Portal de Periódicos da CAPES

Electrogenic Antiport Activities of the Gram-positive Tet Proteins Include a Na+(K+)/K+ Mode That Mediates Net K+ Uptake

1998; Elsevier BV; Volume: 273; Issue: 41 Linguagem: Inglês

10.1074/jbc.273.41.26447

1083-351X

Arthur A. Guffanti, Jianbo Cheng, Terry A. Krulwich,

Lipid Membrane Structure and Behavior

Two Gram-positive Tet proteins, TetA(L) from Bacillus subtilis and TetK from aStaphylococcus aureus plasmid, have previously been suggested to have multiple catalytic modes and roles. These include: tetracycline (Tc)-metal/H+ antiport for both proteins (Yamaguchi, A., Shiina, Y., Fujihira, E., Sawai, T., Noguchi, N., and Sasatsu, M. (1995) FEBS Lett. 365, 193–197; Cheng, J. Guffanti, A. A., Wang, W., Krulwich, T. A., and Bechhofer, D. H. (1996) J. Bacteriol. 178, 2853–2860); Na+(K+)/H+ antiport for both proteins (Cheng et al. (1996)); and an electrical potential-dependent K+ leak mode for TetK and highly truncated segments thereof that can facilitate net K+ uptake (Guay, G. G., Tuckman, M., McNicholas, P., and Rothstein, D. M. (1993) J. Bacteriol.175, 4927–4929). Studies of membrane vesicles from Escherichia coli expressing low levels of complete and 3′-truncated versions of tetA(L) or tetK, now show that the full-length versions of both transporters catalyze electrogenic antiport and that demonstration of electrogenicity depends upon use of a low chloride buffer for the assay. The K+ uptake mode, assayed via 86Rb+ uptake, was also catalyzed by both full-length TetA(L) and TetK. This mode does not represent a potential-dependent leak. Such a leak was not demonstrable in energized membrane vesicles. Rather, Rb+uptake occurred in right-side-out vesicles when the intravesicular space contained either Na+ or K+ but not choline. If an outwardly directed gradient of Na+ or K+ was present, Rb+ uptake occurred without energization in vesicles from cells transformed with a plasmid containing tetA(L) or tetK but not a control plasmid. Experiments in which a comparable exchange was carried out in low chloride buffers to which oxonol was added confirmed that the exchange was electrogenic. Thus, the K+ uptake mode is proposed to be a mode of the electrogenic monovalent cation/H+ antiport activity of TetA(L) and TetK in which K+ takes the place of the external protons. Truncated TetK and TetA(L) failed to catalyze either Tc-metal/H+ or Na+/H+ antiport in energized everted vesicles. Truncated TetK, but not TetA(L), did, however, exhibit modest, electrogenic Na+(K+)/Rb+ exchange as well as a small, potential-dependent leak of Rb+. The C-terminal halves of the TetA(L) and TetK proteins are thus required both for proton-coupled active transport activities of the multifunctional transporter and, perhaps, for minimizing cation leakiness. Two Gram-positive Tet proteins, TetA(L) from Bacillus subtilis and TetK from aStaphylococcus aureus plasmid, have previously been suggested to have multiple catalytic modes and roles. These include: tetracycline (Tc)-metal/H+ antiport for both proteins (Yamaguchi, A., Shiina, Y., Fujihira, E., Sawai, T., Noguchi, N., and Sasatsu, M. (1995) FEBS Lett. 365, 193–197; Cheng, J. Guffanti, A. A., Wang, W., Krulwich, T. A., and Bechhofer, D. H. (1996) J. Bacteriol. 178, 2853–2860); Na+(K+)/H+ antiport for both proteins (Cheng et al. (1996)); and an electrical potential-dependent K+ leak mode for TetK and highly truncated segments thereof that can facilitate net K+ uptake (Guay, G. G., Tuckman, M., McNicholas, P., and Rothstein, D. M. (1993) J. Bacteriol.175, 4927–4929). Studies of membrane vesicles from Escherichia coli expressing low levels of complete and 3′-truncated versions of tetA(L) or tetK, now show that the full-length versions of both transporters catalyze electrogenic antiport and that demonstration of electrogenicity depends upon use of a low chloride buffer for the assay. The K+ uptake mode, assayed via 86Rb+ uptake, was also catalyzed by both full-length TetA(L) and TetK. This mode does not represent a potential-dependent leak. Such a leak was not demonstrable in energized membrane vesicles. Rather, Rb+uptake occurred in right-side-out vesicles when the intravesicular space contained either Na+ or K+ but not choline. If an outwardly directed gradient of Na+ or K+ was present, Rb+ uptake occurred without energization in vesicles from cells transformed with a plasmid containing tetA(L) or tetK but not a control plasmid. Experiments in which a comparable exchange was carried out in low chloride buffers to which oxonol was added confirmed that the exchange was electrogenic. Thus, the K+ uptake mode is proposed to be a mode of the electrogenic monovalent cation/H+ antiport activity of TetA(L) and TetK in which K+ takes the place of the external protons. Truncated TetK and TetA(L) failed to catalyze either Tc-metal/H+ or Na+/H+ antiport in energized everted vesicles. Truncated TetK, but not TetA(L), did, however, exhibit modest, electrogenic Na+(K+)/Rb+ exchange as well as a small, potential-dependent leak of Rb+. The C-terminal halves of the TetA(L) and TetK proteins are thus required both for proton-coupled active transport activities of the multifunctional transporter and, perhaps, for minimizing cation leakiness. tetracycline transmembrane pH gradient, acid out for right-side-out vesicles or cells transmembrane electrical potential, positive out for right-side-out vesicles or cells minimal inhibitory concentration 4-morpholinepropanesulfonic acid. Tc1 enters bacterial cells in a non-carrier dependent fashion that is promoted by a transmembrane pH gradient, acid out (1Yamaguchi A. Ohmori H. Kaneko-Phadera M. Nomura T. Sawai T. Antimicrob. Agents Chemother. 1991; 35: 53-56Crossref PubMed Scopus (60) Google Scholar). The antibiotic thus enters the cell best under neutral and acidic pH conditions and could inhibit cell protein synthesis strongly in sensitive cells in this pH range. Both Gram-positive and Gram-negative Tet efflux proteins catalyze similar exchange reactions which prevent cytoplasmic accumulation of the antibiotic. Tc is actively extruded, as a complex with a divalent cation that bears a single positive charge, in exchange for external H+ (2Yamaguchi A. Udagawa T. Sawai T. J. Biol. Chem. 1990; 265: 4809-4813Abstract Full Text PDF PubMed Google Scholar, 3Yamaguchi A. Shiina Y. Fujihira E. Sawai T. Noguchi N. Sasatsu M. FEBS Lett. 1995; 365: 193-197Crossref PubMed Scopus (29) Google Scholar). The smaller (12-transmembrane segments) Gram-negative Tet proteins and the larger (14-transmembrane segments) Gram-positive Tet proteins share sequence similarity largely in the N-terminal six transmembrane segments regions (4Levy S.B. Antimicrob. Agents Chemother. 1992; 24: 1-3Google Scholar, 5Paulsen I.T. Skurray R.A. Gene (Amst .). 1993; 124: 1-11Crossref PubMed Scopus (104) Google Scholar) but at least some motifs and/or residues in the C-terminal halves of each type of Tc efflux protein cannot be modified without loss of activity (6Fijihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 419: 211-214Crossref PubMed Scopus (8) Google Scholar, 7Yamaguchi A. Akasaka T. Ono N. Someya Y. Kakatani M. Sawai T. J. Biol. Chem. 1992; 267: 7490-7498Abstract Full Text PDF PubMed Google Scholar). Both the Gram-negative and Gram-positive Tet protein families contain examples that have further been shown to complement K+-uptake deficient mutants of Escherichia coli(8Dosch D.C. Salvacion F.F. Epstein W. J. Bacteriol. 1984; 160: 1188-1190Crossref PubMed Google Scholar, 9Griffith J.K. Cuellar D.H. Fordyce C.A. Hutchings K.C. Mondragon A.A. Mol. Membr. Biol. 1994; 11: 271-277Crossref PubMed Scopus (19) Google Scholar, 10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar, 11Guay G.G. Tuckman M. McNicholas P. Rothstein D.M. J. Bacteriol. 1993; 175: 4927-4929Crossref PubMed Google Scholar), but this net K+ uptake mode is not taken as a general property of Tet proteins. It has been attributed to an electrical potential-dependent K+ leak that could also be conferred by truncated forms of proteins that exhibit the property (10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar, 11Guay G.G. Tuckman M. McNicholas P. Rothstein D.M. J. Bacteriol. 1993; 175: 4927-4929Crossref PubMed Google Scholar, 12Nakamura T. Matsuba Y. Ishihara A. Kitagawa T. Suzuki F. Unemoto T. Biol. Pharm. Bull. 1995; 18: 1189-1193Crossref PubMed Scopus (6) Google Scholar). Recently, studies in this laboratory have shown that the chromosomally encoded Bacillus subtilis TetA(L) protein and closely related TetK from a Staphylococcus aureusplasmid catalyze Na+(K+)/H+antiport (13Cheng J. Guffanti A.A. Krulwich T.A. J. Biol. Chem. 1994; 269: 27365-27371Abstract Full Text PDF PubMed Google Scholar, 14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar, 15Cheng J. Guffanti A.A. Wang W. Krulwich T.A. Bechhofer D.H. J. Bacteriol. 1996; 178: 2853-2860Crossref PubMed Google Scholar, 16Cheng J. Hicks D.B. Krulwich T.A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14446-14451Crossref PubMed Scopus (47) Google Scholar, 17Cheng J. Baldwin K. Guffanti A.A. Krulwich T.A. Antimicrob. Agents Chemother. 1996; 40: 852-857Crossref PubMed Google Scholar) in addition to Tc−-Me2+/H+ antiport (2Yamaguchi A. Udagawa T. Sawai T. J. Biol. Chem. 1990; 265: 4809-4813Abstract Full Text PDF PubMed Google Scholar, 13Cheng J. Guffanti A.A. Krulwich T.A. J. Biol. Chem. 1994; 269: 27365-27371Abstract Full Text PDF PubMed Google Scholar, 14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar). These exchanges were evidently electrogenic, as assayed via energy-dependent Tc-cobalt or Na+ uptake by everted vesicles of E. coli that expressed a clonedtetA(L) gene from a weak promoter (14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar). The exchanges were not inhibited by low nigericin concentrations that reduce the ΔpH but were significantly inhibited by valinomycin in the presence of K+, a combination that abolished the ΔΨ generated by respiration (14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar). Consistently, the antiports catalyzed by purified and reconstituted TetA(L) could be energized by an imposed potential (16Cheng J. Hicks D.B. Krulwich T.A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14446-14451Crossref PubMed Scopus (47) Google Scholar). In addition, the important role of TetA(L) in acidifying the cytoplasm of B. subtilis relative to the external medium during growth at alkaline pH would require that the monovalent cation/H+mode be electrogenic (15Cheng J. Guffanti A.A. Wang W. Krulwich T.A. Bechhofer D.H. J. Bacteriol. 1996; 178: 2853-2860Crossref PubMed Google Scholar, 18McNab R.M. Castle A.M. Biophys. J. 1987; 52: 637-647Abstract Full Text PDF PubMed Scopus (33) Google Scholar). Since TetK could substitute for TetA(L) in a mutant of B. subtilis that had a disruptedtetA(L) gene, TetK is presumed to catalyze an electrogenic antiport similar to TetA(L). By contrast, the Tc-metal/H+antiport catalyzed by Gram-negative Tet proteins has been proposed to be electroneutral (19Kaneko M. Yamaguchi A. Sawai T. FEBS Lett. 1985; 193: 194-198Crossref PubMed Scopus (57) Google Scholar). Moreover, Yamaguchi and colleagues (3Yamaguchi A. Shiina Y. Fujihira E. Sawai T. Noguchi N. Sasatsu M. FEBS Lett. 1995; 365: 193-197Crossref PubMed Scopus (29) Google Scholar, 20Hirata T. Wakatabe R. Nielsen J. Someya Y. Fujihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 412: 337-340Crossref PubMed Scopus (12) Google Scholar) have experienced difficulty in demonstrating the Na+/H+ activity of TetK and indicate that the Tc-metal/H+ antiport activity of TetK appeared to be electroneutral in preliminary work. One of the goals of the current study, therefore, was to examine the Tc-metal/H+ and Na+/H+ antiport activities of TetA(L) and TetK side-by-side in comparable preparations and to clarify their electrogenicity versus electroneutrality. The studies have strongly supported the multifunctional and electrogenic nature of both TetA(L) and TetK.A second major goal of the studies was to test an alternate hypothesis to the putative K+ leak in explaining the ability of Tet proteins such as TetK to complement K+ uptake-deficientE. coli. The new hypothesis arises from the discovery that these Tet proteins are electrogenic monovalent cation/H+antiporters, i.e. have a catalytic mode in which cytoplasmic Na+ or K+ is exchanged for a greater number of external H+. If K+ were able to occupy the external H+ sites, then an exchange of cytoplasmic monovalent cation for a greater number of K+ could account for the net uptake of K+ (Fig. 1). Thus, the net uptake of K+ catalyzed by a Tet protein could be one of its normal catalytic modes. Experiments were designed to test this hypothesis with TetK and to examine whether TetA(L), to which this kind of activity had never been attributed, might nonetheless possess a comparable capacity. If the capacity to catalyze net K+ uptake was in fact a function of the monovalent cation antiport mode, then TetA(L) might well demonstrate it. Or, a capacity to catalyze net K+uptake might be restricted to those Tet proteins with both monovalent cation/H+ exchange activity and a particularly high affinity for K+. Both cloned TetA(L) and TetK were shown to restore Na+ exclusion capacity and resistance to a ΔtetA(L) strain of B. subtilis. In such experiments, the Na+ exclusion capacity of TetK, but not of TetA(L), was markedly reduced by the presence of K+. This suggested a higher K+ relative to Na+ affinity for TetK than for TetA(L) (15Cheng J. Guffanti A.A. Wang W. Krulwich T.A. Bechhofer D.H. J. Bacteriol. 1996; 178: 2853-2860Crossref PubMed Google Scholar). The current studies support the hypothesis that net K+ uptake catalyzed by full-length forms of TetA(L) and TetK is a mode of the Na+(K+)/H+ exchange of both proteins.DISCUSSIONThe studies conducted here confirm and extend earlier work indicating that both TetA(L) and TetK are multifunctional antiporters that catalyze electrogenic Tc-metal/H+ and Na+/H+ antiport. The successful demonstration of both of these activities and their electrogenicity clearly depends upon low expression levels of the proteins. Higher levels of expression make both cells and membrane leaky. For experiments in which ionophores are used to assess electrogenicity in an E. coli vesicle system, it is further important to use ionophore concentrations that avoid aberrant exchanges and to reduce the concentration of chloride sufficiently to avoid dissipation of the ΔΨ by the permeant anion alone. The totality of earlier experiments supports the conclusion by Kaneko and co-workers (19Kaneko M. Yamaguchi A. Sawai T. FEBS Lett. 1985; 193: 194-198Crossref PubMed Scopus (57) Google Scholar) that TetA(B) catalyzes a largely electroneutral Tc-metal/H+ antiport. However, these authors themselves indicate some discrepancies in their findings with the conclusion of complete electroneutrality, and the issue might merit re-examination. As discussed below, the specific catalytic properties and possible multifunctional features are important factors in the design of strategies to minimize the interference of antibiotic efflux systems with use of antimicrobial therapies.The truncated versions of TetA(L) and TetK failed to exhibit any of the energy-dependent, proton-coupled activities of the full-length proteins, consistent with the evidence that residues in the C-terminal halves of TetK cannot be mutated without loss of active Tc efflux capacity (6Fijihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 419: 211-214Crossref PubMed Scopus (8) Google Scholar). Nonetheless, there were some modest but reproducible protective effects of the truncated Tet proteins in the whole cell growth complementation experiments (Table I). Possibly the truncated forms retain the capacity to bind Tc, Tc-metal, and monovalent cations, and this accounts for those effects. Such a basis for modest complementation in similar experiments has previously been noted (33Ivey D.M. Guffanti A.A. Shen Z. Kudyan N. Krulwich T.A. J. Bacteriol. 1992; 174: 4878-4884Crossref PubMed Google Scholar).The current studies add a catalytic mode to the repertoire of the Gram-positive Tet proteins, i.e. a mode in which net K+ uptake is achieved via a full catalytic cycle in which more than one K+ is taken up in exchange for a single cytoplasmic Na+ or K+. Clearly, the full-length TetA(L) and TetK do not confer a leakiness upon E. colimembranes to K+ that allows electrogenic K+entry (even down its chemical concentration gradient) in response to energization and establishment of a sizeable Δψ, inside-negative. The generation of a potential, inside-positive, during Na+(K+)/Rb+ exchange by unenergized vesicles is consistent with the operation of the whole catalytic antiport cycle but with the external Rb+ substituting for H+. Were only a partial cycle to be used for the exchange, the Rb+ accumulation would represent counterflow entirely,i.e. with the intravesicular cation transported outward down its gradient, released, and then replaced on the outside with the external Rb+ without use of the “H+” sites. In that case, the exchange should have been electroneutral. The occupation of a cation site by either K+ or H+has similarly been proposed for the complete catalytic cycle of the eukaryotic serotonin transporter (34Keyes S.R. Rudnick G. J. Biol. Chem. 1982; 257: 1172-1176Abstract Full Text PDF PubMed Google Scholar). It will be important to confirm the modest exchange capacity of the truncated TetK (as well as the possible leak) and the lack of a comparable activity by truncated TetA(L) in a purified reconstituted system in which the amount of transporter protein incorporated into the proteoliposome can be made comparable for different versions of the proteins. If Tet-mediated, electrogenic Rb+(K+) uptake depends upon the use of the H+-binding site and translocation pathway by these cations, and if the C-terminal part of the protein is required for proton binding and/or translocation, then even modest net Rb+ accumulation by truncated TetK is unanticipated under non-leaky conditions.The finding that net K+ uptake by full-length Tet proteins is definitely a mode of the normal catalytic functions rather than a leak, is consistent with the robust growth of cells expressing low levels of these proteins. It is notable that TetA(L) behaved qualitatively similar to TetK although it had not earlier been implicated as having the capacity for net K+ uptake. As hypothesized at the start of the study, this capacity may be a correlate of possession by a Tet protein of Na+(K+)/H+ antiporter activity and the extent to which this property occurs broadly among Tet proteins has not been carefully examined. Another question of interest in connection with the net K+ uptake mode is whether it may have a physiological role, e.g. at particular pH values and/or K+ concentrations. It will be of importance to examine the possibility that the Gram-negative TetA(C) (e.g. from pBR322 or pACYC184) might catalyze a similar spectrum of activities to that shown here for the Gram-positive Tet proteins. TetA(C) is among the Tet proteins that can complement K+ uptake-deficient mutants of E. coli (8Dosch D.C. Salvacion F.F. Epstein W. J. Bacteriol. 1984; 160: 1188-1190Crossref PubMed Google Scholar, 9Griffith J.K. Cuellar D.H. Fordyce C.A. Hutchings K.C. Mondragon A.A. Mol. Membr. Biol. 1994; 11: 271-277Crossref PubMed Scopus (19) Google Scholar, 10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar). Moreover, this gene has been shown to have a beneficial effect on the “fitness” of adapted E. coliin the absence of antibiotic (35Lenski R.E. Simpson S.C. Nguyen T.T. J. Bacteriol. 1994; 176: 3140-3147Crossref PubMed Scopus (134) Google Scholar); this could reflect enhanced Na+-resistance and K+ retrieval under some conditions. Whether TetK confers such a benefit on S. aureuswill also be of interest to examine. Considerable current effort is directed toward reducing the prevalence and further spread of antibiotic-resistance genes among pathogenic bacteria or other organisms that might then transfer these genes to pathogens. In assessments of those conditions that will minimize positive selection for antibiotic efflux genes of particular types, e.g. tet genes, the full panoply of roles for the given efflux protein will be important information. For example, it might be important to consider the pH, Na+, and K+concentration to which the organisms are exposed, rather than simply the exposure to Tc, when evaluating strategies for decreasing the prevalence of TetA(L) or TetK. Tc1 enters bacterial cells in a non-carrier dependent fashion that is promoted by a transmembrane pH gradient, acid out (1Yamaguchi A. Ohmori H. Kaneko-Phadera M. Nomura T. Sawai T. Antimicrob. Agents Chemother. 1991; 35: 53-56Crossref PubMed Scopus (60) Google Scholar). The antibiotic thus enters the cell best under neutral and acidic pH conditions and could inhibit cell protein synthesis strongly in sensitive cells in this pH range. Both Gram-positive and Gram-negative Tet efflux proteins catalyze similar exchange reactions which prevent cytoplasmic accumulation of the antibiotic. Tc is actively extruded, as a complex with a divalent cation that bears a single positive charge, in exchange for external H+ (2Yamaguchi A. Udagawa T. Sawai T. J. Biol. Chem. 1990; 265: 4809-4813Abstract Full Text PDF PubMed Google Scholar, 3Yamaguchi A. Shiina Y. Fujihira E. Sawai T. Noguchi N. Sasatsu M. FEBS Lett. 1995; 365: 193-197Crossref PubMed Scopus (29) Google Scholar). The smaller (12-transmembrane segments) Gram-negative Tet proteins and the larger (14-transmembrane segments) Gram-positive Tet proteins share sequence similarity largely in the N-terminal six transmembrane segments regions (4Levy S.B. Antimicrob. Agents Chemother. 1992; 24: 1-3Google Scholar, 5Paulsen I.T. Skurray R.A. Gene (Amst .). 1993; 124: 1-11Crossref PubMed Scopus (104) Google Scholar) but at least some motifs and/or residues in the C-terminal halves of each type of Tc efflux protein cannot be modified without loss of activity (6Fijihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 419: 211-214Crossref PubMed Scopus (8) Google Scholar, 7Yamaguchi A. Akasaka T. Ono N. Someya Y. Kakatani M. Sawai T. J. Biol. Chem. 1992; 267: 7490-7498Abstract Full Text PDF PubMed Google Scholar). Both the Gram-negative and Gram-positive Tet protein families contain examples that have further been shown to complement K+-uptake deficient mutants of Escherichia coli(8Dosch D.C. Salvacion F.F. Epstein W. J. Bacteriol. 1984; 160: 1188-1190Crossref PubMed Google Scholar, 9Griffith J.K. Cuellar D.H. Fordyce C.A. Hutchings K.C. Mondragon A.A. Mol. Membr. Biol. 1994; 11: 271-277Crossref PubMed Scopus (19) Google Scholar, 10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar, 11Guay G.G. Tuckman M. McNicholas P. Rothstein D.M. J. Bacteriol. 1993; 175: 4927-4929Crossref PubMed Google Scholar), but this net K+ uptake mode is not taken as a general property of Tet proteins. It has been attributed to an electrical potential-dependent K+ leak that could also be conferred by truncated forms of proteins that exhibit the property (10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar, 11Guay G.G. Tuckman M. McNicholas P. Rothstein D.M. J. Bacteriol. 1993; 175: 4927-4929Crossref PubMed Google Scholar, 12Nakamura T. Matsuba Y. Ishihara A. Kitagawa T. Suzuki F. Unemoto T. Biol. Pharm. Bull. 1995; 18: 1189-1193Crossref PubMed Scopus (6) Google Scholar). Recently, studies in this laboratory have shown that the chromosomally encoded Bacillus subtilis TetA(L) protein and closely related TetK from a Staphylococcus aureusplasmid catalyze Na+(K+)/H+antiport (13Cheng J. Guffanti A.A. Krulwich T.A. J. Biol. Chem. 1994; 269: 27365-27371Abstract Full Text PDF PubMed Google Scholar, 14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar, 15Cheng J. Guffanti A.A. Wang W. Krulwich T.A. Bechhofer D.H. J. Bacteriol. 1996; 178: 2853-2860Crossref PubMed Google Scholar, 16Cheng J. Hicks D.B. Krulwich T.A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14446-14451Crossref PubMed Scopus (47) Google Scholar, 17Cheng J. Baldwin K. Guffanti A.A. Krulwich T.A. Antimicrob. Agents Chemother. 1996; 40: 852-857Crossref PubMed Google Scholar) in addition to Tc−-Me2+/H+ antiport (2Yamaguchi A. Udagawa T. Sawai T. J. Biol. Chem. 1990; 265: 4809-4813Abstract Full Text PDF PubMed Google Scholar, 13Cheng J. Guffanti A.A. Krulwich T.A. J. Biol. Chem. 1994; 269: 27365-27371Abstract Full Text PDF PubMed Google Scholar, 14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar). These exchanges were evidently electrogenic, as assayed via energy-dependent Tc-cobalt or Na+ uptake by everted vesicles of E. coli that expressed a clonedtetA(L) gene from a weak promoter (14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar). The exchanges were not inhibited by low nigericin concentrations that reduce the ΔpH but were significantly inhibited by valinomycin in the presence of K+, a combination that abolished the ΔΨ generated by respiration (14Guffanti A.A. Krulwich T.A. J. Bacteriol. 1995; 177: 4557-4561Crossref PubMed Google Scholar). Consistently, the antiports catalyzed by purified and reconstituted TetA(L) could be energized by an imposed potential (16Cheng J. Hicks D.B. Krulwich T.A. Proc. Natl. Acad. Sci. U. S. A. 1996; 93: 14446-14451Crossref PubMed Scopus (47) Google Scholar). In addition, the important role of TetA(L) in acidifying the cytoplasm of B. subtilis relative to the external medium during growth at alkaline pH would require that the monovalent cation/H+mode be electrogenic (15Cheng J. Guffanti A.A. Wang W. Krulwich T.A. Bechhofer D.H. J. Bacteriol. 1996; 178: 2853-2860Crossref PubMed Google Scholar, 18McNab R.M. Castle A.M. Biophys. J. 1987; 52: 637-647Abstract Full Text PDF PubMed Scopus (33) Google Scholar). Since TetK could substitute for TetA(L) in a mutant of B. subtilis that had a disruptedtetA(L) gene, TetK is presumed to catalyze an electrogenic antiport similar to TetA(L). By contrast, the Tc-metal/H+antiport catalyzed by Gram-negative Tet proteins has been proposed to be electroneutral (19Kaneko M. Yamaguchi A. Sawai T. FEBS Lett. 1985; 193: 194-198Crossref PubMed Scopus (57) Google Scholar). Moreover, Yamaguchi and colleagues (3Yamaguchi A. Shiina Y. Fujihira E. Sawai T. Noguchi N. Sasatsu M. FEBS Lett. 1995; 365: 193-197Crossref PubMed Scopus (29) Google Scholar, 20Hirata T. Wakatabe R. Nielsen J. Someya Y. Fujihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 412: 337-340Crossref PubMed Scopus (12) Google Scholar) have experienced difficulty in demonstrating the Na+/H+ activity of TetK and indicate that the Tc-metal/H+ antiport activity of TetK appeared to be electroneutral in preliminary work. One of the goals of the current study, therefore, was to examine the Tc-metal/H+ and Na+/H+ antiport activities of TetA(L) and TetK side-by-side in comparable preparations and to clarify their electrogenicity versus electroneutrality. The studies have strongly supported the multifunctional and electrogenic nature of both TetA(L) and TetK. A second major goal of the studies was to test an alternate hypothesis to the putative K+ leak in explaining the ability of Tet proteins such as TetK to complement K+ uptake-deficientE. coli. The new hypothesis arises from the discovery that these Tet proteins are electrogenic monovalent cation/H+antiporters, i.e. have a catalytic mode in which cytoplasmic Na+ or K+ is exchanged for a greater number of external H+. If K+ were able to occupy the external H+ sites, then an exchange of cytoplasmic monovalent cation for a greater number of K+ could account for the net uptake of K+ (Fig. 1). Thus, the net uptake of K+ catalyzed by a Tet protein could be one of its normal catalytic modes. Experiments were designed to test this hypothesis with TetK and to examine whether TetA(L), to which this kind of activity had never been attributed, might nonetheless possess a comparable capacity. If the capacity to catalyze net K+ uptake was in fact a function of the monovalent cation antiport mode, then TetA(L) might well demonstrate it. Or, a capacity to catalyze net K+uptake might be restricted to those Tet proteins with both monovalent cation/H+ exchange activity and a particularly high affinity for K+. Both cloned TetA(L) and TetK were shown to restore Na+ exclusion capacity and resistance to a ΔtetA(L) strain of B. subtilis. In such experiments, the Na+ exclusion capacity of TetK, but not of TetA(L), was markedly reduced by the presence of K+. This suggested a higher K+ relative to Na+ affinity for TetK than for TetA(L) (15Cheng J. Guffanti A.A. Wang W. Krulwich T.A. Bechhofer D.H. J. Bacteriol. 1996; 178: 2853-2860Crossref PubMed Google Scholar). The current studies support the hypothesis that net K+ uptake catalyzed by full-length forms of TetA(L) and TetK is a mode of the Na+(K+)/H+ exchange of both proteins. DISCUSSIONThe studies conducted here confirm and extend earlier work indicating that both TetA(L) and TetK are multifunctional antiporters that catalyze electrogenic Tc-metal/H+ and Na+/H+ antiport. The successful demonstration of both of these activities and their electrogenicity clearly depends upon low expression levels of the proteins. Higher levels of expression make both cells and membrane leaky. For experiments in which ionophores are used to assess electrogenicity in an E. coli vesicle system, it is further important to use ionophore concentrations that avoid aberrant exchanges and to reduce the concentration of chloride sufficiently to avoid dissipation of the ΔΨ by the permeant anion alone. The totality of earlier experiments supports the conclusion by Kaneko and co-workers (19Kaneko M. Yamaguchi A. Sawai T. FEBS Lett. 1985; 193: 194-198Crossref PubMed Scopus (57) Google Scholar) that TetA(B) catalyzes a largely electroneutral Tc-metal/H+ antiport. However, these authors themselves indicate some discrepancies in their findings with the conclusion of complete electroneutrality, and the issue might merit re-examination. As discussed below, the specific catalytic properties and possible multifunctional features are important factors in the design of strategies to minimize the interference of antibiotic efflux systems with use of antimicrobial therapies.The truncated versions of TetA(L) and TetK failed to exhibit any of the energy-dependent, proton-coupled activities of the full-length proteins, consistent with the evidence that residues in the C-terminal halves of TetK cannot be mutated without loss of active Tc efflux capacity (6Fijihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 419: 211-214Crossref PubMed Scopus (8) Google Scholar). Nonetheless, there were some modest but reproducible protective effects of the truncated Tet proteins in the whole cell growth complementation experiments (Table I). Possibly the truncated forms retain the capacity to bind Tc, Tc-metal, and monovalent cations, and this accounts for those effects. Such a basis for modest complementation in similar experiments has previously been noted (33Ivey D.M. Guffanti A.A. Shen Z. Kudyan N. Krulwich T.A. J. Bacteriol. 1992; 174: 4878-4884Crossref PubMed Google Scholar).The current studies add a catalytic mode to the repertoire of the Gram-positive Tet proteins, i.e. a mode in which net K+ uptake is achieved via a full catalytic cycle in which more than one K+ is taken up in exchange for a single cytoplasmic Na+ or K+. Clearly, the full-length TetA(L) and TetK do not confer a leakiness upon E. colimembranes to K+ that allows electrogenic K+entry (even down its chemical concentration gradient) in response to energization and establishment of a sizeable Δψ, inside-negative. The generation of a potential, inside-positive, during Na+(K+)/Rb+ exchange by unenergized vesicles is consistent with the operation of the whole catalytic antiport cycle but with the external Rb+ substituting for H+. Were only a partial cycle to be used for the exchange, the Rb+ accumulation would represent counterflow entirely,i.e. with the intravesicular cation transported outward down its gradient, released, and then replaced on the outside with the external Rb+ without use of the “H+” sites. In that case, the exchange should have been electroneutral. The occupation of a cation site by either K+ or H+has similarly been proposed for the complete catalytic cycle of the eukaryotic serotonin transporter (34Keyes S.R. Rudnick G. J. Biol. Chem. 1982; 257: 1172-1176Abstract Full Text PDF PubMed Google Scholar). It will be important to confirm the modest exchange capacity of the truncated TetK (as well as the possible leak) and the lack of a comparable activity by truncated TetA(L) in a purified reconstituted system in which the amount of transporter protein incorporated into the proteoliposome can be made comparable for different versions of the proteins. If Tet-mediated, electrogenic Rb+(K+) uptake depends upon the use of the H+-binding site and translocation pathway by these cations, and if the C-terminal part of the protein is required for proton binding and/or translocation, then even modest net Rb+ accumulation by truncated TetK is unanticipated under non-leaky conditions.The finding that net K+ uptake by full-length Tet proteins is definitely a mode of the normal catalytic functions rather than a leak, is consistent with the robust growth of cells expressing low levels of these proteins. It is notable that TetA(L) behaved qualitatively similar to TetK although it had not earlier been implicated as having the capacity for net K+ uptake. As hypothesized at the start of the study, this capacity may be a correlate of possession by a Tet protein of Na+(K+)/H+ antiporter activity and the extent to which this property occurs broadly among Tet proteins has not been carefully examined. Another question of interest in connection with the net K+ uptake mode is whether it may have a physiological role, e.g. at particular pH values and/or K+ concentrations. It will be of importance to examine the possibility that the Gram-negative TetA(C) (e.g. from pBR322 or pACYC184) might catalyze a similar spectrum of activities to that shown here for the Gram-positive Tet proteins. TetA(C) is among the Tet proteins that can complement K+ uptake-deficient mutants of E. coli (8Dosch D.C. Salvacion F.F. Epstein W. J. Bacteriol. 1984; 160: 1188-1190Crossref PubMed Google Scholar, 9Griffith J.K. Cuellar D.H. Fordyce C.A. Hutchings K.C. Mondragon A.A. Mol. Membr. Biol. 1994; 11: 271-277Crossref PubMed Scopus (19) Google Scholar, 10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar). Moreover, this gene has been shown to have a beneficial effect on the “fitness” of adapted E. coliin the absence of antibiotic (35Lenski R.E. Simpson S.C. Nguyen T.T. J. Bacteriol. 1994; 176: 3140-3147Crossref PubMed Scopus (134) Google Scholar); this could reflect enhanced Na+-resistance and K+ retrieval under some conditions. Whether TetK confers such a benefit on S. aureuswill also be of interest to examine. Considerable current effort is directed toward reducing the prevalence and further spread of antibiotic-resistance genes among pathogenic bacteria or other organisms that might then transfer these genes to pathogens. In assessments of those conditions that will minimize positive selection for antibiotic efflux genes of particular types, e.g. tet genes, the full panoply of roles for the given efflux protein will be important information. For example, it might be important to consider the pH, Na+, and K+concentration to which the organisms are exposed, rather than simply the exposure to Tc, when evaluating strategies for decreasing the prevalence of TetA(L) or TetK. The studies conducted here confirm and extend earlier work indicating that both TetA(L) and TetK are multifunctional antiporters that catalyze electrogenic Tc-metal/H+ and Na+/H+ antiport. The successful demonstration of both of these activities and their electrogenicity clearly depends upon low expression levels of the proteins. Higher levels of expression make both cells and membrane leaky. For experiments in which ionophores are used to assess electrogenicity in an E. coli vesicle system, it is further important to use ionophore concentrations that avoid aberrant exchanges and to reduce the concentration of chloride sufficiently to avoid dissipation of the ΔΨ by the permeant anion alone. The totality of earlier experiments supports the conclusion by Kaneko and co-workers (19Kaneko M. Yamaguchi A. Sawai T. FEBS Lett. 1985; 193: 194-198Crossref PubMed Scopus (57) Google Scholar) that TetA(B) catalyzes a largely electroneutral Tc-metal/H+ antiport. However, these authors themselves indicate some discrepancies in their findings with the conclusion of complete electroneutrality, and the issue might merit re-examination. As discussed below, the specific catalytic properties and possible multifunctional features are important factors in the design of strategies to minimize the interference of antibiotic efflux systems with use of antimicrobial therapies. The truncated versions of TetA(L) and TetK failed to exhibit any of the energy-dependent, proton-coupled activities of the full-length proteins, consistent with the evidence that residues in the C-terminal halves of TetK cannot be mutated without loss of active Tc efflux capacity (6Fijihira E. Kimura T. Yamaguchi A. FEBS Lett. 1997; 419: 211-214Crossref PubMed Scopus (8) Google Scholar). Nonetheless, there were some modest but reproducible protective effects of the truncated Tet proteins in the whole cell growth complementation experiments (Table I). Possibly the truncated forms retain the capacity to bind Tc, Tc-metal, and monovalent cations, and this accounts for those effects. Such a basis for modest complementation in similar experiments has previously been noted (33Ivey D.M. Guffanti A.A. Shen Z. Kudyan N. Krulwich T.A. J. Bacteriol. 1992; 174: 4878-4884Crossref PubMed Google Scholar). The current studies add a catalytic mode to the repertoire of the Gram-positive Tet proteins, i.e. a mode in which net K+ uptake is achieved via a full catalytic cycle in which more than one K+ is taken up in exchange for a single cytoplasmic Na+ or K+. Clearly, the full-length TetA(L) and TetK do not confer a leakiness upon E. colimembranes to K+ that allows electrogenic K+entry (even down its chemical concentration gradient) in response to energization and establishment of a sizeable Δψ, inside-negative. The generation of a potential, inside-positive, during Na+(K+)/Rb+ exchange by unenergized vesicles is consistent with the operation of the whole catalytic antiport cycle but with the external Rb+ substituting for H+. Were only a partial cycle to be used for the exchange, the Rb+ accumulation would represent counterflow entirely,i.e. with the intravesicular cation transported outward down its gradient, released, and then replaced on the outside with the external Rb+ without use of the “H+” sites. In that case, the exchange should have been electroneutral. The occupation of a cation site by either K+ or H+has similarly been proposed for the complete catalytic cycle of the eukaryotic serotonin transporter (34Keyes S.R. Rudnick G. J. Biol. Chem. 1982; 257: 1172-1176Abstract Full Text PDF PubMed Google Scholar). It will be important to confirm the modest exchange capacity of the truncated TetK (as well as the possible leak) and the lack of a comparable activity by truncated TetA(L) in a purified reconstituted system in which the amount of transporter protein incorporated into the proteoliposome can be made comparable for different versions of the proteins. If Tet-mediated, electrogenic Rb+(K+) uptake depends upon the use of the H+-binding site and translocation pathway by these cations, and if the C-terminal part of the protein is required for proton binding and/or translocation, then even modest net Rb+ accumulation by truncated TetK is unanticipated under non-leaky conditions. The finding that net K+ uptake by full-length Tet proteins is definitely a mode of the normal catalytic functions rather than a leak, is consistent with the robust growth of cells expressing low levels of these proteins. It is notable that TetA(L) behaved qualitatively similar to TetK although it had not earlier been implicated as having the capacity for net K+ uptake. As hypothesized at the start of the study, this capacity may be a correlate of possession by a Tet protein of Na+(K+)/H+ antiporter activity and the extent to which this property occurs broadly among Tet proteins has not been carefully examined. Another question of interest in connection with the net K+ uptake mode is whether it may have a physiological role, e.g. at particular pH values and/or K+ concentrations. It will be of importance to examine the possibility that the Gram-negative TetA(C) (e.g. from pBR322 or pACYC184) might catalyze a similar spectrum of activities to that shown here for the Gram-positive Tet proteins. TetA(C) is among the Tet proteins that can complement K+ uptake-deficient mutants of E. coli (8Dosch D.C. Salvacion F.F. Epstein W. J. Bacteriol. 1984; 160: 1188-1190Crossref PubMed Google Scholar, 9Griffith J.K. Cuellar D.H. Fordyce C.A. Hutchings K.C. Mondragon A.A. Mol. Membr. Biol. 1994; 11: 271-277Crossref PubMed Scopus (19) Google Scholar, 10Griffith J.K. Kogoma T. Corvo D.L. Anderson W.L. Kazim A.L. J. Bacteriol. 1988; 170: 598-604Crossref PubMed Google Scholar). Moreover, this gene has been shown to have a beneficial effect on the “fitness” of adapted E. coliin the absence of antibiotic (35Lenski R.E. Simpson S.C. Nguyen T.T. J. Bacteriol. 1994; 176: 3140-3147Crossref PubMed Scopus (134) Google Scholar); this could reflect enhanced Na+-resistance and K+ retrieval under some conditions. Whether TetK confers such a benefit on S. aureuswill also be of interest to examine. Considerable current effort is directed toward reducing the prevalence and further spread of antibiotic-resistance genes among pathogenic bacteria or other organisms that might then transfer these genes to pathogens. In assessments of those conditions that will minimize positive selection for antibiotic efflux genes of particular types, e.g. tet genes, the full panoply of roles for the given efflux protein will be important information. For example, it might be important to consider the pH, Na+, and K+concentration to which the organisms are exposed, rather than simply the exposure to Tc, when evaluating strategies for decreasing the prevalence of TetA(L) or TetK.