J. Basic Microbiol. 32 (1992) 5, 309-315

(Centro de Biologia Celular y Departamento de Biologia Celular, Escuela de Biologia, Facultad de Ciencias, Universidad Central de Venezuela, Caracas, Venezuela)

The metabolism of gluconate in Escherichia coli: A study in continuous culture NEREIDA B. COELLO and TOMASISTURIZ (Received 13 September 1991/Accepted 4 MUJ 1992)

The gluconate metabolism in Escherichia coli involves duplicate activities of transport and phosphorylation for gluconate. In both cases, these activities can be differentiated in vifro by their different affinities for the substrate. In addition, the two gluconokinases can be differentiated by their heat sensitivities. The technique of continuous culture was used to investigate the influence of the growth rate on this metabolism in an E. coli HfrG6 strain during gluconate-limited growth under conditions of high and low oxygen concentrations. The transport and phosphorylation for gluconate, induced when the cells are cultivated in media with gluconate were differently influenced by the culture dilution rate. These activities were induced under the two conditions investigated; however, the low affinity transport system for gluconate and the thermosensitive gluconokinase were not detected under conditions of high and low oxygen concentrations, respectively. The induction of the dehydratase was favoured under conditions of low oxygen concentration. The experimental data suggest that induction and repression work together to regulate the levels of these activities during gluconate-limited growth conditions. Furthermore, that an effector molecule distinct from gluconate might be involved in the induction of the dehydratase.

The initial metabolism of gluconate in E. coli has proved to be complex. Two sets of enzymes catalysing the gluconate transport and phosphorylation have been described. These two sets of enzymes are specified by two distinctly regulated groups of genes, which are locatated in distinct regions of the bacterial chromosome. The region bioH-asd (75 min) contains the gntT, gntU and gntK genes. These code for high and low affinity transport systems for gluconate and a thermoresistant gluconokinase (its activity is stable for 3 h at 30 "C), respectively. Together these activitities constitute the GntI system. The fdp-vafS region (96 min) , includes the gntS and gntV genes. These code for another high affinity transport system for gluconate and a thermosensitive gluconokinase (it loses 75% or more of its activity if incubated 3 h at 30 "C), respectively, These activities constitute the GntII system. E. coli gntT mutants are severely affected in the utilization of gluconate; moreover, some E. coli bioH-asd deleted mutants are unable to grown on this substrate (NAGELDE ZWAIGet al. 1973, FAIKand KORNBERG 1973 ISTURIZet al. 1979). These and other observations suggested the GntI and GntII systems involved as main and subsidiary respectively, in the capture and phosphorylation ofgluconate (BACHIand KORNBERG 1975a). The GntI and GntII activities convert external gluconate to intracellular 6-phosphogluconate, which can either be decarboxylated to pentose-phosphate by the action of the enzyme 6-phosphogluconate dehydrogenase (EC 1.1.1.44., gene gnd) or metabolized to pyruvate and glyceraldehyde 3-phosphate via the ENTNER-DOUDOROFF pathway (EISENBERG and DOBROGOSZ 1967). The first enzyme involved in the latter pathway, 6phosphogluconate dehydratase (EC 4.2.1.12., gene edd), as well as the GntI and GntII activities are induced when the cells are cultivated in gluconate-containing media (COHEN 1951, FRAENKEL and LEVISOHN 1967, NAGELDE ZWAIGet a/., 1973, POUYSS~GUR et a/. 1974). The second enyzme of the ENTNERDOUDOROFF pathway, 6-phospho-2-dehydrogluconate

310

N. B. COELLO and T. IST~JRIZ

aldolae (EC 4.1.2.14., gene eda) is expressed constitutively at significant levels. This enzyme is possibly induced by its substrate (FRADKINand FRAENKEL 1971). The 75 min region also contains the gntR gene; its product negatively controls the expression of the genes encoding the GntI system and the dehydratase (NAGELDE ZWAIG et al. 1973, ZWAIGet al. 1973) but not that of the genes which encode the GntII system. At present, it is not known how the expression of the GntII system is controlled. The role of the GntI and GntII systems during utilization of gluconate by E . coli is poorly understood. More insight into the function of these systems may be obtained from studies of their regulation. In this respect, it was shown earlier (HUNGet al. 1973) that in aerobic batch cultures (progressively anaerobic because the increase in cell density), the induction of the thennosensitive and thermoresistant gluconokinases occurs respectively during the aerobic and anaerobic growth phases, i.e., sequentially. Nevertheless, later experimental evidence ( I S T ~ R IetZ al. 1979, ISTURIZet al. 1986) indicated that mutants which express either activity are capable of aerobic and anaerobic growth on gluconate. The latter observation and the finding that pyruvate, or some related metabolite, might act as a negative effector on the expression of the GntII system (BACHI and KORNBERC 1975a) suggest that instead of the oxygen concentration per se, it could be the intracellular changes which originate as the gluconate is metabolized, that modulate the inducible expression of the activities involved. In the present study, we have used the technique of continuous cultivation (CALCOT1981), to investigate the influcence of the growth rate (DEAN1972, ABDUL1981) on 'gluconate metabolism in E. coli strain HfrG6 (his but otherwise normal) during gluconate-limited growth under conditions of high and low oxygen concentration. Materials and methods Organisms: The bacterial strains used, are E. coli K12 derivatives. Media: Mineral medium (MM; TANAKA et ul. 1967). plus sodium succinate (2gl-'), histidine (20gg ml-') and thyamin (5 gg ml-') was used in batch cultures. The same medium with doubled concentrations of both micronutrients and growth factors was used in continuous cultures; the concentration of gluconate was 2 g1-l. Casein hydrolysate JCAA) is M M supplemented with 10 gl-' of casein hydrolysate. Growth conditions: Cells to be cultivated in the chemostat (model C30, New Brunswick Scientific Co. Inc., Edison, N.J., with a working volume of 3801111) were pregrown aerobically in MM supplemented with sodium succinate in batch, at 37 "C in volumes of 200 ml in 500 ml flasks, on a gyratory water bath shaker (model G76, New Brunswick Scientific Co., Edison, N. J.) at about 200 cycles min - '. Exponentially growing cells were centrifuged (3000 r.p.m., SORVALL SS34), suspended in 10 ml of MM and inoculated into the culture vessel containing the basal medium without gluconate (about 3 x 10' cells ml-', initial density). The dilution rate was then immediately set at the desired value. Steady state were considered to have been established after six replacement times. In each case, samples of the cells harvested for the appropriate assays were saved to control the phenotype and the presence of possible contaminants. For conditions of high oxygen concenlration in the reactor, the impeller was normally set at 300 r.p.m. and the air input adjusted to obtain 85% air saturation. For conditions of low oxygen concentrations the air input was eliminated and the impeller set at 200 r.p.m. These conditions resulted in less than 20% air saturation. The partial pressure of oxygen was monitored in a dissolved oxygen analyzer. model D.O. 50, New Brunswick Scientific Co., Edison. N.J. Preliminary experiments confirmed the gluconate-limited condition of the cultures. Increases of this substrate in the reservoir in conditions of not limitation by both histidine and thyamin, resulted in corresponding increases in steady state biomass levels. Assay of IU-" CI-gluconate uptake: The high affinity uptake (approximate Km of 2 x M) was assayed according to I S T ~ ~ R IetZ al. (1986). The low affinity transport (approximate Km of 2 x M) was assayed in a similar way except that the cells were incubated with 2 x M of f U - I 4 C ]gluconate instead of 2 x lo-' M. In this case, to approximate the low affinity uptake values

311

Gluconate metabolism in E. coli

in the presence of high affinity activities, the registered values (pmol taken up by lo7 cells min-') were corrected by subtracting from them double the registered values for the corresponding high affinity activities. To estimate the effect of the carbonyl cyanide m-chlorophenyl-hydrazone (CCCP) on the uptake activities, the cells were pretreated during 10 min with this reagent at 30 "C (20 p ~ , final concentration). In each case, the effect is reported as the percentage of inhibition in the incorporation of gluconate as compared with the same assay in cells without treatment. Enzyme assays: Gluconokinase and 6-phosphogluconate dehydratase were assayed by the procedure of FRAENKEL and HORECKER (1964). The gluconokinase heat inactivation is reported in each case as the percentage of the gluconokinase activity lost as determined in 0.2 ml volumes of cells extract incubated at 30 "C during 3 h and compared to that of the same extract without treatment; L.P., kept in cold. Activities are reported as nmol min-' (mg protein)-'. Determination of gluconate in the reactor: Immediately after withdrawing each sample, a 2 ml volume without cells was filtered in the cold through a membrane filter (0.45 pm pore size; MILLIPORE) and frozen at - 10 "C until its utilization. To estimate the gluconate (VM)using the gluconokinase and 6-phosphogluconate dehydrogenase (6-PGDH) reactions, the incubation mixture, in a final volume of 1 ml contained: Tris buffer 50 mM-MgCI, 5 mM pH 7.6; ATP 2 mM; NADP 0.2 mM; 6-PGDH 70 units; gluconokinase 30 units; and sample to be investigated 100 PI. NADPH formation was measured at 340 nm until a maximum value was reached. The concentration of gluconate was calculated from a calibration curve. A unit was defined as the amount of enzyme catalysing the formation of 1 nmol of product per minute under the experimental conditions used. Enzymes: Lactic dehydrogenase was purchased from SIGMA.6-PGDH partially purified (VERONESSE ef al. 1982) was obtained from heat induced cultures of the E. coli lysogen, strain RW 182 [F-, trpR lacZ trpA kdgR A (edd-zwf) A (gnd-his) hc1857st68 h80, hc1857 st 68 h80 dgndhis; WOLFand FRAENKEL 19741. Cell extracts without dehydratase, but containing high gluconokinase activity, were obtained from cultures of strain TI 129 [F-, his A(edd-zwf) gntR strA]. Chemicals: Sodium [U 14C] gluconate, specific activity 5.6 mCi (0.21 GBq) mmol- ', was obtained from AMERSHAM. o-Gluconic acid (potassium salt), 6-phosphogluconic acid (trisodium salt), pyrimidine nucleotides and most other chemicals were purchased from SIGMA.Media were from DIFCO.

Results Gluconcife-timifedgrowth at high oxygen concentration Initially enzyme activities involved in the metabolism of gluconate were measured in E. coli HfrG6 during carbon-limited growth at high oxygen concentration. The dilution rates (D)studied, ranged from 0.044 to 0.746 h-'. At higher dilution rates, the cultures displayed wall growth. Gluconate uptake rates (2 x ~O-'M) and gluconokinase activity, increased with increasing dilution rate (Fig. 1). The low affinity uptake system for gluconate was not detected at any of the D values investigated. At all growth rates, gluconate uptake activity was inhibited by approximately 70% by 20 FM CCCP (not shown in Fig. 1). The gluconokinase activity, which was mostly thermosensitive at low dilution rates (90% heat inactivation at D < 0.1 h-'), became increasingly thermoresistant with increasing D (0% heat inactivated at D 0.746 h-', Fig. 1). Low levels of 6-phosphogluconate dehydratase were observed over the range of D values studied. Activities of the constitutive enzyme 6-phosphogluconate dehydrogenase increased from 45 to 176 nmol min- (mg protein)- ', an effect presumably caused by the increase in growth rates (WOLFet al. 1979).

'

Gluconate-limited growth at low oxygen concentration The expression of enzyme activities involved in the initial steps of gluconate metabolism might be regulated by the intracellular concentrations of certain metabolites. Growth at low dissolved oxygen concentration is likely to influence metabolic concentrations. Therefore, the expression of some key enzyme activities in gluconate metabolism was also studied during gluconate-limited growth of E. coli HfrG6 at low dissolved oxygen concentration.

312

N. B. COELLO and T. I S T ~ I Z

a

9.I

02

0.3

0.4

05

06

07

D(h1-l

Fig. 1 The specific activities of gluconate uptake (0). gluconokinase ( 0 ) .6-phosphogluconate dehydratase (a) and 6-phosphogluconate dehydrogenase (A).as well as the gluconate concentration (m) in the culture vessel, plotted as functions of the dilution rate are shown. The cells were grown in gluconate-limited mediun at high oxygen concentration as dcscribed in Methods. Numbers in parenthesis indicate gluconokinase lability (percentage of activity lost after 30 'C preincubation; see Methods).

Growth at low dissolved oxygen concentrations was studied between D of 0.038 and 0.37 h-', as the cultures washed out at D values close to 0.45 h-' (Fig. 2). All activities studied increased with the dilution rate (Fig. 2), except for the low affinity uptake system for gluconate, which was always detected at about 30 pmol min-' (lo-' cells)-' (not shown in Fig. 2). As seen during growth at high dissolved oxygen concentrations (Fig. l), the uptake of 2 x 1 0 - 5 gluconate ~ was inhibited in each case by approximately 70% by 20 p~ CCCP (not shown in Fig. 2). Gluconokinase activities were about three to four fold higher than during growth at high dissolved oxygen concentrations over the corresponding range of dilution rates (Fig. 2). Gluconokinase activtiy of cells growing at low oxygen concentrations was thermostable (20% and 0% heat inactivated at D of 0.038 to 0.37 h-' respectively; Fig. 2). The activities of the 6-PGDH were about twofold higher than during growth at high dissolved oxygen concentration. In contrast to the regulation pattern observed during growth at high dissolved oxygen concentrations, dehydratase activities increased significantly (12 to 120 nmol min-' (mg protein)-' with increasing dilution rate (Fig. 2). Discussion

Induction and repression during gluconate utilization by E. coii

The results shown in Figs.1 and 2 point out that enzyme activities normally induced when E. coli is cultivated in gluconate containing media, are differently influenced by the culture

Gluconate metabolism in E. coli

313

Fig. 2 The specific activities and the gluconate concentrations in the culture vessel, plotted as in Fig. I are shown. The cells were grown in gluconate-limited mediun a t low oxygen concentration as described in Methods. Numbers in parenthesis and symbols as in Fig. 1. dilution rate during gluconate-limited growth. The inducible character of gluconate metabolism and the repression of the genes which encode the GntII system by pyruvate or some related metabolite (BACHI and KORNBERG 1975a, ISTURIZ et al. 1986). suggest that these different responses might be partly a consequence of a concerted action between two effects. Since under both conditions the residual concentration of external gluconate increases as the growth rate increases and hence probably also intracellular levels of this substrate and metabolites derived from it, both catabolite repression and induction are expected to increase correspondingly. The concerted effect of induction and repression was first indicated by the change in composition shown by the gluconokinase activity as D was increased under conditions of high oxygen concentration (Fig. 1). In this case the corresponding increases of intracellular pools of metabolites (pyruvate among them), could lead to a progressively increasing repression on the synthesis of the thermosensitive gluconokinase, mostly expressed at low Ds, as the progressive induction of the thermoresistant enzyme occurs. Despite the fact that the pattern of activities under low oxygen concentration was different (Fig. 2), the absence of the thermosensitive gluconokinase at practically all the growth rates examined, was also indicative that induction and repression might act together on this particular metabolism. In these conditions, the pyruvate effect could be present at the lower D value investigated to repress significantly the expression of the thermosensitive enzyme.

314

N . B. COELLOand T. ISTfJRIZ

Effector nrolecules in the E. coli gluconate nietaholisrn

As shown in Fig. 1, as D increases. there is a progressive increase in steady state levels of the high affinity transport for gluconate and gluconokinase activity, but not of those of the dehydratase. These three activities increased with increasing D under condition of low oxygen concentrations (Fig. 2 ) , pointing out that the inducible expression of the dehydratase was favoured in this condition. Moreover. under this condition the gluconokinase activities were higher over the corresponding range of dilution rates. Since different amounts of dissolved oxygen in the growth medium. was the unique difference between the experiments shown in Fig. 1 and Fig. 2 , one possibility is that the two distinct observed patterns of expression might be due to different intracellular concentrations of intermediates. These distinct concentrations of metabolites originated as the gluconate is metabolized under the two conditions assayed might partly control the expression of activities involved in this metabolism. In this context, it is of interest that the induction of the dehydratase by its substrate has been suggested (KORNBERG and SOUTAR1973) and that pyruvate seems to act as a negative effector on the expression of the GntII system. Moreover, we have observed that in a E. coli mutant, lacking completely gluconokinase activity, the low as well as the high affinity transports for gluconate are inducibly expressed (ISTCRIZ et al., unpublished results). The analysis of intermediates of interest (c. a., intracellular gluconate, 6phosphogluconate, pyruvate and glyceraldehyde 3-phosphate) under the same conditions used here is required to determine if some of them act as effector molecules in this metabolism. Conclusions The results presented suggest intracellular gluconate and some other intermediate derived from it as effector molecules in the utilization of gluconate by E. coli; likewise, in pointing out induction and repression as mechanisms that might modulate the expression of the activities normally induced when the bacterium is cultivated in gluconate containing media. In this respect. it is of interest that catabolite repression mediated by cyclic AMP has been also reported (BACHI and KORNBERG 1975 b) as having a necessary role in the induction of the components of gluconate catabolism in E. coli. The question is how the above effects complement among them for an efficient utilization of gluconate by the normal strain. To estimate intracellular concentrations of intermediates of interest in the same conditions used here, should help to advance in the knowledge of this complex metabolism. Acknowledgements We thank K. DAWIDOWICZ for his helpful comments and critical reading of the manuscript. This work was supported by the Consejo de Desarrollo Cientifico y Humanistic0 de la Universidad Central de Venezuela (Grants 03. 10. 1970/8 and 03-34-2520/91).

References ABDUL,M., 1981. Regulation of enzyme synthesis as studied in continuous cultures of cells. In: Continuous Cultures of Cells 11, 69-97. CRC Press, Boca Raton, Florida. BACHI. B. and KORNBERG, H. L., 1975a. Genes involved in the uptake and catabolism of gluconate by Escherichia coli. J. Gen. Microbiol., 90,321 -335. BACHI. B. and KORNBERG, H. L., 1975 b. Utilization of gluconate by Eschwichia coli. A role of adenosin 3':5'-cyIic monophosphate in the induction of gluconate catabolism. Biochem. J., 150, 123- 128.

Gluconate metabolism in E. coli

315

CALCOT, P. H., 1981. Theconstruction and operation ofcontinuous culture. In: Continuous Cultures of Cells I., 13-26. CRC Press, Boca Raton, Florida. C@HEN S. S.. 1951. Gluconokinase and the oxidative path of glucose-6-phosphate utilization. J. Biol. Chem., 189. 617-628. DEAN.A. C. R., 1972. Influence of the environment on the control of enzyme synthesis. J. App. Biotechnol., 22, 245 -259. EISENBERG, R. C. and DOBROGOSZ, W. J., 1967. Gluconate metabolism in Escherichiu coli. J. Bacteriol., 93, 941 -949. FAIK,P. and KORNBERG, H. L., 1973. Isolation and properties of E. coli mutants affected in gluconate uptake. FEBS Letters.. 32, 260 -264. FRADKIN. J. E. and FRAENKEL, D. G., 1971. 2-keto-3-deoxygluconate 6-phosphate aldolase mutants of Escherichia coli. J. Bacteriol., 108, 1227- 1283. FRAENKEL, D. G . and HORECKER, B. L., 1964. Pathways of D-glucose metabolism in Salmonella tjphimuriurn. J. Biol. Chem., 239, 2765 -2771. FRAENKEL, D. G. and LEVISOWN, S. R.,1967. Glucose and gluconate metabolism in an Escherichiu coli mutant lacking phosphoglucose isomerase. J. Bacteriol., 93, 1571- 1578. A. and ZWAIG,N., 1970. Evidence for two gluconokinase activities in Escherichia HUNG.A., OROZCO. coli. Bacteriol. Proc. Abs., p. 148. ISTURIZ,T., VITELLI-FLORES, J. and MARDENI, J., 1979. El metabolismo del gluconato en E. coli. Estudio de una mutante delecionada en la regibn bioH-asd del mapa cromosomico. Acta Cient. Venezol., 30, 91 -395. ISTURIZ.T., PALMERO, E. and VITELLI-FLORES, J.. 1986. Mutations affecting gluconate metabolism in Escherichia coli. Genetic mapping of the locus for the thermosensitive gluconokinase J. Gen. Microbiol.. 132, 3209-3219. KORNBERG, H. L. and SOUTAR,A. K., 1973. Utilization of gluconate by E. coli. Induction of gluconatekinase and 6-phosphogluconate dehydratase activities. Biochem. J.. 134. 489 - 498. NAGEL DE ZWAIG,R., ZWAIG,N., ISTURIZ,T. and SANCHEZ, R. S., 1973. Mutations affecting gluconate metabolism in Escherichia coli. J. Bacteriol., 114, 463 -468. POUYSSEGUR, J. M., FAIK,P. and KORNBERG, H. L., 1974. Utilization of gluconate by Escherichiu coli. Uptake of D-gluconate by a mutant impaired in gluconate kinase activity and by membrane vesicles derived therefrom. Biochem. J., 140. I93 -203. TANAKA, S. S., LERNER, A. and LIN,E. C. C., 1967. Replacement of a phosphoenolpyruvate-dependent phosphotransferase by a nicotinamide adenine dinucleotide-linked dehydrogenase for the utilization of mannitol. J. Bacteriol., 93, 642-648. VERONESE, F. M., Boccu, E. and FONTANA, A., 1982. 6-phosphogluconate dehydrogenase from Bacillus stearothermophilus. Meth. Enzymol., 89, 282 - 291. WOLF,R. E. Jr. and FRAENKEL, D. G., 1974. Isolation of specialized transducing bacteriophages for gluconate 6-phosphate (god) of Escherichia coli. J. Bacteriol., 117, 468-474. WOLF, R. E. Jr., PRATHER,D. M. and SHEAF. M., 1979. Growth rate-dependent alteration of 6-phosphogluconate dehydrogenase and glucose 6-phosphate dehydrogenase levels in Escherichia coli K12. J. Bacteriol.. 139, 1093- 1096. ZWAIG,N., NAGELDEZWAIG,R.. ISTURIZ, T. and WECKSLER, M., 1973. Regulatory mutations affecting the gluconate system in Escherichiu coli. J. Bacteriol., 114, 469-473. Mailing address: Prof. T. ISTURIZ,Centro de Biologia Celular y Departamento de Biologia Celular. Escuela de Biologia, Facultad de Ciencias, Universidad Central de Venezuela, Apartado Postal 21201. Caracas, Venezuela