Role of the nhac-encoded Na /H Antiporter of Alkaliphilic Bacillus firmus OF4

Transcription

1 JOURNAL OF BACTERIOLOGY, June 1997, p Vol. 179, No /97/$ Copyright 1997, American Society for Microbiology Role of the nhac-encoded Na /H Antiporter of Alkaliphilic Bacillus firmus OF4 MASAHIRO ITO, ARTHUR A. GUFFANTI, JENNIFER ZEMSKY, D. MACK IVEY, AND TERRY A. KRULWICH* Department of Biochemistry, Mount Sinai School of Medicine of the City University of New York, New York, New York Received 12 March 1997/Accepted 16 April 1997 Application of protoplast transformation and single- and double-crossover mutagenesis protocols to alkaliphilic Bacillus firmus OF4811M (an auxotrophic strain of B. firmus OF4) facilitated the extension of the sequence of the previously cloned nhac gene, which encodes an Na /H antiporter, and the surrounding region. The nhac gene is part of a likely 2-gene operon encompassing nhac and a small gene that was designated nhas; the operon is preceded by novel direct repeats. The predicted alkaliphile NhaC, based on the extended sequence analysis, would be a membrane protein with 462 amino acid residues and 12 transmembrane segments that is highly homologous to the deduced products of homologous genes of unknown function from Bacillus subtilis and Haemophilus influenzae. The full-length version of nhac complemented the Na - sensitive phenotype of an antiporter-deficient mutant strain of Escherichia coli but not the alkali-sensitive growth phenotypes of Na /H -deficient mutants of either alkaliphilic B. firmus OF4811M or B. subtilis. Indeed, NhaC has no required role in alkaliphily, inasmuch as the nhac deletion strain of B. firmus OF4811M, N13, grew well at ph 10.5 at Na concentrations equal to or greater than 10 mm. Even at lower Na concentrations, N13 exhibited only a modest growth defect at ph This was accompanied by a reduced capacity to acidify the cytoplasm relative to the medium compared to the wild-type strain or to N13 complemented by cloned nhac. The most notable deficiency observed in N13 was its poor growth at ph 7.5 and Na concentrations up to 25 mm. During growth at ph 7.5, NhaC is apparently a major component of the relatively high affinity Na /H antiport activity available to extrude the Na and to confer some initial protection in the face of a sudden upshift in external ph, i.e., before full induction of additional antiporters. Consistent with the inference that NhaC is a relatively high affinity, electrogenic Na /H antiporter, N13 exhibited a defect in diffusion potentialenergized efflux of 22 Na from right-side-out membrane vesicles from cells that were preloaded with 2 mm Na and energized at ph 7.5. When the experiment was conducted with vesicles loaded with 25 mm Na, comparable efflux was observed in preparations from all the strains. Seven structurally distinct genes that putatively encode Na /H antiporters have been identified in eubacteria to date, i.e., nhaa (13), nhab (33), and chaa (19) from Escherichia coli, nhac from alkaliphilic Bacillus firmus OF4 (17), napa from Enterococcus hirae (41), teta(l) from Bacillus subtilis (8), and a gene from alkaliphilic Bacillus sp. strain C-125 (16). Many of these already have obvious homologs in other genera (11, 28, 29, 32, 39), and more such homologs are emerging from genome sequencing projects. Three of the antiporter gene products, those of nhaa and nhab from E. coli (34, 38) and teta(l) from B. subtilis (8), have been rigorously shown to encode Na /H antiporters through studies of the purified, reconstituted gene products, and in the last case the protein has been shown to be a multifunctional antiporter that can also catalyze efflux of K and a tetracycline-metal complex. The activities of gene products for other antiporter genes have been inferred from studies of the enhanced antiport activity or ph-dependent Na resistance conferred by the cloned gene in Na - sensitive mutants of either E. coli or the homologous bacterium. The major roles attributed to these gene products in eubacteria are in Na resistance and the capacity to maintain * Corresponding author. Mailing address: Box 1020, Department of Biochemistry, Mount Sinai School of Medicine, 1 Gustave L. Levy Pl., New York, NY Phone: (212) Fax: (212) krulwich@msvax.mssm.edu. Present address: Department of Biological Sciences, University of Arkansas, Fayetteville, AR a cytoplasmic ph that is below the external ph during growth at, or a sudden shift to, an alkaline ph (22, 31, 41). The importance of successfully meeting these physiological challenges is supported by the fragility and pleiotropy of mutants lacking one or more functional antiporter genes (7, 30, 33, 40) and by the likelihood that the genes encoding structurally distinct Na /H antiporters evolved at separate times (31), resulting in the presence of multiple genes of potentially overlapping function in each individual eubacterial species thus far examined. There are several examples of apparent compensation for mutational loss or insufficiency of one antiporter by up-regulation of a gene for a different Na /H antiporter, sometimes facilitated by a mutational change in a regulatory gene, or introduction of a heterologous regulatory gene (7, 13, 33, 36). In addition, many bacteria probably use distinct K /H antiporters for the ph homeostasis function when Na is scarce or the Na /H antiporters are mutationally inactivated (7, 29, 35). However, extremely alkaliphilic Bacillus species exhibit Na specificity for this particularly crucial physiological process (23). In addition to their Na specificity, the capacity of alkaliphilic Bacillus species for ph homeostasis is extraordinary in withstanding sudden, large upward shifts in external ph and in maintaining a cytoplasmic ph that is more than 2 U below the external ph under optimal growth conditions (23, 37). It is therefore of particular interest to characterize the multiple antiporters of a single alkaliphilic strain, their particular roles, and their properties. The putative antiporter gene from Bacil- 3851

2 3852 ITO ET AL. J. BACTERIOL. lus sp. strain C-125 is a good candidate for a major role in alkaliphily, but it has yet to be shown to confer Na /H antiport activity when expressed alone in a homologous or heterologous deletion strain (16). Such a capacity has been shown for the nhac gene from B. firmus OF4 (17), but nothing is yet known about the physiological role of this gene. Interest in nhac is enhanced by the recent finding of homologs in both B. subtilis (GenBank accession no. P54571) and Haemophilus influenzae (11). Comparative studies of the structures and roles of neutrophile and alkaliphile homologs might illuminate part of the basis for alkaliphily. Sequences of the nhac homologs immediately suggested that the sequence that we earlier proposed for the alkaliphile nhac (17, 18) might be incomplete, with a truncation at the N terminus. However, attempts to clone the region upstream of the available alkaliphile nhac fragment were hampered by problems of toxicity in E. coli. Cloning approaches in which the requisite molecular manipulations were carried out within B. firmus OF4 itself, as well as studies of the physiological roles of the gene in the alkaliphile, were limited by the genetic tools available. In the current study, these tools have been expanded by using the protocols developed by Biswas et al. (4) for other gram-positive bacteria. A more complete sequence analysis of nhac, and the chromosomal region encompassing it, is presented here, together with data on the physiological roles of nhac that emerged from studies of a newly constructed nhac deletion strain. MATERIALS AND METHODS Bacterial strains, plasmids, and growth conditions. B. firmus OF4811M, a streptomycin-resistant (Sm r ) methionine auxotroph, was routinely grown either in a medium with a ph of 7.5 or in a medium with a ph of 10.5 with malate as the carbon source (9). The mutant strain of B. firmus OF4811M, DA78, that is negative for growth at ph 10.5 and completely unable to acidify its cytoplasm relative to the medium during a ph shift from 8.5 to 10.5 was grown on a rich medium at ph 8.5 (25). The B. subtilis wild-type strain BD99 and teta(l) deletion strain JC112 were obtained and grown as described previously (7). E. coli EP432, with Na /H antiporter genes nhaa and nhab disrupted, was grown in LBK medium (13) supplemented with NaCl as indicated. The E. coli-b. subtilis shuttle plasmid pyh56 (ColE1 rep pub110 rep, conferring kanamycin resistance [Km r ] in gram-positive bacteria and ampicillin resistance [Ap r ] in gram-negative bacteria) was used as a vector; it was constructed and provided by David Bechhofer. When necessary, kanamycin was added to 50 g/ml, or ampicillin was added to 100 g/ml. Another shuttle plasmid, pbk15, provided by Kevin Zen, conferred Ap r in E. coli and chloramphenicol resistance (Cm r to 5 g/ml) in B. firmus OF4811M. The plasmid pjb10, which contained a truncated version of nhac, was isolated earlier from a pgem3zf( ) library of B. firmus OF4 DNA (17). Plasmid pjx5, also isolated in that earlier study (17), and containing most of nhac and an additional downstream sequence, was used to extend the sequence downstream of nhac. The plasmid pg host4 (pgk12 rep, conferring erythromycin resistance [Erm r ] in gram-positive and -negative bacteria) was obtained from Appligene, Inc., Pleasanton, Calif. Preparation and transformation of protoplasts. Protoplasts were prepared from B. firmus OF4811M by using the protocol described previously for Bacillus sp. strain C-125 (3) and were spread on modified DM-3 medium (3). In initial experiments the medium was adjusted to various ph values with NaOH or HCl. The protoplasts were incubated at 37 C for 3 days, after which the number of colonies was recorded. Colony formation by the protoplasts was strikingly dependent upon the ph of the regeneration medium. About 30% of the protoplasts regenerated at ph 7.0, whereas lower numbers of colonies were found as the ph was increased during regeneration, and no regeneration occurred at ph 6.5. Although optimal regeneration of cell walls apparently occurred at ph 7.0, the colonies formed at this ph were tiny in comparison to those formed at slightly higher ph values, consistent with earlier observations on the effect of ph on the growth of B. firmus OF4 (37). Thus, regeneration was routinely carried out at ph 7.5. Under these conditions, the transformation efficiency was approximately 10 6 transformants per g of plasmid DNA. The method used to transform the protoplasts was slightly modified from that developed by others for B. subtilis (6). Protoplasts were prepared in SMMP medium, consisting of 0.5 M sucrose, 0.02 M maleic acid, and 0.02 M MgCl 2 in double-strength Penassay broth (ph 7.5) (3). Polyethylene glycol 8000 (final concentration, 22.5%) was used to introduce the plasmid DNA into the protoplasts. Kanamycin (200 g/ml) was added to the modified DM-3 medium for selection of transformants. Cloning of sequences upstream of nhac. A 789-bp HindIII fragment derived from the nhac gene was isolated from pjb10 (17) and purified from an agarose gel with the Wizard DNA cleanup system kit (Promega, Madison, Wis.). The fragment was ligated with HindIII-restricted pg host4 to generate the plasmid pgc800. Plasmid pg host4 is sensitive to temperatures above 37 C and has previously been used to obtain single-crossover integrants with as little as 500 bp of homology (4). The recombinant plasmid pgc800 was established in E. coli XL-1 Blue MR (Stratagene, La Jolla, Calif.) and then transformed into B. firmus OF4811M. In order to clone the region upstream of nhac, a transformant strain containing pgc800 was grown overnight at 28 C in the presence of 0.2 g of erythromycin per ml and then diluted and plated at 37 C on complex medium (3) containing 0.2 g of erythromycin per ml (ph 7.5). The frequency of integrations per cell was estimated as the ratio of the number of Erm r cells at 37 C to the number of viable cells at 28 C. Of viable cells, were Erm r, yielding an integrations-per-cell frequency of 1.4%. For subsequent use, the integrants isolated at 37 C were routinely maintained at 37 C on complex medium containing erythromycin. Chromosomal DNA was prepared from B. firmus OF4811M and its derivatives by the method of Wilson (42). Chromosomal DNA prepared from 15 independent single-crossover candidates was subjected to restriction enzyme analysis. Southern hybridization, with the 789-bp HindIII nhac gene fragment as a probe, was then used to confirm candidates for singlecopy integrants at the nhac gene locus (data not shown). One of these integrants, strain CS1, was used for further study. For isolation of the region upstream of nhac, chromosomal DNA from strain CS1 was digested with NsiI and then religated. The ligated DNA was used to transform B. firmus OF4811M and E. coli XL-1 Blue MR to Erm r. Interestingly, only transformants of B. firmus OF4811M could be isolated. This supported the earlier inference that the region upstream of nhac is toxic to E. coli cells, at least when present together with a significant amount of the nhac coding region. Plasmid DNA was prepared from eight independent B. firmus OF4811M transformants by the alkaline sodium dodecyl sulfate method. All of them were shown to have the same size and restriction map. They contained about 1.65 kb of DNA farther upstream from the previously reported 5 end of the nhac gene sequence (18). This region was entirely sequenced on both strands from a plasmid designated pgcup1. The nhac locus was further characterized by extension of the sequence downstream of nhac on the previously isolated clone pjx5, which contains most of the nhac gene and an additional downstream sequence that had not been fully determined (17). All sequencing was conducted in the DNA Core Laboratory of the Mount Sinai School of Medicine with an Applied Biosystems automated DNA sequencer and commercially prepared oligonucleotide primers (Genset Corp., Paris, France). The sequence analysis used the Genetics Computer Group software package (10) on a VAX computer. Cloning of nhac and truncated versions thereof into plasmid vectors. For studies of the antiport activity conferred upon E. coli strains, three versions of nhac of various lengths were amplified by PCR from chromosomal DNA of B. firmus OF4. The gene coding for the entire sequence, beginning with the ATG translational start indicated in Fig. 1, was designated nhac12; a truncated version that was missing 2 putative transmembrane segments (TMS) at the N terminus was designated nhac10 (see Fig. 1); and a truncated version corresponding to that originally isolated and missing 5 putative TMS at the N terminus (see Fig. 1) was designated nhac7. The PCR primers used were NCF12-NCR1 for nhac12, NCF10-NCR1 for nhac10, and NCF7-NCR1 for nhac7. The primer sequences are as follows: NCF12, ATGGAGAAAAAAGCTACCGTTGG; NCF10, ATGATTGATGGCCTGAAGATAGGG; NCF7, ATGTCTCCATTG TCTGACACTAC; and NCR1, CCACTTGCTTTATCCACGTTT. Each PCR product was ligated with HincIII-digested pgem11zf( ), which was then digested with EcoRI and BamHI. The released purified fragment was ligated with pbk15 double digested with EcoRI and BamHI. The genes inserted into pbk15 were behind an ermc promoter, and each protein product would contain 11 additional amino acids (aa) at the N terminus, 8 from ErmC and 3 from the multiple cloning site of pgem11zf( ). The pbk15 plasmids bearing nhac12, nhac10, and nhac7 were designated, respectively, pbkc12, pbkc10, and pbkc7. A different version of nhac that would produce an unmodified product expressed from its own promoter was prepared by PCR amplification from primers indicated in Fig. 1. The forward primer (5 -GGGTGTTGTGGTTTGG TCTA-3 ) corresponded to bp 1372 to 1391 in Fig. 1. The reverse primer (5 -C CACTTGCTTTATCCACGTTT-3 ) corresponded to the complementary sequence of bp 2880 to The PCR product was ligated into HpaI-digested plasmid pyh56 (Ap r in gram-negative bacteria; Km r and Erm r in gram-positive bacteria). The recombinant plasmid containing nhac was designated pnhac. Preparation of membrane vesicles from E. coli EP432 transformants and assays of antiport activity. Cells grown on LBK were used to prepare membrane vesicles by the method of Ambudkar and coworkers (2), as described by Goldberg et al. (13). For assays of Na /H or Ca 2 /H antiport activities, changes in the ph were monitored by quinacrine-hcl fluorescence as described elsewhere (19). Also, as described elsewhere, the buffer for Na /H antiport activity contained Mg 2, while that for Ca 2 /H antiport activity did not (5). Construction of nhac-deleted B. firmus OF4 (N13). PCR was performed on purified B. firmus OF4 chromosomal DNA by using the same two oligonucleotide primers that were employed to prepare the insert for pnhac. The purified PCR product was first ligated into HincII-digested pgem3zf( ) (Ap r ; Promega). The recombinant plasmid was selected for by blue/white screening in E. coli DH5. After isolation, the nhac fragment was released by BamHI-PstI treatment and ligated to the temperature-sensitive plasmid pg host4 cut with the same endonucleases. After isolation, the plasmid was digested with ScaI and a

3 VOL. 179, 1997 nhac OF ALKALIPHILIC B. FIRMUS OF bp ScaI/ScaI fragment was removed from the middle of nhac (Fig. 1). A gene encoding Cm r was ligated to this linear plasmid, resulting in a recombinant plasmid containing a fragment of nhac replaced by the Cm r gene ( nhac::cm r ). After isolation, the recombinant plasmid was introduced into B. firmus OF4 by protoplast transformation, and cell walls were regenerated at 28 C. Transformed cells were resistant to both erythromycin (1 g/ml) and chloramphenicol (3 g/ml). Gene replacement was achieved by the approach of Biswas et al. (4). For isolation of single-crossover integrants, transformants growing at 28 C were diluted and plated onto complex medium containing erythromycin and incubated at 39 C. Of a total of colonies, , or 0.12%, were Erm r at 39 C. This frequency of integration was similar to that in Lactococcus lactis, where the frequency of integration increased with the length of the homologous fragment (4). The single-crossover event, resulting in integration of the entire plasmid in the chromosome, was confirmed by PCR analysis. For the double-crossover event that completes the gene replacement, a single-crossover integrant was grown overnight at 28 C in complex broth containing chloramphenicol. Cells were then diluted and plated onto medium of the same composition and incubated at 39 C. Double-crossover candidates were identified on the basis of Cm r and Erm s ; 8 such candidates were found among 200 total colonies. Both Southern hybridization and PCR analyses confirmed that the gene replacement event had taken place in these candidates. The nhac deletion strain used in these studies was designated N13; preliminary studies indicated that its phenotype was similar to those of the other seven deletion strains isolated. Growth experiments and determinations of doubling time. B. firmus OF4811M/ pyh56, N13/pYH56, and N13/pNhaC were grown on a medium containing 50 mm Na 2 CO 3 (ph 8.5), 1 mm potassium phosphate, and 0.01 mm MgSO 4, plus 0.1% yeast extract, 1% trace salts (9), and 50 mm Na-D,L-malate. An overnight culture grown on this medium, at ph 8.5, was used as an inoculum for determination of growth at ph 7.5 or 10.5 in the presence of various concentrations of NaCl. The basal medium at ph 7.5 consisted of 50 mm potassium phosphate (ph 7.5) and 0.01 mm MgSO 4, plus 0.1% yeast extract, 1% trace salts, and 10 mm potassium malate. The basal medium at ph 10.5 consisted of 50 mm K 2 CO 3 (ph 10.5) and 1 mm potassium phosphate, plus the same additions as at ph 7.5. NaCl was added to various concentrations as indicated. In order to minimize the introduction of Na from the inoculum, the culture was centrifuged for 10 min at 10,000 g, washed, and resuspended in the appropriate ph 7.5 or 10.5 medium. Fifty-milliliter cultures in 250-ml flasks were grown at 30 C in a rotary shaker (200 rpm). Samples were taken at various times, and the A 600 (absorbance at 600 nm) was measured. The rates of growth, in doublings per hour, were calculated from the rates of change in A 600 during the logarithmic phase of growth. Growth studies on B. firmus OF4811M DA78 were conducted with the same media, and those on B. subtilis strains were conducted in malate-containing media at either ph 7 or ph 8.3, as described earlier (7). Assays of 22 Na efflux from right-side-out membrane vesicles prepared from wild-type B. firmus O4 and the nhac deletion strain. Right-side-out membrane vesicles were prepared as described previously (14) by modification of the method of Kaback (20). Cells were grown in 50 mm Na 2 CO 3 medium, ph 8.5, as described above. Protoplasts were lysed by dilution into 25 mm potassium phosphate (ph 7.5) plus 5 mm MgCl 2. After separation of nonvesicular material and washing, the vesicles were frozen and stored at 70 C. For assays of 22 Na efflux, the vesicles were passively loaded at 20 C for2hbytheaddition of 50 Ci of 22 NaCl/ml plus either 2 or 25 mm nonradioactive NaCl. Valinomycin was added to 1 M to the vesicle suspension. Na efflux was initiated by diluting the vesicles 100-fold into either 25 mm potassium phosphate (ph 7.5) plus 5 mm MgCl 2 or 25 mm sodium phosphate (ph 7.5) plus 5 mm MgCl 2. Valinomycin at 1 M was included in the dilution buffers. At various times, 0.5-ml samples were taken, filtered through Millipore GSWP m-pore-size filters, washed with 2 ml of 25 mm potassium phosphate (ph 7.5), dried, and counted by liquid scintillation. Controls for binding, and calculations of internal Na concentration, were conducted as described previously (14). Determination of the cytoplasmic ph after a shift in the external ph from 8.5 to Cells were grown to the mid-logarithmic phase at either ph 7.5 or ph 10.5 in malate-containing medium. They were harvested by centrifugation and resuspended to 10 mg of protein/ml in 50 mm K 2 CO 3, ph 8.5, plus 10 mm potassium malate. They were then rapidly diluted 25-fold into 50 mm K 2 CO 3, ph 10.6, plus 10 mm potassium malate. Where indicated, the buffers at either ph contained various amounts of NaCl. The transmembrane ph gradient ( ph) was determined after 10 min as described elsewhere (24) by the distribution of radiolabelled methylamine. The control for nonspecific binding of the probe contained 10 M gramicidin. Values for the cytoplasmic ph were calculated from duplicate measurements of the outside ph and the ph, in at least six independent determinations. Protein was determined by the method of Lowry et al. (27), by using egg white lysozyme as the standard. Nucleotide sequence accession number. The nucleotide sequence for the nhac gene and the region surrounding it has been deposited in GenBank and given accession no. U RESULTS Sequence analysis of the nhac region and complementation studies using full-length nhac. The nucleotide sequence and the deduced amino acid sequence completed for the nhac gene and the region immediately surrounding it are shown in Fig. 1. The new sequence extends well beyond that shown in the figure, both upstream and downstream of nhac. Novel direct repeats were found upstream of nhac and were designated DRS1 and DRS2 (Fig. 1). The nucleotide identity between DRS1 and DRS2 is 95.7% over a 138-bp overlap. Sequence analysis suggested that it is possible that one or both of the DRS sequences encode a small, highly hydrophobic protein that might span the membrane once or twice and has no obvious homologs or function. No direct repeat sequences were found upstream of the nhac genes from B. subtilis or H. influenzae. The new sequence provided a stronger candidate than that proposed earlier for the nhac translation start site and resulted in a deduced product that was 60 aa longer than previously inferred (Fig. 1). NhaC would be a 462-aa hydrophobic protein (molecular size, 49.5 kda; pi, 6.1), very close in size to the homologs reported from H. influenzae (11) and, especially, B. subtilis (GenBank accession no. P54571). NhaC from B. firmus OF4 is 64.9% similar and 34.6% identical to that of B. subtilis and is 63.0% similar and 34.8% identical to that of H. influenzae. Since the truncated versions of B. firmus OF4 nhac that had been studied earlier had exhibited Na /H antiport activity (17), it was of interest to determine whether the additional sequence at the N terminus added features whose analysis, in isolation, might offer a suggestion about a specific role(s). Analyses using the BLAST algorithm (1) indicated that two different regions of amino acid sequence in this newly sequenced N terminus, each of 19 to 21 residues, showed sequence similarity to regions of the large mouse calcium channel alpha-1 subunit (12); the level of identity was 45% in both matches, and the level of similarity was 75% in one match and 59% in the other. Other calcium channels were similarly matched with these regions of NhaC. An open reading frame that is proposed to have its translational start 37 bp downstream of the end of nhac is also likely to be transcribed together with nhac because of their proximity and the presence of a putative terminator following the downstream gene. This gene was designated nhas. It is predicted to encode a 66-aa hydrophobic protein (molecular size, 7.9 kda; pi, 11.8). NhaS showed significant sequence similarity to a hypothetical protein, YqgO, from B. subtilis (GenBank accession no. P54492), with 26.3% identity and 61.4% similarity in a 57-aa overlap. However, the nhas homolog of B. subtilis is in a completely different chromosomal region from its nhac homolog gene. NhaS also showed similarity to a region encompassing the 3rd and 4th transmembrane helices in the N-terminal half of the eukaryotic Na /K ATPases, e.g., with 26.6% identity and 68.8% similarity to that from Drosophila melanogaster (26). This alignment includes a gap that is created by the presence of a basic-amino-acid- and tyrosine-rich region that is suggestive of the possibility that this gene could have a DNAbinding function. Two truncated forms of nhac, relative to the currently proposed full-length gene, had earlier been shown to complement the nhaa strain of E. coli NM81, as assayed both by growth on Na and via Na /H antiport activity of everted membrane vesicles. The capacity of the full-length form was examined similarly, in the nhaa nhab strain of E. coli, EP432. Upon transformation of this strain with pbkc7, pbkc10, or pbkc12, encoding nhac7, nhac10, and nhac12, respectively, each of the three plasmids allowed the sensitive strain to grow on LBK plus 0.4 M NaCl. E. coli EP432 transformed with control plasmid pbk15 failed to grow on this medium. In addition, all three plasmids conferred modest but reproducible Na /H antiport activity, as measured in a fluorescence assay

4 3854 ITO ET AL. J. BACTERIOL. FIG. 1. The complete nucleotide sequence and deduced amino acid sequence of the B. firmus OF4 nhac region. Inverted repeats that could represent terminators are indicated ( ). Putative Shine-Dalgarno sequences are shown in bold. Upstream of nhac, the two DRS regions are underlined and a putative promoter region is double underlined; preliminary promoter mapping experiments indicate a complexity that has not yet been resolved but includes a promoter in this region (data not shown). Within the nhac gene, 12 putative TMS are underlined and the primers used to generate constructs corresponding to nhac12, nhac10, and nhac7 are indicated with half-barbed arrows. A similar arrow upstream indicates a primer used to amplify a larger nhac construct for preparation of pnhac. The region between the two indicated ScaI sites is that replaced by the antibiotic cassette in the nhac strain N13. in everted membrane vesicles prepared from the transformants of E. coli EP432 and compared to a control transformant. The assays were conducted in buffers in which the control membranes exhibit a small residual antiport, presumably due to K (Ca 2 )/H antiporters that may also account for retention of ph homeostatic capacity; under these conditions, the transformants exhibited an increase of about 100% over the control. In view of the indications from sequence analysis that there might be a function of NhaC in relation to both Na and Ca 2, as had earlier been observed for E. coli chaa (19), an assay of Ca 2 /H antiport activity by the transformants was also conducted, but no differences were observed (data not shown). The results with E. coli EP432 were consistent with NhaC being a Na /H antiporter. However, this double mutant of E. coli does not exhibit an alkali-sensitive phenotype independent of the Na content of the medium (33), and thus complementation of this strain by nhac did not necessarily reflect a capacity to support ph homeostasis at alkaline ph values. The capacity of pnhac to complement mutant strain JC112, a teta(l) deletion strain of B. subtilis that is very alkali sensitive at low Na concentrations as well as sensitive to elevated Na at high phs (7), was therefore examined. A JC112/pNhaC transformant showed no enhancement of growth at ph 8.3, although a wild-type B. subtilis strain exhibited significant enhancement of growth at the alkaline ph. Depending upon the medium at ph 8.3, expression of nhac doubled the growth rate of the wild-type strain, whereas it did not enhance the growth of JC112 (data not shown). The nonalkaliphilic mutant strain of B. firmus OF4811M, DA78, which had been shown to lack the capacity for ph homeostasis (25), was not complemented by pnhac (data not shown). It appeared, then, that pnhac could not complement mutants of Bacillus that were lacking an active form of the genes encoding the Na /H antiporter with a dominant role in ph homeostasis at alkaline phs. This did not, however, rule out a more subtle role for NhaC in this process, or a role at less alkaline ph values at which B. firmus OF4 would still have to extrude the Na entering with its numerous Na /solute symporters (22). Studies were accordingly initiated on B. firmus N13, with a deletion in nhac, in comparison with the wild-type parent strain. The growth phenotype, capacity for ph homeostasis, and diffusion potential 22 Na efflux activity of the nhac strain, B. firmus OF4811M N13. As shown in Fig. 2, the deletion strain N13 grew well at ph 10.5, especially at high NaCl concentrations; the growth rates of the wild-type strain transformed with a control plasmid in these batch cultures were generally somewhat lower than growth rates observed for untransformed wildtype cells in continuous cultures at the same ph value (37). At

5 VOL. 179, 1997 nhac OF ALKALIPHILIC B. FIRMUS OF FIG. 2. Comparison of the doubling times of B. firmus OF4811M/pYH56, N13/pYH56, and N13/pNhaC at phs 7.5 and 10.5 on media with various concentrations of NaCl. Wild-type B. firmus OF4811M and N13 (with nhac deleted) transformed with control plasmid pyh56, and N13 complemented with plasmid pnhac, were grown at either ph 7.5 (A) or ph 10.5 (B), as described in Materials and Methods, in the presence of the indicated concentrations of NaCl. The results are the averages of at least three separate determinations. lower NaCl concentrations, especially at 5 mm NaCl in Fig. 2, there was a pronounced reduction in the growth rate of N13 compared to that of the wild-type strain at ph 10.5, and that difference was completely abolished upon transformation of N13 with pnhac. Lower NaCl concentrations did not support growth well enough to allow an assessment of whether the relative difference between the growth rates of the two strains would be further magnified as the NaCl concentration was reduced even more. Interestingly, the dependence of B. firmus OF4811M on the NaCl concentration of the medium was just as evident at ph 7.5 as at ph 10.5, consistent with the importance of an Na cycle even at near-neutral phs in alkaliphiles that use Na /symport as a major transport mode. In fact, higher concentrations of NaCl were required at ph 7.5 than at ph 10.5 to support optimal (albeit still lower) growth rates. Moreover, at the lowest NaCl concentration used in most experiments at ph 7.5, the N13 strain showed almost no growth, a defect that was complemented by pnhac and was not observed at high NaCl concentrations (Fig. 2). The growth phenotype suggested that NhaC is a relatively high-affinity antiporter whose functions were confined to low NaCl concentrations, with another antiporter(s) taking over at high NaCl concentrations. In order to assay this directly, the efflux of 22 Na was measured upon energization of right-sideout membrane vesicles by imposition of a valinomycin-mediated potassium diffusion potential at ph 7.5. Three vesicle preparations were compared: those from the wild-type strain transformed with a control plasmid and those from N13 that was transformed either with a control plasmid or with pnhac. In several separate experiments, the vesicles were preloaded with either 2 or 25 mm 22 Na in a potassium-containing buffer, before treatment with valinomycin and dilution to establish the diffusion potential. As shown in Fig. 3A, N13 exhibited a significant deficiency in Na efflux from the vesicles preloaded with 2 mm radioactive cation relative to comparable preparations from complemented N13 or the wild-type strain. By contrast, the three vesicle preparations exhibited comparable diffusion potential-mediated efflux when they had been preloaded with 25 mm NaCl (Fig. 3B). That energization of NhaC could be achieved by imposition of a diffusion potential was consistent with the electrogenicity of the antiport, i.e., that its mechanistic coupling ratio involves H inward Na outward. This property would be important in an antiporter functioning to achieve a ph gradient, acid in, during respiratory extrusion of protons. It was of interest to determine whether the growth defect of N13 relative to the wild-type strain at low Na concentrations at ph 10.5 might indeed reflect a defect in ph homeostasis in part, rather than reflecting only a compromise in efflux capacity for Na. After a sudden shift in the external ph from 8.5 to 10.6, the capacity of the three transformants (the wild-type strain transformed with control plasmid pyh56 and N13 transformed either with the control plasmid or with pnhac) to maintain a cytoplasmic ph below the new external ph was assayed. The cells were grown either at ph 10.5 or at ph 7.5 before being equilibrated at ph 8.5 in preparation for the shift. As shown in Table 1, ph 10.5-grown cells of all three transformants exhibited completely comparable capacities for ph homeostasis when the buffer contained 50 mm Na. That capacity diminished for each strain as the Na concentration was reduced to 5 mm and diminished even more when the Na concentration was reduced to 1 mm. Most strikingly though, the falloff in capacity for ph homeostasis at the two lower Na concentrations was especially pronounced in uncomplemented N13 and least pronounced, at 1 mm Na, in N13 complemented with nhac on a multicopy plasmid. When the same determinations were done with ph 7.5-grown cells, at 50 and 5 mm Na only, all three strains were at least a little less capable of acidifying the cytoplasm relative to the new ph upon an alkaline shift, and the relative defect of uncomplemented N13 was significantly accentuated. Interestingly, the wild-type capacity for ph ho- FIG. 3. Diffusion potential-energized efflux of 22 Na from right-side-out membrane vesicles of N13/pYH56, N13/pNhaC, or B. firmus OF4811/pYH56 at two different concentrations of intravesicular Na. Membrane vesicles of B. firmus OF4811M/pYH56 (circles), N13/pYH56 (squares), or N13/pNhaC (triangles) were loaded with either 2 (A) or 25 (B) mm 22 NaCl and pretreated with valinomycin as described under Materials and Methods. They were diluted in the presence of valinomycin into either 25 mm sodium phosphate, ph 7.5, to generate a diffusion potential (solid symbols), or 25 mm potassium phosphate, ph 7.5, as the unenergized control (open symbols). The amount of 22 Na remaining in the vesicles was determined at intervals as described in Materials and Methods.

6 3856 ITO ET AL. J. BACTERIOL. TABLE 1. Internal phs of B. firmus OF4811M/pYH56, N13/pYH56, and N13/pNhaC after a shift of the outside ph from 8.5 to 10.6 ph of medium and strain a Cytoplasmic ph at Na concn b of: 50 mm 5 mm 1 mm 10.5 B. firmus OF4811M/pYH N13/pYH N13/pNhaC B. firmus OF4811M/pYH ND c N13/pYH ND N13/pNhaC ND a Strains were grown in buffer at either ph 10.5 or ph 7.5 as described in Materials and Methods. b Na concentration indicates the amount of Na present in the ph shift buffer. c ND, not determined. meostasis when prior growth was at ph 7.5 closely resembled that of ph 10.5-grown cells of uncomplemented N13. DISCUSSION The current findings support the earlier conclusion that the nhac gene of alkaliphilic B. firmus OF4 encodes an Na /H antiporter. Enhancement of Na /H antiporter activity was observed upon complementation of E. coli EP432. Although this enhancement was modest, it was similar to that observed with other Bacillus antiporters, including one that has been purified and rigorously shown to possess Na /H antiporter activity (8, 15). Most importantly, studies with B. firmus OF4811M itself, in comparison with the nhac strain, N13, provided strong support for NhaC as an Na /H antiporter, i.e., in the correlated reductions in N13, relative to the wildtype strain, in electrogenic 22 Na extrusion and in ph homeostasis upon an upward ph shift. In the presence of other antiporters, e.g., in wild-type B. subtilis and as part of a complement of antiporters in the wild-type alkaliphile, nhac apparently plays a role over a wide range of ph values. Its most apparent role was observed at ph 7.5 in media containing low Na concentrations, where a pronounced defect in growth was observed. Although ph homeostasis is not a major challenge at ph 7.5, the alkaliphile must nonetheless extrude Na that enters in symport with solutes, including the growth substrate malate in the media used here. At ph 10.5, NhaC also plays an apparent role at low Na concentrations; at the alkaline ph, it contributes to ph homeostasis and may also contribute to the capacity of the cell to lower the cytoplasmic Na concentration optimally. However, NhaC is not required for alkaliphily and is not the antiporter with a dominant role in ph homeostasis during growth in Na -replete medium at highly alkaline phs. In fact, even when expressed from a multicopy plasmid, nhac was not able to complement the alkali-sensitive phenotype of a neutrophile that had a deletion in its own dominant Na /H antiporter, the teta(l) deletion strain JC112 of B. subtilis, let alone the nonalkaliphilic DA78 mutant of B. firmus OF4811M. Perhaps an insufficiently high V max and/or H /Na ratio underlies the failure of NhaC to support growth of B. subtilis or B. firmus OF4811M mutants that lack other major antiport systems. In E. coli, where the demands on the ph homeostatic mechanisms are more modest than in the alkaliphile, one of the antiporters that may contribute to this function, NhaA, is electrogenic and exhibits an enormous increase in V max at high phs in the presence of Na that is a combined effect of induction of synthesis and activation (21, 38). NhaB, which can take over in a nhaa mutant strain, is also electrogenic, albeit with a lower coupling ratio than NhaA, and has an even higher V max than activated NhaA (34, 38). A major Na (K )/H antiporter of B. subtilis, the TetA(L) protein, similarly is electrogenic and has a high K m and V max for Na (8, 15). Whereas the current data are most consistent with NhaC being electrogenic, its K m for Na is likely to be relatively low. Both the growth and transport phenotypes of the N13 strain were observed only at low concentrations of Na. Hypothetically, the other product of the nhac operon, NhaS, is a candidate for a cation sensor or an Na -sensitive gene regulator that has a membrane-associated state. This suggestion arises because of both its likely capacity to bind DNA and its sequence similarity to a largely hydrophobic region of the Na /K ATPases that may be involved in monovalent cation binding. However, even fully induced and activated NhaC may not catalyze antiport activity with a high enough V max to establish and maintain the requisite 2-pH-unit ph, acid in, that is found in alkaliphiles such as B. firmus OF4811M (22, 37). The results clearly suggest the presence of more than one additional antiporter in B. firmus OF4811M. At both low and high phs, there must be one or more antiporters that account for the almost complete lack of a phenotype in N13 at high Na concentrations, i.e., a lower-affinity, probably higher-v max antiporter. In cells grown at high phs, there is apparently also at least one additional antiporter functioning in the lower Na concentration range together with NhaC, whereas in ph 7.5- grown cells NhaC is the major contributor to antiport activity at modest Na concentrations. In E. coli, NhaB has been suggested to play a role when the cell confronts a challenge of Na influx under conditions in which NhaA is deleted or is not fully induced or activated by high phs (31, 33). It has also been suggested that ChaA is one of the antiporters playing an auxiliary role, at least in Na exclusion, at elevated phs (30); ChaA and/or K /H antiporters (35) may provide support for ph homeostasis as well. It was notable that the Na requirement for growth of B. firmus OF4811M was higher at ph 7.5 than at ph 10.5; this could reflect ph-dependent properties of the Na /solute symport complement, but none of these has yet been studied at the molecular level. As studies continue to develop a comprehensive model of ph homeostasis in alkaliphilic B. firmus OF4, the specific roles of each antiporter and the properties of the Na uptake systems will all be of importance. ACKNOWLEDGMENTS This work was supported by research grants GM28454 from the National Institutes of Health and MCB from the National Science Foundation. REFERENCES 1. Altschul, S. F., W. Gish, W. Miller, E. F. Myers, and D. J. Lipman Basic alignment search tool. J. Mol. Biol. 215: Ambudkar, S. V., G. W. Zlotnick, and B. P. Rosen Calcium efflux from Escherichia coli: evidence for two systems. J. Biol. Chem. 259: Aono, R., M. Ito, and K. Horikoshi Instability of the protoplast membrane of facultative alkaliphilic Bacillus sp. C-125 at alkaline ph values below the ph optimum for growth. Biochem. J. 285: Biswas, I., A. Gruss, S. D. Ehrlich, and M. Maguin High-efficiency gene inactivation and replacement system for gram-positive bacteria. J. Bacteriol. 175: Brey, R. N., and B. P. Rosen Properties of Escherichia coli mutants altered in calcium/proton antiport activity. J. Bacteriol. 139: Chang, S., and S. N. Cohen High frequency transformation of Bacillus subtilis protoplasts by plasmid DNA. Mol. Gen. Genet. 168:

7 VOL. 179, 1997 nhac OF ALKALIPHILIC B. FIRMUS OF Cheng, J., A. A. Guffanti, W. Wang, T. A. Krulwich, and D. H. Bechhofer Chromosomal teta(l) gene of Bacillus subtilis: regulation of expression and physiology of a teta(l) deletion strain. J. Bacteriol. 178: Cheng, J., D. B. Hicks, and T. A. Krulwich The purified Bacillus subtilis tetracycline efflux protein TetA(L) reconstitutes both tetracyclinecobalt/h and Na (K )/H exchange. Proc. Natl. Acad. Sci. USA 93: Clejan, S., A. A. Guffanti, M. A. Cohen, and T. A. Krulwich Mutation of Bacillus firmus OF4 to duramycin resistance results in substantial replacement of membrane lipid phosphatidylethanolamine by its plasmalogen form. J. Bacteriol. 171: Devereux, J., P. Haeberli, and O. Smithies A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res. 12: Fleischmann, R. D., M. D. Adams, O. White, R. A. Clayton, E. F. Kirkness, A. R. Kerlavage, C. J. Bult, J. F. Tomb, B. A. Dougherty, J. M. Merrick, K. McKenney, G. Sutton, W. FitzHugh, C. A. Fields, J. D. Gocayne, J. D. Scott, R. Shirley, L. I. Liu, A. Glodek, J. M. Kelley, J. F. Weidman, C. A. Philips, T. Spriggs, E. Hedblom, M. D. Cotton, T. R. Utterback, M. C. Hanna, D. T. Nguyen, D. M. Saudek, R. C. Brandon, L. D. Fine, J. L. Frichman, J. L. Fuhrmann, N. S. Geoghagen, C. L. Gnehm, L. A. McDonald, K. V. Small, C. M. Fraser, H. O. Smith, and J. C. Venter Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269: Fletcher, C. F., C. M. Lutz, T. N. O Sullivan, J. D. Shaughnessy, Jr., R. Hawkes, W. N. Frankel, N. G. Copeland, and N. A. Jenkins Absence epilepsy in tottering mutant mice is associated with calcium channel defects. Cell 87: Goldberg, E. B., T. Arbel, J. Chen, R. Karpel, G. A. Mackey, S. Schuldiner, and E. Padan Characterization of a Na /H antiporter gene of Escherichia coli. Proc. Natl. Acad. Sci. USA 84: Guffanti, A. A., and T. A. Krulwich Oxidative phosphorylation by ADP Pi-loaded membrane vesicles of alkaliphilic Bacillus firmus OF4. J. Biol. Chem. 269: Guffanti, A. A., and T. A. Krulwich Tetracycline/H antiport and Na /H antiport catalyzed by the Bacillus subtilis TetA(L) transporter expressed in Escherichia coli. J. Bacteriol. 177: Hamamoto, T., M. Hashimoto, M. Hino, M. Kitada, Y. Seto, T. Kudo, and K. Horikoshi Characterization of a gene responsible for the Na /H antiporter system of alkalophilic Bacillus species strain C-125. Mol. Microbiol. 14: Ivey, D. M., A. A. Guffanti, J. S. Bossewitch, E. Padan, and T. A. Krulwich Molecular cloning and sequencing of a gene from alkaliphilic Bacillus firmus OF4 that functionally complements an Escherichia coli strain carrying a deletion in the nhaa Na /H antiporter gene. J. Biol. Chem. 266: Ivey, D. M., A. A. Guffanti, and T. A. Krulwich The Na cycle in alkaliphilic Bacillus species, p In E. P. Bakker (ed.), Alkali cation transport systems in prokaryotes. CRC Press, Inc., Boca Raton, Fla. 19. Ivey, D. M., A. A. Guffanti, J. Zemsky, E. Pinner, R. Karpel, S. Schuldiner, E. Padan, and T. A. Krulwich Cloning and characterization of a putative Ca 2 /H antiporter gene from Escherichia coli upon functional complementation of Na /H antiporter-deficient strains by the overexpressed gene. J. Biol. Chem. 268: Kaback, H. R Bacterial membranes. Methods Enzymol. 22: Karpel, R., T. Alon, G. Glaser, S. Schuldiner, and E. Padan Expression of a sodium proton antiporter (NhaA) in Escherichia coli is induced by Na and Li ions. J. Biol. Chem. 266: Krulwich, T. A Alkaliphiles: basic molecular problems of ph tolerance and bioenergetics. Mol. Microbiol. 15: Krulwich, T. A., J. Cheng, and A. A. Guffanti The role of monovalent cation/proton antiporters in Na -resistance and ph homeostasis in Bacillus: an alkaliphile vs a neutralophile. J. Exp. Biol. 196: Krulwich, T. A., J. G. Federbush, and A. A. Guffanti Presence of a non-metabolizable solute that is translocated with Na enhances Na -dependent ph homeostasis in alkalophilic Bacillus. J. Biol. Chem. 260: Krulwich, T. A., M. Ito, R. Gilmour, M. G. Sturr, A. A. Guffanti, and D. B. Hicks Energetic problems of extremely alkaliphilic aerobes. Biochim. Biophys. Acta 1275: Lebovitz, R. M., K. Takeyasu, and D. M. Fambrough Molecular characterization and expression of the (Na K)-ATPase alpha-subunit in Drosophila melanogaster. EMBO J. 8: Lowry, O. H., N. J. Rosebrough, A. L. Farr, and R. J. Randall Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193: Nakamura, T., Y. Domano, E. Itaya, K. Tsukamoto, T. Tsuchiya, and T. Unemoto Cloning and sequencing of an Na /H antiporter gene from the marine bacterium Vibrio alginolyticus. Biochim. Biophys. Acta 1190: Nakamura, T., H. Enomoto, and T. Unemoto Cloning and sequencing of the nhab gene encoding an Na /H antiporter from Vibrio alginolyticus. Biochim. Biophys. Acta 1275: Ohyama, T., K. Igarashi, and H. Kobayashi Physiological role of the chaa gene in sodium and calcium circulations at a high ph in Escherichia coli. J. Bacteriol. 176: Padan, E., and S. Schuldiner Na /H antiporters, molecular devices that couple the Na and H circulation in cells. J. Bioenerg. Biomembr. 25: Pinner, E., O. Carmel, H. Bercovier, S. Sela, E. Padan, and S. Schuldiner Cloning, sequencing and expression of the nhaa and nhar genes from Salmonella enteritidis. Arch. Microbiol. 157: Pinner, E., Y. Kotler, E. Padan, and S. Schuldiner Physiological role of NhaB, a specific Na /H antiporter in Escherichia coli. J. Biol. Chem. 268: Pinner, E., E. Padan, and S. Schuldiner Kinetic properties of NhaB, ana /H antiporter from Escherichia coli. J. Biol. Chem. 269: Plack, R. H., Jr., and B. P. Rosen Cation/proton antiport systems in Escherichia coli: absence of potassium/proton antiporter activity in a phsensitive mutant. J. Biol. Chem. 255: Sturr, M. G., A. J. Ablooglu, and T. A. Krulwich. A Bacillus subtilis locus encoding several gene products affecting transport of cations. Gene, in press. 37. Sturr, M. G., A. A. Guffanti, and T. A. Krulwich Growth and bioenergetics of alkaliphilic Bacillus firmus OF4 in continuous culture at high ph. J. Bacteriol. 176: Taglicht, D., E. Padan, and S. Schuldiner Overproduction and purification of a functional Na /H antiporter coded by nhaa(ant) from Escherichia coli. J. Biol. Chem. 266: Tani, K., T. Watanabe, H. Matsuda, M. Nasu, and M. Kondo Cloning and sequencing of the spore germination gene of Bacillus megaterium ATCC 12872: similarities to the NaH-antiporter gene of Enterococcus hirae. Microbiol. Immunol. 40: Thanlen, P., T. Tsuchiya, and E. B. Goldberg Characterization and mapping of a major Na /H antiporter gene of Escherichia coli. J. Bacteriol. 173: Waser, M., D. Hess-Bienz, K. Davies, and M. Solioz Cloning and disruption of a putative Na/H-antiporter gene of Enterococcus hirae. J. Biol. Chem. 267: Wilson, K Preparation of genomic DNA from bacteria, unit 2.4. In F. M. Ausubel, R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl (ed.), Current protocols in molecular biology, vol. 1. John Wiley & Sons, NewYork, N.Y.