15krulwich.pm

J. Mol. Microbiol. Biotechnol. (2001) 3(2): 237-246.
Multifunctionality of Tetracycline Efflux Functions 237 JMMB Symposium
Functions of Tetracycline Efflux Proteins
that Do Not Involve Tetracycline
Terry A. Krulwich*, Jie Jin, Arthur A. Guffanti, and David
transmembrane segments (TMS) and are encoded in the H. Bechhofer
chromosome or on plasmids found predominantly in Gram-positive prokaryotes (Speer et al., 1992; Roberts, 1996; Department of Biochemistry and Molecular Biology, Mount McMurry and Levy, 2000). The Group 1 Tet proteins, by Sinai School of Medicine, 1 Gustave L. Levy Place, New contrast, are predominantly found in Gram-negative prokaryotes and have 12-TMS. Both the 12-TMS and 14-TMS Tet proteins have significant deduced similarity toprokaryotic MDRs of the secondary antiporter type, e.g.
Abstract
Bmr and Emr, respectively. Like such MDRs, the Tet effluxproteins are members of the Major Facilitator Superfamily Tet(L) and Tet(K) are specific antibiotic-resistance
(MFS) of transport proteins (Marger and Saier, 1993). The determinants. They catalyze efflux of a tetracycline(Tc)-
Tet proteins have been further classified together with divalent metal complex in exchange for protons, as
MDRs and other drug efflux proteins in two families within do other Tet efflux proteins. These Tet proteins also
the MFS, based on topological similarities rather than catalyze Na+ and K+ exchange for protons. Each of the
clustering by specific substrate, i.e. “Drug:H + Antiporter “cytoplasmic substrates”, Na+, K+ and the Tc-metal ion
(14-Spanner) or DHA14” and “Drug:H + Antiporter (12- complex, can also be exchanged for K+, a catalytic
Spanner) or DHA 12” families (Paulsen et al., 1996; Pao mode that accounts for the long-recognized K+ uptake
et al., 1998). Tet efflux proteins of both the DHA 12 and capacity conferred by some Tet proteins. The multiple
DHA 14 families confer tetracycline (Tc)-resistance by virtue catalytic modes of Tet(L) and Tet(K) provide potential
of their capacity to actively translocate a complex of Tc new avenues for development of inhibitors of these
and a divalent metal ion that apparently bears a single net efflux systems as well as avenues for exploration of
positive charge; Co2+ is generally the most effective in structure-function relationships. The multiple catalytic
supporting transport, but Mg2+, Mn2+, Cd2+ and Ca2+ can modes of Tet(L), which is chromosomally encoded in
also complex with Tc to form a substrate (Yamaguchi et Bacillus subtilis, also correspond to diverse physiolog-
al., 1990; Yamaguchi et al., 1995; Guffanti and Krulwich, ical roles, including roles in antibiotic-, Na+-, and alkali-
resistance as well as K+ acquisition. The use of K+ as
Different groups and even classes of Tet efflux proteins an external coupling ion may contribute not only to
exhibit different patterns with respect to the capacity to the organism’s K+ uptake capacity but also to its ability
translocate diverse Tc analogues (Speer et al., 1992; Guay to exclude Na+ and Tc at elevated pH values. Regulation
and Rothstein, 1993). Given this sort of “classical” of the chromosomal tetL gene by Tc has been proposed
specificity and the absence of any known antibiotic to involve a translational re-initiation mechanism that
substrates for Tet proteins other than Tc and its analogues, is novel for an antibiotic-resistance gene and increases
the Tet efflux proteins cannot be described as bona fide Tet expression seven-fold. Other elements of tetL
MDRs. Strategies for inactivating these problematic expression and its regulation are already evident,
resistance determinants have accordingly focused on including gene amplification and use of multiple
elements related to tet gene regulation and proliferation in promoters. However, further studies are required to
response to exposure to Tc and on structure-based inhibitor clarify the full panoply of regulatory mechanisms, and
development using Tc as the starting point (Nelson et al., their integration to ensure different levels of tetL
1993, 1994; Nelson and Levy, 1999; Schnappinger and expression that are optimal for its different functions.
Hillen, 1996; Sum et al., 1998). But if not true MDRs, the It will also be of interest to investigate the implications
Tet(L) and Tet(K) proteins are clearly multifunctional of Tet(L) and Tet(K) multifunctionality on the
antiporters that are involved in resistance to multiple emergence and persistence of these antibiotic-
stresses (Padan and Krulwich, 2000). These stresses, resistance genes.
apart from antibiotic stress (Alekshun and Levy, 2000),include Na+ stress, alkali stress, and the challenge of K+ Introduction
insufficiency. Studies in our laboratory have shown thatTet(L) and Tet(K) utilize inorganic monovalent cations, Na+ Tet(L) and Tet(K)1, which are generally close to 60%identical to each other, are the major classes of tetracyclineefflux proteins (Tet proteins) among the Group 2 Tet proteins 1The nomenclature suggested by Levy et al. (1999) for tetracycline (McMurry and Levy, 2000). These proteins have 14- resistance determinants will be used in this article. Thus Tet(L) and Tet(K)will designate the proteins, and tetL and tetK will designate the genes ofclass L and K tetracycline resistance that were earlier named with indicationsof the allele, e.g. TetA(L) (Levy et al., 1989). For historical reasons (Williams *For correspondence. Email terry.krulwich@mssm.edu; and Smith, 1979), the B. subtilis tetL chromosomal locus was originally Tel. (212) 241-7280; Fax. (212) 996-7214.
designated tetB and is referred to as such in the genome annotation.
Figure 1. Catalytic modes of Tet(L) and Tet(K).
and K+, as alternate cytoplasmic substrates to the Tc- - surrounding the possibility that selective pressures other Me2+ complex, i.e. Tet(L) and Tet(K) act as Na+(K+)/H+ than antibiotics may influence their retention and antiporters (Cheng et al., 1994, 1996a,b,c; Guffanti and Krulwich, 1995), even though divalent inorganic cationsare not transported except in complex with Tc (Yamaguchi Catalytic Modes of Tet(L) and Tet(K)
et al., 1990; Cheng et al., 1996c). These proteins also havebeen shown to use K+ as well as H+ as the coupling ion The three catalytic modes of Tet(L) and Tet(K) are depicted whose entry down its electrochemical potential energizes in Figure 1. Evidence for Mode I, the Tetracycline Efflux the efflux of the cytoplasmic substrate (Guffanti et al., 1998).
mode, was anticipated by work with more intensively These additional catalytic modes, which will be detailed studied Tet proteins and was confirmed by Yamaguchi et further below, have in common the use of cationic al. (1995) and Guffanti and Krulwich (1995). Modes II and substrates with single net charge, but the cytoplasmic III, the Monovalent cation/H+ Antiport and Monovalent substrates range from a Tc- -Me2+ complex that has regions cation or Tc- -Me2+/K+ Antiport (the Net K+ Uptake Mode), with highly non-polar features to highly hydrated respectively, were unanticipated, although Ives and Bott monovalent cations. The substrate diversity of Tet(L) and (1990) speculated that tetL was likely to have a Tc-unrelated Tet(K) opens up some of the same questions that are function of physiological significance. Chromosomally central to function of those MDRs whose substrates encoded Tet(L) from Bacillus subtilis was identified as a encompass diverse structures and both non-polar and polar possible Na+/H+ antiporter after a mutant strain with a compounds. The multiple catalytic modes of Tet(L) and disruption in the tetL promoter region was isolated from Tet(K), and their important physiological roles, also open transposition libraries on the basis of Na+- and alkali- up new approaches to structure-function analyses and to sensitivity (Cheng et al., 1994; Krulwich et al., 1994). Na+/ development of Tet protein inhibitors. The multiple roles H+ antiport activity, in addition to Tc- -Co2+ antiport, was have several further implications. These include the subsequently demonstrated in assays of everted possible evolutionary relationship between these resistance membrane vesicles of Escherichia coli strains expressing determinants and house-keeping antiporters, and issues either tetL or tetK (Cheng et al., 1994, 1996a; Guffanti and Table 1. Cytoplasmic concentrations of Na+ after 15 h of growth of transformants of B. subtilis wild type and tetL deletion strain JC112 in malate mediumat pH 8.3 containing 100 mM Na+ and either 1 mM or 100 mM K+ Cytoplasmic Na+ (mM +/- standard deviation) Data reproduced from Cheng et al., 1996b with permission.
Multifunctionality of Tetracycline Efflux Functions 239 Krulwich, 1995; Guffanti et al., 1998) and of pACYC184, and Tet(K), in particular, had been reported to proteoliposomes reconstituted with purified, C-terminally, complement K+ uptake-deficient strains of E. coli (Dosch hexa-histidine tagged Tet(L) (Cheng et al., 1996c). In the et al., 1984; Griffith et al., 1988; Guay et al., 1993). It had latter assays, 86Rb+ translocation in response to an been suggested that the presence of one of these Tet imposed pH gradient — an assay of K+/H+ antiport — was proteins in the membrane, especially as expressed from a also shown. Several observations indicated that the Tet- multicopy plasmid, led to an electrogenic leak of K+ across mediated Tc- -Me2+/H+ and Na+(K+)/H+ antiports (Modes I the membrane (Guay et al., 1993), although Guay and and II) are both electrogenic, i.e. the y:x ratio in Figure 1 is Rothstein (1993) also noted the possibility that the capacity greater than unity. If electrogenic in this manner, each for K+ uptake could relate to the H+ or hydronium uptake turnover results in a greater number of the coupling ions associated with antiport energization. The N-terminal entering a right-side-out system than the number of Na+, domain of the Tet protein was particularly implicated in the K+ or Tc- -Me2+ complexes effluxing. Net positive charge promotion of K+ influx (Guay et al., 1993; Griffith et al., would thus be translocated inward during each turnover of 1988, 1994; Nakamura et al., 1995). Since Tet(K) the transporter, accounting for the observation that abolition possessed a strong K+/H+ antiport capacity, with H+/K+ >1, of the transmembrane electrical potential (∆Ψ) markedly we hypothesized that rather than a leak, the K+ uptake inhibits these antiports (Guffanti and Krulwich, 1995; mode might be a mode of the antiport (Mode III) in which Guffanti et al., 1998) and, in proteoliposomes, an imposed K+ serves as the external coupling ion (Guffanti et al., 1998).
transliposomal electrical potential can energize antiport In the electrogenic exchange, even when K+ was also the (Cheng et al., 1996c). The actual stoichiometry of Tet(L) cytoplasmic substrate, net K+ uptake would result.
and Tet(K)-mediated antiports has yet to be determined.
Moreover, a controlled mode of the K+ antiport rather than In the assays conducted thus far, both K+ and Li+ have a K+ leak fit better with the observation that K+ uptake- been found to inhibit Na+/H+ antiport but cross inhibition deficient mutants of E. coli grew robustly when transformed between Na+ and the Tc- -Me2+ complex has not yet been with plasmids expressing tet(K). This would not be expected demonstrated (Guffanti and Krulwich, 1995; Cheng et al., if the cost of K+ acquisition was an un-regulated 1996c). Lack of cross-inhibition might reflect the existence electrogenic leak that would dissipate the ∆Ψ in an adverse of distinct binding domains for the different substrates.
manner. Strong support for the Mode III, Net K+ Uptake However, we would not yet rule out the possibility that the Mode of the Tet antiport, was obtained in assays of right- optimal assay conditions for demonstrating cross-inhibition side-out vesicles from E. coli TK2420 (K+ uptake deficient) have not yet been developed. This is an important issue to expressing Tet(K) or Tet(L). An outwardly-directed gradient resolve and is of particular interest given the report of of Na+, K+, or Tc- -Me2+ drove electrogenic entry of 86Rb+; multiple drug interaction sites in the lactococcal MDR, LmrP the Rb+ accumulation was not supported by intravesicular (Putman et al., 1999). Further definition of the pH profiles choline, and not supported by Tet-less vesicles (Guffanti of each substrate exclusion activity should also be of et al., 1998; J. Jin, unpublished data). In a second interest in delineating how the combination of modes works.
experimental protocol, in which intravesicular choline or Initial findings indicate a greater Tc efflux activity at near K+ and extravesicular K+(Rb+) were at equal neutral pH and a greater Na+ efflux activity at elevated pH concentrations, energy-dependent Rb+ accumulation did not occur upon generation of the transmembrane ∆p in It will also be of interest to quantify the relative Km the choline-containing vesicles, as would have been values of Tet(L) and Tet(K) for K+ and Na+ in the Na+(K+)/ predicted for a leak, but depended upon the presence of a H+ antiport. A relatively higher K+ than Na+ preference for Tet protein substrate inside the vesicles (Guffanti et al., Tet(K), compared to Tet(L), was indicated by the results of 1998). In both experimental protocols, the K+ uptake mode a “Na + exclusion assay” from whole cells of tetL deletion was more active in Tet(K) than in Tet(L) vesicles although strain JC112 of B. subtilis (Cheng et al., 1996b). As shown the activity of the latter was clear. It has not been shown in Table 1, cells of wild type B. subtilis, with a normal conclusively that K+ ions can serve as the sole coupling chromosomal tetL and transformed with a control plasmid, ions, rather than moving inward together with protons.
excluded 22Na+ effectively during growth in medium However, recent data on the pH dependence of the Net K+ containing 100 mM Na+ and either 1 mM or 100 mM K+. In Uptake Mode suggest that K+ can replace the H+ coupling the presence of the higher K+ concentration, only a modest elevation of Na+ accumulation in the cytoplasm was Important questions are raised by the finding of the observed relative to that found in the lower K+ medium , third, Net K+ Uptake Mode, apart from physiological i.e. 43 vs. 32 mM. The tetL deletion strain, transformed questions that will be considered below. Among these with the control plasmid, exhibited greatly reduced capacity questions are the following. First, does this mode extend to exclude Na+. Upon transformation with a recombinant the capacity of Tet(L) and Tet(K) proteins to support Tc plasmid expressing tetL, the parental pattern was restored and Na+ exclusion in the alkaline range of pH, driven by to the mutant strain. By contrast, upon transformation with the ∆Ψ, but independent of the reduced ∆pH (see Figure a recombinant plasmid expressing tetK, effective Na+ 1)? This question could be explored using recently isolated exclusion was only observed in the low K+ medium.
mutants with a specific reduction in this Mode. Second, do Once the Monovalent cation/H+ Antiport Mode (Figure the original indications that expression of only the N- 1, Mode II) of both Tet(L) and Tet(K) had been established, terminal one-quarter to half of Tet proteins can complement we hypothesized that a long observed capacity of some K+ uptake-deficient E. coli strains reflect a capacity of Tet proteins to support K+ uptake might be another mode truncated Tet proteins to catalyze some antiport modes? of the Tet antiport activities. Tet(C), e.g. from pBR322 or Can truncated forms catalyze electrogenic monovalent Cytoplasm
Membrane D1 L
Periplasm
Figure 2. Topological model of Tet(L) showing motifs summarized by Paulsen et al. (1996). The model fits the deduced sequence of the chromosomally-encoded B. subtilis Tet(L) to the topological data obtained experimentally for Tet(K) (Ginn et al., 1997; Hirata et al., 1998).
cation/K+ exchange or are the truncated forms really just Monovalent cation/H+ Antiport and the Net K+ Uptake leaking K+? An initial set of experiments on one set of highly Modes reflects the sharing of common structural features truncated Tet(L) and Tet(K) forms did not indicate a capacity for the cytoplasmic and coupling ions in different for antiport (Guffanti et al., 1998), but the strength of the conformations of the transporter. This has been indicated earlier observations makes it worthwhile to undertake more for the SMR protein EmrE (Yerushalmi and Schuldiner, extensive studies of different truncated forms if examples 2000). Finally, it will be of interest to determine whether of such forms can be identifed that are substantially the robust K+ uptake supported by Tet(C) may also reflect assembled into the membrane. Third, it will be of great the capacity of this Group 1 Tet protein for antiport modes interest to probe whether the apparently greater K+ II and III. The Tc- -Co2+/H+ antiport catalyzed by TetB, the translocation by Tet(K) relative to Tet(L) in both the Tn10-encoded Tet efflux protein, which is a 12-TMS Group Table 2. Tc-, Na+- and alkali-related phenotypes of B. subtilis tetL mutant types typified by strains JC112 and JC112C B. pH Homeostasis in Wild type and JC112 Transformants Cytoplasmic pH, 10 min after pH 7.5 -> 8.5 shift Data for parts A and B, shown with standard deviations, were reproduced with permission from the publisher from Wang et al., 2000 and Cheng et al., 1996b,respectively.
Multifunctionality of Tetracycline Efflux Functions 241 1 Tet protein like Tet(C), has been proposed to catalyze an mutants that lack Na+ efflux capacity also lack Tc- -Me2+ electroneutral exchange (Yamaguchi et al., 1991a). If the efflux capacity and vice versa. Given hypotheses about exchanges catalyzed by the 12-TMS Tet proteins are specific motifs, more targeted questions can also be posed.
indeed electroneutral, they would not be expected to be For example, Ginn et al. (2000) recently reported on a able to achieve net K+ uptake via a K+/K+ exchange, but detailed analysis of the Motif C of Tet(K), using site-directed could do so with a Tc- -Co2+/K+ exchange. The possibility mutagenesis of individual residues and studying their of an electrogenic K+/H+(K+) antiport capacity for Tet(C) effects upon the Tetracycline Efflux Mode of transport. The would also be worth examining directly since K+ uptake is capacity for Net K+ Uptake was assayed in a growth assay observed in the absence of antibiotic in pBR322 and only, as a general assessment of transporter functionality pACYC184 transformed cells (Dosch et al., 1984; Griffith to supplement data from direct membrane incorporation et al., 1988; Nakamura et al., 1995).
assays. Because Motif C has been proposed to beimportant in the coupling aspects of the DHA 14 Opportunities for Structure-Function Studies
mechanism, i.e. to be an “antiporter-associated motif”(Varela et al., 1995), we have recently undertaken studies Motifs that are common to DHA 12 and/or DHA 14 type of selected residues within this same motif, in Tet(L) and transporters in general have been reviewed in detail by Tet(K), but carried out quantitative vesicle assays of the Paulsen et al. (1996). A topological model of the Net K+ Uptake Mode. Thus far, these studies strongly chromosomally encoded Tet(L) from B. subtilis, showing support the conclusion that important determinants of the the motif regions in this particular protein, is presented in capacity for K+ uptake, and the greater capacity in Tet(K) Figure 2; it was drawn using the experimental data obtained vs. Tet(L), reside in Motif C in the region surrounding a for Tet(K) topology by Ginn et al. (1997) and Hirata et al.
highly conserved proline, P156 (J. Jin, unpublished data).
(1998). Extensive studies of mutations across the whole It will be of particular interest to determine whether single- molecule of 12-TMS Tet(B), especially by Yamaguchi and site mutants of Tet(L) that are Tet(K)-like with respect to colleagues, have provided a wealth of information. This the high activity of the K+ Uptake Mode also exhibit the includes data about the dispensability of particular residues greater K+:Na+ preference ascribed to Tet(K) relative to for Tc translocation, indications of residues and regions that may be involved in antibiotic substrate binding, and The other arena in which the catalytic activities that evidence for the involvement of particular TMS in forming do not involve Tc may be useful is that of Tet protein the perimeter of a water-filled channel (Iwaki et al., 2000; inhibitors. There are indications that amiloride inhibits Kimura-Someya et al., 2000; Kubo et al., 2000, and see Tet(L)-mediated Na+ fluxes (Guffanti and Krulwich, 1995).
McMurry and Levy, 2000). Hirata et al. (1998), noting that It will be of interest to more fully examine the Na+ and/or Tet(B) and Tet(K) share at least some important residues K+ related activities of these proteins as possible new of comparable charge in analogous positions, suggested avenues for inhibition of the transporter as a whole.
the possibility that the fundamental 3-dimensional structureof these 12- and 14-TMS Tet proteins will turn out to be Physiological Roles and Regulation of 14-TMS Tet
similar, but direct structural information is not yet available.
Proteins
Although specific information about regions and residuesin Tet(K) has begun to emerge with respect to the Tc- - The physiological roles of B. subtilis Tet(L), that could also Me2+ Efflux Mode (Fujihara et al., 1996; Ginn et al., 2000), be supported by Tet(K), are shown in Figure 1. These roles this information is still too fragmentary to assess the extent were inferred from the phenotypes of mutants with an of the correspondence between key residues and regions insertion of a chloramphenicol-resistance cassette in place of the 14- and 12-TMS Tet proteins. Nor has the oligomeric of the tetL gene (Cheng et al., 1996b; Wang et al., 2000).
state of 14-TMS Tet proteins been established, a property Ives and Bott (1990) had earlier been unable to isolate that can be important for developing precise mechanistic tetL deletion mutants, which led them to propose, as already models. Studies of separate expression of two equal-sized noted, that this locus probably had an important domains from closely related 12-TMS Tet proteins had physiological role. Perhaps sensitivities that were not yet earlier shown apparent domain interactions that led to anticipated, especially to [Na+], complicated there deletion activity and that could involve intermolecular interactions attempts. Even our successfully isolated mutant strains are (Rubin and Levy, 1990, 1991). More recently, McMurry and complicated by a pleiotropy in which two dominant types Levy (1995) showed that the N-terminal half of Tet(B) have been noted in a single isolation protocol (Cheng et contains a “dimerization domain” and presented a general al., 1996b; Wang et al., 2000). One type, which is less model for intra and intermolecular interactions. Dimers and frequently found and the more severely affected, is typified higher oligomers of several secondary transporters from by the mutant strain JC112. The more frequently found eukaryotes have been proposed (e.g. Casey and and less affected mutant type is typified by strain JC112C.
Reithmeier, 1991; Fafournoux et al., 1994; Kilic and The Tc- and Na+-sensitivities of these mutants, and the restoration of the wild type level of resistance by re- As the structure-function relationships in 14-TMS Tet introduction of a functional tetL gene, are shown in Table proteins are further probed, there will be value added if 2A. JC112C exhibits wild type resistance to Na+, even at the studies encompass effects on the other modes of alkaline pH, and exhibits a Tc-sensitivity that is intermediate transport, i.e. utilizing cytoplasmic substrates other than a Tc- -Me2+ complex or using K+ instead of H+. At the simplest The frequency with which tetL deletion is achieved, level, it will be of interest to examine the extent to which i.e. with no lower frequency than a comparable disruption B.subtilis Strains
Figure 3. Dependence of the growth of wild type B. subtilis and two phenotypic types of tetL deletion mutants on the [K+] of the medium. The data arereplotted from Wang et al. (2000).
of the dispensable amyE locus, makes it unlikely that the proteins that occurred as part of the putative suppression mutational loss of tetL is rapidly lethal in B. subtilis.
However, there may be strong enough selective pressure The control of tetB expression from Tn10, like the to promote the appearance of suppressor mutations, as genes encoding several other 12-TMS Tet proteins, is has been observed in a multiply Na+/H+ antiporter-deficient mediated by a repressor that is expressed from a divergent strain of E. coli (Harel-Bronstein et al., 1995). We have promoter (Bertrand et al., 1983). By contrast, plasmid and hypothesized that there are different paths of suppression, chromosomally encoded Tet(K) and Tet(L) proteins have of differing efficacy, that may be adopted in such been presumed to utilize largely, if not entirely, translational circumstances (Wang et al., 2000). The transporter(s) that attenuation control mechanisms because of the presence accounts for the better “compensation” of the JC1 12C of a leader sequence upstream of the structural gene mutant type has not yet been identified. However, (Hoshino et al., 1985; Khan and Novick, 1983; Mojumdar expression of genes encoding two different transport and Khan, 1988). Using transcriptional and translational systems that are putatively involved with monovalent cation fusions, increases in B. subtilis tetL expression up to 20- homeostasis in B. subtilis were found to be elevated in the fold have been found in response to Tc. Using diverse tetL deletion mutants. This elevation was not reversed by mutations in the leader region, we found that a significant re-introduction of a functional tetL gene (Wang et al., 2000) portion of this Tc-regulated expression was best accounted and is not observed in mutants in another locus, e.g. amyE for by a translational re-initiation model that had not earlier locus, prepared in parallel. Clarification of the events that been proposed for an antibiotic-resistance gene occur upon tetL disruption in B. subtilis will be of interest (Stasinopoulos et al., 1998). Preliminary data indicate that as a potential avenue into the apparently complex systems Tc also leads to a 3-fold increase in tetL mRNA stability by physiology of this multifunctional gene.
a mechanism that has yet to be elucidated (S.
Using the JC112 mutant, in which Na+- and alkali- Stasinopoulos, unpublished data). Mojumdar and Khan sensitivity are evident, a role for Tet(L) in both Na+- and (1988) had noted both an increase in tetK mRNA and a K+-dependent pH homeostasis was shown in pH shift higher increase in Tet protein upon induction of the tetK experiments (Table 2B) (Cheng et al., 1996b). Tet(L) is gene of pT181 by Tc in B. subtilis minicells. Additional apparently physiologically important in cytoplasmic pH elements that may, under some circumstances, contribute regulation, a role that is supported by the poor growth of to changes in Tet(L) levels were revealed by second-site JC112 at alkaline pH (Cheng et al., 1996b). Both tetL mutant mutations that restored wild type or higher levels of Tc- strains also exhibit some growth deficit at pH 7.0 that may resistance to B. subtilis strains that had been rendered partly reflect a physiological role for the Net K+ Uptake sensitive by deletion of the polynucleotide phosphorylase Mode of Tet(L). As shown in Figure 3, the two mutant strains gene. These mutations included: 4-5 fold tetL gene are particularly deficient in their ability to grow at low [K+] amplification; and a single nucleotide change in the tetL (Wang et al., 2000). JC112, especially, exhibits some promoter region that resulted in 10-fold higher transcription residual growth deficiency at pH 7 even when a functional from a normally low basal level (Bechhofer and tetL is restored and no antibiotic or monovalent cation Stasinopoulos, 1998). Two promoters for the monocistronic stresses are imposed. This residual deficit may reflect a tetL gene have been found in B. subtilis (Cheng et al., secondary effect of the elevation of multiple membrane 1996b; Bechhofer and Stasinopoulos, 1998). However, Multifunctionality of Tetracycline Efflux Functions 243 Figure 4. Tetracycline uptake and accumulation and its anticipated reduction by either Na+/H+ antiport or the more specific and effective Tc– -Me2+/H+ antiport:part of the functional rationale for conjoining these activities.
transcriptional regulation that might account for the particular have been proposed to have arisen on two or observation of Na+- and alkali-sensitivity of the original three separate occasions, presumably from other transport transposition mutant, which has an insertion in the promoter proteins (Sheridan and Chopra, 1991). The 12-TMS Tet region (Cheng et al., 1994), has not yet been found. Nor is and other DHA 12 proteins are thought to have arisen by the basis for the low basal level of tetL expression gene duplication from an ancestral 6-TMS transporter understood. It is likely that the levels of tetL expression (Levy, 1992; Paulsen and Skurray, 1993). Whereas the that are needed to exclude Tc, which can partition across 14-TMS proteins such as Tet(L) and Tet(K) might have the membrane (see section below), may be too high for involved the fusion of the putative 6-TMS ancestor with an optimal growth of B. subtilis under other conditions, 8-TMS interloper, features of the C-terminal 6-TMS of such including the moderate stresses and challenges of Na+, proteins have led to the alternate proposal that the DHA alkali and K+ insufficiency in which Tet(L) function is 14 proteins resulted from the incorporation of two new, involved. Different, more temperate mechanisms of control central TMS in between the halves of a 12-TMS efflux may obtain in response to these stresses, and it will be of protein (Griffith et al., 1992; Paulsen and Skurray, 1993).
Perhaps the multifunctional Tet proteins, which may include 12-TMS Tet(C) (Griffith et al., 1994) as well as the Implications of 14-TMS Tet Multifunctionality for
14-TMS Tet(L) and Tet(K), evolved from structurally related Theories of Origin and for Conferral of Competitive
transport proteins that have monovalent cation/H+ antiport activity. The genome of B. subtilis contains numerous geneswhose products have strong sequence similarity and similar The antibiotic-producing organisms are an obvious source predicted topologies to Tet(L), as ascertained using BLAST of “the ancestral” antibiotic resistance determinants. To (Altschul et al., 1990) for example. The best matches the extent that production of antibiotics serves an offensive among dozens of such gene products lack specific residues function in the ecosystem for the producer, these organisms that are believed to be essential for Tc efflux, e.g. E152.
must effectively efflux the antibiotic (Saunders, 1984; Perhaps one or more of these proteins nonetheless has Cundliffe, 1989). However, sequence analyses of diverse some modest Tc efflux capacity and/or have monovalent types of multi-drug and specific drug efflux proteins (Saier cation-related functions. We have hypothesized (Cheng et al., 1998) suggested that the capacity for drug efflux et al., 1996a) that there is an inherent “logic” to the emerged in each category of membrane transport proteins conjoining of this antiport capacity with that which excludes from proteins that had their own physiological roles, Tc. As illustrated in Figure 4, Tc is proposed to be although this was a rare event. Subsequently, the primordial predominantly charged at near neutral pH values but the drug transporter was maintained over a long time period uncharged component would partition across the during which the substrate specificity underwent membrane in a carrier-independent manner (Nikaido and considerable change and diversification. Tet proteins in Thanassi, 1993; Sigler et al., 2000). Yamaguchi et al.
(1991b) presented evidence that protonation of the et al., 1986; Udo and Grubb, 1996). Simpson et al. (2000) abundant Tc form that bears net negative charge results in recently studied the competitive fitness of S. aureus that carrier-independent, ∆pH-dependent Tc accumulation.
expressed tetK from plasmid pT181 vs. chromosomally Even though it has been suggested that inward fluxes of integrated forms of tetK that were expressed from different both Tc forms are likely to be slow (Sigler et al., 2000), Tc promoter configurations. They found that the would accumulate more and be most toxic at pH values in chromosomally integrated form, in spite of a lower gene the neutral and acidic range, at which the ∆pH component dosage, exhibited higher Tc resistance and higher of the ∆p was maximal. Moreover, enhanced expression competitive fitness than the S. aureus carrying pT181. The of monovalent cation/H+ antiporters, e.g. the Nha shown enhancement of competitive fitness was only observed in in Figure 4, would reduce Tc toxicity even without any the presence of Tc in these studies, but it would be of capacity for Tc efflux since they would reduce the ∆pH. A interest to re-examine the different constructs under Tet protein that both lowered the ∆pH and catalyzed active conditions that included Na+-, alkali- or low K+ challenge.
extrusion of the Tc- -Me2+ complex would, of course, be farmore effective for Tc exclusion. Having acquired the added Acknowledgement
capacity to catalyze Tc- -Me2+/H+(K+) antiport, a Tet protein The work in the authors’ laboratories was supported by research grant that evolved from a Na+ and/or K+/H+(K+) antiporter might GM52837 from the National Institute of General Medical Sciences.
at least sometimes retain all these catalytic capacities, as References
observed in Tet(L) and Tet(K). Even if all the substrates ofsuch multifunctional antiporters turn out to utilize a common Alekshun, M.N., and S.B. Levy. 2000. Bacterial drug resistance: response cytoplasmic binding site on the Tet protein, different pH- to survival threats. In: Bacterial Stress Responses. (G. Storz and R.
dependence for binding of the different substrates could Hengge-Aronis, eds) pp. 323-366, ASM Press, Washington, D.C.
Altschul, S.F., W. Gish, W. Miller, E.W. Myers, and D.J. Lipman. 1990. Basic prioritize the most physiologically important catalytic mode local alignment search tool. J. Mol. Biol. 215: 403-410 under particular conditions of pH. The Tc exclusion activity Amano, H. , C. L. Ives, K. F. Bott, and K. Shishido. 1991. A limited number (Mode I) is most important in the neutral to acidic pH range of Bacillus subtilis strains carry a tetracycline-resistance determinant at asite close to the origin of replication. Biochim. Biophys. Acta. 1088: 251- where Tc is most toxic, whereas the Na+- and alkali- resistance functions (Mode II and III) would be most Bechhofer, D.H., and S.J. Stasinopoulos. 1998. tetA(L) mutants of a important in the alkaline range (Padan and Krulwich, 2000).
tetracycline-sensitive strain of Bacillus subtilis with the polynucleotide Might the multiple functions of Tet proteins also provide phosphorylase gene deleted. J. Bacteriol. 180: 3470-3473 Bertrand, K.P., K. Postle, L.V. Wray, and W.S. Reznickoff. 1983. Overlapping selective pressure for retention of tet genes independently divergent promoters control expression of Tn10 tetracycline resistance.
of whether Tc was in abundance in the environment? This is an important issue to consider in light of the findings Bouma, J.E., and R.E. Lenski. 1988. Evolution of a bacteria/plasmid with Tet(L) and Tet(K) and the current focus on reducing Casey, J.R. and R.A. Reithmeier (1991) Analysis of the oligomeric state of antibiotic levels in the environment as a means of reducing Band 3, the anion transport protein of the human erythrocyte membrane, the prevalence of resistance determinants (Levy, 1997).
by size exclusion high performance liquid chromatography. Oligomeric In this connection, a finding of Lenski and colleagues stability and origin of heterogeneity. J. Biol. Chem. 266: 15726-15737 Cheng, J., K. Baldwin, A.A. Guffanti, and T.A. Krulwich. 1996a. Na+/H+ (Bouma and Lenski, 1988; Lenski et al., 1994; Lenski, 1997) antiport activity conferred by Bacillus subtilis tetA(L), a 5' truncation product is intriguing. These investigators studied the competitive of tetA(L), and related plasmid genes upon Escherichia coli. Antimicrob.
fitness of plasmid-bearing E. coli that had “co-evolved” Cheng J., A.A. Guffanti, and T. A. Krulwich. 1994. The chromosomal over 500 generations with pACYC184. The question posed tetracycline resistance locus of Bacillus subtilis encodes a Na+/H+ was whether, after prolonged co-existence mandated by antiporter that is physiologically important at elevated pH. J. Biol. Chem.
the presence of Tc or chloramphenicol, the separated, co- evolved host or plasmid would show a different response Cheng, J., A.A. Guffanti, W. Wang, T.A. Krulwich, and D.H. Bechhofer.
1996b. Chromosomal tetA(L) gene of Bacillus subtilis: regulation of than naive host or plasmid in a new engagement. Normally expression and physiology of a tetA(L) deletion strain. J. Bacteriol. 178: a plasmid imposes a burden on a host in the absence of antibiotic selection. The co-evolved plasmid exhibited no Cheng, J., D.B. Hicks, and T. A. Krulwich. 1996c. The purified Bacillus subtilis tetracycline efflux protein TetA(L) reconstitutes both tetracycline-cobalt/ change in the imposition of this burden when introduced H+ and Na+/H+ exchange. Proc. Natl. Acad. Sci. USA. 93: 14446-14451 into a new “naive” host E. coli. By contrast, the co-evolved Cundliffe, E. 1989. How antibiotic-producing organisms avoid suicide. Ann.
E. coli host that had been cured of the plasmid no longer showed a burden when fresh, naive pACYC184 was Dosch, D.C., F.F. Salvacion, and W. Epstein. 1984. Tetracycline element of pBR322 mediates potassium transport. J. Bacteriol. 160: 1188-1190 reintroduced (Bouma and Lenski, 1988). In fact, a benefit Fafournoux, P., J. Noel, and J. Pouyssegur. 1994. Evidence that the Na+/ was observed. Subsequent studies showed that this benefit H+ exchanger isoforms NHE1 and NHE3 exist as stable dimers in was specifically related to the tetC gene of the plasmid membranes with a high degree of specificity for homodimers. J. Biol.
Chem. 269: 2589-2596 (Lenski et al., 1994). Since tetC, in particular, introduces Fujihira, E., T. Kimura, Y. Shiina, and A. Yamaguchi. 1996. Transmembrane capacities for K+ uptake as well as Tc-resistance, the glutamic acid residues play essential roles in the metal-tetracycline/H+ possibility is raised that this and the even more antiporter of Staphylococcus aureus. FEBS Lett. 391: 243-246 multifunctional Tet(K) and Tet(L) proteins would offer Gillespie, M.T., J.W. May, and R.A. Skurray. 1986. Detection of an integrated tetracycline resistance plasmid in the chromosome of methicillin-resistant antibiotic-unrelated advantages to even a modestly Staphylococcus aureus. J. Gen. Microbiol. 132: 1723-1728 adapted host, depending upon the environmental Ginn, S.L., M.H. Brown, and R.A. Skurray. 1997. Membrane topology of conditions. There are tetL naive strains of B. subtilis that the metal-tetracycline/H+ antiporter TetA(K) from Staphylococcus aureus.
J. Bacteriol. 179: 3786-3789
could be used to examine this issue (Amano et al., 1991).
Ginn, S.L., M.H. Brown, and R.A. Skurray. 2000. The TetA(K) Tetracycline/ In this connection, it is notable that tetK has been found to H+ antiporter from Staphylococcus aureus: mutagenesis and functional incorporate into the chromosome of S. aureus (Gillespie analysis of motif C. J. Bacteriol. 182: 1492-1498 Multifunctionality of Tetracycline Efflux Functions 245 Griffith, J.K., M.E. Baker, D.A. Rouch, M.G. Page, R.A. Skurray, I.T. Paulsen, Levy, S.B., L.M. McMurry, V. Burdett, P. Courvalin, W. Hillen, M.C. Roberts, K.F. Chater, S.A. Baldwin, and P.J. Henderson. 1992. Membrane transport and D.E. Taylor. 1989. Nomenclature for tetracycline resistance proteins: implications of sequence comparisons. Curr. Opin. Cell Biol. 4: determinants. Antimicrob. Agents Chemother. 33: 1373-1374 Marger, M.D. and M.H. Saier, Jr. 1993. A major superfamily of Griffith, J.K., D.H. Cuellar, C.A. Fordyce, K.C. Hutchings and A.A.
transmembrane facilitators that catalyse uniport, symport and antiport.
Mondragon. 1994. Structure and function of the class C tetracycline/H+ antiporter: three independent groups of phenotypes are conferred by McMurry, L.M., and S.B. Levy. 1995. The NH2-terminal half of the Tn10 tetracycline efflux protein TetA contains a dimerization domain. J. Biol.
Griffith, J.K. T. Kogoma, D.L. Corvo, W.L. Anderson, and A.L. Kazim. 1988.
An N-terminal domain of the tetracycline efflux protein increases McMurry, L.M, and S.B. Levy. 2000. Tetracycline resistance in Gram-positive susceptibility to aminoglycosides and complements potassium uptake bacteria. In: Gram Positive Pathogens (V.A. Fischetti, R.P. Novick, J.J.
defects in Escherichia coli. J. Bacteriol. 170: 598-604 Ferretti, D.A. Portnoy, J.I. Rood, eds.) pp 660-677, ASM Press, Guay, G.G., and D.M. Rothstein. 1993. Expression of the tetK gene from Staphylococcus aureus in Escherichia coli: comparison of substrate Mojumdar, M., and S.A. Khan. 1988. Characterization of the tetracycline specificities of TetA(B), TetA(C), and TetK efflux proteins. Antimicrob.
resistance gene of plasmid pT181 of Staphylococcus aureus. J. Bacteriol.
Guay, G.G., M. Tuckman, P. McNicholas, and D.M. Rothstein. 1993.
Nakamura, T., Y. Matsura, A. Ishihara, T. Kitagawa, F. Suzuki, and T.
Expression of the tet(K) gene from Staphylococcus aureus mediates the Unemoto. 1995. N-terminal quarter part of tetracycline transporter from transport of potassium in Escherichia coli. J. Bacteriol. 175: 4927-4929 pACYC184 complements K+ uptake activity in K+ uptake-deficient mutants Guffanti, A.A., and T.A. Krulwich. 1995. Tetracycline/H+ antiport and Na+/ of Escherichia coli and Vibrio alginolyticus. Biol. Pharm. Bull. 18: 1189- H+ antiport catalyzed by the Bacillus subtilis TetA(L) transporter expressed in Escherichia coli. J. Bacteriol. 177: 4557-4561.
Nelson, M.L., and S.B. Levy. 1999. Reversal of tetracycline resistance Guffanti, A.A., J. Cheng, and T.A. Krulwich. 1998. Electrogenic antiport mediated by different bacterial tetracycline resistance determinants by activities of the gram-positive Tet proteins include a Na+(K+)/K+ mode an inhibitor of the Tet(B) antiport protein. Antimicrob. Agents Chemother.
that mediates net K+ uptake. J. Biol. Chem. 273: 26447-26454 Harel-Bronstein, M., P. Dibrov, Y. Olami, E. Pinner, S. Schuldiner, and E.
Nelson, M.L., B.H. Park, J.S. Andrew, V.A. Georgian, R.C. Thomas, and Padan. 1995. MH1, a second- site revertant of an Escherichia coli mutant S.B. Levy. 1993. Inhibition of the tetracycline efflux antiport protein by 13- lacking Na+/H+ antiporters (∆nhaA∆nhaB), regains Na+ resistance and a thio-substituted-5-hydroxy-6-deoxy-tetracycline. J. Med. Chem. 36: 370- capacity to excrete Na+ in a ∆µH+-dependent fashion. J. Biol. Chem. 270: Nelson, M.L., B.H. Park, and S.B. Levy. 1994. Molecular requirements for Hirata, T., E. Fujihira, T. Kimura-Someya, and A. Yamaguchi. 1998.
the inhibition of the tetracycline antiport protein and the effect of potent Membrane topology of the Staphylococcal tetracycline efflux protein Tet(K) inhibitors on the growth of tetracycline-resistant bacteria. J. Med. chem.
determined by antibacterial resistance gene fusions. J. Biochem. 124: Nikaido, H., and D.G. Thanassi. 1993. Penetration of lipophilic agents with Hoshino, T., T. Ideda, N. Tomizuka, and K. Furukawa. 1985. Nucleotide multiple protonation sites into bacterial cells: tetracyclines and sequence of the tetracycline resistance gene of pTHT15, a thermophilic fluoroquinolones as examples. Antimicrob. Agents Chemother. 37: 1393- Bacillus plasmid: comparison with staphylococcal TcR controls. Gene 37: Padan, E., and T.A. Krulwich. 2000. Sodium Stress. In: Bacterial Stress Ives, C.L. and K.F. Bott. 1989. Cloned Bacillus subtilis chromosomal DNA Responses (G. Storz and R Hengge-Aronis, eds) pp. 117-130, ASM Press, mediates tetracycline resistance when present in multiple copies. J.
Pao, S.S., I.T. Paulsen, and M.H. Saier, Jr. 1998. Major Facilitator Iwaki, S., N. Tamura, T. Kimura-Someya, S. Nada, and A. Yamaguchi. 2000.
Superfamily. Microbiol. Molec. Biol. Rev. 62: 1-34.
Cysteine-scanning mutagenesis of transmembrane segments 4 and 5 of Paulsen, I.T., M.H. Brown, and R.A. Skurray. 1996. Proton-dependent the Tn10-encoded metal-Tetracycline/H+ antiporter reveals a permeability multidrug efflux systems. Microbiol. Rev. 60: 575-608 barrier in the middle of a transmembrane water-filled channel. J. Biol.
Paulsen, I.T. and R.A. Skurray. 1993. Topology, structure and evolution of two families of proteins involved in antibiotic and antiseptic resistance in Khan, S.A., and R.P. Novick. 1983. Complete nucleotide sequence of pT181, eukaryotes and prokaryotes — an analysis. Gene 124: 1-11 a tetracycline-resistance plasmid from Staphylococcus aureus. Plasmid Putman, M., L.A. Koole, H.W. van Veen, and W.N. Konings. 1999. The secondary multidrug transporter LmrP contains multiple drug interaction Kilic, F. and G. Rudnick. 2000. Oligomerization of the serotonin transporter and its functional consequences. Proc. Natl. Acad. Sci. USA 97: 3106- Roberts, M.C. 1996. Tetracycline resistance determinants: mechanisms of action, regulation of expression genetic mobility, and distribution. FEMS Kimura-Someya, T., S. Iwaki, S. Konishi, N. Tamura, Y. Kubo, and A.
Yamaguchi. 2000. Cycsteine-scanning mutagenesis around Rubin, R.A., and S.B. Levy. 1990. Interdomain hybrid Tet proteins confer transmembrane segments 1 and 11 and their flanking loop regions of tetracycline resistance only when they are derived from closely related Tn10-encoded metal-Tetracycline/H+ antiporter. J. Biol. Chem. 275: 18692- members of the tet gene family. J. Bacteriol. 172: 2303-2312 Rubin, R.A., and S.B. Levy. 1991. Tet protein domains interact productively Krulwich, T.A., J. Cheng, and A.A. Guffanti. 1994. The role of monovalent to mediate tetracycline resistance when present on separate polypeptides.
cation/proton antiporters in Na+-resistance and pH homeostasis in Bacillus: an alkaliphile versus a neutralophile. J. Exp. Biol. 196: 457-470 Saier, M.H., Jr., I.T. Paulsen, M.K. Sliwinski, S.S. Pao, R.A. Skurray, and H.
Kubo, Y., S. Konishi, T. Kawabe, S. Nada, and A. Yamaguchi. 2000. Proximity Nikaido. 1998. Evolutionary origins of multidrug and drug-specific efflux of periplasmic loops in the metal-Tetracycline/H+ antiporter of Escherichia coli observed on site-directed cross-linking. J. Biol. Chem. 275: 5270- Saunders, J.R. 1984. Genetics and evolution of antibiotic resistance. Brit.
Lenski, R.E. 1997. The cost of antibiotic resistance from the perspective of Schnappinger, D. and Hillen, W. 1996. Tetracyclines: antibiotic action, uptake the bacterium. In: Antibiotic resistance: origins, evolution, selection and and resistance mechanisms. Arch. Microbiol. 165: 359-369 spread. Wiley, Chichester (Ciba Foundation Symposium 207) pp 131- Sigler, A., P. Schubert, W. Hillen, and M. Niederweis. 2000. Permeation of tetracyclines through membranes of liposomes and Escherichia coli. Eur.
Lenski, R.E., S.C. Simpson, T.T. Nguyen. 1994. Genetic analysis of a plasmid-encoded host genotype- specific enhancement of bacterial fitness.
Simpson, A.E., R.A. Skurray, and N. Firth. 2000. An IS257-derived hybrid promoter directs transcription of a tetA(K) tetracycline resistance gene in Levy, S.B. 1992. Active efflux mechanisms for antimicrobial resistance.
the Staphylococcus aureus chromosomal mec region. J. Bacteriol. 182: Antimicrob. Agents Chemother. 36: 695-703 Levy, S.B. 1997. Antibiotic resistance: an ecological imbalance. In: Antibiotic Sheridan, R.P., and I. Chopra. 1991. Origin of tetracycline efflux proteins: resistance: origins, evolution, selection and spread. Wiley, Chichester conclusions from nucleotide sequence analysis. Mol. Microbiol. 5: 895- Levy, S.B., L.M. McMurry, T.M. Barbosa, V. Burdett, P. Courvalin, W. Hillen, Speer, B.S., N.B. Shoemaker, and A.A. Salyers. 1992. Bacterial resistance M.C. Roberts, J.I. Rood, and D.E. Taylor. 1999. Nomenclature for new to tetracycline: mechanisms, transfer, and clinical significance. Clin.
tetracycline resistance determinants. Antimicrob. Agents Chemother. 43: Stasinopoulos, S.J., G.A. Farr, and D.H. Bechhofer. 1998. Bacillus subtilis tetA(L) gene expression: evidence for regulation by translationalreinitiation. Mol. Microbiol. 30: 923-932 Sum, P.-E., F.-W. Sum, and S.J. Projan. 1998. Recent developments in tetracycline antibiotics. Curr. Pharmaceut. Design 4: 119-132 Udo, E.E., and W.B. Grubb. 1996. A phage-mediated transfer of chromosomally integrated tetracycline resistance plasmid inStaphylococcus aureus. Curr. Microbiol. 32: 286-290 Varela, M.F., C.E. Sansom, and J.K. Griffith. 1995. Mutational analysis and molecular modelling of an amino acid sequence motif conserved inantiporters but not symporters in a transporter superfamily. Mol. Membr.
Biol. 12: 313-319 Wang, W., A.A. Guffanti, Y. Wei, M. Ito, and T.A. Krulwich. 2000. Two types of Bacillus subtilis tetA(L) deletion strains reveal the physiologicalimportance of TetA(L) in K+ acquisition as well as in Na+, alkali, andtetracycline resistance. J. Bacteriol. 182: 2088-2095 Williams, G., and I. Smith. 1979. Chromosomal mutations causing resistance to tetracycline in Bacillus subtilis. Mol. Gen. Genet. 177: 23-29 Yamaguchi, A., Y. Iwasaki-Ohba, N. Ono, M. Kaneko-Ohdera, and T. Sawai.
1991a. Stoichiometry of metal-tetracycline/H+ antiport mediated bytransposon Tn10-encoded tetracycline resistance protein in Escherichiacoli. FEBS Lett. 282: 415-418 Yamaguchi, A., H. Ohmori, M. Kaneko-Ohdera, T. Nomura, and T. Sawai.
1991b. ∆pH-dependent accumulation of tetracyline in Escherichia coli.
Antimicrob. Agents Chemother. 35: 53-56 Yamaguchi, A., Y. Shiina, E. Fujihira, T. Sawai, N. Noguchi, and M. Sasatsu.
1995. The tetracycline efflux protein encoded by the tet(K) gene fromStaphylococcus aureus is a metal-tetracycline/H+ antiporter. FEBS Lett.
365: 193-197 Yamaguchi, A., T. Udagawa, and T. Sawai. 1990. Transport of divalent cations with tetracycline as mediated by the transposon Tn10-encodedtetracycline resistance protein. J. Biol. Chem. 265: 4809-4813 Yerushalmi, H. and S. Schuldiner. 2000. A common binding site for substrates and protons in EmrE, an ion-coupled multidrug transporter.
FEBS Lett. 476: 93-97.

Source: http://www.horizonpress.com/backlist/jmmb/v/v3/v3n2/15.pdf