Since by spraying with p-anisidine-HCl or ammoniacal similar results were obtained with analogous silver nitrate. artificial electron acceptors, P. pseudotuberculosis Chemical determinations. Pentose and heptulose probably possessed a respiratory NADH2 de- were assayed by the orcinol reaction (16), and the hydrogenase, whereas the existence of a NADPH2 cysteine-H2SO4 method was used for the determina- dehydrogenase appeared unlikely. tion of hexose (1). Triose phosphate was measured Nevertheless, NADPH2 was rapidly oxidized as alkali-labile organic phosphate by addition of 0.5 by NAD, and the reverse reaction could be Ml of 1 N NaOH to 0.1 ml of sample; after incubation for 15 min at 25 C, the samples were neutralized and followed in the presence of endogenous gluta- estimated by the method of Fiske and SubbaRow thione reductase and added oxidized gluta- (11). Protein was assayed by the method of Lowry thione. This activity was not influenced by et al. (21). adenosine-2'-monophosphate or other nucleo- Reagents. All purified enzymes, coenzymes, sugars tides tested, and at least one component of the and phosphorylated intermediates were purchased system was removed from solution by centrifuga- from the Sigma Chemical Co. (St. Louis, Mo.), tion at 144,000 X g for 60 min or by passage with the exception of 2-keto-3-deoxyphosphogluco- through a Millipore filter (Table 1). Similar re- nate which was received through the courtesy of sults were obtained from P. pestis and E. coli, R. P. Mortlock. Radioactive compounds were obtained from the Volk Radiochemical Co. (Skokie, although the difference between specific activities Ill.). in the latter species did not approach that of the RESULTS Pasteurella species (Table 2). In contrast, the rate of electron exchange catalyzed by the soluble Preparations of broken V+ cells utilized about nucleotide-dependent NAD(P) transhydrogenase six times as much oxygen in the presence of of P. fluorescens (18) was similar to that noted excess NADH2 (0.061 ,umole of oxygen atoms for the particulate activities in P. pseudotubercu- per mg of protein per min) as was used when losis and P. pestis. Further study was directed NADPH2 served as reductant (0.009 ,umole per toward defining the significance, if any, of these mg of protein per min). The organisms possessed transhydrogenase systems with regard to the an active NADH2 oxidase system, whereas regulation of hexose metabolism. Radioactive glucose or gluconate was added TABLE 1. Evidence for the presence of NADH2 to suspensions of resting cells, and the distribu- oxidase and pyridine nucleotide trans- tion of 14C in metabolic pyruvate was determined. hydrogenase systems in normal and The carboxyl carbon of pyruvate originating centrifuged (144,000 X g, 60 min) extracts ofglucose-adapted cells of from glucose-3,4-14C (in glucose-adapted cells) V+ Pasteurella pseudotuberculosisa or gluconate-1-14C (in gluconate-adapted cells) was heavily labeled (Table 3), which indicated Specific activity that these compounds were fermented via the Addition to reaction mixture TABLE 2. Determination of NAD(P) transhy- Normal cedntri drogenase in Pseudomonasfluorescens etatextract and analogous activities in Pasteurella pestis and Escherichia colia NADH2 ...... 0.074 0.012 NADPH2...... 0.004 0.004 Specific activity NADPH2, NAD .. 0.054 Addition to reaction mixture NADH2, GSSG.. 0.090 0.027 NADPH2, GSSG . . 0.538 1.519 rescens P. pestis E. coli NADH2, GSSG, NADP.. 0.235 0.031 NADH2 ...... 0.126 0.052 0.178 a The reaction mixture contained 100 ,umoles of NADPH2. .0...... 0.016 0.002 0.010 Tris buffer (pH 7.6), 0.4 mg of protein (normal NADPH2, NAD...... 0.087 0.030 0.039 extract) or 0.51 mg of protein (centrifuged ex- NADH2, GSSG ...... 0.140 0.060 0.195 tract), excess glutathione reductase, and, where NADPH2, GSSG ...... 0.607 0.521 0.538 indicated, 0.5 ,.mole of NADH2 or NADPH2, NADH2, GSSG, NADP 0.385 0.215 0.246 0.1 jAmole of NAD or NADP, and 10 /Amoles of GSSG (oxidized glutathione) in a volume of 3.0 a The reaction mixture was identical to that ml. described for Table 1. VOL. 95, 1968 CARBOHYDRATE METABOLISM IN P. PSEUDOTUBERCULOSIS 1701 TABLE 3. Distribution of radioactivity in pyruvate formed during the fermentation of glucose-J-'4C, glucose-3 ,4-14C and gluconate-1-l4C

Percentage of recovered radioactivity Pyruvate Organism CarbohydrateCarbohydrateiningrowth______a_growt Fermentation substrate" formed Total Carboxyl (pumoles/ml) pyruvate carbon ofb

Pasterella pseudotubercu- Glucose Glucose-1-14C 31 3 7.3 losis Glucose Glucose-3,4-14C 39 34 7.5 Gluconate Gluconate-1-_4C 24 19 8.0 Gluconate Glucose-l-14C 38 8 7.4 Escherichia coli Glucose Glucose-1-14C 29 4 6.8 Glucose Glucose-3,4-14C 27 22 6.9 Gluconate Gluconate-1-_4C 34 26 6.9 Gluconate Glucose-1-'4C 32 7 7.3 Pseudomonas fluorescens Glucose Glucose-1-14C 17 13 5.8 Gluconate Gluconate-1-'4C 13 10 4.2 a Initial radioactivity for glucose-1-'4C, glucose-3,4-'4C, and gluconate-1-14C was 4.1 X 104, 3.9 X 104I and 4.1 X 104 count/min per ml, respectively. bMeasured as 14CO2 liberated after ceric sulfate degradation.

Embden-Meyerhof and Entner-Doudoroff paths, The velocities obtained during a typical de- respectively. Similar results were obtained in termination of phosphogluconate dehydrase and control experiments performed with E. coli. 2-keto-3-deoxyphosphogluconate aldolase in ex- P. fluorescens yielded carboxyl-labeled pyruvate tracts of gluconate adapted cells are shown in after fermentation of both glucose-J1-4C and Fig. 1. The aldolase appeared constitutive; gluconate-J-'4C. specific activities of 0.139 and 0.128 were ob- To obtain an approximate estimation of the tained for extracts of glucose- and gluconate- percentage participation of the Embden-Meyer- adapted cells, respectively. In contrast, the hof, Entner-Doudoroff, and hexose-monophos- dehydrase was not detected in glucose-adapted phate paths, "'CO2 released from glucose-1-"4C cells, whereas a specific activity of 0.029 was and glucose-6-'4C (in glucose-adapted cells) recorded for gluconate-adapted cells. Similarly, was collected over periods of 30 min. As judged little or no gluconokinase was found in glucose- by differences in rates of evolution (33), a value adapted cells; this enzyme, however, yielded a not exceeding 20% participation was obtained specific activity of 0.304 in extracts of organisms for the hexose-monophosphate path in E. coli cultivated with gluconate. and V+ and V- cells of P. pseudotuberculosis. The enzymes of the hexose-monophosphate Glucose- and gluconate-adapted cells released pathway did not appear to be under regulatory "'CO2 from glucose-1-"4C at similar rates, whereas control. The rate of the formation of glyceralde- gluconate was not significantly metabolized by hyde phosphate from ribose-5-phosphate (via glucose-adapted cells. Gluconate-adapted or- the action of ribulosephosphate-3-epimerase, ganisms released 4(CO2 from gluconate-1-14C ribulose-phosphate isomerase, and transketolase) at a rate which was essentially identical to that was not altered by cultivation with 5 different obtained with glucose-J-'IC. sources of energy (Fig. 2). The products formed These results indicated that the ability of P. during the destruction of ribose-5-phosphate by pseudotuberculosis to maintain NADP in the an extract of glucose-adapted cells are given in oxidized state did not, in itself, permit the Table 4. The generation of hexose, coupled with catabolism of hexose to occur via the nongly- the appearance and disappearance of sedo- colytic paths. This restriction, however, may have heptulose as the system approached equilibrium, been caused by a metabolic block, perhaps indicated the presence of transaldolase. Essen- analogous to that in P. pestis. Thus, the presence tially identical velocities were obtained when this of the individual enzymes of the nonglycolytic experiment was repeated with extracts of glu- paths was demonstrated by direct or indirect conate- and xylose-adapted cells. Attempts to determination. demonstrate the presence of phosphoketolase in 1702 BRUBAKER J. BACTERIOL.

0 a 0 cm a 0i

0 1 2 3 4 5 MINUTES FIG. 1. Determination of phosphogluconate de- 0 1 2 3 4 5 hydrase and 2-keto-3-deoxy-phosphogluconate aldolase in an extract of gluconate-adapted V+ cells of Pas- MINUTES teurella pseudotuberculosis. The reaction mixture, in FIG. 2. Formation of glyceraldehyde phosphate a total volume of 1.0 ml, contained 75 jAmoles of Tris from ribose-5-phosphate by extracts of V+ cells culti- (pH 8.5, 3 ,umoles of reduced glutathione, 0.5 ,umole vated with glucose, galactose, gluconate, xylose, and of NADH2, 0.14 mg of protein, excess lactic dehy- glycerol. The reaction mixture, in a volume of 3.0 ml, drogenase, and either 5 jmoles of potassium phospho- contained 225 ,umoles of Tris (pH 8.5), 10 ,imoles of gluconate (LI) or 5 j,moles of potassium 2-keto-3- L-cysteine (pH 7.0), 2 lumoles ofMgCI2, 30 ,imoles OJ deoxyphosphogluconate (A); 0 represents the sodium arsenate (pH 8.5), bacterial extract (I mg), corresponding velocity obtained with potassium py- and 10 ,umoles ofribose-5-phosphate. Velocities for all ruvate. extracts fell within the shaded area. P. pseudotuberculosis by formation of hydrox- amates were not successful. Significant hexosedi- Table 5. Velocities obtained during qualitative phosphatase could not be detected in extracts of assays of the latter are shown in Fig. 3. Of special V+ or V- cells grown with glucose, gluconate, or interest was the finding that the specific activity glycerol. This determination, based on the of glucokinase in V+ cells was generally about generation of NADPH2 in the presence of added four times greater than that of V- cells. fructose diphosphate, glucosephosphate isomerase, and glucose-6-phosphate dehydrogenase, DISCUSSION yielded a specific activity of 0.018 for a similar The factors which control the metabolic fate extract of glycerol-adapted cells of E. coli. of hexose in bacteria are not completely under- Specific activities between 0.163 and 0.106 were stood. Salmonella typhimurium (12) and E. coli obtained for glucose-6-phosphate dehydrogenase (10, 13) utilized the Embden-Meyerhof path for in extracts of V+ and V- cells cultivated with a fermentation of glucose, whereas the Entner- variety of carbohydrates, and a slightly lower Doudoroff path was used during growth on range was recorded for phosphogluconate de- gluconate. Evidence for the existence of this hydrogenase. catabolic pattern in other organisms has been All glycolytic enzymes were detected and reported (13, 30). In contrast, many pseudo- their specific activities, with the exceptions of monads and related bacteria ferment both glu- phosphoglycerate kinase, phosphoglyceromutase, cose and gluconate via the Entner-Doudoroff and phosphopyruvate hydratase, are given in path; these species exhibit a glycolytic block, VOL. 95, 1968 CARBOHYDRATE METABOLISM IN P. PSEUDOTUBERCULOSIS 1703 TABLE 4. Products formed during degradation of ribose-5-phosphate by an extract of glucose-adapted cells of V+ Pasteurella pseudotuberculosis' Time of incubation Component determinedb 0 min 5 min 10 min 20 min

Pentose...... 10.0 7.1 6.0 4.5 Triose phosphate ..... 0 1.2 1.1 1.0 Hexosec ...... 0 1.4 1.8 1.0 Heptulosed ...... 0 0.5 0.6 0.2 .200 a Reaction mixtures contained 50 ,umoles of ribose-5-phosphate, 12.5 ,umoles of MgC12, 250 ,umoles of Tris (pH 7.7), and approximately 5.0 mg of protein per 5.0 ml. Samples were removed for assay at the times indicated. .100 b Pentose equivalents (jumoles of carbon/5). cApproximately 90% fructose and 10% glucose, as judged by chromatographic assay. d Identified as sedoheptulose by chromatographic assay. .000 0 2 4 6 8 TABLE 5. Glycolytic enzymes in extracts ofglucose- MINUTES adapted cells of Pasteurella pseudotuberculosis FIG. 3. Determination of phosphoglycerate kinase, phosphoglyceromutase, and phosphopyruvate hydratase Specific activity in an extract ofglucose-adapted V+ cells ofPasteurella Enzyme pseudotuberculosis. The reaction mixture, in a total V+ cells V- cells volume of 3.0 ml, contained 225 ,umoles of Tris (pH 7.4), 10 ,moles ofA TP, 2 Mnoles of MgCl2, 10 j,moles of L-cysteine (pH 7.0), 0.5 ,umole of NADH2, 0.21 Glucokinase.0.220 0.052 mg of protein, excess glyceraldehydephosphate dehy- Phosphoglucomutase...... 0.720 0.684 drogenase, and 0.1 ml of a 0.1 M (or saturated) solu- Glucosephosphate isomerase 0.9650 1 .065 tion of either potassium 3-phosphoglycerate (0), Phosphofructokinase . 0.297 0.324 potassium 2-phosphoglycerate (A), or potassium Aldolase.0.193 0.229 phosphoeniolpyruvate (O ). The latter reaction reflects Triosephosphate isomerase... 0.286 0.317 dismutation phosphoenolpyruvate to diphospho- Glyceraldehydephosphate de- of hydrogenase. 2.120 2.255 glycerate and pyruvate. Pyruvate kinase.0.256 0.278 Lactic dehydrogenase. 0.226 0.314 showed that both substrates induced phospho- Pyruvate dehydrogenase. 0.018 0.021 gluconate dehydrase in P. fluorescens and sug- Phosphate acetyltransferase. 0.227 0.239 gested that gluconate, arising from glucose via Acetate kinase.0.052 0.074 the action of glucose oxidase, was the actual Determined as triose phosphate. inducer. Eagon (8) noted that cell-free extracts of pseudomonads and related nonglycolytic bac- probably at the level of phosphofructokinase teria contained NAD(P) transhydrogenase or (19). However, the percentage of participation respiratory NADPH2 dehydrogenase, whereas of the Entner-Doudoroff or hexose-monophos- similar preparations of glycolytic bacteria did phate paths was not increased in mutants of S. not. On this basis, Eagon proposed that the typhimurium (12) or E. coli (13) possessing restricted catabolism of hexose via the nongly- analogous deficiencies of glucosephosphate isom- colytic paths of the latter species was caused by erase; the growth of these mutants on glucose, an inability to maintain a sufficient pool of but not gluconate, was significantly slower than NADP to permit saturation of glucose-6-phos- that of the prototrophs. In both cases, gluconoki- phase dehydrogenase. In accord with this hy- nase and phosphogluconate dehydrogenase were pothesis was the finding that methylene blue and markedly induced by gluconate but not by pure oxygen promoted a dramatic and probably glucose. But Eisenberg and Dobrogosz (10) analogous increase in participation of the hexose- 1704 BRUBAKER J. BACTERIOL, monophosphate path in erythrocytes (4) and Doudoroff path can be attributed to a deficiency species of Bacillus (26), respectively. Moreover, of phosphogluconate dehydrase. However, even levels of key enzymes controlling the flow of in- when this enzyme was induced, as in gluconate- termediates and factors regulating the size of the adapted cells, hexose was almost exclusively NADP pool may prove to be equally important catabolyzed by the Embden-Meyerhof path in determining the catabolic fate of hexose. (Table 3). Similar results, previously reported It is well established that NADH2 is generally for S. typhimurium (12) and E. coli (10), may involved in catabolic reduction, whereas NADPH2 reflect a respective high and low affinity of phos- usually supplies reducing power for biosynthetic phogluconate dehydrogenase and phosphoglu- reactions (17). The presence of a NADH2 de- conate dehydrase for their common substrate hydrogenase in P. pseudotuberculosis is con- (12); the same situation may exist in the case of sistent with this concept and was suggested by P. pseudotuberculosis. It should be noted, how- the determinations given in Table 1. However, ever, that the soluble NAD(P) transhydrogenase the unexpected finding that this organism could of P. fluorescens can directly reoxidize NADPH2 rapidly oxidize NADPH2 in the presence of NAD generated by the cytoplasmic NADP-linked indicated that the former could indirectly initiate dehydrogenases. Less efficient reoxidation might the process of respiration. As a result of this be expected in P. pseudotuberculosis, where at exchange of electrons, it seemed reasonable to least one component of the transhydrogenase assume that the level of NADP in P. pseudo- activity is particulate in nature and may, there- tuberculosis might be sufficiently high to favor fore, be associated with the cytoplasmic mem- use of the nonglycolytic paths during growth on brane. hexose. The low value obtained for the percentage of ACKNOWLEDGMENT participation of the hexose-monophosphate path This investigation was supported by Public Health in P. pseudotuberculosis probably reflects an Service Graduate Training Grant GM-603 from the essentially anabolic function similar to that National Institute of General Medical Sciences. described in E. coli (13); nearly identical results were recorded in this study for both organisms. LITERATURE CITED The ability to form glyceraldehyde phosphate 1. ASHWELL, G. 1957. Colorimetric analysis of (Fig. 2) or hexose, heptulose, or triose phosphates sugars, p. 73-105. In S. P. Colowick and N. 0. (Table 4) from ribose-5-phosphate did not seem Kaplan [ed.], Methods in enzymology, vol. 3. to be influenced by the source of energy used Academic Press, Inc., New York. during cultivation. One factor that may have 2. BOWMAN, J. E., R. R. BRUBAKER, H. FRISCHER, limited the of AND P. E. CARSON. 1967. Characterization of participation the hexose-mono- enterobacteria by starch-gel electrophoresis of phosphate path in P. pseudotuberculosis was the glucose-6-phosphate dehydrogenase and phos- apparent deficiency of hexosediphosphatase, an phogluconate dehydrogenase. J. Bacteriol. enzyme necessary for the complete cyclic oxida- 94:544-551. tion of hexose. The inability to observe hexosedi- 3. BRAY, G. A. 1960. A simple efficient liquid scin- phosphatase probably reflects an instability that tillator for counting aqueous solutions in a is not uncommon for this enzyme, which, ac- liquid scintillation counter. Anal. Biochem. cording to current knowledge, is also essential 1:279-285. for the biosynthesis of hexose and pentose during 4. BRIN, M., AND R. H. YONEMOTO. 1958. Stimula- growth on triose. tion of the glucose oxidative pathway in human During the oxidation of glucose erythrocytes by methylene blue. J. Biol. Chem. it is still possible, even in the absence of hexosedi- 230:307-317. phosphatase, for accumulated triose phosphate 5. BRUBAKER, R. R. 1967. Growth of Pasteurella to enter the glycolytic path. Fraenkel and pseudotuberculosis in simulated intra- and ex- Horecker (12) suggested that the capacity of the tracellular environments. J. Infect. Diseases hexose-monophosphate path in S. typhimurium 117:403-417. was constant and 6. BURROWS, T. W., AND G. A. BACON. 1960. V and limited by the turnover of W antigens in strains of Pasteurella pseudo- phosphogluconate dehydrogenase. This may also tuberculosis. Brit. J. Exptl. Pathol. 39:278-291. be true in P. pseudotuberculosis. 7. COLOWICK, S. P., AND N. 0. KAPLAN [ed.]. 1955. The ability of P. pseudotuberculosis to main- Methods in enzymology, vol. 1. Academic tain a large pool of NADP did not permit the Press, Inc., New York. fermentation of hexose to occur via the Entner- 8. EAGON, R. G. 1963. Rate limiting effects of pyri- Doudoroff path, although the organism dine nucleotides on carbohydrate catabolic possessed pathways of microorganisms. Biochem. the requisite structural genes. The inability of Biophys. Res. Commun. 12:274-279. glucose-adapted cells to utilize the Entner- 9. EAGON, R. G. 1962. Pyridine nucleotide-linked VOL. 95, 1968 CARBOHYDRATE METABOLISM IN P. PSEUDOTUBERCULOSIS 1705

reactions of Pseudomonas natriegens. J. Bac- menit with the Folin phenol reagent. J. Biol. teriol. 84:819-821. Chem. 193:265-275. 10. EISENBERG, R. C., AND W. J. DOBROGOSZ. 1967. 22. MALPRESS, F. H., AND A. B. MORRISON. 1949. Gluconate metabolism in Escherichia coli. J. Use of pyridine in the deionization of solutions Bacteriol. 93:941-949. for paper chromatography. Nature 164:963. 11. FiSKE, C. H., AND Y. SUBBAROW. 1925. The 23. MATSUHASHI, S., AND J. L. STROMINGER. 1967. colorimetric determination of phosphorus. J. Isolation of thymidine diphosphate D-glucose, Biol. Chem. 66:375-450. thymidine diphosphate D-galactose, and thymi- 12. FRAENKEL, D. G., AND B. L. HORECKER. 1964. dine diphosphate 4-acetamido-4,6-dideoxy-D- Pathways of D-glucose metabolism in Sal- galactose from Pasteurella pseudotuberculosis. monella typhimurium: A study of a mutant J. Bacteriol. 93:2017-2019. lacking phosphoglucose isomerase. J. Biol. 24. MORTLOCK, R. P. 1962. Gluconate metabolism of Chem. 239:2765-2771. Pasteurella pestis. J. Bacteriol. 84:53-59. 13. FRAENKEL, D. G., AND S. R. LEVISOHN. 1967. 25. MORTLOCK, R. P., AND R. R. BRUBAKER. 1962. Glucose and gluconate metabolism in an Glucose - 6 - phosphate dehydrogenase and Escherichia coli mutant lacking phosphoglucose 6-phosphogluconate dehydrogenase activities isomerase. J. Bacteriol. 93:1571-1578. of Pasteurella pestis and Pasteurella pseudo- 14. GLOCK, G. E., AND P. McLEAN. 1953. Further tuberculosis. J. Bacteriol. 84:1122-1123. studies on the properties and assay of glucose- 26. PEPPER, R. E., AND R. N. COSTILOW. 1964. 6-phosphate dehydrogenase of rat liver. Glucose catabolism by Bacillus popilliae and Biochem. J. 55:400-408. Bacillus lentimorbus. J. Bacteriol. 87:303-310. 15. HIGUCHI, K., AND C. E. CARLIN. 1958. Studies on 27. RrrrER, D. B., AND R. K. GERLOFF. 1966. De- the nutrition and physiology of Pasteurella oxyribonucleic acid hybridization among some pestis. II. A defined medium for the growth species of the genus Pasteurella. J. Bacteriol. of Pasteurella pestis. J. Bacteriol. 75:409-413. 92:1838-1839. 16. HORECKER, B. L. 1957. The orcinol reaction for 28. RUGOSA, M., M. I. KRICHEVSKY, AND F. S. BISHOP. mixtures of pentose and heptulose, p. 105-107. 1965. Truncated glycolytic system in Veil- In S. P. Colowick and N. 0. Kaplan [ed.], lonella. J. Bacteriol. 90:164-171. Methods in enzymology, vol. 3. Academic 29. SANTER, M., AND S. AJL. 1955. Metabolic re- Press, Inc., New York. actions of Pasteurella pestis. III. The fermenta- 17. KAPLAN, N. 0. 1963. Symposium on multiple tion of glucose. J. Bacteriol. 69:298-302. forms of enzymes and control mechanisms. 30. SOKATCH, J. T., AND J. C. GUNSALUS. 1957. I. Multiple forms of enzymes. Bacteriol. Rev. Aldonic acid metabolism. I. Pathway of carbon 27:155-169. in an inducible gluconate fermentation by 18. KAPLAN, N. 0., S. P. COLOWICK, E. F. NEUFELD, Streptococcus faecalis. J. Bacteriol. 73:452-460. AND M. M. CIOTTI. 1953. Pyridine nucleotide 31. TINELLJ, R., A. M. MICHELSON, AND J. L. transhydrogenase. IV. Effect of adenylic acid STROMINGER. 1963. Epimerization of thymi- a on the bacterial transhydrogenase. J. Biol. dine diphosphate glucose in bacterial extracts. J. Bacteriol. 86:246-251. Chem. 205:17-29. 32. VANDERMARK, P. J., AND W. A. WOOD. 1956. 19. KATZNELSON, H. 1958. Metabolism of phyto- The pathways of glucose dissimilation by pathogenic bacteria. IT. Metabolism of carbo- Microbacterium lacticum. J. Bacteriol. 71:385- hydrate by cell-free extracts. J. Bacteriol. 392. 75:540-543. 33. WANG, C. H., I. STERN, C. M. GILMOUR, S. 20. KREBS, H. A., AND W. A. JOHNSON. 1937. Me- KLUNGSOYR, D. J. REED, J. J. BIALY, B. E. tabolism of ketonic acids in animal tissues. CHRISTENSON, AND V. H. CHELDELIN. 1958. Biochem. J. 31:645-660. Comparative study of glucose catabolism by 21. LOWRY, 0. H., N. J. ROSEBROUGH, A. L. FARR, the radiorespirometric method. J. Bacteriol. AND R. J. RANDALL. 1951. Protein measure- 76:207-216.