Significance of the Glutamate-139 Residue of the V-Type Na⁺-ATPase NtpK Subunit in Catalytic Turnover Linked with Salt Tolerance of Enterococcus hirae

Abstract

A Glu139Asp mutant of the NtpK subunit (kE139D) of Enterococcus hirae vacuolar-type ATPase (V-ATPase) lost tolerance to sodium but not to lithium at pH 10. Purified kE139D V-ATPase retained relatively high specific activity and affinity for the lithium ion compared to the sodium ion. The kE139 residue of V-ATPase is indispensable for its enzymatic activity that is linked with the salt tolerance of enterococci.

TEXT

Vacuolar-type ATPase (V-ATPase) functions as a proton pump in the acidic organelles and plasma membranes of eukaryotic and prokaryotic cells (19, 20). V-ATPase is composed of two structural domains: the hydrophilic V₁ moiety and the membrane-embedded V_o moiety. The V_o moiety is responsible for translocation of H⁺, which couples to ATP hydrolysis on the V₁ moiety. This catalytic reaction is explained by the “rotation catalysis mechanism” (2).

Enterococcus hirae V-ATPase is unique in transporting Na⁺ and Li⁺ rather than H⁺ and is indispensable for the salt tolerance of E. hirae at alkaline pHs (11). E. hirae V-ATPase requires all of its nine subunits (NtpFIKECGABD) (15) for its activity (5). The NtpK subunit forms the rotor ring of the V_o moiety, which is responsible for ion translocation (17), and the Glu139 residue of NtpK, which constitutes the ion binding site (18), is crucial for its catalytic mechanism (21). Notably, E. hirae V-ATPase recognizes Na⁺ and Li⁺ with nearly equal affinities. The K_m values for Na⁺ and Li⁺ of ATP hydrolysis of the purified enzyme are 20 μM and 60 μM, respectively (15). The crystal structures of the Na⁺-bound NtpK ring and the Li⁺-bound NtpK ring have been obtained (17, 18). The overall structure of the Li⁺-bound NtpK is almost identical to that of Na⁺-bound NtpK. The coupling ion (Na⁺, Li⁺) is surrounded by five oxygen atoms contributed by residues Thr64, Gln65, Gln110, Glu139, and Leu61. The bound ion is occluded by the side chain of Glu139 (PDB code: 2BL2, 2CYD). Here we further investigated the catalytic properties of E. hirae V-ATPase carrying mutations in NtpK Glu139 and the effects on bacterial growth at high salt concentrations.

Strains and plasmids used in this study are shown in Table 1. Mutant enzymes were constructed using the wild-type ntpK gene in the shuttle vector pHEex (6) as the template (pHEexK) for PCR-generated mutations. Figure 1 shows the growth of NtpK mutant strains at an alkaline pH in high concentrations of Na⁺ or Li⁺. Cells were cultured at 37°C in NaTY or KTY standard complex medium (8). When the cultures reached an optical density at 600 nm (OD₆₀₀) of 0.05, the pH was shifted to 10 by the addition of 50 mM Na₂CO₃ to NaTY medium or 50 mM K₂CO₃ and 100 mM LiCl to KTY medium. The concentrations of Na⁺ or Li⁺ in these media were approximately 220 mM and 100 mM, respectively. An ntpK disrupted mutant, NKD, is sensitive to Na⁺ at alkaline pHs (5). NKD harboring the plasmid pHEex (control) failed to grow in 220 mM Na⁺ or in 100 mM Li⁺ at pH 10, thus confirming that NtpK is required for this organism's tolerance to Li⁺ as well as Na⁺ at an alkaline pH. Transforming NKD with pHEexK (NKD/kE139) restored salt tolerance, as reported previously (5). An NKD strain harboring pHEexK(E139Q) (NKD/kE139Q) did not grow well in either high-Na⁺ or high-Li⁺ medium at pH 10. NKD harboring pHEexK(E139D) (NKD/kE139D) was also sensitive to high concentrations of Na⁺, but we found that NKD/kE139D grew well in high Li⁺ medium. NKD/kE139D (0.3/h) grew at a rate similar to that of the wild type in high-Li⁺ medium (0.35/h) (Fig. 1B). These results indicate that the NtpK subunit E139D (kE139D) mutation had a limited effect on Li⁺ sensitivity.

Table 1.

Strains and plasmids used in this study

Strain or plasmid	Relevant feature	Reference
Strains
E. hirae ATCC 9790	Wild type, obtained from the American Type Culture Collection
NKD	An ntpK disrupted strain derived from ATCC 9790	5
25DKD	An ntpK disrupted strain derived from 25D (9)	This study
NKD/kE139	NKD harboring pHEexK	This study
NKD/kE139D	NKD harboring pHEexK(E139D)	This study
NKD/kE139Q	NKD harboring pHEexK(E139Q)	This study
NKD/pHEex	NKD harboring pHEex, used as a control	This study
Plasmids
pHEex		6
pHEexK	Wild-type ntpK gene in the shuttle vector pHEex	This study
pHEexK(E139D)		This study
pHEexK(E139Q)		This study

Fig. 1.

Salt sensitivity of kE139 mutants. (A and B) Growth in high-salt media at an alkaline pH. Strains NKD/pHEex (control, open circles), NKD/kE139 (wild type, filled circles), NKD/kE139Q (filled triangles), or NKD/kE139D (open triangles) were cultured in NaTY medium containing 50 mM Na₂CO₃ (pH 10) (white bars) or KTY medium containing 50 mM K₂CO₃ and 100 mM LiCl (pH 10) (black bars), and growth was monitored by measuring the turbidity at a wavelength of 600 nm. pH was shifted at the time indicated by arrows. (B) Growth rates were determined by the change in optical densities between 0.1 and 0.2. The data shown are the means ± SD of results of triplicate experiments.

There are two pumping systems for extrusion of Na⁺ and Li⁺ in E. hirae: V-ATPase and the Na⁺(Li⁺)/H⁺ antiporter driven by the electrochemical proton gradient (proton potential) (7). Because the generation of proton potential of E. hirae is not significant at an alkaline pH (10), extrusion of Na⁺ and Li⁺ at an alkaline pH must be catalyzed primarily by V-ATPase. In fact, strain NKD does not exhibit proton potential-independent Na⁺ extrusion at an alkaline pH (5, 21). Because extrusion of Na⁺ or Li⁺ from living cells is a prerequisite for tolerance to toxic concentrations of these ions (3, 4), the kE139D mutant probably extrudes Li⁺ at an alkaline pH. NKD/pHEex (control), NKD/kE139 (wild type), NKD/kE139D, and NKD/kE139Q were grown overnight in a high-Na⁺ medium to induce V-ATPase production (16) and washed, and then cells were resuspended in extrusion buffer [100 mM Na⁺–2-(cyclohexylamine)ethanesulfonic acid (CHES; pH 10) containing 200 mM KCl for Na⁺ extrusion or 50 mM Li⁺-CHES (pH 10) containing 200 mM KCl for Li⁺ extrusion] to a cell density equivalent to 1 mg/ml protein. Proton potential was dissipated by adding 20 μM valinomycin and 20 μM carbonylcyanide m-chlorophenylhydrazone (CCCP), and extrusion reactions were initiated by the addition of 10 mM glucose. At intervals, cell samples were obtained and filtered through Nuclepore polycarbonate membrane filters (pore size, 0.4 μm; Costar Scientific Co., Cambridge, Mass.). The Li⁺ or Na⁺ contents in the cells were determined by atomic absorption after extraction with hot 5% trichloroacetic acid. Western blot analysis of cell lysates with anti-V-ATPase serum revealed that mutant V-ATPase levels were equivalent among these strains (data not shown). Introducing kE139D or kE139Q did not restore the Na⁺ extrusion activity of NKD (Fig. 2 A). Li⁺ extrusion activity was not detected in NKD/kE139Q but was clearly retained in NKD/kE139D (Fig. 2B), consistent with its salt-tolerant growth (Fig. 1). Li⁺ extrusion was slower for the kE139D mutant than for the wild type. Because the activity of Na⁺ or Li⁺ extrusion from cells under these conditions reflects ion pumping by V-ATPase (6, 9), these results suggest that the kE139D replacement had little effect on Li⁺ transport by E. hirae V-ATPase, in contrast to the effect on Na⁺ transport. Thus, the V-ATPase containing the kE139D mutation was changed so that it discriminated between Na⁺ and Li⁺ as the coupling ion.

Fig. 2.

Na⁺ and Li⁺ transport activity of kE139 mutants. Ion extrusion from whole cells. The strains NKD/pHEex (control), NKD/kE139 (wild type), NKD/kE139Q, and NKD/kE139D were cultured in NaTY medium and then loaded with Na⁺ (A) or Li⁺ (B) as described in the text. Extrusion of Na⁺ or Li⁺ was initiated by the addition of 10 mM glucose at 0 min (closed symbols).

To further characterize the kE139D mutation in catalytic properties of V-ATPase, we measured the ATP hydrolytic activity of the purified kE139D enzyme. E. hirae possesses H⁺-translocating F-type ATPase (F-ATPase) (1) as well as V-ATPase. To avoid contamination with F-ATPase during purification of V-ATPase, an ntpK deletion mutant was obtained from the F-ATPase-defective strain 25D (9). The ntpK gene of strain 25D was replaced with a chloramphenicol resistance cassette (5). The resulting mutant, 25DKD, was transformed by plasmid pHEexK or pHEexK(E139D).

V-ATPase of strain 25DKD, harboring either pHEexK or pHEexK(E139D), was purified by anion-exchange and gel filtration chromatographies (15). Figure 3 shows the dose dependency of the ATP hydrolytic activities for the Na⁺ or Li⁺ ion of purified V-ATPases. SDS-PAGE analysis of purified V-ATPases revealed that the polypeptide compositions of wild-type and kE139D enzymes were almost the same, although a few small polypeptides differing from the nine V-ATPase subunits were found (Fig. 3B).

Fig. 3.

Na⁺ and Li⁺ dependence of purified V-ATPase containing the kE139D mutation. (A) V-ATPase was purified (15) from 25DKD/pHEexK (wild type) or 25DKD/pHEexK(E139D) (kE139D) cultured in NaTY medium containing 0.5 M NaCl. The effects of NaCl (open circles) or LiCl (filled circles) on ATP hydrolytic activities of V-ATPases were examined. (B) SDS-PAGE analysis of purified V-ATPase (wild type and kE139D mutant).

The ATPase activity was assayed as described previously (14). The ATP hydrolytic activities of purified V-ATPases were tightly coupled with the Na⁺ or Li⁺ ion, and the kinetics of ATP hydrolysis showed a biphasic pattern (15). Since a high K_m value (low affinity) was not observed when the ATP concentration was low, the two affinity values determined for E. hirae V-ATPase were thought to correspond to the enzymatic reaction of V-ATPase (K_m2, low affinity) and the binding of Na⁺ or Li⁺ to the NtpK ring (K_m1, high affinity) (13). The apparent K_m values of the two phases, 29 ± 6.3 μM and 2.5 ± 1.4 mM for Na⁺ and 40 ± 6.3 μM and 2.0 ± 1.3 mM for Li⁺, respectively, were estimated for the wild-type enzyme, being consistent with the values for purified V-ATPase reported previously (Fig. 3A; Table 2) (15). The wild-type enzyme thus recognizes Na⁺ slightly better than Li⁺ (15).

Table 2.

Kinetics values for ion dependence in ATP hydrolysis of purified V-ATPase mutants^a

Mutation	Ion	Km for Na+ or Li+		Vmax (μmol/min/mg protein)
		Km1 (μM)	Km2 (mM)
None (wild type)	Na+	29 ± 6.3	2.5 ± 1.4	5.6 ± 0.2
	Li+	40 ± 6.3	2.0 ± 1.3	5.6 ± 0.6
kE139D	Na+	100 ± 28	5.1 ± 2.6	2.2 ± 0.4
	Li+	67 ± 21	5.5 ± 2.6	4.5 ± 0.5

^aThe ATP hydrolytic activities of purified mutant enzymes were measured in the presence of various concentrations of NaCl or LiCl with 2 mM ATP as the substrate. The values were averaged from the results of four independent experiments.

The kE139D mutant enzyme responded differently to Na⁺ and Li⁺. The apparent K_m values for Na⁺ (100 ± 28 μM and 5.1 ± 2.6 mM) of the kE139D enzyme were much higher than those of the wild-type enzyme. V_max values of the kE139D enzyme were reduced to approximately 34% of that of the wild-type enzyme. Although the Na⁺-stimulated ATPase activity of the membrane vesicles prepared from NKD/kE139D was negligible in our previous study (21), we observed Na⁺-stimulated ATPase activity with the purified enzyme of the kE139D mutant (Fig. 3A). It is possible that the low activity of the kE139D mutant was masked by the high background activity derived from F-ATPase in the membrane vesicles. It is interesting that the kE139D enzyme appears uncoupled from Na⁺ but not from Li⁺. However, the Na⁺ extrusion observed in this study was the net extrusion rate, reflecting Na⁺ influx and efflux from whole cells. Thus, the lower efflux rate of the kE139D mutant enzyme could be overcome by a higher influx rate, and then apparently no extrusion activity would be observed. Further experiments using purified and reconstituted enzyme are required for the precise determination of the Na⁺ transport activity of the kE139D enzyme. In contrast, the apparent K_m values for Li⁺ (67 ± 21 μM and 5.5 ± 2.6 mM) of the kE139D enzyme were nearly equal to those of the wild-type enzyme, and the V_max values of the mutant were approximately 70% of those of the wild-type enzyme. These results suggest that the arm length of the important carboxyl group at the 139th amino acid residue of NtpK is critical for Na⁺-coupled catalytic turnover of E. hirae V-ATPase. The ion-transporting mechanism of V-ATPase is explained by the “two half-channel” model (18). The clockwise rotation of the NtpK ring driven by ATP hydrolysis brings a bound Na⁺ ion into the interface of NtpI and the NtpK ring. At this site, an electrostatic interaction between NtpI R573 (12) and NtpK E139 disrupts the hydrogen bond network in the ion binding site of the NtpK ring. This results in the release of the bound ion into the periplasm via a half-channel in NtpI. Further rotation caused by ATP hydrolysis may disrupt the R573-E139 interaction, resulting in the binding of a cytoplasmic Na⁺/Li⁺ ion. The release of the ion may possibly be assisted by electrostatic repulsion by NtpI R573. We suggest that the effect of the alteration of the side chain length on the catalytic reaction of this enzyme is more remarkable for Na⁺ than for Li⁺. From the crystal structures of the Na⁺-bound NtpK ring and Li⁺-bound NtpK ring, the cavity size of the ion binding site formed by these oxygen atoms accommodates Na⁺ (diameter, 2.04 Å) better than Li⁺ (diameter, 1.52 Å) (17, 18), and Na⁺ should bind more tightly to its binding pocket than Li⁺. Therefore, Li⁺ might be released more easily from its binding pocket to a half-channel, without an electrostatic interaction between E139 and R573.

The rates of influx and efflux of Na⁺ and Li⁺ balance their intracellular concentrations. Although their influx pathways have not been elucidated, the efflux of these ions by E. hirae at an alkaline pH is attributed mainly to V-ATPase (6, 9). In this study, we demonstrated an essential role of the kE139 residue in the turnover rate of E. hirae V-ATPase with the Na⁺ ion, linked to the Na⁺ tolerance of this bacterium at an alkaline pH. A partial decrease in the turnover rate of V-ATPase resulting from the replacement of Glu with Asp rendered cells unable to sustain Na⁺ homeostasis of E. hirae at an alkaline pH.

The genome database suggests that this type of V-ATPase is distributed as the ion pump among a variety of eubacteria, such as Streptococcus pyogenes, Clostridium perfringens, and Chlamydia trachomatis. In all of these NtpK subunits, the Glu residue, not the Asp residue, is conserved. The conservation of the 139th glutamic acid residue of the NtpK subunit for the full activity of V-ATPase is thus important for the physiology of several organisms in adapting to a saline environment.