i-galactoside Transport in Bacterial Membrane

Transcription

1 Proceedings of the National Academy of Sciences Vol. 66, No. 4, pp , August 1970 i-galactoside Transport in Bacterial Membrane Preparations: Energy Coupling via Membrane-Bound D-Lactic Dehydrogenase Eugene M. Barnes, Jr.* and H. R. Kabackt NATIONAL HEART AND LUNG INSTITUTE, NATIONAL INSTITUTES OF HEALTH, BETHESDA, MARYLAND Communicated by E. R. Stadtman, May 19, 1970 Abstract. The transport of f3-galactosides by isolated membrane preparations from Escherichia coli strains containing a functional y gene is markedly stimulated by the conversion of D-lactate to pyruvate. The addition of D-lactate to these membrane preparations produces a 19-fold increase in the initial rate of uptake and a 10-fold stimulation of the steady-state level of intramembranal lactose or thiomethylgalactoside. Succinate, DL-a-hydroxybutyrate, and L- lactate partially replace D-lactate, but are much less effective; ATP and P-enolpyruvate, in addition to a number of other metabolites and cofactors, do not stimulate lactose transport by the vesicles. Lactose uptake by the membrane preparations in the presence of D-lactate requires oxygen, and is blocked by electron transport inhibitors and proton conductors; however, uptake is not significantly inhibited by high concentrations of arsenate or oligomycin. Furthermore, the P-enolpyruvate-P-transferase system is not involved in,3-galactoside transport by the E. coli membrane vesicles. The findings indicate that the 3-galactoside uptake system is coupled to the membrane-bound D-lactic dehydrogenase via an electron transport chain but does not involve oxidative phosphorylation. Although the,3-galactoside transport system of Escherichia coli has been examined in very great detail, the mechanism of the coupling of metabolic energy to active f3-galactoside transport remains poorly understood. Scarborough, Rumley, and Kennedy' suggested an involvement of ATP in the lactose transport system of E. coli. However, recent studies by Pavlasova and Harold2 on anaerobic methyl-l-thio-$-d-galactoside (TMG) uptake indicate that uncouplers of oxidative phosphorylation block TMG accumulation but do not alter ATP levels. Fox and Kennedy demonstrated the existence of a "permease" protein (the M protein) which was a product of the y gene.3 The subsequent suggestion of a role for the P-enolpyruvate-P-transferase system in TMG uptake4 in E. coli raised the possibility that the M protein might be an inactivated Enzyme II. This topic has been discussed in detail in a recent review.5 Recent studies in this laboratory have described the coupling of a membranebound D-lactic dehydrogenase to amino acid transport in isolated membrane preparations from E. coli.6 This paper reports a similar coupling of the 83-galac- 1190

2 VOL. 66, 1970 BIOCHEMISTRY: BARNES AND KABACK 1191 toside transport system to the membrane-bound D-lactic dehydrogenase and describes the general properties of the,3-galactoside accumulation system in isolated bacterial membranes. Evidence is presented which indicates that electron transport, in the absence of oxidative phosphorylation, is required for,3- galactoside uptake by the membrane vesicles. Furthermore, evidence is also presented which demonstrates that the P-enolpyruvate-P-transferase system is not involved in this transport mechanism. Methods. Whole cells: E. coli ML (i-z-y+a+) was grown on medium A7 containing 0.4% succinate or glycerol. The ML 30 strain (i+z+y+a+) was grown on the same glycerol minimal medium and where indicated 0.5 mm isopropyl-f-d-thiogalactopyranoside (IPTG) was added as inducer. E. coli GN-2 (i-z+y+a+, Enzyme I-) was grown as described previously.8 Membrane preparations: Membranes were l)reparedl by methods already described.8-10 Uptake studies: The assay methods for methyl-a-d-glucopyranoside (a-ig) uptake by isolated membrane preparations have been reported.8 The assay for 0-galactoside uptake was identical to that described for amino acids.6 Enzyme assays: D-Lactic dehydrogenase and succinic dehydrogenase activities in the membrane preparations were assayed by a procedure reported earlier.6 Identification of accumulated lactose or TMG: Membrane samples that had accumulated lactose-'4c or TMG-14C were extracted8 or applied directly to silica gel G thin-layer plates (Mann Biochemicals, N.Y.) which were then developed with chloroform:methanol:water (60:70:26, v/v/v). The plates were radioautographed and the Rf values of the radioactive spots compared to those of known standards. Estimation of concentration ratio: The ratio of intravesicular to external lactose concentrations was estimated by a method already described." Materials. Lactose-1-'4C was obtained from Amersham/Searle and a-methyl- '4C-thiogalactoside was a product of New England Nuclear Corp. All chemicals used in these experiments were of reagent grade and were obtained from the usual commercial sources. Results. Effect of D-lactate on lactose uptake: The effect of D-lactate on the initial rate of lactose uptake and the steady-state level of accumulation by isolated membranes from E. coli ML is shown in Figure 1. The addition of D-lactate stimulated the initial rate of transport 19-fold over controls incubated without D-lactate. The steady-state level of lactose accumulation was increased 10-fold by D-lactate. In 5 min membranes in the presence of D-lactate accumulated lactose to an intravesicular concentration about 30 times higher than that of the medium. It can also be seen that the addition of DNP results in the rapid loss of approximately 90% of the accumulated radioactivity. Over the time course indicated in Figure 1 more than 95% of the radioactivity accumulated in the membranes was recovered as unchanged lactose, and there was no detectible lactose-p at any of the times sampled. The rapid exit of lactose-1-'4c when excess unlabeled lactose was added is consistent with this observation. Conversion of D-lactate to pyruvate: As indicated in Figure 2, membrane preparations of AML rapidly converted D-lactate to pyruvate in a nearly stoichiometric fashion. In these experiments, D-lactate-U-'4C was employed as substrate and pyruvate was the only detectible radioactive product on thin-layer chromatograms. The loss of about 20% of the total carbon after 20 min may have been caused by partial decarboxylation of the pyruvate formed.

3 FIG BIOCHEMISTRY: BARNES AND KABACK Puoc. N. A. S D-Loctote Total S TIME(min0.) a- 8 ~~~~~~~~~~~~~4 Eo aa 0 (rnnvte 0 2~~~* to,50mm poasiu 3hsht c-p trol.) n 0m ansimslae fe i TIME (min) TIME (min) (Left) FIG. 1. -Effectof D-lactate on lactose uptake by E. coli ML membrane preparations. Aliquots (50 Al of membranes from ML (grown on succinate medium) containing 0.41 mg protein were diluted to a final volume of 100,A containing, in final concentrations, 11D-atae 50 mm potassium phosphate (ph 6.6), and 10 mm magnesium sulfate. After a 2-mm incubation at25sc, lithium D-lactate was added (where indicated) at a final concentration of 20 mm, and immediately thereafter lactose-1-14c (10.2 mci/mmol) was added to yield a final concentration of 0.2 mm. The incubations were continued at 250C for the indicated times and then terminated and assayed as described previously.6 The arrow indicates the time of addition of 10mMa C-lactose(-A-)or1mMdinitrophenol (ma.). Controlincubationscontained nol D-lactate. (Right) FIG. 2.-Metabolism Of D-lactate by ML membrane preparations. The experiment shown was conducted as described in Methods. Energy source specificity for lactose uptake: The effect of various metabolites and cofactors on the steady-state level of lactose accumulation in membrane vesicles is shown in Table 1. Of the compounds tested, only D-lactate (line 2), DL-a-hydroxybutyrate (line 3) (a known substrate for D-lactic dehydrogenase), succinate (line 4), and L-lactate (line 5) increased lactose transport above endogenous levels. As shown above (Fig. 2), 3 ed-lactate. the membrane preparations convertedda lactate to pyruvate. Although the data 'ere are not shown, the same membrane prep- DL- a-hydroxybutyrote IW ~~~~~~~19),fumarate (line 26), and a-ketobuty- E ~~ ~ ~~LLcoe rate (line 31) failed to support transport. A + t + ~~The initial rates of lactose uptake as a 0 CONENRAIO (mm) arations also stoichiometrically converted shown0 L-lactate to pyruvate and succinate to 0* Succinote fumarate. Furthermore, pyruvate (line function of increasing concentrations of energy sources is illustrated in Figure 3. FIG. 3.-Effect of energy sources on the As indicated, at concentrations which rate of lactose uptake by ML yilmamlintlrteofuakd membranes. Incubation mixtures were yilmamlintlrteofuak,- the same as described in the legend of Fig. lactate is clearly the most effective com- 1 except that the energy sources indicated pound. DL-a-hydroxybutyrate and sucwere added at the final concentrations intweerlivllsstmuaoy shown and lactose uptake was' terminated cint eerltvl essiuaoy after 1 min. At limiting concentrations, however, sue-

4 VOL. 66, 1970 BIOCHEMISTRY: BARNES AND KABACK 1 19:,) TABLE 1. Effect of various energy sources on lactose uptake by isolated membrane preparations from E. coli ML Lactose uptake (nmoles/mg membrane protein/10 min) Energy source (20 mm) 1 None 2 1-Lactate 3 DIa-Hydroxybutylate 4 Succinate 5 i=lactate 6 Glucose 7 6-P-Gluconate 8 Glucose-6-P 9 Glucose-1-P 10 Fructose-6-P 11 Fructose-i-P 12 Fructose-1,6-P2 13 a-glycerol-p 14 Dihydroxyacetone-P 15 3-P-Glycerate 16 1,2-PrGlycerate 17 2-P-Glycerate 18 P-enolpyruvate 19 Pyruvate Energy source (20 mm) 20 Acetate 21 Acetyl-CoA 22 Citrate 23 Isocitrate 24 cis-aconitate 25 a-ketoglutarate 26 Fumarate 27 Malate 28 Oxaloacetate 29 -y-hydroxybutyrate 30 j8-hydroxybutyrate 31 a-ketobutyrate 32 ATP 33 CTP 34 3',5'-AMP 35 UDP-Glucose 36 NADH 37 NADPH 38 Acetyl-P 39 Carbamyl-P Lactose uptake (nmoles/mg membrane protein/10 min) Effect of various energy sources on lactose uptake by isolated membrane preparations from E. coli ML Incubations were identical to those described in the legend of Fig. 1 except the energy sources listed replaced D-lactate and lactose uptake was terminated after 10 min. cinate was slightly more effective than D-lactate. L-Lactate did not produce any significant stimulation. Although not shown, the presence of NAD or NADP in addition to D-lactate did not cause any additional stimulation of lactose uptake. Induction of D-lactate-coupled g-galactoside uptake: As shown in Figure 4, membrane preparations from uninduced E. coli ML 30 took up very little lactose, nor was lactose uptake by these membranes stimulated by the addition of D-lactate. Membranes from ML (y constitutive) grown on the same glycerol medium rapidly concentrated lactose in the presence of D-lactate. Membranes prepared from IPTG-induced ML 30 rapidly concentrated TMG in the presence of D-lactate as shown in Figure 5. Again, as with lactose uptake (Fig. 4), membranes from uninduced ML 30 took up very little TMG and D- lactate had no effect. The absence of D-lactate-coupled lactose or TMG uptake in uninduced ML 30 membranes is not caused by a defect in D-lactic dehydrogenase since induced and uninduced ML 30 membranes had the same l)- lactic dehydrogenase activity and concentrated proline to the same extent in the presence of D-lactate. Dependence of uptake rates on external g-galactoside concentration: Initial rates of lactose uptake by isolated membranes from ML as a function of increasing external lactose concentration is shown in Figure 6. The initial rates of lactose uptake in the presence of D-lactate exhibit saturation kinetics and a Michaelis constant (K.) of 0.19 mm. Similar data for TMG uptake were also obtained (Fig. 7), yielding a K. of 0.51 mm.

5 1194 BIOCHEMISTRY: BARNES A ND KABACK PROC. N. A. S. _ 12 o -~~~~~~ 10 ~~~~~~~~~y+, D-Loctate D-ocot y0 y+, D- Lactate E ~~~~~~~~~~~~~ a0 I-~~~ 62 ~~~~~~E3 0 E TIME (min) TIME (min) (Left) FIG;. 4.-Effect of D-lactate on lactose uptake by membranes prepared from ML and uninduced MAL 30 grown on glycerol medium. Incubation mixtures were the same as described in the legend of Fig. 1 except that the incubations contained 0.43 mgmembraneprotein from ML 30 (-A-, -A-) or 0.52 mg membrane protein from ML (-0---,--) grown on glycerol medium. Closed symbols indicate omission of D-lactate. (Right) FIG. 5.-Effect of D-lactate on TMG uptake by IPTG-induced and uninduced ML 30 membranes. Incubations were the same as described in the legend of Fig. 1 except that 0.2 mm '4C-TMG (8.7 mci/mmol) replaced lactose-1-'4c and 0.43 mg membrane protein from uninduced ML 30 (-A-, -A-) or 0.48 mg from IPTG-induced MIL 30 (-0-,-0-) was present. Closed symbols indicate omission of D-lactate. Competition by sugars for lactose entry and exchange: The data presented in Table 2 indicate the relative capacity of various sugars to compete for the lactose uptake system of the membrane vesicles. When unlabeled sugars (0.2 or 2.0 mm) were added at the same time as lactose-1-'4c (0.2 mmn), only,b-galactosides, melibiose, and galactose caused significant inhibition of lactose entry (columns 2 and 3). Similarly, only,b-galactosides, melibiose, and galactose were effective in displacing lactose from preloaded membranes (column 4). TMG uptake in GN-2 membranes: Previous work from this laboratory8 demonstrated that membrane preparations from E. coli GN-2, a mutant lacking enzyme I of the P-transferase system,'2 were unable to vectorially phosphorylate a-mg even in the presence of high concentrations of P-enolpyruvate. The data in Figure 8 indicate that GN-2 membranes, despite their inability to transport a-mg, rapidly concentrated TMG in the presence of D-lactate. As shown, GN-2 membranes exhibited a slightly higher intial rate of TMG uptake than induced ML 30 membranes (Fig. 5). The latter vectorially phosphor-lated ax-mg normally (data not shown). It is especially noteworthy that D-lactate did not stimulate a-nmg uptake by any of the isolated membranes. Concentrations of P-enolpyruvate up to 0.1 M (which gave optimal rates of a-mg uptake) did not stimulate lactose or TMG uptake, nor was lactose-p or TMG-P detected in these experiments. Finally, membranes prepared from E. coli ML failed to exhibit phosphatase activity towards TMdG-P, and the addition of lactose to ML membranes incubated in the presence of 32P-enolpyruvate did not accelerate the appearance of 32ioas might be expected if a lactose-p P-hydrolase were involved in this system.

6 VOL. 66, 1970 BIOCHEMISTRY: BARNES AND KABACK r IV / wo 0 0 a 0 <-Js. 5 " LACTOSE (mm) I/S FIG. 6.-Effect of external lactose concentration on initial rates of lactose uptake by ML membranes. Incubations were the same as indicated in the legend of Fig. 1 except that lactose-1-'4c was present at the concentrations shown and all incubations contained 20 mm D-lactate. Lactose uptake was terminated after 30 sec. The inset is a double reciprocal plot of the data shown. weff.c tl 9 2 E E Is TMG (mm) FIG. 7.-Effect of external TMG concentration on initial rates of TMG uptake by ML membranes. Ineubations were the same as described in the legend of Fig. 6 except that TMG-'4C (8.7 mci/mmol) replaced lactose. Effect of metabolic state on lactose uptake: As indicated in Table 3, D-lactatecoupled lactose uptake by ML membranes was inhibited 94% by exclusion of oxygen (line 2). Furthermore, the electron transport inhibitors azide TABLE 2. Competition membranes. 12C-Sugar TDG Melibiose ONPG IPTG TMG Galactose Maltose Sucrose Glucose Mannose Mannitol Fructose by various sugars for lactose uptake and exchange by ML Lactose Accumulated (% control) Competitive --Competitive Entry-- displacement 0.2 mm 2.0 mm 2.0 mm 2C-sugar "?C-sugar '2C-sugar a H Competition by various sugars for lactose uptake and exchange by ML membranes. Incubations were the same as described in the legend of Fig. 1 except that the "C-sugar was added at the same time as lactose-1-14c in the competitive entry experiments and uptake was for 1 min. For the competitive displacement experiments, the membranes were preloaded for 5 min with lactose- 1-14C, then the "2C-sugar shown was added, and incubations were continued for 2 min. All incubations contained 20 mm D-lactate. The values for lactose accumulated in the membranes are expressed as percentages of control values with no added "C-sugar.

7 ,1196 BIOCHEMISTRY: BARNES AND KABACK PROC. N. A. S. 530 "E T 3 D 1~~ 0E E "' 2 0 w I- 0 C 0( 5 10 TIME (min) Lactate (line 3), amytal (line 4), cyanide (line 5), antimycin A (line 6), and 2-heptyl4-hydroxyquinoline-N-oxide (line 7) also effectively blocked lactose accumulation. The inhibition by cyanide and azide is consistent with the observed anaerobic inhibition and indicates the involvement of oxygen as a terminal electron acceptor. Although not shown, anaerobic incubations inhibited the membrane D-lactate dehydrogenase activity by 80%. Finally, DNP (line 8), carbonyl cyanide trifluoromethoxyphenylhydro zone (line 9), and valinomycin (line 10) were all effective inhibitors of lactose transport. FIG. 8.-Effect of D-14actate on These compounds, well-known uncouplers of ox- TMG uptake by GN-2 mem- idative phosphorylation, are all effective proton branes. Incubations iwere the conductors.13 same as described in the legend of Fig. 5 except that mg of As also shown in Table 3, arsenate inhibited membrane protein fro: m GN-2 D-lactate-coupled lactose uptake by only 30% was employed. despite concentrations as high as 50 mm (line 11). Experiments performed in phosphate-free media produced a similar degree of inhibition. The transport system was similarly insensitive to oligomycin (line 12). Lactose transport by the membranes was inhibited by the sulfhydryl reagents N-ethylmaleimide and p-chloromercuribenzoate (lines 13 and 14). This is consistent with similar findings with whole cells and the M protein. 14 -Discussion. The data presented here demonstrate the 13-galactoside transport in E.. coli membrane preparations is coupled to a membrane-bound D-lactic dehydrogenase via a respiratory chain. A similar system mediating the transport of a wide variety of amino acids has recently been reported.6 The effect of D-lactate, DL-a-hydroxybutyrate, and succinate on lactose or TMG uptake by the membrane preparations clearly does not involve the P- enolpyruvate-p-transferase system. This conclusion is supported by the observation that P-enolpyruvate failed to stimulate lactose transport in membrane preparations which readily utilized P-enolpyruvate for the transport of a-mgnor did D-lactate substitute for P-enolpyruvate in this system. Furthermore, membranes prepared from E. coli GN-2 (defective in enzyme I of the P-transferase system) which were completely unable to catalyze the vectorial phosphorylation of a-mg were fully capable of supporting D-lactate-coupled TMG accumulation. Although the precise nature of the coupling of D-lactic dehydrogenase and succinic dehydrogenase to fl-galactoside transport is uncertain, the inhibitor studies presented in Table 3 provide a strong indication that an electron transfer chain mediates this coupling. Furthermore, the effect of D-lactate or succinate on 5-galactoside transport is apparently not exerted through the production of stable high-energy phosphate compounds. The lack of sensitivity of f3-galacto- 'side transport to arsenate or oligomycin, the failure of added ATP to stimulate

8 VOL. 66, 1970 BIOCHEMISTRY: BARNES AND KABACK 1197 TABLE 3. Effect of anaerobic incubation and metabolic inhibitors on lactose uptake by ML membranes. Inhibitor Inhibition of Condition during concentration lactose uptake incubation (M) (%) 1 Aerobic.. 2 Anaerobic 94 3 Sodium azide Sodium amytal a Sodium cyanide Antimycin A 2 X 10-' 18 2X HOQNO 2 X X Dinitrophenol CCCP Valinomycin 2 X X Sodium arsenate X Oligomycin 2 X N-ethylmaleimide p-chloromercuribenzoate Effect of anaerobic incubation and metabolic inhibitors on lactose uptake by ML membranes. All incubations were performed as indicated in the legend of Fig. 1 except that the membrane vesicles were exposed to the condition or inhibitor shown for 15 min prior to addition of ulactate and 14C-lactose. The membranes were then allowed to take up lactose for 10 min. Results are expressed as percentage inhibition of the lactose concentrating ability of a 250C, aerobic control (12.5 nmoles/mg membrane protein). All incubations contained 20 mm D-lactate. The anaerobic experiments were performed in a nitrogen atmosphere containing less than 50 ppm oxygen. Antimycin A, HOQNO, valinomycin, and oligomycin were added as aliquots of dimethylsulfoxide solutions. lactose transport, and the observation that similar membrane preparations cannot conduct oxidative phosphorylation"5 argue strongly against the involvement of respiration-linked phosphorylation in f3-galactoside transport. In light of these observations, a possible coupling of electron transfer to active fl-galactoside transport via oxidation-reduction of a membrane "carrier" protein (e.g., the M protein) with resultant conformation changes is an attractive hypothesis. However, further experimentation is necessary before such a theory can be seriously considered. Abbreviations: DNP, 2,4-dinitrophenol; IPTG, isopropyl-g8-dthiogalactopyranoside; a- MG, methyl-a-d-glucopyranoside; ONPG, o-nitrophenyl-,3-d-galactopyranoside; TDG, jsdgalactosyl-l-thio-fi-u-galactopyranoside; HOQNO, 2-heptyl4-hydroxyquinoline-N-oxide; TMG, methyl-l-thio-,3-u-galactoside. * Postdoctoral fellow of the American Cancer Society (PF-545). t Associate member of The Roche Institute of Molecular Biology, Nutley, N.J , and a Requests for re- visiting scientist in the National Heart and Lung Institute until July 1, prints should be sent to Dr. H. R. Kaback at the New Jersey address.

9 1198 BIOChEMISTRY: BARNEPhS A ND KABACK Pitoc. N. A. S. Scarborough, G. A., Al. K. Rumley, and E. P. Kennedy, Proc. Nat. Acad. Sci. USA, 60, 951 (1968). 2 Pavlasova, E., and F. M. Harold, J. Bacteriol., 98, 198 (1969). 3 Fox, C. F., J. R. Carter, E. P. Kennedy, Proc. Nat. Acad. Sci. USA, 57, 698 (1967). 4Kundig, W., F. D. Kundig, B. E. Anderson, and S. Roseman, J. Biol. Chem., 241, 3243 (1966). r Kaback, H. R., Ann. Rev. Biochem., 39, in press. 6 Kaback, H. R., and L. S. Milner, Proc. Nat. Acad. Sci. USA, 66, 1008 (1970). 7Davis, B. D., and E. S. Mingioli, J. Bacteriol., 60, 17 (1950). 8 Kaback, H. R.., J. Biol. Chem., 243, 3711 (1968). 9 Kaback, H. R., Proc. Nat. Acad. Sci. USA, 63, 724 (1969). 10 Kaback, H. R., Methods in Enzymology, ed. W. B. Jakoby (New York: Academic Press, in press). X1 Kaback, H. it., and E. R. Stadtman, Proc. Nat. Acad. Sci. USA, 55, 920 (1966). 12Tanaka, S., Fraenkel, D. G., and Lin, E. C. C., Biochem. Biophys. Res. Commun., 27, 63 (1967). 18 Thompson, T. E., and F. A. Henn, in Membranes of Mitochondria and Chloroplasts, ed. E. Racker (New York: Van Nostrand Reinhold Co., 1970), p Fox, C. F., and E. P. Kennedy, Proc. Nat. Acad. Sci. USA, 54, 891 (1965). Is Kleiii, W. L., A. S. Dharms, and P. D. Boyer, Fed. Proc., 29, 341 (1970).