Tet(L) and tet(K) tetracycline-divalent metal/H+ antiporters: characterization of multiple catalytic modes and a mutagenesis approach to differences in their efflux substrate and coupling ion preferences

Abstract

The Tet(L) protein encoded in the Bacillus subtilis chromosome and the closely related Tet(K) protein from Staphylococcus aureus plasmids are multifunctional antiporters that have three cytoplasmic efflux substrates: a tetracycline-divalent metal (TC-Me(2+)) complex that bears a net single positive charge, Na+, and K+. Tet(L) and Tet(K) had been shown to couple efflux of each of these substrates to influx of H+ as the coupling ion. In this study, competitive cross-inhibition between K+ and other cytoplasmic efflux substrates was demonstrated. Tet(L) and Tet(K) had also been shown to use K+ as an alternate coupling ion in support of Na+ or K+ efflux. Here they were shown to couple TC-Me(2+) efflux to K+ uptake as well, exhibiting greater use of K+ as a coupling ion as the external pH increased. The substrate and coupling ion preferences of the two Tet proteins differed, especially in the higher preference of Tet(K) than Tet(L) for K+, both as a cytoplasmic efflux substrate and as an external coupling ion. Site-directed mutagenesis was employed to test the hypothesis that some feature of the putative "antiporter motif," motif C, of Tet proteins would be involved in these characteristic preferences. Mutation of the A157 in Tet(L) to a hydroxyamino acid resulted in a more Tet(K)-like K+ preference both as coupling ion and efflux substrate. A reciprocal S157A mutant of Tet(K) exhibited reduced K+ preference. Competitive inhibition among substrates and the parallel effects of the single mutation upon K+ preference, as both an efflux substrate and coupling ion, are compatible with a model in which a single translocation pathway through the Tet(L) and Tet(K) transporters is used both for the cytoplasmic efflux substrates and for the coupling ions, in an alternating fashion. However, the effects of the A157 and other mutations of Tet(L) indicate that even if there are a shared binding site and translocation pathway, some elements of that pathway are used by all substrates and others are important only for particular substrates.

FIG. 1.

The motif C regions of Tet(L) and Tet(K). The motif C (41) consensus sequence is shown in alignment with the sequences in the B. subtilis chromosomally encoded Tet(L) and Tet(K) regions. (A) Topological model of the whole Tet(L) protein, based on the data for Tet(K) (12, 22), with TMS V indicated by the rectangle and the position of motif C shaded in gray. (B) Alignment of the Tet(L) and Tet(K) motif C. (C) The motif C regions of Tet(L) and Tet(K), showing the site-directed mutants constructed and characterized in this study.

FIG. 2.

Capacity of Tet(L) and Tet(K) to support ⁸⁶Rb⁺ uptake by RSO vesicles of E. coli TK2420 upon generation of an outwardly directed gradient of TC-Co²⁺. The assay, conducted as described in Materials and Methods, was initiated by dilution of RSO vesicles that were preloaded with TC-Co²⁺ at pH 7.5 into buffers at the pH values indicated. Control experiments were conducted with vesicles that were incubated at pH 7.5, but not loaded with TC-Co²⁺ prior to dilution. These controls did not exhibit accumulation observed in the TC-Co²⁺-preloaded vesicles, as is shown for the control from the pH 8.3 experiment (▵). Toluene-treated vesicles were used as the binding control for this experiment.

FIG. 3.

Effect of K⁺ on the rate of Tet(L)- or Tet(K)-mediated [³H]TC efflux from RSO vesicles of E. coli TK2420. The experimental preparations were similar to those used in the experiments depicted in Fig. 2, except that the extravesicular pH was 8.3, the intravesicular TC was tritiated and at a somewhat lower concentration, and, when present at 1 mM in the outside buffer (right panel), nonradioactive KCl alone rather than an ⁸⁶Rb⁺-KCl mix was added as a coupling ion.

FIG. 4.

Double reciprocal plots of experiments showing the effect of added K⁺ on TC and Na⁺ uptake by everted vesicles of E. coli Tet(K) vesicles. K⁺, Na⁺, and TC-Co²⁺ were serving as efflux substrates on the “cytoplasmic” side (outside of the everted system) of the vesicles. There was no intravesicular K⁺. Uptake of [³H]TC (A) or ²²Na⁺ (B) was assayed in everted vesicles of transformants of E. coli DH5α and E. coli NM81, respectively, expressing tet(K). Assays were carried out as described in Materials and Methods, in the presence (○) or absence (•) of added extravesicular K⁺ at 5 mM (A) or 10 mM (B). Reciprocal plots of the data were plotted by using time points in the linear range (up to 1 min) after correction by subtraction of values for transport in the presence of CCCP. The results shown are the average of at least five separate determinations, and the error bars represent the standard deviation.

FIG. 5.

Energy and intravesicular cation dependence of ⁸⁶Rb⁺ uptake, as a coupling ion, by RSO vesicles of E. coli TK2420 transformed with various tet plasmids. Vesicles were passively loaded with either 100 μM choline-Cl (control, nonefflux substrate) or KCl (efflux substrate). Uptake was initiated by diluting 25 μl of vesicles into 500 μl of 10 mM Tris-HCl (pH 7.5), containing a final concentration of 100 μM ⁸⁶Rb⁺-KCl. To half of the reaction mixtures, 10 mM Tris-d-lactate was added to energize those vesicles. Samples were taken at the times indicated and treated as described in Materials and Methods. The initial velocities, v_i, in these determinations are shown for the energized vesicles that contained intravesicular K⁺ (efflux substrate) and are the average of at least four separate determinations. The error bars show the standard deviation of the values. The subscript “in” denotes intravesicular location.

FIG. 6.

Transport activities of the Tet(K) S157A mutant compared with those of wild-type Tet(L) and Tet(K) and the A157S mutant of Tet(L). (A) ⁸⁶Rb⁺-K⁺ uptake (as coupling ion) by right-side-out vesicles was assayed as described in the legend to Fig. 5 in the presence of intravesicular K⁺ (as efflux substrate) and added d-lactate as energy source. The data shown here were corrected for binding. (B) [³H]TC uptake (as efflux substrate) in exchange for H⁺ was assayed as described in Materials and Methods, and the values shown are corrected for the TC bound by the vesicles of each type in CCCP-treated preparations. (C) ²²Na⁺ uptake (as efflux substrate) in exchange for H⁺ was assayed in a BTP buffer-based reaction mixture as described in Materials and Methods, and the data were corrected for the binding in a CCCP-treated control for each construct. Since the bacterial mutant used for these assays has residual Na⁺/H⁺ antiporter activity, the activity for the vector control was also subtracted. Although not shown, other control assays of the same types shown for earlier assays were conducted and yielded values similar to those in the earlier assays.