Defect in the Formation of 70S Ribosomes Caused by Lack of Ribosomal Protein L34 Can Be Suppressed by Magnesium

Abstract

To elucidate the biological functions of the ribosomal protein L34, which is encoded by the rpmH gene, the rpmH deletion mutant of Bacillus subtilis and two suppressor mutants were characterized. Although the ΔrpmH mutant exhibited a severe slow-growth phenotype, additional mutations in the yhdP or mgtE gene restored the growth rate of the ΔrpmH strain. Either the disruption of yhdP, which is thought to be involved in the efflux of Mg²⁺, or overexpression of mgtE, which plays a major role in the import of Mg²⁺, could suppress defects in both the formation of the 70S ribosome and growth caused by the absence of L34. Interestingly, the Mg²⁺ content was lower in the ΔrpmH cells than in the wild type, and the Mg²⁺ content in the ΔrpmH cells was restored by either the disruption of yhdP or overexpression of mgtE. In vitro experiments on subunit association demonstrated that 50S subunits that lacked L34 could form 70S ribosomes only at a high concentration of Mg²⁺. These results showed that L34 is required for efficient 70S ribosome formation and that L34 function can be restored partially by Mg²⁺. In addition, the Mg²⁺ content was consistently lower in mutants that contained significantly reduced amounts of the 70S ribosome, such as the ΔrplA (L1) and ΔrplW (L23) strains and mutant strains with a reduced number of copies of the rrn operon. Thus, the results indicated that the cellular Mg²⁺ content is influenced by the amount of 70S ribosomes.

INTRODUCTION

The eubacterial ribosome (70S), which plays a central role in protein synthesis, is composed of a small (30S) subunit and a large (50S) subunit. The small subunit is comprised of the 16S rRNA and more than 20 proteins, whereas the large subunit is comprised of the 23S and 5S rRNAs and more than 30 proteins (1, 2). The molecular mechanisms of translation have been elucidated in detail by the convergence of various approaches, including crystal structure analysis (3–8). The individual functions of several ribosomal proteins have also been elucidated by biochemical and genetic analyses, including reconstitution and mutational analysis. For example, ribosomal protein L2 plays important roles in the assembly of the ribosomal subunits, binding of the tRNA to the A and P sites, peptidyltransferase activity, and formation of the peptide bond (9–13). However, in general, disruption of the genes that encode the ribosomal proteins has been avoided as a means of identifying protein function, because these genes, which are highly conserved in bacteria, have been considered essential for cell proliferation (14). Nonetheless, recently, it was found that 22 of the 54 Escherichia coli genes for ribosomal proteins could be deleted on an individual basis (15, 16). We have also shown that out of the 57 ribosomal-protein-encoding genes that have been annotated in the Gram-positive bacterium Bacillus subtilis, at least 22 genes are not individually essential for cell proliferation (17). The rpmH gene encoding ribosomal protein L34, which is a component of the large subunit, is one of the genes that can be disrupted. Lack of L34 causes a severe defect in the formation of the 70S ribosome and a reduced growth rate (17). The rpmH genes of many bacteria were cloned over 2 decades ago by researchers working on replication initiation, because the gene is located near the replication origin in many bacteria (18–22). It is known that L34 assembles into the 50S subunit at a late stage of its formation (23), and it has also been suggested that L34 participates in the regulation of the polyamine biosynthetic pathway (24). However, the functions of L34 within the ribosome, such as its involvement in protein synthesis, large subunit assembly, and 70S formation, remain unclear.

Magnesium is important for the maintenance of ribosome structure and for translation. The in vitro association of the 30S and 50S ribosomal subunits to form intact 70S ribosomes depends strongly on the concentration of Mg²⁺ ions within the cell (25–27). Mg²⁺ is also required for both stabilization of the secondary structure of rRNA and binding of the ribosomal proteins to the rRNA (28–30). Furthermore, Mg²⁺ stabilizes the codon-anticodon interaction in the A site and influences the binding of ribosome recycling factor (RRF) to the ribosome (31–34). In addition to its involvement in the ribosome, Mg²⁺ has a crucial role in numerous biological processes and cellular functions, such as the activation and catalytic reactions of hundreds of enzymes, utilization of ATP, and maintenance of genomic stability (35, 36).

In the present study, we isolated two suppressor mutants, the ΔrpmH srh1 and ΔrpmH srh2 mutants, from the ΔrpmH mutant strain and characterized these mutants to elucidate the biological functions of the ribosomal protein L34 in B. subtilis. The studies on the suppression mechanism of the srh1 and srh2 suppressor mutations indicated that the function of L34, which is involved not only in normal assembly of the 50S subunit but also in efficient formation of the 70S ribosome, can be complemented partially by Mg²⁺ ions. Furthermore, we found that the content of Mg²⁺ in the cell is influenced by the amount of 70S ribosomes. On the basis of our findings, we also discuss the importance of Mg²⁺ ions in the ribosome.

MATERIALS AND METHODS

Media and culture conditions.

LB medium (37) and LB agar were used. The culture conditions and media for preparing competent cells have been described previously (38). When required, 0.5 μg ml⁻¹ erythromycin, 5 μg ml⁻¹ kanamycin, 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG), and 100 mM or 200 mM MgSO₄ were added to the medium. Ampicillin (50 μg ml⁻¹) was used to select E. coli strains. To observe growth under Mg²⁺-deficient conditions, cells were grown in minimal medium with Casamino Acids (MCA medium) that contained 1 mM MgSO₄ at 37°C with shaking to early exponential phase (optical density at 600 nm [OD₆₀₀] of ∼0.2) and harvested. The harvested cells were washed with MCA medium without MgSO₄, reinoculated into MCA medium that contained 1 mM or 10 μM MgSO₄, and grown at 37°C with shaking. MCA medium is based on Spizizen's minimal glucose medium (39). The composition of MCA medium is as follows: 6 g of KH₂PO₄, 14 g of K₂HPO₄, 2 g of (NH₄)₂SO₄, 1 g of trisodium citrate · 2H₂O per liter of distilled and deionized water, 0.5% glucose, and 0.1% Casamino Acids (Difco). It should be noted that 0.1% Casamino Acids is equivalent to 6 μM Mg².

Bacterial strains and plasmids.

All the B. subtilis strains used in the study were isogenic with B. subtilis strain 168 trpC2. The ΔrpmH::cat strain, which was constructed by replacing the open reading frame of the rpmH gene with a cat gene that lacked any promoter or Rho-independent terminator sequence, was described previously (17). The plasmid pDGrpmH, which carries an rpmH gene controlled by an IPTG-inducible Pspac promoter, was constructed as follows. The rpmH gene and the terminator region of the rpmE gene were amplified using the primers rpmHF and rpmHR and primers rpmEterF and rpmEterR, respectively (see Table S1 in the supplemental material for the nucleotide sequences of all the primers used in this study). These fragments were then used simultaneously as the template for PCR amplification with the primers rpmHF and rpmEterR. The resulting fragment was digested with HindIII and SalI and cloned into pDG148 (40). The plasmid pDGyhdP, which carries the yhdP gene controlled by an IPTG-inducible Pspac promoter, was constructed as follows. The yhdP gene and the terminator region of the rpmE gene were amplified using the primers yhdPF and yhdPR and primers rpmEterFYP and rpmEterR2, respectively. These fragments were then used simultaneously as the template for PCR amplification with the primers yhdPF and rpmEterR2. The resulting fragments were digested with SalI and SphI and cloned into pDG148. Similarly, to construct the plasmid pDGmgtE, which carries the mgtE gene without a riboswitch (41) controlled by the Pspac promoter, the mgtE gene and the terminator region of the rpmE gene were amplified using the primers mgtEF and mgtER and primers rpmEterFME and rpmEterR2, respectively. These fragments were then used simultaneously as the template for PCR amplification with primers yhdPF and rpmEterR2. The resulting fragments were digested with SalI and SphI and cloned into pDG148. To replace the yhdP gene with an erythromycin resistance gene (erm), oligonucleotide primers were used to amplify the upstream (yhdPuF and yhdPuR) and downstream (yhdPdF and yhdPdR) regions of the yhdP gene. Next, the erm gene of pTC3 (formally named pAE41 [42]) was amplified by PCR using the primers ermF and ermR. Another PCR amplification, in which all three above-mentioned amplified fragments were added as the DNA template, was performed using the primers yhdPuF and yhdPdR. The resulting fragment was used to transform strain 168 (trpC2), giving rise to the erythromycin-resistant transformant (ΔyhdP::erm trpC2). Chromosomal DNA extracted from the ΔrpmH::cat mutant was used to transform the ΔyhdP::erm strain, and the transformants were selected on the basis of a chloramphenicol-resistant phenotype (ΔrpmH::cat ΔyhdP::erm trpC2). The strain whose genome encoded MgtE tagged with Strep-tag II (WSHPQFEK) was constructed as follows. The region upstream of the aprE gene was amplified with the primers aprEUF and aprEURmg and the region downstream was amplified with the primers aprEDF and aprEDR using chromosomal DNA from B. subtilis strain 168 as the template. The mgtE gene, including the promoter and riboswitch, was amplified with primers mgtEstF and mgtEstR, which also encoded the Strep-tag II, using chromosomal DNA from strain 168 as the template. The kanamycin resistance gene from pDG148 was amplified by PCR using the primers kanF and kanR. The four fragments obtained (i.e., the upstream and downstream regions of aprE [fragments A and B, respectively] and the mgtE region and the kanamycin resistance gene [fragments C and D, respectively]) were used as the templates for two rounds of PCR amplification. First, fragments A and C were used simultaneously as the templates for amplification with primers aprEUF and mgtEstR, and fragments B and D were used simultaneously as the templates for amplification with primers kanF and aprEDR. Next, the two fragments obtained were used simultaneously as the templates for amplification with primers aprEUF and aprEDR. Finally, the resulting product was used to transform strain 168, and kanamycin-resistant transformants were selected (aprE::mgtE-Streptag kan trpC2). Chromosomal DNA extracted from the ΔrpmH::cat strain was used to transform the aprE::mgtE-Streptag kan mutant, and the transformants were selected on the basis of a chloramphenicol-resistant phenotype (ΔrpmH::cat aprE::mgtE-Streptag kan trpC2). Strains whose genomes encoded YhdP tagged with the Strep-tag II in the wild-type background (aprE::yhdP-Streptag kan trpC2) or in the ΔrpmH background (ΔrpmH::cat aprE::yhdP-Streptag kan trpC2) were constructed in a similar manner. The region upstream of the aprE gene was amplified with the primers aprEUF and aprEURyh, and the region downstream was amplified with the primers aprEDF and aprEDR. The yhdP gene, including the promoter, was amplified with the primers yhdPstF and yhdPstR. The kanamycin resistance gene was amplified by PCR using the primers kanF and kanR.

Isolation and identification of suppressor mutations in the ΔrpmH mutant.

To obtain spontaneous suppressor mutations in the ΔrpmH mutant that restored growth, an overnight culture of the ΔrpmH strain was diluted and plated onto LB solid medium and then incubated overnight at 37°C. Two suppressor mutants that formed larger colonies than did the parental strain were designated the ΔrpmH srh1 and ΔrpmH srh2 mutants. The ΔrpmH srh1 colonies were larger than those of the ΔrpmH parental strain but smaller than those of the ΔrpmH srh2 strain. The ΔrpmH srh2 colonies were almost the same size as those of the wild type. The frequency of ΔrpmH srh1-type suppressor mutants was 0.16%, whereas the frequency of ΔrpmH srh2-type mutants was 0.006%. To identify the suppressor mutations, sequencing libraries were prepared using the NEBNext DNA Sample Prep reagent set 1 (New England BioLabs) according to the manufacturer's protocols. Briefly, 3 μg of genomic DNA was fragmented to an average length of 300 bp using the Covaris S2 system (Covaris, Woburn, MA). The fragmented DNA was repaired, a single A nucleotide was ligated to the 3′ end, Illumina Index PE adapters (Illumina, San Diego, CA) were ligated to the fragments, and fragments between 350 and 450 bp were selected with a 1.5% Pippin Prep gel (Sage Science, Beverly, MA). The size-selected sample was amplified by PCR for 12 cycles with the primers InPE1.0, InPE2.0, and an index primer, which contained a unique-index tag for each individual sample. The final product was validated by using an Agilent Bioanalyzer 2100 (Agilent, Santa Clara, CA). The pooled libraries were sequenced on an Illumina Genome Analyzer IIx in accordance with the manufacturer's instructions, to generate 100-bp paired-end reads and 6-bp index tags. Details of the output data are given in Table S2 in the supplemental material. Sequence reads from each sample were mapped onto the B. subtilis 168 reference genome (accession no. NC_000964.3), using the BWA software package (ver. 0.5.1) (44) with default parameters. Possible variations (single nucleotide polymorphism [SNP]/indel) were listed using the SAMtools software (ver. 0.1.9) (43). To identify authentic variations, we applied the following filtering criteria to the candidates: (i) coverage at the nonreference allele of at least 5, (ii) indels had to meet an SNP quality threshold of 50 and substitutions had to meet an SNP quality threshold of 20 (SAMtools assigns SNP quality, which is the Phred-scaled probability that the consensus is identical to the reference), (iii) variations had to meet a mapping quality of 30 (SAMtools assigns mapping quality, which is the Phred-scaled probability that the read alignment is wrong), and (iv) percentage of sequence reads showing the variant allele had to exceed 55%. The filtered lists of variations were annotated by COVA (comparison of variants and functional annotation) (http://sourceforge.net/projects/cova/). To identify structural variations, we used the BreakDancer software package (ver. 0.0.1r81) (45) with default parameters. Regions of interest were inspected on the Integrative Genomics Viewer (IGV) (46).

Measurement of Mg²⁺ content in a cell.

Bacillus subtilis cells were grown in LB medium at 37°C with shaking to exponential phase (OD₆₀₀, ∼0.4) and harvested. Simultaneously, viable cells were counted by plating the culture on LB agar plates and then incubating them at 37°C for 24 h. The cells were resuspended in lysis buffer (10 mM Tris-HCl, pH 8.0, 1 mg ml⁻¹ lysozyme) and disrupted sufficiently by sonication, and then the pH of the crude extract was adjusted to approximately 3.0 with hydrochloric acid in order to denature the proteins. After vortexing for 10 min followed by incubation at room temperature for 10 min, the crude extract was centrifuged for 10 min at 10,000 × g. The amount of Mg²⁺ in the supernatant was measured with a Metallo assay kit for magnesium (Metallogenics) in accordance with the manufacturer's instructions. The Mg²⁺ content per cell was calculated by dividing the amount of Mg²⁺ in the crude extract by the number of viable cells. The concentration of Mg²⁺ was calculated by assuming that a B. subtilis cell is a cylinder (radius, 0.4 μm; length, 4 μm).

Sucrose density gradient sedimentation analysis.

Bacillus subtilis cells were grown in LB medium at 37°C with shaking to exponential phase (OD₆₀₀, ∼0.4) and harvested. The sucrose density gradient sedimentation analysis was performed as described previously (47). Briefly, the cells were disrupted by passage through a French pressure cell and cell debris was removed by centrifugation. Aliquots of extract were layered onto 10 to 40% sucrose density gradients and centrifuged at 4°C for 17.5 h at 65,000 × g (Hitachi P40ST rotor). Samples were collected with a piston gradient fractionator (BioComP), and absorbance profiles were monitored at 254 nm using a Bio-Mini UV monitor (Atto, Japan). When normalizing the applied volume by the total absorbance at 260 nm, 10 A₂₆₀ units of crude extract per tube were used (see Fig. 4). When normalizing the applied volume by the amount of chromosomal DNA (see Fig. 6B), chromosomal DNA was extracted from 1 ml of each culture that was used for density gradient sedimentation analysis and quantified by spectrophotometric assay. The OD₆₀₀ of each sample was measured before (designated Db) and after (designated Da) passage through the French press, and the efficiency of cell lysis (designated E) for each sample was calculated from the following equation: E = 1 − Da/Db. The volume of wild-type extract used for the assay was determined by measuring the A₂₆₀ of the extract obtained from the wild-type cells and calculating the volume of wild-type extract that contained 10 A₂₆₀ units (designated A_wt). The volume of extract for each mutant to be used for the assay (designated A_mut) was determined from the following equation: A_mut = A_wt × (C_wt × E_wt)/(C_mut × E_mut), where C_wt and C_mut are the total amounts of chromosomal DNA extracted from the wild-type and mutant cells, respectively, and E_wt and E_mut are the efficiency of cell lysis for the wild-type and mutant cultures, respectively.

FIG 4.

Defect in 70S ribosome formation in the absence of L34 and its suppression by the disruption of yhdP and overexpression of mgtE. Crude cell extracts were sedimented through a 10 to 40% sucrose gradient as described in Materials and Methods. The 30S, 50S, and 70S peaks are indicated in each individual profile. The term “/pDGmgtE” indicates the overexpression of mgtE in the mutant cells. wt, wild type; Abs, absorbance.

FIG 6.

Reduction of the Mg²⁺ content in mutants that contained reduced amounts of 70S ribosomes. (A) The Mg²⁺ content per cell during exponential phase was measured as described in Materials and Methods. The means of three independent experiments are shown. Error bars indicate standard deviations. RIK539 harboring only the rrnA operon within the genome is indicated as 1 rrn. RIK1754 harboring only the rrnA and rrnI operons is indicated as 2 rrn. RIK1437 harboring only the rrnA, rrnI, and rrnO operons is indicated as 3 rrn. RIK1463 harboring only the rrnA, rrnI, rrnO, and rrnE operons is indicated as 4 rrn. These mutants harboring reduced numbers of the rrn operon have been described by Yano et al. (62). (B) Crude cell extracts, whose applied volumes were normalized as described in Materials and Methods, were sedimented through a 10 to 40% sucrose gradient. The 30S, 50S, and 70S peaks are indicated in each individual profile. wt, wild type; Abs, absorbance.

Analysis of 70S ribosome formation.

30S and 50S ribosomal subunits were prepared from wild-type and ΔrpmH mutant cells grown in LB at 37°C. The preparation of small and large subunits and association experiments were performed as described previously with a slight modification (48). Briefly, the S30 fraction prepared from wild-type cells cultivated to exponential phase (OD₆₀₀, ∼0.4) was centrifuged at 65,000 × g at 4°C for 17 h through a 10 to 40% sucrose gradient in buffer B (20 mM Tris-HCl, pH 7.6, 100 mM CH₃COONH₄, 0.1 mM dithiothreitol [DTT], and 2 mM phenylmethylsulfonyl fluoride [PMSF]) that contained 1 mM (CH₃COO)₂Mg in a Hitachi P28SA rotor to separate the 30S and 50S subunits. In the case of the S30 fraction prepared from ΔrpmH mutant cells cultivated to exponential phase (OD₆₀₀, ∼0.4), buffer B that contained 10 mM (CH₃COO)₂Mg was used but otherwise the centrifugation was carried out under the same conditions. The subunits were diluted (1:2) with buffer B that contained 10 mM (CH₃COO)₂Mg, sedimented at 65,000 × g for 21 h at 4°C in a Hitachi P40ST rotor, and suspended in buffer B that contained 10 mM (CH₃COO)₂Mg. The purified small subunits (A₂₆₀ = 1) and large subunits (A₂₆₀ = 2.5) were mixed in buffer B that contained an appropriate concentration of (CH₃COO)₂Mg and incubated for 15 min at 37°C. The mixture was centrifuged at 55,000 × g for 17.5 h at 4°C through a 10 to 40% sucrose gradient in buffer B that contained an appropriate concentration of (CH₃COO)₂Mg in a Hitachi P40ST rotor. Absorbance profiles were monitored at 254 nm using a piston gradient fractionator (Biocomp) and a Bio-Mini UV monitor (Atto).

Preparation of crude ribosomes and RFHR 2-D gel electrophoresis.

Cells were grown in LB medium at 37°C and harvested in exponential phase (OD₆₀₀, ∼0.4). Crude ribosomes were obtained as described previously (49). The radical free and highly reducing (RFHR) two-dimensional (2-D) gel electrophoresis (50) was performed essentially in accordance with the published procedures (49).

Western blot analysis.

Aliquots (30 μg of protein) of crude extracts that had been prepared from cells were loaded onto a sodium dodecyl sulfate-polyacrylamide gel (12%) and transferred to a polyvinylidene difluoride (PVDF) membrane (Millipore). Strep-Tactin alkaline phosphatase (AP) conjugate (IBA), which can detect Strep-tag II fusion proteins, was used at a dilution of 1:4,000. Immunodetection procedures were carried out in accordance with the manufacturer's instructions.

Nucleotide sequence accession number.

The raw sequence reads used in this study are available at the DDBJ Sequence Read Archive (DRA) under accession no. DRA002237.

RESULTS

Isolation of suppressor mutants from ΔrpmH and identification of the corresponding mutations.

In the ΔrpmH::cat mutant, the growth rate was significantly lower than that in the wild type, which was mainly due to a defect in 70S ribosome formation followed by a reduction of translation activity in the cell (17) (Fig. 1; see also Fig. 4). To investigate the possibility that insertion of the cat gene into the rpmH gene, which is located near the replication origin, affects the growth rate by inhibiting the initiation of DNA replication, genetic complementation tests were performed. We expected the expression level of rpmH to be very high, because rpmH is located near the replication origin and encodes a ribosomal protein that is expressed abundantly in cells. Consequently, we introduced a multicopy plasmid that contained the rpmH gene under the control of an IPTG-inducible Pspac promoter into the ΔrpmH mutant to ensure a sufficiently high level of expression of rpmH in the complementation test. As expected, the induction of rpmH expression restored the growth rate of the ΔrpmH mutant to that of the wild type (Fig. 1). This result demonstrated that the severe slow-growth phenotype observed in the ΔrpmH mutant could be attributed solely to the lack of ribosomal protein L34.

FIG 1.

Genetic complementation of the ΔrpmH mutation. Cells were grown in LB at 37°C, and the optical density at 600 nm was measured. wt, wild type.

Two suppressor mutants were isolated spontaneously from the ΔrpmH mutant strain and designated ΔrpmH srh1 (suppressor of rpmH) and ΔrpmH srh2. The slow growth caused by lack of L34 was moderately corrected in the ΔrpmH srh1 strain, whereas the ΔrpmH srh2 mutant showed almost the same growth rate as that of the wild type (Fig. 2A). To identify the suppressor mutations, differences in sequence (SNP/indel mutations) between the ΔrpmH mutant, the ΔrpmH srh1 or ΔrpmH srh2 strain, and the B. subtilis 168 reference genome (accession no. NC_000964.3) were detected by whole-genome sequencing. When detected sequence differences in the suppressor mutants were compared with the sequence of the parental strain (ΔrpmH), it was apparent that the yhdP gene in the ΔrpmH srh1 strain contained a 175-bp duplication, whereas the ΔrpmH srh2 mutant harbored two nucleotide substitutions that converted Asp343 to Ala in the YhdP protein and Ala266 to Val in the MgtE protein (formerly called YkoK). It should be noted that no other changes in the genome were detected in the suppressor mutants. Given that the frequency of ΔrpmH srh2-type mutant colonies was much lower than that of ΔrpmH srh1-type mutant colonies (0.006% versus 0.016%), it is probable that ΔrpmH srh2-type mutant colonies arose from ΔrpmH srh1-type mutant strains through a second spontaneous mutation. YhdP shows homology with CorB and CorC of Salmonella enterica serovar Typhimurium (24% and 27% identity, respectively), which have been predicted to be efflux pumps for Mg²⁺ (51). Detailed studies on MgtE, including crystal structure analyses in Thermus thermophilus, have demonstrated that MgtE is an Mg²⁺ transporter that plays a central role in Mg²⁺ homeostasis (52–54). Recently, it has also been reported that MgtE in B. subtilis, which shows 33% sequence identity with MgtE of T. thermophilus, provides the primary route of magnesium import, although YfjQ, a homologue of the magnesium/cobalt transporter CorA, is also able to import Mg²⁺ (74).

FIG 2.

Suppression of the reduced growth rate observed in the ΔrpmH mutant. Cells were grown in LB at 37°C, and the optical density at 600 nm was measured. wt, wild type.

Disruption of yhdP and overexpression of mgtE restored the reduction in the cellular Mg²⁺ content caused by lack of L34.

To confirm that the mutations detected in the yhdP and mgtE genes did indeed suppress the defects observed in the ΔrpmH mutant, the yhdP and mgtE genes were each integrated into a multicopy plasmid separately, and the resulting plasmids were introduced into the ΔrpmH mutant. When these genes were induced in the absence of L34, overexpression of YhdP exacerbated the defect in the growth rate, whereas overexpression of MgtE partially restored the growth rate (Fig. 2B). This observation prompted us to disrupt yhdP in the ΔrpmH strain. As expected, elimination of YhdP in the ΔrpmH mutant partially restored the growth rate (Fig. 2B). Furthermore, when yhdP was disrupted and mgtE was overexpressed simultaneously in the ΔrpmH strain, the growth rate was almost the same as that of the wild type (Fig. 2B). It should be noted that neither overexpression of yhdP or mgtE nor disruption of yhdP affected the growth of wild-type cells (see Fig. S1 in the supplemental material). In combination with the observations that overexpression of MgtE, a Mg²⁺ transporter, and disruption of YhdP, a homologue of the presumed Mg²⁺ efflux pumps CorB and CorC, restored the growth rate of the ΔrpmH mutant, these data strongly suggested that increasing the cellular Mg²⁺ concentration suppresses the defects caused by the lack of L34. To test this hypothesis, the Mg²⁺ content in the mutants and wild-type cells was determined using xylidyl blue I, which forms a chelate complex with Mg²⁺. The Mg²⁺ ions that were chelated in enzymes and ribosomes were probably also detected by this method, because the cells were disrupted sufficiently by sonication and proteins were denatured by acid treatment, as described in Materials and Methods. Figure 3 shows the amount of Mg²⁺ per cell in the wild type and in each mutant. The concentration of Mg²⁺ in a wild-type cell was estimated to be 73 mM by assuming that a B. subtilis cell is a cylinder (radius, 0.4 μm; length, 4 μm). This result agrees with the previous observation that the concentration of Mg²⁺ ions, including Mg²⁺ ions chelated in proteins and nucleic acids, in an E. coli cell is 90 to 110 mM (55). Surprisingly, the Mg²⁺ content in the ΔrpmH cells was significantly lower than that in the wild-type cells. The reduction in Mg²⁺ concentration that was observed in the absence of L34 could be abrogated significantly, although not completely, by the srh1 suppressor mutation or by the disruption of yhdP or overexpression of mgtE (Fig. 3). When yhdP was disrupted and mgtE was overexpressed simultaneously in the ΔrpmH mutant, the Mg²⁺ content was almost the same as that of the wild type. Sufficient restoration of the Mg²⁺ content was also observed in the ΔrpmH srh2 mutant (Fig. 3).

FIG 3.

Observed reduction of Mg²⁺ content in the ΔrpmH cells and its restoration by the suppressor mutations as well as by disruption of yhdP and overexpression of mgtE. The Mg²⁺ contents per cell in exponential phase, which were measured as described in Materials and Methods, are shown. The term “/pDGmgtE” indicates the overexpression of mgtE in the mutant cells. The means of three independent experiments are shown. Error bars indicate standard deviations.

Restoration of the Mg²⁺ concentration suppressed the defect in 70S ribosome formation caused by lack of L34.

The elimination of ribosomal protein L34 caused a remarkable reduction in the efficiency of 70S ribosome formation, accompanied by the abnormal accumulation of both 30S and 50S subunits (Fig. 4). It is known that the association of 30S and 50S ribosomal subunits requires Mg²⁺ ions (25–27) and that Mg²⁺ stabilizes the secondary structure of rRNA and the binding of the ribosomal proteins to the rRNA (28–30). These results raise the possibility that restoration of the Mg²⁺ concentration by the disruption of yhdP and overexpression of mgtE suppresses the defect in 70S ribosome formation that is observed in the ΔrpmH mutant. To test this possibility, the ribosome profiles in the ΔrpmH mutants after restoration of the cellular Mg²⁺ content by the disruption of yhdP and/or overexpression of mgtE were examined. Overexpression of MgtE or disruption of YhdP in the ΔrpmH mutant partially suppressed the defect in 70S ribosome formation caused by the lack of L34. In contrast, when the mgtE gene was overexpressed and the yhdP gene was deleted simultaneously in the ΔrpmH mutant, 70S ribosome formation occurred almost as efficiently as in the wild type (Fig. 4). To confirm that the restoration of 70S ribosome formation that was observed in vivo could be attributed to increases in the concentration of Mg²⁺, in vitro experiments on 70S ribosome formation using ribosomal subunits prepared from the ΔrpmH mutant were performed at different Mg²⁺ concentrations (Fig. 5). When the Mg²⁺ concentration in the gradient buffer was 1 mM, 70S ribosomes could not be formed even when the subunits were prepared from wild-type B. subtilis. In the presence of 10 mM Mg²⁺, 30S and 50S subunits prepared from the wild type could form 70S ribosomes, whereas those from the ΔrpmH mutant could not. However, in gradient buffer that contained 100 mM Mg²⁺, a small peak of 70S ribosomes was observed even when the subunits were prepared from the ΔrpmH mutant. These results indicate that 50S subunits that lack L34 can interact with 30S subunits and form 70S ribosomes only at high concentrations of Mg²⁺. Thus, it is probable that increased levels of cellular Mg²⁺ suppress the defect in the formation of 70S ribosomes and the accompanying severe slow-growth phenotype that is caused by lack of ribosomal protein L34.

FIG 5.

50S subunits lacking L34 form 70S ribosomes in vitro at high concentrations of Mg²⁺. 30S and 50S subunits were prepared from the ΔrpmH mutant and wild type, and association experiments were performed at different Mg²⁺ concentrations. The 30S, 50S, and 70S peaks are indicated in each individual profile. wt, wild type; Abs, absorbance.

Correlation between the amount of 70S ribosomes and the content of Mg²⁺ in the cell.

Although we assumed that lack of L34 affected the expression of yhdP and mgtE because decreased levels of Mg²⁺ were observed in the ΔrpmH mutant cells, the transcriptional activities of these genes, as well as the amounts of these proteins in the cell, did not differ significantly between the wild type and the ΔrpmH mutant (see Table S3 and Fig. S2 in the supplemental material). Consequently, we expected that addition of Mg²⁺ to the medium would suppress the defects observed in the absence of L34. However, the growth rate of the ΔrpmH mutant was not restored even if Mg²⁺ was added to the medium at a final concentration of 200 mM (see Fig. S3). Thus, an increase in the environmental concentration of Mg²⁺ could not restore the reduced Mg²⁺ content that resulted from a lack of L34. The content of free Mg²⁺ ions in the cell is maintained at an essentially invariant level by tight regulation of uptake and efflux (56, 57). It has been demonstrated that the cytosolic cystathionine β-synthase (CBS) domain of MgtE in T. thermophilus binds specifically to Mg²⁺ and strictly controls the homeostasis of Mg²⁺ in the cell through a conformational change in the transmembrane domain (52–54). Moreover, the expression of mgtE is controlled by a Mg²⁺-responsive riboswitch, which is located in the 5′ leader region of the mgtE gene in B. subtilis (41). If the concentration of free Mg²⁺ ions in ΔrpmH cells is also tightly controlled by the regulation of mgtE through the Mg²⁺ riboswitch and by the activity of MgtE, it can be assumed that the reduction in Mg²⁺ content that occurs in the absence of L34 is due to a deficiency in the capacity of the cell to accumulate Mg²⁺. It is known that the number of ribosomes per cell can vary between 3,000 and 70,000 depending on the growth rate (58). Given the strong evidence that free Mg²⁺ is regulated tightly to between 1 and 5 mM (59, 60), and assuming that the ribosome-bound Mg²⁺ represents a substantial fraction of the total Mg²⁺, then a variation in the number of 70S ribosomes will have a significant effect on the total Mg²⁺ concentration, which is normally around 110 mM. To test this hypothesis, we determined the Mg²⁺ content in the ΔrplA (L1) and ΔrplW (L23) mutants, in which the efficiency of 70S ribosome formation is reduced significantly, and found that the lack of ribosomal protein L1 or L23 also decreased the Mg²⁺ content in the cells, as seen in the ΔrpmH mutant (Fig. 6A). Indeed, the numbers of 70S ribosomes and polysomes in cells lacking one of these ribosomal proteins were decreased markedly compared with those in wild-type cells when the applied volumes of crude extracts used for the experiment were normalized to the amount of chromosomal DNA as described in Materials and Methods (Fig. 6B). The areas of the peaks that corresponded to the 70S ribosomes and polysomes were 47%, 42%, and 32% of the wild-type area in the ΔrplA (L1), ΔrplW (L23), and ΔrpmH (L34) mutants, respectively. In addition, the total amount of 30S subunits plus 50S subunits plus 70S ribosomes, including polysomes, was decreased slightly in these mutants. When the 50S and 30S subunits were included with the 70S ribosomes and polysomes, the total areas of the peaks were 89%, 77%, and 80% of the wild-type area in the ΔrplA, ΔrplW, and ΔrpmH mutants, respectively. Previously, we have constructed mutants that harbor one to nine copies of the rrn operon in their genome (the wild type contains 10 copies) and confirmed that the number of ribosomes is reduced in these mutants, especially in those that harbor only one or two rrn operons (61, 62). In the present study, we determined the cellular Mg²⁺ content in these mutants and found that the Mg²⁺ content decreased as the number of rrn operons decreased (Fig. 6A). The decrease in the number of rrn operons correlated with the reduction in the amount of ribosomes as well as the reduction in the growth rate (61, 62). These results suggest that the Mg²⁺ content in the cell is influenced by the amount of ribosomes present, especially 70S ribosomes and polysomes.

To investigate the effect on cell proliferation of the decrease in 70S ribosomes and concomitant reduction in Mg²⁺ content that was caused by the lack of L34, growth of the ΔrpmH mutant was observed under Mg²⁺-deficient conditions (Fig. 7). In the MCA medium to which 1 mM MgSO₄ had been added, both the wild type and the ΔrpmH mutant grew continuously, whereas cell proliferation was limited in the MCA medium to which only 10 μM Mg²⁺ had been added. It should be noted that the standard MCA medium contained 6 μM Mg²⁺. Under the Mg²⁺-deficient conditions, a 5.6-fold increase in the OD₆₀₀ value of the wild-type culture was observed whereas the OD₆₀₀ value of the ΔrpmH culture increased only 1.9-fold. These results suggest that the initial amount of 70S ribosomes and Mg²⁺ in the cell was an important factor in limiting cell proliferation under Mg²⁺-deficient conditions.

FIG 7.

Effect of the lack of L34 on cell proliferation under Mg²⁺-deficient conditions. Precultured cells were grown at 37°C in minimal medium that contained 1 mM Mg²⁺ or 10 μM Mg²⁺, and the optical density at 600 nm was measured (see details in Materials and Methods). wt, wild type.

DISCUSSION

The rpmH gene encodes the ribosomal protein L34. Although rpmH genes have been cloned from many bacteria owing to their genomic location close to the replication origin (18–22), the function of L34 within the ribosome is not fully understood. In the present study, we characterized the ΔrpmH mutant of B. subtilis and two suppressor mutants and demonstrated that ribosomal protein L34 is essential for efficient formation of the 70S ribosome. We also showed that its function can be complemented partially by Mg²⁺. In addition, we obtained results that suggest that the number of 70S ribosomes in a cell influences the cellular Mg²⁺ content.

The defects in the formation of the 70S ribosome and the associated slow-growth phenotype that were observed in the absence of L34 were suppressed by disruption of the yhdP gene and overexpression of the mgtE gene, which increased the Mg²⁺ content in the cell. Therefore, we predict that the detected insertion and missense mutations in yhdP inhibit Mg²⁺ efflux, whereas the missense mutation in mgtE promotes Mg²⁺ import. MgtE, a major importer of Mg²⁺ in B. subtilis, has been studied in detail, including crystal structure analyses, in T. thermophilus. It has been suggested that binding of Mg²⁺ to the cytosolic CBS domain of the MgtE dimer alters the orientation of the plug helices, which comprise long helices that connect the CBS and transmembrane domains, and limits the import of Mg²⁺ into the cell by controlling the gating of the ion-conducting pore in the transmembrane domains (52–54). The predicted structure of B. subtilis MgtE suggests that Ala266, which is mutated to Val in the ΔrpmH srh2 mutant, is located in the connecting region between the plug helix and transmembrane domain (see Fig. S4 in the supplemental material). Therefore, it is likely that the A226V mutation affects the change in orientation of the plug helix and associated closing of the transmembrane domains that occur upon Mg²⁺ binding and facilitate the transport of Mg²⁺ into the cell. It is difficult to infer the effect that the mutations in YhdP have on the conformation of the protein because no crystal structures of YhdP homologues are available. However, the mutations in the yhdP gene that were found in both the ΔrpmH srh1 and ΔrpmH srh2 suppressor mutants must destroy the function of YhdP because they suppressed the defects caused by lack of L34, similarly to disruption of yhdP, which restored the Mg²⁺ content in the ΔrpmH cell. These results, together with the observation that overexpression of yhdP exacerbated the low growth rate in the ΔrpmH mutant, strongly suggest that YhdP is an efflux pump for Mg²⁺ in B. subtilis.

We found that L34 is essential for efficient interaction between the 30S and 50S subunits, even though the location of L34 within the ribosome is far from the region of interaction of the two subunits (8). In the 50S fraction from the ΔrpmH mutant, the amount of ribosomal protein L16 was reduced significantly in addition to the lack of L34 (see Fig. S5A in the supplemental material). We expect that the lack of L34 reduces the binding affinity of L16 for the 50S subunit, probably due to a conformational change in the 50S subunit. In the ribosome, L16 is located near the region of interaction of the two subunits and is inserted between two stems of the 23S rRNA, helix 38 and helix 89 (8, 63). Given that helix 38 is known to form an intersubunit bridge in the 70S structure (8, 63), lack of L16 probably causes a defect in the formation of the 70S ribosome. In fact, it has been reported that incorporation of L16 induces a strong conformational change in 50S particles and influences the association of the ribosomal subunits (64, 65). Similarly, we have reported recently that the H142L mutation in ribosomal protein L2 not only reduces the efficiency of formation of the 70S ribosome but also leads to a deficiency of L16 in the 50S subunit (48). Therefore, it seems that the defect in formation of the 70S ribosome in the ΔrpmH mutant is probably due to conformational changes in the 50S subunit that are caused by lack of L34 followed by a reduction in the binding affinity of L16.

A key question is how the restoration of Mg²⁺ content suppressed the defect in 70S ribosome formation caused by lack of L34. One probable reason is stabilization of the 50S subunit. Mg²⁺ stabilizes the binding of ribosomal proteins to the rRNA and the secondary structure of rRNA in the ribosome (28–30). Moreover, it has been demonstrated that Mg²⁺ concentration controls the dynamics of the ribosome exquisitely through altering ribosome stability and flexibility (66). Therefore, it seems likely that the restoration of Mg²⁺ content in the cell enables the normal rRNA structure to form and L16 to bind to the 50S subunit, which improves the conformational stability of 50S subunits that lack L34 and in turn restores formation of the 70S ribosome. In fact, the ribosome fractions that were prepared from the ΔrpmH mutant after restoration of the Mg²⁺ content by disruption of YhdP and overexpression of MgtE contained considerable amounts of L16 (see Fig. S5B in the supplemental material). Another possible reason for why Mg²⁺ improves the efficiency of 70S ribosome formation in the absence of L34 is increased stabilization of the intersubunit bridges. It is known that the in vitro association of the 30S and 50S ribosomal subunits depends strongly on the Mg²⁺ concentration (25–27) and that at least 5 of the 12 subunit bridges that exist between different ribosomal components are stabilized by Mg²⁺ (8, 67). Therefore, in the ΔrpmH mutant, restoration of the Mg²⁺ content probably improved the efficiency of 70S ribosome formation by stabilizing both the conformation of the 50S subunit and the intersubunit bridges (Fig. 8). It should be noted that the decrease in growth rate of the ΔrplW mutant, in which the efficiency of 70S ribosome formation is decreased significantly, was partially suppressed by the overexpression of MgtE, probably due to the restoration of 70S ribosome formation (see Fig. S6A in the supplemental material). In contrast, the growth rate of the ΔrplA mutant was not restored by overexpression of MgtE (see Fig. S6B in the supplemental material). The fact that the growth rate was not restored even when 70S formation was restored by an increase in Mg²⁺ can be explained by the fact that ribosomal protein L1, which is localized to the stalk region near the E site (8, 68), plays a critical role in protein synthesis in terms of the translocation of the newly deacylated tRNA from the P to the E site (69). At a high concentration of Mg²⁺, 50S subunits that lacked L34 could form 70S ribosomes even in the in vitro association experiments (Fig. 5). The concentration of free Mg²⁺ in E. coli and Salmonella enterica, not including Mg²⁺ ions chelated in proteins and nucleic acids, has been estimated experimentally to be between 1 and 5 mM (59, 60). Given that the Mg²⁺ concentration that allowed formation of the 70S peak from subunits prepared from the ΔrpmH mutant under in vitro conditions was extremely high (100 mM) compared with the concentration of free Mg²⁺ in the cell, the 50S subunit presumably interacted with the 30S subunit through intersubunit bridges that do not depend on L34 or L16 in vitro. However, these 70S ribosomes, whose conformation can be maintained only under limited conditions, may be unstable. 70S ribosomes were not formed sufficiently in vitro, even at the extremely high (100 mM) concentration of Mg²⁺, because ribosomal protein L16 was not supplied to ribosome subunits prepared from the ΔrpmH mutant. However, the restoration of 70S ribosome formation that was observed when the mgtE gene was overexpressed in the ΔrpmH mutant with the yhdP deletion was probably due to elevation of the free Mg²⁺ concentration in the cell.

FIG 8.

Mechanism of suppression in mutant lacking L34 through disruption of yhdP and overexpression of mgtE. (A) Defect in formation of 70S ribosome followed by reduction of Mg²⁺ content in the absence of L34. (B) Restoration of Mg²⁺ content and of 70S ribosome formation by disruption of yhdP and overexpression of mgtE in the ΔrpmH mutant. See the text for details.

In the ΔrpmH mutant, a reduction in the cellular Mg²⁺ content was also observed (Fig. 3). The probable reasons for this reduction are (i) a decrease in the amount of 70S ribosomes, (ii) a decrease in the amount of protein and RNA other than ribosomes that can chelate Mg²⁺, and (iii) a reduction in the free Mg²⁺ content in the cell. The ribosome has the highest capacity of any macromolecule in the cell for binding Mg²⁺, with more than 170 Mg²⁺ ions per complex (70). These 170 Mg²⁺ ions bound to the 70S ribosome were detected with particularly high B factors and thus did not include Mg²⁺ ions that were associated loosely or only through outer-sphere interactions. Therefore, the 70S ribosome probably contains many more Mg²⁺ ions than reported previously. Given that E. coli growing in exponential phase contains approximately 70,000 ribosomes per cell (58), a large amount of Mg²⁺ is probably sequestered in the 70S ribosomes. Indeed, we found that, in addition to the ΔrpmH mutant, the Mg²⁺ content was also reduced in other mutants that contained significantly decreased amounts of 70S ribosomes, such as disruptants of ribosomal-protein genes and strains with a reduced number of rrn operons (Fig. 6). The difference in the areas of the peaks between the ΔrpmH mutant and the wild type was greater when only the 70S ribosomes and polysomes were considered (the mutant was 32% of the wild-type value) than when the 30S and 50S subunits were included as well (the mutant was 80% of the wild-type value). Therefore, it is likely that the decrease in cellular Mg²⁺ content observed in the mutant analyzed in the present study is caused by a reduction in the level of 70S ribosomes and polysomes rather than a reduction in the total amount of available 30S and 50S subunits. Given that Mg²⁺ is also involved in the stabilization of both tRNA binding and intersubunit bridges in translating 70S ribosomes (32–34, 67, 71), the total amount of Mg²⁺ that is contained in 70S ribosomes and polysomes is greater than that contained in the equivalent amount of individual ribosome subunits. In addition, the binding capacity of 50S subunits that lack L34 and L16 for Mg²⁺ is probably lower than that of intact 50S subunits. Consequently, we predict that the decrease in the amount of Mg²⁺ ions is much greater than that due simply to the reduction in number of ribosome subunits that is caused by the lack of L34. Moreover, it has been reported previously that the amount of Mg²⁺ in the ribosome fraction is influenced by the concentration of Mg²⁺ in the culture medium (72) and that ribosome flexibility is controlled by Mg²⁺ concentration (66). Thus, the amount of Mg²⁺ in the 70S ribosome might vary. The second probable reason why the absence of L34 caused a reduction in the Mg²⁺ content is a decrease in the amount of protein and RNA in general that can chelate Mg²⁺. A decrease in the amount of 70S ribosomes and the consequent reduction in translation activity in a cell causes a reduction in the growth rate and in the total amount of protein and RNA, including that which can chelate Mg²⁺. The reduction in these chelating agents might decrease the cellular Mg²⁺ content. The final probable reason is a decrease in the free Mg²⁺ content in the cell. However, as described previously, the concentration of free Mg²⁺ is controlled tightly (59, 60). Given that the level of expression of MgtE, a major transporter of Mg²⁺ in B. subtilis, was not changed significantly in the ΔrpmH mutant compared with the wild type (see Table S3 and Fig. S2 in the supplemental material), the concentration of free Mg²⁺ in ΔrpmH cells might not decrease drastically. However, the possibility remains that a defect occurs in an alternative route of Mg²⁺ transport, for example, via YfjQ. Although we cannot fully explain at present the mechanism by which the absence of L34 causes a reduction in the Mg²⁺ content, the restoration of Mg²⁺ content in the ΔrpmH mutant by the disruption of yhdP and overexpression of mgtE might be caused by an increase in the free Mg²⁺ concentration followed by a subsequent increase in the amount of 70S ribosomes, as illustrated in Fig. 8.

When sufficient cellular reserves of 70S ribosomes and Mg²⁺ are available, the cells can adapt to environmental changes such as a decrease in Mg²⁺ concentration to some extent. When the Mg²⁺ concentration in the culture medium was decreased, the wild-type cells could proliferate three times as much as the ΔrpmH mutant, which did not contain sufficient 70S ribosomes and Mg²⁺ (Fig. 7). Given that the synthesis of 70S ribosomes, which requires Mg²⁺, is probably difficult under Mg²⁺-deficient conditions, the amount of 70S ribosomes and Mg²⁺ in the cell is reduced further by cell division under such conditions. In fact, the depletion of ribosomes is observed under Mg²⁺-deficient conditions in E. coli (73). Therefore, ribosomal proteins such as L34 play important roles in ensuring not only efficient cell proliferation but also cell survival in an environment with a limited concentration of Mg²⁺.

The finding that the function of ribosomal protein L34, which is required for efficient 70S ribosome formation, can be complemented partially by Mg²⁺ reemphasizes the importance of Mg²⁺ for the conformational stability of the ribosome. In addition, we obtained results that suggest that the amount of 70S ribosomes influences the Mg²⁺ content in the cell. Further investigations should reveal whether, under Mg²⁺-deficient conditions, the 70S ribosome, which binds a large number of Mg²⁺ ions, can donate Mg²⁺ to enzymes that require the ion for activation.