Role of Na' and Li' in Thiomethylgalactoside Transport by the Melibiose Transport System of Escherichia coli

Transcription

1 JOURNAL OF BACTERIOLOGY, Apr. 1978, p /78/ $0.200/0 Copyright 1978 American Society for Microbiology Vol. 134, No. 1 Printed in U.S.A. Role of Na' and Li' in Thiomethylgalactoside Transport by the Melibiose Transport System of Escherichia coli JANE LOPILATO, TOMOFUSA TSUCHIYA, AND T. HASTINGS WILSON* Department of Physiology, Harvard Medical School, Boston, Massachusetts Received for publication 7 November 1977 Thiomethyl-f?-galactoside (TMG) accumulation via the melibiose transport system was studied in lactose transport-negative strains of Escherichia coli. TMG uptake by either intact cells or membrane vesicles was markedly stimulated by Na+ or Li' between ph 5.5 and 8. The Km for uptake of TMG was approximately 0.2 mm at an external Na+ concentration of 5 mm (ph 7). The a-galactosides, melibiose, methyl-a-galactoside, and o-nitrophenyl-a-galactoside had a high affinity for this system whereas lactose, maltose and glucose had none. Evidence is presented for Li+-TMG or Na+-TMG cotransport. The mechanism for membrane transport of several sugars and amino acids involves membrane carriers that couple the movement of monovalent cations and these substrates. The energy stored as an electrochemical potential difference of the cation across the cell membrane provides the driving force for accumulation of the substrate via the carrier. Animal cells use the sodium ion for cotransport across the plasma membrane, whereas microorganisms frequently use the hydrogen ion for this purpose. There is, however, increasing evidence that microorganisms are more versatile than originally supposed and use Na+ for cotransport of certain substances. Studies with a marine pseudomonad led MacLeod and his collaborators (5, 6, 34, 35, 37, 38) to suggest that Na-sugar and Na-amino acid cotransport account for the marked Na+ dependency of transport in this organism. The addition of Na+ to various bacteria stimulates the transport of sugars (6, 36, 39, 41) and amino acids (5-9, 11, 13, 16-24, 27, 33-35, 37, 38, 40, 42, 44); addition of Li' stimulates proline transport in Escherichia coli (14, 15). The cotransport of Na+ and thiomethylgalactoside (TMG) via the melibiose transport system in Salmonella typhimurium was first demonstrated by Stock and Roseman (36) in They showed Na+ stimulation of TMG uptake and TMG stimulation of Na+ uptake. These studies have been recently extended by Tokuda and Kaback (39) who demonstrated TMG-Na+ (Li') cotransport in membrane vesicles from this organism. Lanyi and coworkers (18-21) have provided strong evidence that the light-driven amino acid transport of Halobacterium halobium is via a mechanism of Na+-amino acid cotransport. When exposed to light, the purple membrane extrudes protons resulting in a membrane potential (inside negative) and a ph gradient (outside acid). Na+ is extruded in exchange for protons via a specific Na+-H+ exchange system. Na+ then enters the cell down its electrochemical potential difference with amino acids via specific cotransport (symport) mechanisms. In E. coli, although many amino acids are taken up with protons, glutamic acid is accumulated by an Na+ cotransport process (7-9, 13, 22, 24, 40, 42). The purpose of this paper is to describe investigations on the melibiose transport system of E. coli. A brief description of some of the findings has been reported (41). The data are consistent with the view that there is obligatory coupling between the movement of Na+ and TMG into the cell. MATERIALS AND METHODS Bacteria. Strain W3133 (from S. Luria) is a lactosedeleted strain with a temperature-sensitive melibiose transport system. It grows on melibiose at 30 but not 37 C (Table 1). When this strain was grown for several days in melibiose at 370C a mutant (W3133-2), which possessed a temperature-stable melibiose transport system, grew. An a-galactosidase-negative melibiose transport-positive strain (RAl1) was obtained from strain W with nitrosoguanidine mutagenesis (1) followed by ampicillin treatment selection in the presence of melibiose. Strains 7-6 and NR70-1 are lactose transport-negative (lacy) derivatives of strain 7 (from E. C. C. Lin [10]) and NR70 (from B. Rosen [29]), respectively. NR70 is an ATPase-negative derivative of strain 7. All of these strains (7, 7-6, NR70, NR70-1) possess a melibiose transport system that is stable at 370C. The lacy mutants were obtained by the method of Miiller-Hill et al. (25) with o-nitrophenyl-f,-thiogalactopyranoside. Strain W3133 was grown at 300C on medium 63 (4) supplemented with 10 mm melibiose, 0.2% Casamino Acids, and 0.5 jtg of B, per ml. Strains 7-6 and NR70-1 were grown in the same medium at 37 C. Strains 147

2 148 LOPILATO, TSUCHIYA, AND WILSON TABLE 1. Characteristics of three strains Detennination Strain W3133 Strain W Strain RAll Growth on melibiose at 37 C - + Growth on melibiose at 30 C + + TMG accumulationa Uninduced Induced a-galactosidaseb Uninduced Induced J. BACTERtIOL. Genotype me1a+b(ts) mela+b+ me1ab+ a TMG concentration ratio in/out at 30 s. Medium concentration contained mm TMG and 5 mm Na+. b Enzyme activity expressed as micromoles of p-nitrophenol split per minute per gram (wet weight) of cells. Assay performed by the method of Burstein and Kepes (3). W and RAll were grown at 37 C in medium 63 supplemented with 10 mm melibiose, 1% pancreatic digest of casein (tryptone; Difco), and 0.5 /Ig of B1 per ml. Cells were harvested after approximately three doublings. Transport assays. To measure transport in strains W3133, W3133-2, and RAll, cells were washed twice in 100 mm potassium phosphate buffer (ph 7) and were suspended in the same solution. ["4C]TMG was added to cells (approximately 0.75 mg [wet wt] per ml) suspended in 100 mm potassium phosphate buffer (ph 7) to give a final TMG concentration of mm (2.5,uCi/ml). Samples (0.5 ml) were taken at intervals, filtered on membrane filters (0.65-,m pore size; Millipore Corp.), and washed with 100 mm potassium phosphate buffer (ph 7) at room temperature. Various concentrations of NaCl and LiCI were added to the reaction mixture where appropriate. To test the effect of ph, potassium phosphate buffers at ph 6 and ph 8 were used in the reaction mixture. The effect of several metabolic inhibitors was determined by adding the inhibitor to the reaction mixture and preincubating the cells with the inhibitor for 10 min before adding the ['4C]TMG. The intracellular TMG concentration was calculated assuming that 1 ml of a suspension with an absorbance of 100 Klett units (no. 42 filter) contained 0.6 pl of intracellular water (43). A 1-ml portion of such a suspension contains 0.81 mg (wet weight) of cells and approximately 1 x 101 cells per ml. Transport induced by 4p.u+ in energy-depleted cells. In one series of experiments TMG transport was driven by an artificially produced electrochemical potential difference for Li' (AU4i). Logarithmically grown cells of strain NR70-1 were washed and energy depleted by exposure to 5 mm dinitrophenol for 1 h at 37 C (2). Energy-depleted cells were divided into four groups and each was washed three times in one of four different solutions at room temperature. Two were washed in 50 mm K2SO4 and 50 mm K phosphate, one at ph 6 and the second at ph 8. The third and fourth groups were washed in 50 mm Li2SO4 and 50 mm K phosphate (ph 6 or 8). This period of washing usually lasted about 75 min and approximate equilibration of the internal and external compartments was presumed to occur. After the three washes, the cells were suspended in their appropriate wash solution to a concentration of 146 mg (wet weight) per ml. Carbonylcyanide-p-trifluoromethoxyphenyl-hydrazone (CCFP) and KCN were added to give final concentrations of 5,uM and 2 mm, respectively. Ten minutes after this addition, 20 Al of the concentrated cell suspension was diluted 100-fold into a solution (2 ml) consisting of 50 mm Li2SO4, 50 mm K phosphate (ph 8), 5,M CCFP, 5 mm KCN, and 0.1 mm ['4C]TMG. Samples (0.2 ml) were taken at intervals, filtered, washed, and counted. Membrane vesicles and transport with vesicles. Membrane vesicles were prepared from strain 7-6 according to a method by Kaback (12). Transport assay with vesicles was also performed according to the method of Kaback (12). Chemicals. Phenylethyl-fi-D-galactoside, phenyl-,b-d-galactoside, TMG, a-d-melibiose, a-lactose, D-galactose, 1-O-methyl-/?-D-galactopyranoside, 1-0- methyl-a-d-galactopyranoside, p-nitrophenyl-a-d-galactopyranoside, and o-nitrophenyl-a-d-galactopyranoside were obtained from Sigma Chemical Co. D-Galactopyranosyl -,8- D - thiogalactopyranoside, o - nitrophenyl-,-d-galactopyranoside, D-raffinose pentahydrate, isopropyl-,b-d-thiogalactopyranoside, and iodoacetic acid were obtained from Schwarz-Mann. Carbonylcyanide-m-chlorophenylhydrazone (CCCP) and phenazine methosulfate were from Calbiochem; N- methyl-n'-nitro-n-nitrosoguanidine was obtained from Aldrich, and ampicillin was obtained from Bristol Laboratories. o-nitrophenyl-,-d-thiogalactoside was obtained from Cyclo Chemical. ['4C]Thiomethyl-figalactoside was obtained from New England Nuclear. CCFP was a gift from E. P. Kennedy. RESULTS E. coli possesses two distinct pathways for entry of melibiose, the lactose transport system (coded for by the lacy gene) and the melibiose transport system (coded for by the melb gene). All studies reported here use lactose transportnegative (lacy) cells to exclude the Y gene product. One of the primary strains in this study was W3133, which contains a deletion through the lacz and lacy genes. Although this cell grows on melibiose as a sole source of carbon

3 VOL. 134, 1978 and energy at 300C, it fails to grow on this substrate at 370C due to a temperature-sensitive transport protein, as described by Prestidge and Pardee (28). A mutant, W3133-2, which grew well at 370C and was thus more convenient for routine study than the parental strain, was isolated. Table 1 compares these two strains with an a-galactosidase-negative strain, RAll, derived from W W and RAl were found to be partially constitutive. Effect of cations. When melibiose-induced cells were incubated with radioactive TMG in potassium-containing buffer, the sugar accumulated within the cell to a concentration 15 times that in the incubation medium (Fig. 1). Addition of Na+ to the medium stimulated the initial rate of sugar entry as well as the steady state achieved. The intracellular concentration of TMG was approximately 290 times that in the external medium when cells were incubated for 10 min in the presence of 15 mm Na+. Cells - 7 E J -J 4j 3 I~3 zi 15mM No+I FIG. 1. Effect ofna+ concentration on TMG transport. Induced cells (-) of strain W3133 (0.2 mg [dry weight] of cells per ml) were incubated in medium 63 containing ['4CJTMG (0.1 m.m) with different concentrations of added NaCL. Uninduced cells (0) were assayed in the presence of 5 mm Na+ and no Na+. (Similar very low values were obtained with and without Na+) Intracellular TMG concentration in uninduced cells was less than that in the medium MELIBIOSE TRANSPORT IN E. COLI 149 exposed to 100 mm Na+ showed less than maximum sugar accumulation, giving a steady-state value of 5 mm TMG (Fig. 1). Uninduced cells showed no accumulation above that in the external medium in either the presence or absence of added Na+. Cells grown in the presence or absence of Na+ showed no apparent difference in their capacity to accumulate TMG. The effect of TMG concentration was studied at two different concentrations of Na+ (Fig. 2). The apparent Km for TMG was 0.6 mm at 0.2 mm Na+ and 0.21 mm at 5 mm Na+. With no added Na+ the rate of sugar entry was too slow to accurately determine the Km. Lithium ion was found to give a stimulation of TMG uptake similar to that observed with Na+. Figure 3 shows the effect of Li+ on TMG transport in strain W3133. The cations found to be ineffective in stimulating sugar uptake by RAll and W3133 include K', Rb+, Cs', NH4', and choline+. It was important to show that only one transport system was involved for TMG in these experiments. The first line of evidence against multiple transport systems was that cells grown in the absence of melibiose failed to transport TMG in the presence (or absence) of either Na+ or Li' (Fig. 1). The second type of evidence was genetic. Fifteen transport-negative mutants were isolated with the penicillin technique (see above). Most of these showed neither a-galactosidase nor transport when grown in the presence of melibiose; one mutant showed enzyme activity but no transport. In all of these mutants no TMG transport was observed in the presence or absence of Na+ or Li'. Effect of ph. The effect of ph on the Na+ stimulation of TMG transport was studied in three strains (Table 2). The steady-state level of accumulation of TMG was higher at ph 8 than at ph 7 or ph 6. Accumulation in the absence of Na+ was elevated at the higher ph values in all but one case. Effect of inhibitors. N-ethylmaleimide and p-chloromercuribenzoate were found to be potent inhibitors of transport in both the presence and absence of Na+ (Table 3). lodoacetic acid at a concentration of 1 mm had little or no effect. The proton conductor CCCP was a particularly effective inhibitor (Fig. 4). At a concentration of 16,uM, CCCP gave complete inhibition of TMG accumulation in the presence of Na+. This inhibitor also blocked accumulation of sugar in cells incubated in the absence of Na+ (not shown). In another series of experiments (Table 4) neither KCN nor KN3 alone gave a very strong inhibition of transport although together they blocked 90% of the normal TMG uptake. Electrochemical potential difference of

4 150 LOPILATO, TSUCHIYA, AND WILSON 60i-,.0 J. BACTERIOL. S 50o 40 V 3C 20[ mM No+ Km *0.60 mm..- e-.5 m No+ - * KmO0.21 MM 10., o I TMG (mm) FIG. 2. Kinetics of TMG transport at two concentrations of Na+. Induced cells of strain RAI1 were added to a reaction mixture containing medium 63, 0.2 mm NaCl or 5 mm NaCl, and various concentrations of ['4CJTMG. A 15-s point was taken. The final cell density was 0.88 mg (wet weight) per ml. E 7 -J5 w 4 4- z HA9 2 Y5mM Li 3-0' FI.3 feto Li+ 1ocetat5moMGtas O TIME (min) FIG. 3. Effect of Li' concentration on TMG transport. Cells of strain W3133 were added to a reaction mixture consisting of medium 63, mm ['4C]TMG, and LiCi at the concentration indicated. Enough KCI was added to bring the final combined concentrations of LiCl + KCI to 100 mm. The final cell density was 0.96 mg (wet weight) per ml. Li+ drives TMG accumulation. The sodium (or lithium) stimulation might be due to either an indirect effect as a "cofactor" for a H+-TMG cotransport or a direct effect in a Na+-TMG (or Li+-TMG) cotransport. To distinguish between these two possibilities, conditions were estab- TABLE 2. Effect ofph on the melibiose transport systema Effect of Intracellular TMG concn after 5 min (mm) ph on: W3133 W RAll ph 6 Control mm Na ph 7 Control mm Na ph 8 Control mm Na a Cells were incubated in 100 mm K+ phosphate buffer at varying ph values containing mm [14C]TMG plus or minus 5 mm NaCl. lished in which the proton and the other cation moved in opposite directions. Lithium was tested for its capacity to interact with the melibiose carrier (Fig. 5). In one experiment (Fig. 5B) a membrane potential (inside negative) was generated by a proton diffusion potential, whereas the Li' concentration was equal on the two sides. This was accomplished by equilibrating energy-depleted cells at ph 6 for 75 min. The proton ionophore CCFP was added and the cells were incubated for an additional 10 min. The concentrated cell suspension (146 mg [wet weight] of cells per ml) was then diluted 100-fold into a medium at ph 8. The outward movement of protons via CCFP would be expected to generate a large membrane potential, inside negative. Such a potential would provide an electrical driving force for the inward movement of Li'. To eliminate a chemical driving force on the Li' in these experiments, equal concentrations of this ion inside and outside were arranged as follows: 50 mm Li2SO4 was included in the prein-

5 VOL. 134, 1978 MELIBIOSE TRANSPORT IN E. COLI 151 TABLE 3. Effect of various inhibitors on the melibiose transport systema % Inhibition Inhibitor W3133 W RAll mm monensin 0 Not tested Not tested mm monensin + 5 mm Na+ 45 Not tested Not tested mm CCCP Not tested mm CCCP + 5 mm Na mm pcmbb mm pcmbb+ 5 mm Na mm NEMW mm NEMb + 5 mm Na mm iodoacetic acid mm iodoacetic acid + 5 mm Na a Cells (0.75 mg [wet weight] per ml) were incubated in 100 mm KPO4 buffer (ph 7) with or without one of the inhibitors and incubated for 10 min at room temperature. [14C]TMG was then added to give a final concentration of mm. A 0.5-ml sample was taken at 5 min. The procedure was repeated with 5 mm NaCl added to the reaction mixture. b pcmb, p-chloromercuribenzoate; NEM, N-ethylmaleimide. E 0 4 -j cr J -J z 91 5mM Na+ FIG. 4. Effect of various concentrations of CCCP. CeUs of strain W3133 were added to a reaction mixture consisting of medium 63, 5 mm NaCI and mm TMG the presence or absence of CCCP. The final cell density was mg (wet weight) per ml. cubation medium to preload the cells with Li' and, subsequently, the cells were diluted into a final incubation medium containing 50 mm Li2SO4. A representative experiment of this type TABLE 4. Effect of inhibitors of oxidative phosphorylationa Inhibitor TMG uptake (mm at 30 s) Control mm KCN mm KN mM KCN + 5mM KN uM CCFP 0.14 a Cells of W3133 were preincubated with inhibitors for 10 min in a medium containing 5 mm NaCl and 0.1 M morpholinopropane sulfonic acid neutralized to ph 7.0 with tetramethylammonium hydroxide. [14C]TMG was added to give a final concentration of 0.1 mm. on energy-depleted cells of NR70-1 is given in Fig. 5B. During the first minute the cells accumulated TMG to a concentration 16 times higher than that in the external medium. The sugar concentration then declined, presumably due to the fall in membrane potential resulting from electrogenic entry of Li+ and the reduction in ph gradient. In another experiment (Fig. 5A) conditions were arranged to provide both electrical and chemical forces for the inward movement of Li+. To accomplish this, cells were preincubated at ph 6 with CCFP but without preloading the cells with Li+. Dilution of the cells into medium containing 50 mm Li2SO4 at ph 8 resulted in a 22-fold accumulation of TMG (Fig. 5A). A chemical gradient alone for Li+ was produced by preincubating the cells at ph 8 in the absence of Li+. Dilution of cells into a medium containing Li+ resulted in a ninefold sugar accumulation (Fig. 50).

6 152 LOPIUATO, TSUCHIYA, AND WILSON E 0 I 4 -J -J 4 -Iz ph6 ph8 L i +Li ph phb8 (H 8 ph ph8 K \ n,conc in medium Li FIG. 5. TMG accumulation induced by artificial driving forces in the presence of Li'. A membrane potential was induced by providing a ph gradient across the membrane (inside acid) in the presence of CCFP. Energy-depleted cells of NR70-1 were preincubated for 75 min in buffer at ph 8 or 6 with or without Li' to give the desired intracellular ionic concentrations. The chemical gradient for Li' was established by adding lithium-free cells to the external medium containing Li'. The ionic pattern is indicated in the figure. See text for experimental details. Finally, the control experiment was to preincubate cells at ph 8 with 50 mm Li2SO4 and dilute the mixture into the same medium. In the absence of either a chemical or electrical driving force on the Li+, no TMG accumulation was observed (Fig. 5D). Attempts were made to vary the membrane potential by preincubating the cells in buffers of different ph values in the presence of CCFP. The cells were then diluted into a buffer at ph 8 in the presence of CCFP. Cells preincubated at ph 6 showed a 15-fold accumulation of sugar when diluted into a medium at ph 8 (Fig. 6); cells preincubated in buffer at higher ph values showed less accumulation. In the absence of a ph gradient, only a slight accumulation was observed probably due to a small residual respiration by these energy-depleted cells. The effect of varying the chemical potential difference of ia+ in the absence of membrane potential was tested. The higher the external concentration of lithium, the greater was the TMG accumulation (Fig. 7). Membrane vesicles. In an attempt to simplify the system, studies were carried out on membrane vesicles prepared by the method of Kaback (12). A representative experiment is given in Fig. 8. Vesicles prepared from strain 7-6 were incubated in the presence ofk phosphate buffer (ph 7) plus phenazine methosulfate (PMS) and ascorbate with or without added Na+ or Li+. A marked stimulation of TMG uptake was observed in the presence of Na+ or Li+. The effect of these two cations on the kinetics of sugar uptake is given in Fig. 9. Both Na+ and Li' increase the affinity of the carrier for TMG without having a significant effect on the maximal rate. The Km for TMG uptake and in the presence of 20 mm Li+ was 0.31 mm; in 10 mm Na+ it was 0.55 mm; with neither cation it was 1.2 mm. The effect of ph on the TMG accumulation in the presence of ascorbate and PMS is given in Table 5. The activity was relatively constant from ph 8 to 6.5; at lower ph values accumulation was reduced. A variety of carbohydrates were tested as possible inhibitors of the melibiose transport system (Table 6). Strong inhibition was observed with the following a-galactosides: melibiose, o- nitrophenyl-a-d-galactopyranoside, a-methylgalactoside, and raffinose. Among the,-galac- E 1.6r - I 4 & phz J _j0.8ap- wr 0.6pHe O.5 z ApHz0 0.4 FIG6.EfcAoayn membaneptnilo '0.2 J. BACTERIOL FIG. 6. Effect of varying membrane potential on TMG accumulation. Energy-depleted cells of NR70-1 were preincubated for 75min in 50 mm K phosphate buffer at the various ph values (6.0, 6.5, 7.0, 7.5, 8.0) in the presence of 50 mm Li2SO4. CCFP was added and cells (146 mg [wet weight] per ml) were diluted 100-fold into 50 mm K phosphate buffer (ph 8) containing 50 mm Li2SO4. The ApH values were calculated assumingph equilibrium inside and outside of cells during the preincubation period.

7 VOL. 134, 1978 MELIBIOSE TRANSPORT IN E. COLI 153 FIG. 7. Effect of Li' concentration on TMG accumulation. Energy-depleted cells of NR70-1 were preincubated for 75 min in a solution containing 50 mm Kphosphate (ph 8) and 50mM KS,04. Cells (146 mg [wet weight] per ml) were then diluted 1(X-fold into solutions containing 50 mm K phosphate (ph 8) plus various concentrations of Li2S04. tosides, phenylethyl-,b-galactosidase and thiodigalactoside were the most inhibitory. Lactose had no effect. DISCUSSION The melibiose transport system in E. coli was first studied by Pardee (26) and later studied in more detail by Prestidge and Pardee in 1965 (28). They found that melibiose induced both the lactose and melibiose systems, whereas galactinol and melibiitol induced only the melibiose system. An interesting finding was that the melibiose transport system was temperature sensitive, being inactive at 370C although fully active at 300C. These findings explained the observation that growth of K-12 strains on melibiose at 370C led to the induction of the lac operon plus a-galactosidase; the sugar entered via the lactose transport system and was subsequently split by a-galactosidase. The melibiose transport system of E. coli has been further studied by Rotman et al. (30) and by Burstein and Kepes (3), and its genetics has been investigated by Schmitt (31, 32). The cation requirement for melibiose trans- FIG. 8. Stimulation of TMG transport by Na+ or Li+ in vesicles. Vesicles ofstrain 7-6 (170 pg ofprotein per ml) were preincubated for 10 min in a solution containing 50 mm potassium phosphate (ph 7) and 10 mm MgSO4. Ascorbate (20 mm) and phenazine methosulfate (0.1 mm) were then added as an energy source. [14CJTMG was added at a final concentration of 0.1 mm and 0.2-ml samples were taken. Control (0); + 10 mm NaCl (0); + 10 mm LiCi (U) /S (mm)-' FIG. 9. Determination of kinetic parameters of TMG transport with vesicles. Vesicles of strain 7-6 were preincubated for 10 min in a solution containing 50 mm K phosphate (ph 7) and 10 mm MgSO4. Ascorbate (20 mm) and phenazine methosulfate (0.1 mm) were added as an energy source. Various concentrations of [14C]TMG were added. Samples (0.2 ml) were taken at 15 and 30 s, and a linear rate was obtained. The final vesicle density was 270 pg of protein per ml.

8 154 LOPILATO, TSUCHIYA, AND WILSON port was first studied in S. typhimurium by Stock and Roseman (36) who noted that Na+ or Li' could stimulate TMG uptake. The lactose transport system is absent in this organism so that the complication of a second carrier was eliminated. These authors found that addition of TMG under suitable conditions stimulated Na+ entry into the cell. They concluded that there was obligatory coupling between Na+ or Li' and TMG. Recently Tokuda and Kaback (39) have extended these findings to membrane vesicles of S. typhimurium. TMG-dependent Na+ uptake was observed when an outward K+ diffusion potential was induced across the vesicle with valinomycin. The stoichiometry between Na+ and TMG was 1:1 at low ph values, although higher ratios were obtained when the ph was raised to 8. The present work provides strong evidence for a similar cation-cotransport system in E. coli. The data in Fig. 5 eliminate the possibility that the mechanism was H+-TMG cotransport with Li' as a cofactor. The purpose of the experiment was to arrange conditions such that protons and Li' were moving in opposite directions. The strategy was to allow protons to set up a diffusion potential with CCFP as the proton conductor. In an experiment with ph 6 inside and ph 8 outside, protons would be expected to escape from the cell via CCFP and generate a membrane potential (inside negative). This would presumably provide an electrical driving force for the inward movement of Lit. Cotransport of TMG and Li' occurred under conditions in which proton movement was outward. An experiment of this type with Na+ was published previously (41). Further evidence in support of the Na+-TMG cotransport hypothesis comes from experiments in which the external Na+ was measured with TABLE 5. Effect ofph on TMG transport with vesiclesa ph TMG uptake (mm at 30 s) a Vesicles of strain 7-6 (200 yg of protein per ml) were preincubated in 50 mm potassium phosphate buffer at the indicated ph, 10 mm NaCl, 10 mm MgSO4, 20 mm ascorbate, and 0.1 mm phenazine methosulfate. ['4C]TMG was added to give a final concentration of 0.1 mm, and 0.2-ml samples were taken. J. BACTERIOL. TABLE 6. Effect of analogs on TMG transport by vesiclesa Analog TMG uptake (mm at 30 s) % Control Experiment 1 Control Melibiose TDGb Lactose Maltose Galactose Glucose Experiment 2 Control M-a-Galb Raffinose a-onpg' Experiment 3 Control M-fl-Galb PhEt-f?Galb fl-onpgb ,8-IPTGb a Vesicles of strain 7-6 (200,ug of protein per ml) were preincubated in 50 mm potassium phosphate (ph 7)-10 mm MgSO4-10 mm NaCl (ph 7.0) at room temperature for 10 min. Ascorbate and phenazine methosulfate were added to give final concentrations of 20 mm and 0.1 mm, respectively. The sugar analogs (1 mm final concentration) were added, and transport was initiated by the addition of ['4C]TMG to give a final concentration of 0.1 mm. Samples of 0.2 ml were taken at 30 and 60 s. b TDG, D-Galactopyranosyl-,f-D-thiogalactopyranoside; M-a-Gal, 1-O-methyl-a-D-galactopyranoside; a-onpg, o-nitrophenyl-a-d-galactopyranoside; PhEtfl-Gal, phenylethyl-,b-d-galactoside;,f-onpg, o-nitrophenyl-,8-d-galactopyranoside; fl-iptg, isopropyl-f,- D-thiogalactopyranoside. an Na+ electrode. Addition of TMG to melibiose-induced cells resulted in a prompt fall in Na+ concentration in the medium (T. Tsuchiya and T. H. Wilson, unpublished data). When TMG was added to unbuffered cells of the same strain the external ph fell, indicating release of H' from the cell (41). These two experiments are consistent with the view that TMG enters the cell with Na+ giving rise to a membrane potential (inside positive) with a resulting exit of H'. It appears that although E. coli uses protonsubstrate cotransport for membrane transport of a variety of sugars, amino acids, and ions, it uses Na+-substrate cotransport for glutamic acid and melibiose. ACKNOWLEDGMENTS We thank S. Luria for strain W3133, E. C. C. Lin for strain 7, and B. Rosen for strain NR70. This research was supported by Public Health Service

9 VOL. 134, 1978 grant no. AM from the National Institute of Arthritis, Metabolism and Digestive Diseases. LITERATURE CITED 1. Adelberg, E. A., AL Mandel, and G. C. C. Chen Optimal conditions for mutagenesis by N-methyl-N-nitroso-N-nitrosoguanidine in Escherichia coli K12. Biochem. Biophys. Res. Commun. 18: Berger, E. A Different mechnim of energy coupling for the active transport of proline and glutamine in Eswherichia coli. Proc. Natl. Acad. Sci. U.S.A. 70: Burstein, C., and A. Kepes The a-galactosidase from Escherichia coli K12. Biochim. Biophys. Acta 230: Cohen, G. N., and H. V. Rickenberg Concentration specifique reversible des amino acides chez Ewcherichia coil Ann. Inst. Pasteur 91: Drapeau, G. R., and R. A. MacLeod Na+ dependent active transport of a-aminoisobutyric acid into cells of a marine pseudomonad. Biochem. Biophys. Res. Commun. 12: Drapeau, G. R., T. L Matula, and R. A. MacLeod Nutrition and metabolism of marine bacteria. XV. Relation of Na+-activated twansport to the Na+ requirement of a marine pseudomonad for growth. J. Bacteriol. 92: Frank, L, and L. Hopkins Sodium-stimulated transport of glutamate in Escherichia coli. J. Bacteriol. 100: Halpern, Y. S., H. Barash, S. Dover, and K. Druck Sodium and potassium requirements for active transport of glutamate by Ewcherichia coli K-12. J. Bacteriol. 114: Hasan, S. M., and T. Tsuchiya Glutamate transport driven by an electrochemical gradient of sodium ion in membrane vesicles of Ewcherichia coli Biochem. Biophys. Res. Commun. 78: Hayashi, S., J. P. Koch, and E. C. C. Lin Active transport of L-a-glycerophosphate in Escherichia coli J. Biol. Chem. 239: Hirata, H., F. Kosmakos, and A. F. Brodie Active transport of proline in membrane preparations from Mycobacterium phkei. J. Biol. Chem. 249: Kaback, H R Bacterial membranes. Methods Enizymol. 22: Kahane, S., M. Marcus, H. Barash, S. Halpern, and H. R. Kaback Sodium-dependent glutamate transport in membrane vesicles of Ewcherichia coli K- 12. FEBS Lett. 56: Kawasaki, T., and Y. Kayama Effect of lithium on proline transport by whole cells of Ewcherichia coli Biochem. Biophys Res. Commun. 55: Kayama, Y., and T. Kawasaki Stimulatory effect of lithium ion on proline transport by whole cells of Escherichia coli. J. Bacteriol. 128: Kitada, M., and K. Horikoshi Sodium ion-stimulated a-[1-'4c]aminoisobutyric acid uptake in alkalophilic BaciUus species. J. Bacteriol. 131: Koyama, N., A. Kiyomiya, and Y. Nooh Na+dependent uptake of amino acids by an alkalophilic BaciUus. FEBS Lett. 72: Lanyi, J. K., V. Y. Drayton, and RI E. MacDonald Light-induced glutamate transport in Halobacterium halobium envelope vesicles. I. Kinetics of the light dependent and the sodium-gradient-dependent uptake. Biochemistry 16: Lanyi, J. K., R. Renthal, and R. E. MacDonald Light-induced glutamate transport in Halobacterium halobium envelope vesicles. II. Evidence that the driving force is a light-dependent sodium gradient. Bio- MELIBIOSE TRANSPORT IN E. COLI 155 chemistry 15: MacDonald, R. E., R. V. Greene, and J. K. Lanyi Light-activated amino acid transport systems in Halobacterium halobium envelope vesicles: role of chemical and electrical gradients. Biochemistry 16: MacDonald, R. E., and J. K. Lanyi Light-induced leucine transport in Halobacterium halobium envelope vesicles: a chemiosmotic system. Biochemistry 14: MacDonald, R. E., J. K. Lanyi, and R. V. Greene Sodium-stimulated glutamate uptake in membrane vesicles of Escherichia coli: the role of ion gradients. Proc. Natl. Acad. Sci. U.S.A. 74: MacLeod, R. A., P. Thurman, and H. J. Rogers Comparative transport activity of intact cells, membrane vesicles, and mesosomes of BaciUus licheniformis. J. Bacteriol. 113: Miner, K. M., and L. Frank Sodium-stimulated glutamate transport in osmotically shocked cells and membrane vesicles of Escherichia coli. J. Bacteriol. 117: Muller-Hill, B., L. Crapo, and W. Gilbert Mutants that make more lac repressor. Proc. Natl. Acad. Sci. U.S.A. 59: Pardee, A An inducible mechanism for accumulation of melibiose in Escherichia coli. J. Bacteriol. 73: Pearce, S. M., V. A. Hildebrandt, and T. Lee Third system for neutral amino acid transport in a marine pseudomonad. J. Bacteriol. 130: Prestidge, L S., and A. B. Pardee A second permease for methyl-thio-fl-d-galactoside in Escherichia coli. Biochim. Biophys. Acta 100: Rosen, B. P ,B-Galactoside transport and proton movements in an adenosine triphosphatase-deficient mutant of Ewcherichia col. Biochem. Biophys. Res. Commun. 53: Rotman, B., A. K. Ganesan, and R. Guzman Transport systems for galactose and galactosides in Escherichia coli. H. Substrate and inducer specificities. J. Mol. Biol. 36: Schmitt, R Analysis of melibiose mutants deficient in a-galactosidase and thiomethylgalactoside permease II in Escherichia coli K-12. J. Bacteriol. 96: Schmitt, R., and B. Rotman a-galactosidase activity in cell-free extracts of Escherichia coli. Biochem. Biophys. Res. Commun. 22: Shiio, I., R. Miyajima, and N. Kashima Na+dependent transport of threonine in Brevibacterium flavum. J. Biochem. 73: Sprott, G. D., J. P. Drozdowski, E. L. Martin, and R. A. MacLeod Kinetics of Na+-dependent amino acid transport using cells and membrane vesicles of a marine pseudomonad. Can. J. Microbiol. 21: Sprott, G. D., and R. A. MacLeod Na+-dependent amino acid transport in isolated membrane vesicles of a marine pseudomonad energized by electron donors. Biochem. Biophys. Res. Commun. 47: Stock, J., and S. Roseman A sodium-dependent sugar cotransport system in bacteria. Biochem. Biophys. Res. Commun. 44: Thompson, J., and R. A. MacLeod Functions of Na+ and K' in the active transport of a-aminoisobutyric acid in a marine pseudomonad. J. Biol. Chem. 246: Thompson, J., and R. A. MacLeod Na+ and K' gradient and a-amino-isobutyric acid transport in a marine pseudomonad. J. Biol. Chem. 248: Tokuda, H., and H. R. Kaback Sodium-depend- transport in membrane vesicles isolated from Salmonella typhimurium. Biochemistry 16: ent methyl 1-thio-ft-D-galactopyranoside

10 156 LOPILATO, TSUCHIYA, AND WILSON 40. Tuschiya, T., S. M. Hasan, and J. Raven Glutamate transport driven by an electrochemical gradient of sodium ions in Escherichia coli. J. Bacteriol. 131: Tuschiya, T., J. Raven, and T. H. Wilson Cotransport of Na+ and methyl-,8-d-thiogalactopyranoside mediated by the melibiose transport system of Escherichia coli. Biochem. Biophys. Res. Commun. 76: Willis, R. C., and C. E. Furlong Interactions of a glutamate-asparate binding protein with the glutamate J. BACTERIOL. transport system of Escherichia coli. J. Biol. Chem. 260: Winkler, H. H., and T. H. Wilson The role of energy coupling in the transport of /8-galactosides by Escherichia coli. J. Biol. Chem. 241: Wong, P. T. S., J. Thompson, and R. A. MacLeod Nutrition and metabolism of a marine bacteria. XVII. Ion-dependent retention of a-aminoisobutyric acid and its relation to Na+-dependent transport in a marine pseudomonad. J. Biol. Chem. 244: