The Lysine 299 Residue Endows the Multisubunit Mrp1 Antiporter with Dominant Roles in Na⁺ Resistance and pH Homeostasis in Corynebacterium glutamicum

ABSTRACT

Corynebacterium glutamicum is generally regarded as a moderately salt- and alkali-tolerant industrial organism. However, relatively little is known about the molecular mechanisms underlying these specific adaptations. Here, we found that the Mrp1 antiporter played crucial roles in conferring both environmental Na⁺ resistance and alkali tolerance whereas the Mrp2 antiporter was necessary in coping with high-KCl stress at alkaline pH. Furthermore, the Δmrp1 Δmrp2 double mutant showed the most-severe growth retardation and failed to grow under high-salt or alkaline conditions. Consistent with growth properties, the Na⁺/H⁺ antiporters of C. glutamicum were differentially expressed in response to specific salt or alkaline stress, and an alkaline stimulus particularly induced transcript levels of the Mrp-type antiporters. When the major Mrp1 antiporter was overwhelmed, C. glutamicum might employ alternative coordinate strategies to regulate antiport activities. Site-directed mutagenesis demonstrated that several conserved residues were required for optimal Na⁺ resistance, such as Mrp1A K²⁹⁹, Mrp1C I⁷⁶, Mrp1A H²³⁰, and Mrp1D E¹³⁶. Moreover, the chromosomal replacement of lysine 299 in the Mrp1A subunit resulted in a higher intracellular Na⁺ level and a more alkaline intracellular pH value, thereby causing a remarkable growth attenuation. Homology modeling of the Mrp1 subcomplex suggested two possible ion translocation pathways, and lysine 299 might exert its effect by affecting the stability and flexibility of the cytoplasm-facing channel in the Mrp1A subunit. Overall, these findings will provide new clues to the understanding of salt-alkali adaptation during C. glutamicum stress acclimatization.

IMPORTANCE The capacity to adapt to harsh environments is crucial for bacterial survival and product yields, including industrially useful Corynebacterium glutamicum. Although C. glutamicum exhibits a marked resistance to salt-alkaline stress, the possible mechanism for these adaptations is still unclear. Here, we present the physiological functions and expression patterns of C. glutamicum putative Na⁺/H⁺ antiporters and conserved residues of Mrp1 subunits, which respond to different salt and alkaline stresses. We found that the Mrp-type antiporters, particularly the Mrp1 antiporter, played a predominant role in maintaining intracellular nontoxic Na⁺ levels and alkaline pH homeostasis. Loss of the major Mrp1 antiporter had a profound effect on gene expression of other antiporters under salt or alkaline conditions. The lysine 299 residue may play its essential roles in conferring salt and alkaline tolerance by affecting the ion translocation channel of the Mrp1A subunit. These findings will contribute to a better understanding of Na⁺/H⁺ antiporters in sodium antiport and pH regulation.

INTRODUCTION

Bacteria (or prokaryotes) constantly encounter diverse environmental stresses throughout their lives, which for some organisms include high-salt and alkaline-pH conditions (1). Bacteria have developed numerous efficient strategies to maintain cellular homeostasis and adapt to these external stress changes (2–6). One of these physiological strategies is governed by the monovalent cation/proton antiporters, commonly referred to as Na⁺/H⁺ or K⁺/H⁺ antiporters (7). These antiporters are widely distributed in prokaryotes and unicellular eukaryotes and generally catalyze the efflux of intracellular cations such as Na⁺, Li⁺, K⁺, and NH₄⁺ in exchange for external protons (8–10). Energized by the membrane potential, Na⁺/H⁺ antiporters are essential for the establishment of a sodium electrochemical gradient that drives Na⁺-coupled solute uptake from the surrounding environments (11–13). In bacteria, the Na⁺/H⁺ antiporters have been categorized into multiple different families according to the sequence-based transporter classification system (3, 7, 14), including the cation/proton antiporter 1 (CPA1) to CPA3 families, the major facilitator superfamily (MFS), the calcium/cation antiporter (CaCA) family, and other as-yet-unclassified Na⁺/H⁺ antiporters. Most characterized Na⁺/H⁺ antiporters are individual hydrophobic gene products, whereas the Mrp-type antiporters, belonging to the CPA3 family, function as hetero-oligomeric complexes (10). Interestingly, most bacteria contain at least 5 to 9 distinct Na⁺/H⁺ antiporters; however, some organisms that are exposed to numerous environmental stresses exhibit even higher numbers of antiporters (7). Although the physiological characteristics of individual Na⁺/H⁺ antiporters have been extensively investigated, the reason for which a bacterium contains so many different types of Na⁺/H⁺ antiporters needs to be further elaborated. Krulwich et al. (7) hypothesized that a particular antiporter might have a dominant role in confronting a specific condition, e.g., a high-NaCl stress, while multiple types of Na⁺/H⁺ antiporters with overlapping roles give the organism more opportunity to cover all contingencies. Thus, detailed information about the specific properties and physiological roles of multiple Na⁺/H⁺ antiporters will provide new insights into the strategies employed by various bacteria to cope with environmental challenges.

The Mrp-type antiporter system is widely distributed among physiologically diverse prokaryotes, and most organisms contain only one Mrp-type system (10). However, several bacteria, such as certain strains derived from Corynebacterium, Staphylococcus, and Sinorhizobium species, have dual Mrp-type systems (7). The Mrp operon typically contains six (mrpACDEFG) or seven (mrpABCDEFG) genes encoding hydrophobic proteins, and all of them are required for optimal Mrp-dependent sodium and alkali resistance (10, 15, 16). However, the sole MrpA subunit derived from the archaeon Methanosarcina acetivorans is sufficient for antiport activity, challenging the current dogma for Mrp complexes (17). The Mrp-type system generally functions as a secondary active transporter that is energized by an imposed transmembrane potential generated from proton-pumping respiratory complexes or ATP hydrolysis. Many studies demonstrate that the Mrp-type antiporter supports dominant roles in conferring tolerance to high-salt and alkaline stresses in some bacteria, in contrast to other types of Na⁺/H⁺ antiporters (10, 18, 19). One reason might be that the oligomeric structure of the Mrp-type system presents a larger proton-gathering surface, which facilitates the capture and flow of protons under high-pH conditions (10). In addition, three Mrp gene products, MrpA, MrpC, and MrpD, exhibit a striking resemblance to the membrane-embedded NuoL, NuoK, and NuoM/N subunits of the NADH:quinone oxidoreductase (complex I) of the bacterial respiratory chains, providing another coherent explanation that there is a higher transmembrane potential in the presence of the Mrp-type antiporter than in its absence (20–22).

The structural complexity of the Mrp-type antiporter has made it difficult to fully discern the antiport mechanisms (10). Nevertheless, some attempts have been made to investigate the potential roles of single subunits or conserved motifs in the hetero-oligomeric Mrp-type antiporters. Two previous studies have revealed that both the G³⁹³ site of BhMrpA protein and the G⁸² site of BhMrpC protein are required for the alkaliphily of alkaliphilic Bacillus halodurans C-125 (23, 24). Site-directed mutagenesis studies of Bacillus subtilis Mrp antiporter have suggested that several acidic residues in BsMrpA and BsMrpD subunits are required for normal antiport activities instead of Mrp complex formation (25, 26). In alkaliphilic Bacillus pseudofirmus OF4, the BpMrpE subunit is required for normal membrane levels of other Mrp proteins and the formation of a stable active Mrp complex (27). Further studies about the effects of point mutations in 28 different sites throughout the BpMrp subunits indicated the importance of multiple amino acid residues in Na⁺/H⁺ antiport activity, sodium exclusion, or BpMrp complex formation, including the positions BpMrpA K²⁹⁹, BpMrpA H⁷⁰⁰, BpMrpD E¹³⁷, BpMrpE P¹¹⁴, and BpMrpG P⁸¹ (28). Thus, a more-detailed characterization of the Mrp-type antiporters will undoubtedly expand our knowledge of the antiport mechanism.

The Gram-positive soil bacterium Corynebacterium glutamicum is widely referred to as an industrial workhorse for amino acid production (29). However, the bacteria are usually confronted with numerous metabolic challenges and stress situations throughout industrial fermentation, leading to negative effects on biomass and product yields (30–33). Previous reports have suggested that C. glutamicum is a moderately salt- and alkali-tolerant organism with optimal growth at neutral to alkaline pH (31, 34). C. glutamicum genome sequences reveal that there are four putative Na⁺/H⁺ antiporters, Mrp1, Mrp2, NhaP, and ChaA, belonging to the CPA3, CPA1, and CaCA families of secondary transporters, respectively (3). Our previous experiments suggest that only the Mrp1 antiporter displays obvious Na⁺(Li⁺)/H⁺ antiport activities with low apparent K_m values, which enable Mrp1 to effectively rescue growth defects of the Na⁺-sensitive Escherichia coli strain KNabc, whereas other antiporters exhibit weaker antiport activities than Mrp1 for monovalent cations (34). The in vivo roles and mechanisms of these Na⁺/H⁺ antiporters conferring physiological adaptation are still unclear. In this study, we report on efforts to explore the underlying mechanisms by which C. glutamicum Na⁺/H⁺ antiporters affect physiology in the presence of complex salt and alkali stresses and attempt to identify specific conserved sites necessary for the optimal activity of the Mrp1 antiporter.

RESULTS

The Mrp-type antiporters are involved in resistance to high salt and adaption to alkaline pH.

C. glutamicum is considered to be a moderately salt-tolerant organism that exhibits optimal growth at neutral to alkaline pH (31, 34). According to the sequence-based transporter classification system, C. glutamicum harbors at least four putative Na⁺/H⁺ antiporters, including Mrp1, Mrp2, NhaP, and ChaA, as shown in Fig. 1A. Sequence analyses reveal that both Mrp1 and Mrp2 are encoded by 6 individual genes in each operon, whereas the other two antiporters are encoded by a single gene. Four single-antiporter-deficient mutants (Δmrp1, Δmrp2, ΔnhaP, and ΔchaA mutants) and a double-antiporter-deficient mutant (Δmrp1 Δmrp2 double mutant) were constructed to explore the physiological roles of these antiporters in C. glutamicum.

FIG 1.

Roles of C. glutamicum Na⁺/H⁺ antiporters in response to high salt and alkaline pH. (A) Schematic diagram of the four putative Na⁺/H⁺ antiporters in C. glutamicum ATCC 13032. The gene arrangements of these operon clusters are also shown. (B) Effects of pH, NaCl, and KCl on growth of C. glutamicum wild-type strain or its putative Na⁺/H⁺ antiporter-deficient mutants. Each panel shows the growth curves of the corresponding strains in the indicated LBO medium buffered with BTP to different pH values. A concentration of 0.6 M NaCl or 0.6 M KCl was added into the medium. The growth curves are presented as means ± standard deviations (SD) from three independent experiments. (C) Effects of Na⁺/H⁺ antiporters on the growth of potassium uptake-deficient E. coli TK2420 cells. The growth of TK2420 recombinants in Ko medium with various KCl concentrations was determined by measuring the optical density at 600 nm (OD₆₀₀) after 24 h of incubation. Asterisks indicate significant differences between E. coli control with empty plasmid and its derivatives (**, P < 0.01).

As shown in Fig. 1B, under no-stress conditions at pH 7.0, only the Δmrp1 and Δmrp1 Δmrp2 mutants showed moderate growth defects, and the complementation of the specific mrp1 gene was able to partially restore the defective phenotypes of the Δmrp1 mutant. At alkaline pH, the Δmrp1 mutant displayed more-serious growth attenuation with a pH higher than 8.0 whereas the Δmrp2 mutant did not show growth differences compared with the wild-type strain. The Δmrp1 Δmrp2 double mutant lost almost completely the ability to grow under highly alkaline conditions. Under high-NaCl conditions, deletion of mrp1 had an obvious negative effect on bacterial growth, which was exacerbated by increased pH values. However, the Δmrp2 mutant showed growth properties similar to those of the wild-type strain. The Δmrp1 Δmrp2 double mutant almost completely lost the ability to grow in the presence of NaCl stress. When 0.6 M KCl was added instead, deletion of mrp2 resulted in a significant growth defect at pH 7.0, which was also exacerbated by increased pH values. Although deletion of mrp1 had a minor effect on cell growth, the Δmrp1 Δmrp2 double mutant displayed a more-serious growth defect than did each single mutant. Furthermore, as shown in Fig. S1 in the supplemental material, deletion of nhaP or chaA had no significant influence on cell growth under either alkaline-pH or high-salt conditions.

Our previous findings indicated that both C. glutamicum Mrp-type and NhaP antiporters exhibit detectable K⁺/H⁺ antiport activities (34). Thus, we further employed the potassium transport-deficient E. coli strain TK2420 to investigate their potential roles. E. coli TK2420, lacking three major K⁺ uptake systems (Trk, Kup, and Kdp), exhibits severe growth retardation under K⁺-limiting or highly osmotic conditions (35). As shown in Fig. 1C, when grown in Ko minimal medium (see Materials and Methods) containing 115 mM NaCl as the osmoticum, E. coli TK2420 derivatives failed to grow when KCl concentration was at or below 10 mM. Although an increased K⁺ concentration improved the growth of most E. coli TK2420 derivatives, the cells carrying C. glutamicum Mrp2 or NhaP antiporter still showed an obvious growth deficiency even in the presence of 100 mM KCl. The growth attenuation meant that the intracellular K⁺ content in these two mutants was still lower than the minimal requirement for normal growth. Interestingly, the E. coli TK2420 isolates carrying E. coli nhaA or C. glutamicum mrp1 showed enhanced growth behaviors compared to the control strain in the presence of 20 mM KCl. This might be attributed to a decrease in Na⁺ osmotic pressure caused by these antiporters in the preferable Na⁺ efflux from the cytoplasm.

Loss of Mrp-type antiporters affects intracellular Na⁺ and pH homeostasis.

To investigate the potential mechanism underlying growth defects, we examined the intracellular Na⁺ content and cytoplasmic pH (pH_i) levels in the Mrp-type antiporter deletion mutants. As shown in Fig. 2A, the Δmrp1 mutant showed an elevated intracellular Na⁺ concentration in comparison to wild-type cells, whereas the simultaneous deletion of mrp1 and mrp2 further exacerbated intracellular Na⁺ accumulation. Figure 2B showed that the Δmrp1 mutant exhibited a slightly more alkaline pH_i level than that of wild-type control under high-alkaline conditions. The Δmrp1 Δmrp2 double mutant was unable to maintain its pH_i homeostasis under alkaline conditions, showing no difference between internal and external pH values. These data were consistent with growth phenotypes of the markerless deletions, suggesting that C. glutamicum Mrp-type antiporters, especially the Mrp1 antiporter, played a predominant role in conferring Na⁺ resistance and alkali tolerance.

FIG 2.

Deletion of Mrp-type antiporters affected intracellular Na⁺ (A) and cytoplasmic pH (B) levels. Intracellular Na⁺ content was determined by ICP-OES analysis, and intracellular pH values were measured by using the pH-sensitive fluorescent probe BCECF as described in Materials and Methods. The data are presented as means ± SD from three independent experiments. Asterisks indicate significant differences between C. glutamicum wild-type and mrp1 derivative mutants by a two-tailed unpaired Student's t test (*, P < 0.05; **, P < 0.01).

C. glutamicum Na⁺/H⁺ antiporters are differentially expressed in the alkaline and salt stress response.

In order to achieve a greater understanding of the physiological roles of these Na⁺/H⁺ antiporters in C. glutamicum, the absolute expression levels of these four antiporter genes in response to alkaline and salt stresses were determined (Table 1). Given that the mrp-type operons of C. glutamicum contain 6 different adjacent genes, we chose two representative genes (mrpA and mrpG) to evaluate the transcription patterns of whole operons. Under normal, neutral-pH conditions, the mrp2 operon showed higher transcript levels than three other genes (mrp1, nhaP, and chaA), reaching approximately 304 copies per ng total cellular RNA. Compared to the noninducing condition at pH 7.0, an alkaline stimulus induced transcription of Mrp-type antiporters but not nhaP and chaA to high levels, further supporting a pivotal role of Mrp-type antiporters in the maintenance of pH homeostasis under alkaline conditions. When confronted with high-NaCl or -KCl challenges, C. glutamicum responded with significant elevations in the expression levels of Mrp-type and NhaP antiporters, consistent with previous findings that the Mrp-type and NhaP antiporters show clear Na⁺(K⁺)/H⁺ antiport activity. Although biochemical and growth assays for ChaA antiporter revealed no obvious antiport activity (data not shown), the transcript levels of chaA gene in response to NaCl stress were also slightly induced, suggesting a possible minor role in conferring Na⁺ resistance under specific conditions.

TABLE 1.

Absolute transcript levels of C. glutamicum Na⁺/H⁺ antiporters under different stress conditions^a

Stress condition	Copies/ng total RNA
	mrp1A	mrp1G	mrp2A	mrp2G	nhaP	chaA	gyrB
pH 7.0	147 ± 11	107 ± 17	304 ± 42	163 ± 45	187 ± 29	234 ± 8	695 ± 57
pH 9.0	371 ± 47**	581 ± 23**	1112 ± 138**	717 ± 129**	228 ± 51	222 ± 31	741 ± 48
1.2 M NaCl	314 ± 44**	325 ± 42**	1401 ± 174**	726 ± 169**	847 ± 114**	368 ± 3*	818 ± 102
0.8 M KCl	258 ± 36*	265 ± 31*	618 ± 125**	347 ± 37*	518 ± 36**	184 ± 9	767 ± 63

^aValues are means ± SD. Asterisks indicate significant differences between the normal (pH 7.0) condition and environmental stresses by two-tailed unpaired Student's t test (*, P < 0.05; **, P < 0.01).

The above-described assays exhibited that double deletion of mrp1 and mrp2 led to a more severe reduction of cell growth under tested conditions. One explanation for this phenomenon is that C. glutamicum might have alternative regulatory strategies, such as a functional redundancy and/or compensatory upregulation mechanism, to rescue negative effects caused by the single mrp1 disruption. To test this hypothesis, relative transcript levels of the three other antiporters in the Δmrp1 mutant versus the wild-type strain were analyzed under high-salt and alkaline-pH conditions. As shown in Fig. 3, loss of the mrp1 gene had no significant influence on the expression levels of the three other antiporters under normal, neutral-pH conditions. Salt stress elevated the transcript levels of both mrp2 and nhaP when the mrp1 gene was absent, whereas the expression of the chaA antiporter was not affected. Furthermore, the relative levels of gene expression for all three other antiporters in the Δmrp1 mutant were clearly higher than those of the wild-type strain under alkaline conditions.

FIG 3.

Disruption of mrp1 affected the transcript levels of other Na⁺/H⁺ antiporters. Relative transcript levels of target genes were quantified by real-time PCR and normalized to the 16S rRNA reference gene. The data are presented as means ± SD from three independent experiments. Asterisks indicate significant differences between C. glutamicum wild-type and mrp1-deficient mutant strains under indicated stress conditions by a two-tailed unpaired Student's t test (*, P < 0.05; **, P < 0.01).

Overexpression of the mrp1 gene improves salt tolerance under specific stress conditions.

Given the importance of Mrp-type antiporters in efficient Na⁺ resistance, we attempted to improve the salt tolerance of a C. glutamicum strain by altering its gene expression. The native promoter of either the mrp1 or mrp2 gene was exchanged by C. glutamicum promoters with different strengths, including a strong promoter of the sod gene (encoding superoxide dismutase) and a relatively weak promoter of the ilvA gene (encoding threonine dehydratase) (36). As shown in Fig. 4, although the replacement of the native mrp1 promoter by the Psod or PilvA promoter significantly improved growth performance in response to high-LiCl stress (another substrate molecule for the Mrp-type antiporter), there was no obvious effect on cell growth in the presence of added NaCl at pH 7.0. However, the overexpression of mrp1 slightly improved growth ability under mixed salt and alkaline stress conditions. Moreover, the overexpression of the mrp2 gene under the control of the PilvA promoter had little influence on growth phenotypes, whereas the exchange of the strong Psod promoter resulted in an unexpected deleterious effect under salt stress conditions.

FIG 4.

Effect of overexpression of Mrp-type antiporters by different promoters on salt resistance. Tenfold serial dilutions of each cell suspension were spotted on the indicated plates and incubated at 32°C for 3 days before being photographed.

Site-specific mutagenesis of C. glutamicum Mrp1 antiporter reveals several conserved residues conferring Na⁺ resistance.

We further performed multiple-sequence alignments among C. glutamicum Mrp1 subunits and other Mrp-related subunits, attempting to investigate the residues that are required for Na⁺ resistance. The Mrp antiporter-like NADH dehydrogenase subunits NuoL, NuoK, and NuoM/N from E. coli and Thermus thermophilus were also aligned. Nine different conserved sites were selected for point mutations and are shown in Fig. 5A and Fig. S2 in the supplemental material. Six highly conserved residues across all tested Mrp-type subunits were chosen, including Mrp1A K²⁹⁹, Mrp1C I⁷⁶, Mrp1D E¹³⁶, Mrp1E P⁶¹, Mrp1F R³⁵, and Mrp1G R²⁸. These conserved residues were replaced by amino acids showing similar chemical or structural properties. The other three additional replacements were chosen based on the combination of conserved motif and functional prediction. The Mrp1A H²³⁰ residue was inferred to be a potential quinone-binding site based on a published report showing that the corresponding histidine 224 site of E. coli NuoN might participate in proton translocation (37). The Mrp1A H⁴⁹⁹ residue was deemed to be a potential quinone-binding site, meeting the criterion of a putative quinone-binding motif (⁴⁹⁵LX₃HX₃T⁵⁰³) proposed for the respiratory complex by Fisher and Rich (38). Mrp1A G³⁷⁸ was included because of the finding that the corresponding BhMrp1A G³⁹³ residue of alkaliphilic Bacillus halodurans C-125 was required for Na⁺/H⁺ antiporter activity (23).

FIG 5.

Diverse point mutations of the Mrp1 antiporter attenuated its capacity to rescue the growth defects of the Na⁺-sensitive E. coli strain KNabc. (A) Schematic representation of amino acid positions corresponding to single site mutations. CgMrp1 and CgMrp2 were from C. glutamicum ATCC 13032, BpMrp was from Bacillus pseudofirmus OF4, BhMrp was from Bacillus halodurans C-125, BsMrp was from Bacillus subtilis 168, SaMnh was from Staphylococcus aureus Newman, PaMrp was from Pseudomonas aeruginosa PAO1, SfPha2 was from Sinorhizobium fredii, HyMrp was from Halomonas sp. Y2, VcMrpA was from Vibrio cholerae O395, EcNuoL/M/N/K were from Escherichia coli MG1655, TtNuoL/M/N/K were from Thermus thermophilus HB8. The accession number of each homolog is shown in Fig. S2 in the supplemental material. (B, C) Growth of E. coli KNabc strains carrying the original pMW118-mrp1 plasmid or its site-mutated derivatives in LBK medium supplemented with 200 mM or 400 mM NaCl. The data are presented as mean values from three independent experiments, and the standard deviations are less than 10% of the mean values.

The complementation assay of Na⁺-sensitive E. coli strain KNabc was performed to determine the relative importance of these residues. E. coli strain KNabc, lacking three major Na⁺/H⁺ antiporters, is highly sensitive to the presence of Na⁺ and fails to grow at 200 mM NaCl (34). As shown in Fig. 5B, in the presence of 200 mM NaCl, several point mutants showed remarkably impaired growth in comparison with the control strain carrying the original mrp1 operon. In particular, the mrp1A K299H mutated plasmid did not allow E. coli KNabc growth with NaCl treatment. The mrp1C I76F mutated plasmid led to an approximately 50% reduction in the capacity to confer Na⁺ resistance. In addition, two other point mutants, mrp1A H230K and mrp1D E136D, also exhibited reduced capacity to rescue growth deficiency. As shown in Fig. 5C, at the higher NaCl concentrations (up to 400 mM), the growth defects of these mutants were more severe, and nearly all point mutants showed clear growth retardation, suggesting that most residues had different degrees of effects on bacterial growth in response to high-salt stresses.

Lys 299 was essential for Na⁺ resistance and pH homeostasis.

To further characterize the residues required for Mrp1 activity, four point mutations were introduced into the chromosomal gene of interest, and growth assays were carried out under different stress conditions. As shown in Fig. 6A, the C. glutamicum Mrp1A K299H mutant strain exhibited remarkable growth defects under high-NaCl or alkaline conditions. These defects were more severe than in the Δmrp1 single mutant but quite similar to those in the Δmrp1 Δmrp2 double mutant. Complementation assays confirmed that the reverse mutation of the Mrp1A K299H strain completely restored all phenotypes to wild-type levels under tested conditions (see Fig. S3 in the supplemental material). As also shown in Fig. 6A, increased pH exacerbated the growth defects with the addition of high NaCl. The Mrp1D E136D mutant strain had growth phenotypes similar to those of the Mrp1C I76F mutant strain, which almost completely lost the capacity to grow in the presence of 0.6 M NaCl at pH 9.0 or 1.0 M NaCl at pH 8.0. The Mrp1A H230K mutant strain displayed relatively high NaCl tolerance, showing a defective phenotype only when the concentration of NaCl reached 1.0 M at pH 9.0.

FIG 6.

Effects of point mutations in the Mrp1 antiporter on Na⁺ resistance and alkali tolerance in vivo. (A) Growth assay of C. glutamicum mutant cells under different stress conditions. Tenfold serial dilutions of each cell suspension were spotted on the indicated plates and incubated at 32°C for 3 days before being photographed. (B) Intracellular pH levels were measured by using the pH-sensitive fluorescent probe BCECF as described in Materials and Methods. (C) Intracellular Na⁺ content was determined by ICP-OES analysis as described in Materials and Methods and presented as nanograms/10⁸ cells. Asterisks indicate significant differences between C. glutamicum wild-type and Mrp1-derivative mutant strains by a two-tailed unpaired Student's t test (*, P < 0.05; **, P < 0.01).

Consistent with growth phenotypes, the Mrp1A K299H mutant strain showed a significantly elevated pH_i compared to the wild-type strain at alkaline pH, implying that it lost its capacity to maintain pH_i homeostasis (Fig. 6B). Although the point mutation of Mrp1A H²³⁰ residue had no obvious effect on cell growth in the range of pH 7.0 to 9.0, the pH_i of the mutant strain is more alkaline when exposed to pH 9.0, suggesting a potential contribution of the residue to proton transfer. In addition, two other site mutants showed pH_i levels equal to those of the wild-type strain. As shown in Fig. 6C, the Mrp1A K299H mutant strain had the highest intracellular Na⁺ content, approximately 3-fold higher than that of the wild-type strain, when confronted with NaCl treatments at either neutral or alkaline pH. Site mutations in Mrp1C I⁷⁶ and Mrp1D E¹³⁶ also resulted in an increased accumulation of intracellular Na⁺, which was almost the same as that of the Δmrp1 mutant. However, the replacement of histidine with lysine in Mrp1A H²³⁰ had no significant effect on intracellular Na⁺ content.

The effects of these point mutations on gene expression of Na⁺/H⁺ antiporters were also determined. As shown in Fig. S4A in the supplemental material, under normal, neutral-pH conditions, the mutations of Mrp1A and Mrp1C subunits showed expression patterns similar to those of the wild-type strain, whereas residue replacement in Mrp1D E¹³⁶ slightly increased the expression levels of mrp2 and nhaP. Under high-NaCl or alkaline-pH conditions, the residue replacements in Mrp1A K²⁹⁹ or Mrp1D E¹³⁶ enhanced gene expression of mrp1, mrp2, and nhaP to different degrees (Fig. S4B and C). The increased transcription of these Na⁺/H⁺ antiporters still failed to rescue growth defects of the Mrp1A K²⁹⁹H mutant, ruling out the possibility that the effects of Mrp1 point mutations on Na⁺/H⁺ antiport activity were exerted at the transcriptional levels.

The homology molecular model of the Mrp1 antiporter provided a possible ion translocation mechanism.

The modeled structure of the Mrp1 antiporter was constructed based on the homologous structure of T. thermophilus respiratory complex I. As shown in Fig. S5 in the supplemental material, Mrp1A, Mrp1C, and MrpD form a functional antiporter unit that is quite similar to the NuoL-NuoJ-NuoK membrane domain of respiratory complex I. Based on sequence alignment and structure prediction, the Mrp1A or Mrp1D model reveals a putative ion translocation pathway (Fig. 7). The pathway is continuous from the cytoplasm to the periplasm and has two half-closed symmetry-related channels. The cytoplasm-facing channel of Mrp1A subunit is surrounded by S113, E142, S148, T170, and K223. The periplasm-facing channel of Mrp1A subunit contains S305, S308, H331, S393, K394, E395, T425, Y428, and S429. Several highly conserved charged or polar residues exist to connect these two channels, including H248, T306, H335, and K339, which are required for intact ion translocation pathway. The Mrp1D subunit also has a similar ion translocation pathway. The cytoplasm-facing channel of the Mrp1D subunit is formed by Y108, E136, S142, Y143, N165, S169, and K215. The periplasm-facing channel of the Mrp1D subunit contains Q298, Y327, S377, G389, K390, S416, and S422. The channels are connected by many essential polar residues, including G242, T245, and K246. The modeled structure of the Mrp1 subcomplex provides new insights into the underlying mechanisms of ion antiport and pH regulation.

FIG 7.

The homology molecular model of Mrp1 subcomplex reveals two possible ion translocation pathways. Polar residues forming the putative transmembrane channel are shown as stick models, with carbons shown in orange for the cytoplasmic-facing cavity, in green for the periplasmic-facing cavity, and in magenta for the connecting zone. The selected conserved residues of the Mrp1 antiporter for point mutations in this study were also shown as sticks, with carbon shown in yellow. The proposed ion translocation channels are indicated by blue arrows. The question mark (?) indicates that the substrate binding or transport direction is unclear.

The lysine 299 residue is located near the middle of transmembrane segment 8 (TM8) and TM9 in the Mrp1A subunit and forms an intermolecular hydrogen bond with methionine 237 in the loop between TM6 and TM7 (Fig. 7, left). Given that the key residue H248 involved in the coupling of Na⁺ or proton translocation is located at TM7 (39, 40), we speculate that loss of the hydrogen bonding interaction might affect conformational changes and the intrinsic flexibility of the Mrp1A subunit, causing an unanticipated ion leakage during salt and/or alkaline stress responses. Glutamate 136 is located at the cytoplasm-facing channel of the Mrp1D subunit (Fig. 7, right). The growth defects of the Mrp1D E¹³⁶D mutant might be attributed to the disturbance of the ion translocation pathway caused by residue substitution (39). In addition, histidine 230 lies in the TM6 of the Mrp1A subunit, and isoleucine 76 is located at a highly conserved transmembrane region of the Mrp1C subunit. However, the potential roles of these two residues in the catalytic activity of Mrp1 antiporter are still unclear.

DISCUSSION

Industrial organisms, including C. glutamicum, are often subjected to diverse stress challenges during industrial fermentation, which could be caused by nutrient limitations or environmental stresses, such as alterations in external pH, osmolality, and temperature and oxygen availability, as well as increased, toxic salt concentrations. Apart from their function in the accumulation of so-called compatible solutes in the cytoplasm, the Na⁺/H⁺ antiporters also play an important role in enhancing bacterial stress tolerance (4, 7, 30). In this study, our results demonstrate that the Mrp-type antiporters, especially the Mrp1 antiporter, play a predominant role in C. glutamicum Na⁺ resistance and alkali tolerance and that the Mrp2 antiporter is critical for mediating cellular K⁺ tolerance at alkaline pH. These observations are consistent with previous findings in many Bacillus alkaliphiles (7, 18, 19). In addition, the Δmrp1 Δmrp2 double mutant exhibited more-severe growth defects than did each single mutant. These growth properties are consistent with the previous finding that the Mrp2 antiporter also has a small but clear Na⁺/H⁺ antiport activity (34), indicating that the Mrp2 antiporter might function as a great substitute in response to high-NaCl or alkaline-pH challenges, particularly when the Mrp1 antiporter is overwhelmed. In addition, growth experiments performed with E. coli mutants suggest that the Mrp2 and NhaP antiporters might be major contributors to intracellular K⁺ efflux, while growth assays performed with C. glutamicum mutants reveal that the Mrp-type antiporters, in particular the Mrp2 antiporter, were primary resistance determinants under high-KCl stresses. One possible explanation for this discrepancy is that the Mrp-type antiporters play a leading role in reducing the cytoplasmic toxic KCl concentration, which could mask the adverse effects due to the absence of the NhaP antiporter. As for the Mrp1 antiporter, its heterologous expression has no significant impact on the growth of potassium transport-deficient E. coli strain TK2420, which could be attributed to the fact that the Mrp1 antiporter clearly exhibits both Na⁺/H⁺ and K⁺/H⁺ antiport activities, leading to the alleviation of intracellular Na⁺ toxicity (34). Our results show the physiological importance of the Mrp-type antiporters in C. glutamicum stress adaptation.

Although functional characterization of the Na⁺/H⁺ antiporters has made great progress over the past few years, relatively little work has focused on the gene expression patterns of these Na⁺/H⁺ antiporters (2, 3, 7). Several reports have shown that alkaline-pH or salt stresses affect the expression of certain Na⁺/H⁺ antiporters, such as the E. coli nhaA gene and the Bacillus subtilis mrp operon (41–43). In accord with growth phenotypes, our study also indicates that C. glutamicum Na⁺/H⁺ antiporters are differentially expressed under high-salt or alkaline-pH stress conditions. Further investigations confirmed that the loss of the Mrp1 antiporter has a profound effect on gene expression levels of other antiporters when exposed to high-salt or alkaline-pH stress. The Δmrp1 mutant exhibited a higher intracellular Na⁺ content under salt-stress conditions than the wild-type strain and a slightly elevated pH_i level at alkaline pH. Thus, we propose a simple model about the physiological roles of C. glutamicum Na⁺/H⁺ antiporters in the adaptation to environmental stresses (Fig. 8). In wild-type cells, salt stress typically results in the accumulation of intracellular Na⁺ or K⁺, thereby differentially increasing the transcription of the mrp-type and nhaP genes through as-yet-uncharacterized regulatory pathways, which eventually contribute to the pumping of toxic Na⁺ or K⁺ out of cells. Interestingly, NaCl stress also has a slight effect on the expression of the chaA gene. Additionally, an alkaline stimulus leads to a disturbance of cytoplasmic pH homeostasis and then significantly upregulates transcription levels of the mrp-type gene in C. glutamicum. In the Δmrp1 cells, there might be an alternative regulatory strategy to rescue negative effects caused by the absence of the Mrp1 antiporter. As shown in Fig. 2, deletion of the Mrp1 antiporter clearly led to intracellular Na⁺ accumulation under NaCl stress and caused a more-alkaline pH_i at a high pH (i.e., pH 9.0). The increase of intracellular Na⁺ and pH_i might directly and/or indirectly trigger higher expression levels of mrp2 and nhaP in response to salt stress or elevate transcript levels of the three other antiporters to various degrees to protect from alkaline damages. In short, the diverse expression patterns and overlapping roles of C. glutamicum Na⁺/H⁺ antiporters would help the strain cope with multiple environmental challenges.

FIG 8.

A simple model for C. glutamicum putative Na⁺/H⁺ antiporters in the adaptation to environmental NaCl and alkaline stresses in wild-type (A) and mrp1-deficient (B) strains. Solid blue or purple arrows indicate separate, positive regulation or activation, and dotted blue or purple arrows indicate unknown or as-yet-uncharacterized regulatory pathways. Solid pink arrows show secondary effects caused by the mrp1 deficiency. The orange arrows represent the transport of Na⁺, the brown arrows represent the transport of K⁺, and the green arrows represent the transport of proton. +, positive regulation; X, gene deficiency.

Previous reports have revealed that there is a close evolutionary link between the Mrp antiporter and complex I subunits (17, 39). The subunits MrpA and MrpD of the Mrp antiporter share similar ion translocation pathways with complex I subunits NuoL and NuoM, respectively. Sazanov et al. (39) proposed a model in which both MrpA and MrpD subunits of Bacillus subtilis Mrp antiporter harbor proton transport channels, whereas the interface of MrpA and MrpD forms a Na⁺ transport pathway. Highly conserved glutamic acid residues at the MrpA-MrpD interface function as putative cation binding sites. In addition, experimental evidence suggested that the individual NuoL subunit of complex I exerts an obvious Na⁺/H⁺ antiporter activity and may be able to restore the defective phenotypes of an mrpA-deficient strain (44–46). In the absence of the NuoL subunit, E. coli complex I still translocates H⁺ but does not transport Na⁺, suggesting its dual relation with H⁺ and Na⁺ as the coupling ions (47, 48). In this study, we presented a homology model of the subunits of Mrp antiporters using T. thermophilus respiratory complex I as the template, and they showed ion translocation pathways very similar to those in the model of the B. subtilis Mrp antiporter. Many key residues, including E132, K223, H248, and K394 of the MrpA subunit and E136, K215, K246, and K390 of the MrpD subunit, are also located in or near ion translocation pathways. However, although the MrpA subunit from the C. glutamicum Mrp1 antiporter reveals a possible ion channel, more-direct experimental evidence is needed to determine whether the pathway transports Na⁺.

The important roles of lysine residues in antiport activity in the Mrp-type and proton-pumping complex have also been explored in multiple previous studies (28, 37, 39, 49). In this study, we found that the lysine 299 residue of the Mrp1A subunit plays essential roles in Na⁺ resistance and alkali tolerance. Although the Mrp1A K299H mutant strain showed obvious increased transcription of mrp1, mrp2, and nhaP, the mutant still lost the capacity to survive under high-NaCl or alkaline conditions, ruling out the effects of compensatory regulatory strategies. We speculate that the lysine 299 residue might be involved in Na⁺ or proton transfer. The replacement of lysine with histidine might affect the conformational freedom of the ion translocation pathway, resulting in an unanticipated ion leakage of the Mrp1A subunit. This site mutation of lysine 299 renders the Mrp1A subunit nonfunctional and thus results in a deleterious effect on the roles of the Mrp-type antiporter. Further work will be needed to fully address this hypothesis. In general, although several conserved residues among the Mrp1 complex have been shown to be important for optimal Na⁺ resistance, the precise molecular mechanism is still not clear. Further elaboration of this underlying mechanism will provide new insights into our understanding of the relationships between amino acid residue and functional properties of the Mrp-type antiporters.

MATERIALS AND METHODS

Strains and growth conditions.

Bacterial strains and plasmids used in this study are listed in Table 2. E. coli DH5α or HST02 was used as the host cell for general cloning. E. coli KNabc, lacking three major Na⁺/H⁺ antiporters (NhaA, NhaB, and ChaA), was used as the Na⁺/H⁺ antiporter-deficient background strain (50). E. coli TK2420, lacking three major K⁺ uptake systems (Trk, Kup, and Kdp) was used as the K⁺ transport-deficient background strain (35). C. glutamicum ATCC 13032 was used as the wild-type strain in the functional analysis and as the parental strain for gene disruption. The plasmids used for heterologous expression in E. coli cells were pMW118 derivatives. E. coli KNabc and its derivatives were routinely grown at 37°C in LBK medium (0.5% yeast extract, 1% tryptone, 0.6% KCl) supplemented with indicated NaCl concentrations. E. coli TK2420 and its derivatives were routinely grown at 37°C in Ko medium [46 mM Na₂HPO₄, 23 mM NaH₂PO₄, 8 mM (NH₄)₂SO₄, 0.4 mM MgSO₄, 6 μM FeSO₄, 1 mM sodium citrate, 1 mg liter⁻¹ thiamine, 2 g liter⁻¹ glucose] supplemented with indicated KCl concentrations. For growth assay experiments, C. glutamicum was pregrown in A-rich medium [2 g liter⁻¹ yeast extract, 7 g liter⁻¹ Casamino Acids, 2 g liter⁻¹ urea, 7 g liter⁻¹ (NH₄)₂SO₄, 0.5 g liter⁻¹ KH₂PO₄, 0.5 g liter⁻¹ K₂H₂PO₄, 0.5 g liter⁻¹ MgSO₄·7H₂O, 6 mg liter⁻¹ Fe₂SO₄·7H₂O, 4.2 mg liter⁻¹ Mn₂SO₄·H₂O, 0.2 mg liter⁻¹ biotin, 0.2 mg liter⁻¹ thiamine] with 4% glucose at 32°C (51), washed, and resuspended to fresh LBO medium (0.5% yeast extract, 1% tryptone) containing indicated NaCl or KCl concentrations and buffered with 50 mM bis-Tris-propane (BTP) to different pH values. When appropriate, antibiotics were added to a final concentration of 100 μg ml⁻¹ (ampicillin), 25 μg ml⁻¹ (kanamycin), or 5 μg ml⁻¹ (chloramphenicol), and 10 μM IPTG (isopropyl-β-d-thiogalactopyranoside) was used for induction of gene expression.

TABLE 2.

Plasmids and strains used in this study

Strain construction and complementation of markerless deletions.

The primers used in this study are listed in Table 3. C. glutamicum mutant strains were achieved by a two-step homologous recombination using the temperature-sensitive plasmid pCRD206 as described previously (51). To avoid polar effects, markerless chromosomal in-frame deletions were constructed. For example, the Δmrp1 mutant was constructed as follows: the mrp1A upstream flanking region, including the first 10 codons of mrp1A gene, and the mrp1G downstream flanking region, including the last eight codons of mrp1G gene, were amplified with the Phusion high-fidelity DNA polymerase (Thermo Scientific, USA) using the primer pairs mrp1-1-For/mrp1-2-Rev and mrp1-3-For/mrp1-4-Rev, respectively. These two fragments were then fused by overlap extension PCR using the primer pairs mrp1-1-For/mrp1-4-Rev. The final PCR product was digested with BamHI and XbaI, ligated into the same sites of pCRD206, and directly transformed into E. coli HST02 host cells to yield the pCRD206-mrp1 plasmid. The resulting plasmid was then transformed into C. glutamicum ATCC 13032 by the electroporation method. The Δmrp1 deletion mutant was obtained through the first temperature selection and the second sucrose selection steps. The correct mutants were confirmed by colony PCR with the primer pairs mrp1-UF/mrp1-DR. In addition, the Δmrp1 Δmrp2 double mutant was obtained by transformation of the pCRD206-mrp2 plasmid into the C. glutamicum Δmrp1 single mutant, and the screening of double-crossover candidates was performed through the above-describe procedure. Other C. glutamicum strains containing gene disruption, point mutation, or reverse mutation at the chromosomal level were also generated according to a similar strategy.

TABLE 3.

Primers used for gene disruption, point mutation, and promoter exchange

Chromosomal replacement of native mrp1 promoters by the strong sod promoter was performed as follows: a part of the upstream sequence of the mrp1 gene containing an overlapping sequence with the sod promoter at the 3′ end was amplified with the primers mrp1-1-For and mrp1-Psod-2-Rev, a part of the mrp1 gene containing the start codon and overlapping sequence with the sod promoter at the 5′ end was amplified with the primers mrp1-Psod-5-For and mrp1-6-Rev, and the sod promoter containing the overlapping sequence was amplified using primers Psod-3-For and mrp1-Psod-4-Rev. These three PCR products were mixed together and used as a template for a fusion fragment with the primers mrp1-1-For and mrp1-6-Rev. The final PCR product was digested with BamHI and XbaI and ligated into the same sites of pCRD206 to yield the pCRD206-Psod-mrp1 plasmid. This resulting plasmid was then transformed into C. glutamicum ATCC 13032 by the electroporation method. The Psod-mrp1 strain was obtained by a two-step homologous recombination. Similar procedures were employed to generate other promoter replacements in C. glutamicum.

The plasmids for complementation of the markerless deletions were built using the shuttle vector pXMJ19. The mrp1 full-length encoding fragment was amplified from genomic DNA template with the primer pairs mrp1-5comp and mrp1-3comp. The PCR product was digested with XbaI and SmaI and ligated into the same sites of pXMJ19 to generate the inducible expression vector pXMJ19-mrp1. The pXMJ19-mrp2 vector was generated according to a similar strategy. The inducible expression vector was transformed into the Δmrp1 or Δmrp2 mutant to obtain the complemented strain. All the constructs in this study were confirmed by DNA sequencing.

Growth experiments.

The cultures of C. glutamicum were grown for 16 h at 32°C with shaking at 200 rpm prior to growth experiments. Overnight cultures were then washed and resuspended in fresh LBO medium to an optical density at 600 nm (OD₆₀₀) of 0.2. A 10-μl volume of cell suspension was inoculated into 190 μl of corresponding medium in 96-well microtiter plates, and the plates were incubated with shaking at 800 rpm at 32°C in a Microtron shaking incubator (INFORS-HT, Switzerland). OD₆₀₀ readings were collected every 2 h. Growth curves use averages from at least three independent experiments done in duplicate repeats.

The potassium uptake-deficient E. coli TK2420 cells carrying the empty vector pMW118 or its derivatives were pregrown overnight in LB medium at 37°C. The cultures were harvested, washed, and resuspended in standard minimal Ko medium with an OD₆₀₀ of 0.2. A 100-μl volume of cell suspension was inoculated into 4 ml Ko medium containing the indicated KCl concentrations. Growth was determined by measuring the optical density after 24 h of incubation in a 37°C shaking incubator.

Nucleic acid extraction and purification.

Chromosomal DNA isolation from C. glutamicum cells was performed using the E.Z.N.A. bacterial DNA kit (Omega Bio-tek, USA) according to the manufacturer's instructions. Before RNA preparation, cells were harvested, snap-frozen in liquid nitrogen, and immediately stored at −80°C. The pellets were resuspended in 100 μl of Tris-EDTA (TE) buffer (10 mM Tris-Cl, pH 7.5, 1 mM EDTA) with 10 mg ml⁻¹ lysozyme and mechanically disrupted with glass beads by three 2-min pulses in the Biospec Mini-beadbeater. Total RNA from indicated C. glutamicum cells was isolated using the RNAprep pure cell/bacteria kit and DNase I on-column treatment (Tiangen Biotech, China). The quality and quantity of isolated total RNA were assessed using the Nanodrop ND-1000 spectrophotometer (Thermo Scientific, USA) and formaldehyde denaturing agarose gel electrophoresis. The isolated total RNA was used as the template for the amplification of PCR controls to rule out DNA contamination (data not shown). For the cDNA synthesis procedure, 1 μg of total RNA was reverse transcribed with the RevertAid first-strand cDNA synthesis kit (Thermo Scientific, USA); the first-strand cDNA can be directly used as the template in the subsequent PCR assay.

qPCRs.

Absolute and relative quantitative real-time reverse transcription-PCRs (qPCRs) were performed to calculate the expression levels of target genes. Overnight cultures of C. glutamicum strains were pregrown to mid-exponential phase (OD₆₀₀ ≈ 5.0) in rich medium at 32°C. Then, cells were harvested separately, shifted to their corresponding medium buffered with 50 mM BTP, and incubated for a further 2 h before RNA extraction. For the relative qPCR analysis, the reaction mixtures were prepared with the SYBR green real-time PCR master mix (Toyobo, Japan) according to the manufacturer's instructions. The primers used in this procedure are listed in Table 4. The relative qPCR was usually performed in triplicate samples and repeated in three independent experiments using an Applied Biosystems 7500 fast real-time PCR system (Thermo Scientific, USA). Thermocycling conditions were as follows: 1 min at 95°C; 40 cycles of 95°C for 15 s, 60°C for 15 s, and 72°C for 45 s; a final step of 95°C for 15 s, 60°C for 15 s, and 95°C for 15 s. The relative fold changes in gene expression were calculated according to the delta delta threshold cycle method after normalization using 16S rRNA gene or gyrB as the reference gene (52). Similar results were obtained for the two different reference genes, and 16S rRNA gene was used as the reference gene for normalization in the following results. For the absolute qPCR analysis, the standards for target and reference gene transcripts were obtained by PCR using C. glutamicum chromosomal DNA as the template, and the primers used in this procedure are listed in Table 4. After gel extraction, the concentrations of standards were measured in nanogram per microliter using the Nanodrop spectrophotometer and converted to copies per microliter based on the molecular weight of PCR products. Individual standards were diluted to produce stocks at 10⁹ copies μl⁻¹ and then combined to produce a master mix of standards for all genes of interest at 10⁸ copies μl⁻¹. The standards for absolute quantitation were generated for concentrations ranging from 10⁷ to 10¹ copies μl⁻¹ by 10-fold serial dilutions. Standard curves were constructed by plotting threshold cycle values against log template concentrations of each gene. The absolute transcript copies of target genes were analyzed based on standard curves by the qPCR assays.

TABLE 4.

Primers used for qPCR analyses

Site-directed mutagenesis.

Site-directed mutagenesis of the conserved amino acid residues from the Mrp1 antiporter was generated by PCR using oligonucleotide primer pairs with the desired mismatching nucleotides (53). The primers used in this study are listed in Table 5. The pMW118-mrp1A-H²³⁰K mutated plasmid was constructed as follows. The pMW118-mrp1 plasmid, containing all six genes of mrp1ACDEFG, was used as the parental template. The primer pairs Mrp1A-H230K-5F/Mrp1A-H230K-3R were designed with mutated bases at the center of the oligonucleotides. The PCR products were digested with DpnI to remove methylated parental plasmid and transformed into E. coli DH5α to screen correct transformants. Similar strategies were performed for the construction of other site-mutated plasmids. All constructs were confirmed by DNA sequencing, and the correct mutations were separately transformed into E. coli KNabc host cells for further experiments.

TABLE 5.

Primers used for site-directed mutagenesis

^aThe mutated sites are shown in bold and underlined.

Measurement of cytoplasmic pH and intracellular Na⁺ content.

Cytoplasmic pH was determined by using the pH-sensitive fluorescent probe 2′,7′-bis-(2-carboxyethyl)-5(6)-carboxyfluorescein (BCECF)-acetoxymethyl ester as previously described with some modification (54). The precultures of indicated strains were harvested at the mid-exponential growth phase (OD₆₀₀ ≈ 5.0), washed and resuspended in LBO medium buffered to the indicated pH values, and then incubated with 2.0 μM BCECF-acetoxymethyl ester (Sigma-Aldrich, Germany) at 32°C for 30 min in the dark. For analysis of intracellular pH (pH_i) values, cells were washed with fresh medium and 100 μl of cell suspension was transferred to 96-well microplates. The fluorescence was measured using the SpectraMax M5 Microplate Reader at 535-nm emission wavelength after the dual excitation at 440 nm and 490 nm. The pH-dependent spectral shifts exhibited by BCECF allow the calibration of the pH response in terms of the ratio of fluorescence intensities measured at two different excitation wavelengths. An in situ calibration curve for the BCECF excitation ratio was performed by incubating cells at various external pH values between pH 6.0 and pH 9.0, in the presence of a mixture of 150 μM KCl, 10 μM nigericin, 50 μM carbonyl cyanide-3-chlorophenylhydrazone, 20 μg ml⁻¹ valinomycin, and 50 μg ml⁻¹ gramicidin in order to equilibrate the intracellular pH and external pH values. The cytoplasmic pH was estimated using a standard calibration curve obtained from the normalized emission values correlated with the controlled external pH values.

Intracellular Na⁺ content was quantified by the inductively coupled plasma optical emission spectrometer (ICP-OES) analysis as described previously (55). Briefly, C. glutamicum cells were harvested at the mid-exponential growth phase (OD₆₀₀ ≈ 5.0), resuspended directly in an equal volume of LBO–0.6 M NaCl medium at the indicated pH, and incubated for a further 60 min at 32°C. The strains were then pelleted by centrifugation, washed six times with distilled-deionized water to remove any exogenous sodium in the medium, and finally digested with a mixture of HNO₃-HClO₄ (3:1) solution. The quantification of Na⁺ content in the indicated sample was analyzed using a PerkinElmer Optima 8300 ICP-OES and is shown in nanograms per 10⁸ cells. The total number of cells in the sample was calculated by counting the colonies in the gradient dilution plates.

Homology molecular modeling.

Multiple-sequence alignments among C. glutamicum Mrp1 subunits and other Mrp homologs derived from alkaliphiles, neutralophiles, and pathogens were performed by Clustal Omega and the ESPript 3.0 server (56, 57). The molecular models of N-terminal (Mrp1A_N) and C-terminal (Mrp1A_C) domains of Mrp1A antiporter were built using the crystal structures of homologous subunits NuoL (PDB ID 4HEA) and NuoJ (PDB ID 4HEA), respectively, from Thermus thermophilus respiratory complex I as the templates (20). The Mrp1C model was generated using a template based on the homologous structure of the NuoK subunit (PDB ID 4HEA) from T. thermophilus respiratory complex I, and the Mrp1D model was obtained using templates of homologous subunit NuoM/N (PDB ID 3RKO) from E. coli respiratory complex I (40).

The homology models were built using the Phyre 2 molecular modeling server (58) and checked with the RAMPAGE server (59). The combinations of the ø and ψ angles of residues in favored, allowed, and outlier regions of Ramachandran plot were qualified (see Table S1 and Fig. S6 in the supplemental material), and the assessment indicated that the homology models were reliable for structural analyses. The tentative structure of the Mrp1 subcomplex was constructed by merging independent models of Mrp1A_N, Mrp1A_C, Mrp1C, and Mrp1D using the PyMOL molecular graphics system based on the template of T. thermophilus NuoLJKMN model (PDB ID 4HEA).