- Research article
- Open Access
Involvement of the YneS/YgiH and PlsX proteins in phospholipid biosynthesis in both Bacillus subtilis and Escherichia coli
BMC Microbiology volume 7, Article number: 69 (2007)
Phospholipid biosynthesis commences with the acylation of glycerol-3-phosphate (G3P) to form 1-acyl-G3P. This step is catalyzed by the PlsB protein in Escherichia coli. The gene encoding this protein has not been identified, however, in the majority of bacterial genome sequences, including that of Bacillus subtilis. Recently, a new two-step pathway catalyzed by PlsX and PlsY proteins for the initiation of phospholipid formation in Streptococcus pneumoniae has been reported.
In B. subtilis, 271 genes have been reported to be indispensable, when inactivated singly, for growth in LB medium. Among these, 11 genes encode proteins with unknown functions. As part of a genetic study to identify the functions of these genes, we show here that the B. subtilis ortholog of S. pneumoniae PlsY, YneS, is required for G3P acyltransferase activity, together with PlsX. The B. subtilis genome lacks plsB, and we show in vivo that the PlsX/Y pathway is indeed essential for the growth of bacteria lacking plsB. Interestingly, in addition to plsB, E. coli possesses plsX and the plsY ortholog, ygiH. We therefore explored the functional relationship between PlsB, PlsX and YgiH in E. coli, and found that plsB is essential for E. coli growth, indicating that PlsB plays an important role in 1-acyl-G3P synthesis in E. coli. We also found, however, that the simultaneous inactivation of plsX and ygiH was impossible, revealing important roles for PlsX and YgiH in E. coli growth.
Both plsX and yneS are essential for 1-acyl-G3P synthesis in B. subtilis, in agreement with recent reports on their biochemical functions. In E. coli, PlsB plays a principal role in 1-acyl-G3P synthesis and is also essential for bacterial growth. PlsX and YgiH also, however, play important roles in E. coli growth, possibly by regulating the intracellular concentration of acyl-ACP. These proteins are therefore important targets for development of new antibacterial agents.
Phospholipids are major components of the cell membrane. Glycerol-3-phosphate (G3P) forms the backbone of all phospholipid molecules . Phospholipid biosynthesis begins with two steps of G3P acylation, leading to the formation of phosphatidic acid (PA). In bacteria, PA is converted to CDP-diacylglycerol, the precursor of the three major phospholipids, phosphatidylethanolamine (PE), phosphatidylglycerol (PG), and cardiolipin (CA)  (Figure 1). Earlier biochemical and genetic analyses disclosed that G3P and 1-acylglycerol-phosphate (1-acyl-G3P) acyltransferase are encoded by the plsB and plsC genes, respectively, in Escherichia coli [3–5]. Interestingly, while plsC is universally conserved in eubacteria, plsB has been identified in only a limited number of species, mainly belonging to the gamma proteobacterial group. Another gene homologous to E. coli plsC, possibly encoding G3P acyltransferase, plsD, has been cloned from the Clostridium butyricum genome, based on its ability to complement plsB defects in E. coli. However, the role of plsD in C. butyricum cells is currently unclear .
E. coli PlsB utilizes acyl-ACP (Acyl Carrier Protein) and acyl-CoA as acyl donors to synthesize 1-acyl-G3P . Recently, Lu et al. [8, 9] have reported a new two-step pathway that utilizes a novel fatty acid intermediate for the initiation of phospholipid formation in Streptococcus pneumoniae. They demonstrated biochemically that PlsX produced a unique activated fatty acid by catalyzing the synthesis of fatty acyl-phosphate from acyl-ACP, and then showed that PlsY transferred the fatty acid moiety from acyl-phosphate (acyl-PO4) to G3P. The plsX gene is widely conserved in eubacteria and it had been suggested that this gene was involved in fatty acid and/or phospholipid synthesis in E. coli, although the exact role of the gene remained unknown . In B. subtilis, PlsX is essential for growth  and PlsX expression is controlled by FapR, a protein that regulates fatty acid and phospholipid biosynthesis genes . Although the plsY gene, encoding a membrane protein around 200 amino acids long, is also widely conserved in eubacteria [8, 9], the plsY gene function had been completely unknown.
In the gram-positive spore-forming bacterium,Bacillus subtilis, we showed that 271 genes, including plsC (yhdO) and plsX, when inactivated singly, were indispensable for growth in LB medium . Among these genes, 11 encoded proteins with unknown functions, and these proteins are therefore targets for development of new antibacterial agents. As part of a genetic study to identify the functions of these genes, here we show that the B. subtilis ortholog of S. pneumoniae PlsY, YneS, is required for G3P acyltransferase activity, together with PlsX. The B. subtilis genome lacks plsB, and our results demonstrate in vivo that the PlsX/Y pathway is essential for the growth of bacteria lacking plsB. Interestingly, in addition to plsB, E. coli possesses plsX and the plsY ortholog, ygiH. Thus, we explored the functional relationship between PlsB, PlsX and YgiH in E. coli, and found that, although PlsB plays the principal role in 1-acyl-G3P synthesis, PlsX and YgiH are also important for optimal E. coli growth.
Isolation of a B. subtilis yneS-ts mutant
To explore the functions of essential genes of unknown function, we adopted a strategy to isolate temperature-sensitive (ts) mutants of these genes and then to seek extragenic suppressors of the ts genes. To generate a ts mutation in yneS, we initially introduced a nonfunctional chloramphenicol-resistant gene (cat) downstream of yneS, and transformed cells with the yneS-functional cat fragment mutagenized by PCR in vitro. Chloramphenicol-resistant transformants were selected at 30°C, and colonies that were unable to grow above 42°C were analyzed. In this manner, we isolated a ts mutant of yneS (strain MY103) that had a one-base substitution altering Val190 to Glu in the C-terminal cytoplasmic region of the protein.
Mapping of the suppressor mutation site in the yneS-ts sup-1 strain
Next, we isolated a spontaneous extragenic suppressor mutant of the ts phenotype (strain MY105, yneS-ts sup-1, Figure 2) by cultivation at the restrictive temperature. Transformation of wild-type strain 168 cells with chromosomal DNA of MY105 led to the ts growth of more than 90% of chloramphenicol-resistant transformants, indicating that the sup-1 mutation was not linked to the yneS gene. MY105 cells lost sporulation ability, and the Spo- phenotype was maintained in MY107 cells (Pspac::yneS sup-1) in which the yneS-ts gene was replaced with wild-type yneS under control of the IPTG-dependent Pspac promoter. A DNA fragment (ca. 8 kb) that complemented the Spo- phenotype of MY107 in trans was isolated from a B. subtilis genomic DNA library in which an Mbo I partial digest was ligated with the phage vector, φCM, as described previously . The 8 kb fragment was cut into 2.5 kb and 5 kb fragments. The 5 kb fragment with the cat gene of the phage vector was cloned into pBR322, while the 2.5 kb fragment was ligated into pCA191 containing the cat gene . MY107 cells were transformed with these plasmids for integration into the chromosome via single crossovers. The results indicated that the 5 kb fragment allowed recovery of the ability to sporulate, while the 2.5 kb fragment did not. Sequencing of the 5 kb fragment revealed three genes. These were the 3'-terminal region of glpF, encoding the glycerol uptake facilitator, and full-length glpK and glpD, encoding glycerol kinase and glycerol-3-phosphate dehydrogenase, respectively. Because glpF and glpK constitute an operon, and because the cloned 8 kb fragment did not contain the promoter sequence , glpK was possibly not expressed from the fragment cloned in φCM. The glpD promoter between glpK and glpD has, however, been identified . These results suggested that the mutation(s) responsible for the Spo- phenotype mapped in the glpD gene. To confirm this theory, full-length wild type glpD (in pCA191glpD-FL), the 5'-terminal half of the gene (in pCA191glpD-N), and the 3'-terminal half (in pCA191glpD-C), were separately cloned into pCA191. The resulting constructs were integrated into the MY107 chromosome by single crossovers and the sporulation abilities of transformants were examined. The 5'-terminal region of glpD, and the full-length gene, conveyed the ability to sporulate, but the 3'-terminal glpD region did not. Sequencing of the 1.2 kb 5'-terminal half of glpD in MY103 (yneS-ts) revealed a single base (G) insertion between nucleotides C 492 and G 493, which resulted in inactivation of the gene. This result was consistent with the previous finding that glpD inactivation impairs sporulation ability .
We further confirmed that glpD inactivation suppressed the yneS-ts phenotype by insertion of a plasmid, pMutinNC , carrying an internal segment of the glpD gene, into the glpD gene of yneS-ts cells (strain MY109, yneS-ts, ΔglpD::pMutin NC, Figure 2).
Increases in intracellular G3P concentration suppress the yneS-ts phenotype
The intracellular G3P concentration is known to be elevated in glpD mutants . We speculated that increases in the G3P level might suppress the yneS-ts phenotype. Indeed, we found that supplementation of the growth medium (LB) with 0.1% (w/v) G3P, or glycerol, complemented the yneS-ts phenotype (Figure 2). Growth of MY105 (yneS-ts sup-1) and MY109 (yneS-ts, ΔglpD::pMutinNC) cells was impaired compared to glpD+ cells (wild type and yneS-ts cells), when glycerol or G3P was supplemented in the growth medium. This phenotype was also observed in MY108 (ΔglpD::pMutinNC) cells (data not shown). Given that accumulation of G3P is known to result in abnormal septation and inhibition of sporulation [18, 19], an increase in the intracellular concentration of G3P would explain the growth impairment of cells without glpD. It should be noted that G3P and glycerol did not complement the growth defect resulting from YneS depletion in the IPTG-dependent yneS mutant cells, strain YNESp  (data not shown). The YneS-ts mutant protein will retain a reduced activity at high temperature, although the activity is insufficient to support cell growth. Residual activity would be enhanced by increases in the intracellular concentrations of G3P or glycerol.
Phospholipid synthesis in B. subtilis plsC-ts, yneS-ts, and IPTG-dependent plsX mutants, under restrictive conditions
G3P is a substrate for G3P acyltransferase, and inhibition of G3P acyltransferase activity as a result of simultaneous mutations in plsB and plsX in E. coli led to G3P auxotrophy . Accordingly, we examined the involvement of YneS in 1-acyl-G3P synthesis by analyzing phospholipid synthesis in yneS-ts cells at the restrictive temperature. We additionally constructed mutations in genes possibly involved in phospholipid synthesis. These included a ts mutation in plsC, in which 7 C-terminal residues were deleted (strain MY112), and an IPTG-dependent plsX mutation in strain MY111.
To determine phospholipid biosynthesis rates, B. subtilis cells were pulse-labeled with 32 [Pi] for 5 min. Phospholipids were extracted using the method of Bligh and Dyer , and separated by two-dimensional thin layer chromatography . We detected PA, PG, PE, and CL in wild-type cells growing at either 30°C or 42°C (Figure 3A). When plsC-ts cells were labeled at 30°C, a pattern similar to that of wild-type cells was obtained. At the restrictive temperature, however, 1-acyl-G3P accumulation was observed, and phospholipid synthesis did not take place (Figure 3B). This is consistent with the known 1-acyl-G3P acyltransferase activity of PlsC. In contract, no phospholipids were detectable in yneS-ts cells following a shift to the restrictive temperature (Figure 3C). Our data indicate that YneS inactivation results in blockage of the first step in phospholipid synthesis in B. subtilis cells. Furthermore, PlsX depletion in IPTG-dependent mutant cells induces a similar change in phospholipid composition (Figure 3D). The results are consistent with the two-step mechanism of 1-acyl-G3P synthesis reported by Lu et al. , and show that the PlsX/Y pathway is essential for growth of B. subtilis lacking plsB.
Functional relationship between plsB, plsX and yneS/ygiH in E. coli
It is generally believed that E. coli plsB is essential for bacterial growth, although experimental evidence supporting this hypothesis has yet to be obtained, and transposon mutagenesis experiments suggested that plsX is dispensable in E. coli . The ortholog of S. pneumoniae plsY and B. subtilis yneS in E. coli, ygiH, is yet another dispensable gene . Accordingly, we determined the functional relationship between plsB, plsX, and ygiH, in E. coli.
We initially attempted to systematically inactivate E. coli genes either singly or in combination. For this purpose, we replaced the plsX and ygiH genes with a kanamycin-resistance gene (kan) using the method of Datsenko and Wanner . Inactivation of plsB was performed in the presence of plasmid-borne plsB. Phage P1 transduction frequencies of each allele into cells of various genetic backgrounds were measured to avoid possible effects of secondary mutations arising during gene inactivation (the kan cassette was removed, using FLP recombinase, in recipient cells for P1 transduction). The results, summarized in Table 1, show that plsB is essential for E. coli growth, while single deletions of either plsX or ygiH do not affect cell growth. These results suggested that PlsB plays a principal role in G3P acyltransferase activity in E. coli. Additionally, to examine whether plsB function might be complemented by overexpression of other genes, we introduced derivatives of plasmid pSTV28 or pSTV29 (12–15 copies per cell) harboring E. coli plsX and ygiH, and B. subtilis plsX and yneS into recipient strains for P1 transduction, without restoration of plsB function. Unexpectedly, although single deletions of plsX or ygiH did not affect cell growth, we found that the simultaneous inactivation of plsX and ygiH was impossible (Table 2), indicating that plsX and ygiH also have important roles in E. coli growth. Next, we expressed B. subtilis plsX and yneS genes in E. coli, and also the corresponding E. coli genes (as controls). The lethal phenotype of the plsX and ygiH double deletion was suppressed by B. subtilis plsX or yneS, indicating that these genes were able to complement the plsX and ygiH E. coli gene activities. In addition, a plsX and ygiH double deletion was possible when plsB was overexpressed from pSTV29plsB. To obtain further insights into the functional relationships between plsB, plsX and ygiH, we next introduced a plasmid harboring a mutated plsB gene (plsB26) into E. coli, and then attempted to inactivate plsB, plsX or ygiH (Table 3). The plsB26 gene has a point mutation that changes Gly1045 to Ala, reducing the affinity of PlsB26 for G3P . In the presence of plasmid-borne plsB26, genomic plsB could be inactivated, indicating that PlsB26 protein expressed from the plasmid-borne gene could support E. coli growth. The plasmid-borne gene could not, however, complement the lethality of the plsX and ygiH double deletion. Furthermore, while native plsB could be inactivated in the presence of pSVT29plsB26 in cells lacking functional ygiH, plsB inactivation became impossible in cells with inactivated plsX, indicating that PlsX activity becomes indispensable for the growth of cells when the plsB activity is supplied by a plasmid copy of plsB26.
The essential nature of plsB, and the lethality of the plsX and ygiH double deletion, were further confirmed using inducible forms of these genes. To this end, we placed the coding sequences of plsB, plsX and ygiH under the control of an IAA (3β-indoleacrylic acid)-inducible trp promoter (Pw), integrated the genes into the chromosome as described in Methods, and then deleted the native copies of either plsB alone, or both plsX and ygiH. As expected, strains MEC201 (Pw-plsB, ΔplsB), MEC306 (Pw-plsX, ΔplsX, ΔygiH) and MEC307 (Pw-ygiH, ΔplsX, ΔygiH) cells displayed IAA-dependent growth (Figure 4A and 4B).
The PlsB and PlsX enzymes utilize acyl-ACP as acyl-donor, and abnormal accumulation of acyl-ACP has been shown to result in growth impairment . To examine acyl-ACP accumulation in cells with inactivated plsX and ygiH genes, we introduced a derivative of pUC18, harboring the E. coli tesA gene encoding thioesterase I , into cells of strains MEC306 and MEC307. Interestingly, the growth defects of these strains in the absence of IAA were suppressed by the additional supply of thioesterase I (TesA hydrolyses acyl-ACP) (Figure 4C). We have not yet succeeded in demonstrating the expected TesA-dependent suppression of the inability to achieve plsX inactivation when plsB activity is supplied by a plasmid encoding plsB26. This is because it is difficult to construct an appropriate strain in which to examine the hypothesis.
Complementation of the PlsB activity by a combination of PlsX and YneS/YgiH
As we found that the combination of PlsX and YneS conferred the ability to synthesize 1-acyl-G3P upon B. subtilis, we expected that simultaneous overexpression of these two proteins would complement a plsB mutation in E. coli. To test this hypothesis, we introduced various combinations of plasmids harboring plsX and yneS/ygiH into E. coli cells. In these experiments, we employed derivatives of plasmid pMW118 (ca. 5 copies per cell), whose replication is compatible with that of pSTV28/29 derivatives, when more than one cloned gene was to be expressed. We successfully complemented the E. coli plsB mutation by expression of B. subtilis plsX and yneS (Table 4). The complementation of the E. coli PlsB activity by the B. subtilis PlsX and YneS combination was also demonstrated using an IAA-inducible plsB gene (Figure 4A). Other gene combinations, including those with E. coli genes, did not complement plsB. Plasmid pMW118ygiH complemented the growth defect of MEC306 and MEC307 cells in the absence of IAA (data not shown), indicating that functional YgiH is expressed from pMW118ygiH. Although we could not determine the expression levels of the various proteins due to the lack of appropriate antibodies, the specific activities of E. coli PlsX and YgiH may be lower than those of their B. subtilis counterparts.
Our genetic studies have shown that PlsX and YneS (now renamed PlsY) are essential for G3P acyltransferase activity in B. subtilis cells, and that the combination of the two B. subtilis proteins complements PlsB activity in E. coli cells. Our findings are consistent with the two-step mechanism of 1-acyl-G3P synthesis reported by Lu et al. , and we have shown that the PlsX/Y pathway is essential for the growth of B. subtilis lacking plsB. The combination of plsX and plsY is widely conserved in eubacteria, but most eubacteria lack plsB , strongly suggesting that plsX/Y will be essential genes in these bacteria.
A number of bacteria, belonging mainly to the gamma group of proteobacteria, possess both plsB and plsX/plsY, however. Another important finding reported here is that, in such cases, the relationship between the two 1-acyl-G3P synthesis pathways may be variable and complex. Although plsB has been reported to be dispensable in Pseudomonas aeruginosa , we found that the gene is indispensable in E. coli, indicating that PlsB is principally responsible for 1-acyl-G3P synthesis in E. coli. On the other hand, we found that simultaneous inactivation of both PlsX and PlsY is impossible, and that this lethality was complemented by overexpression of PlsB. As single deletions in either plsX or ygiH did not affect cell growth, the lethality of the plsX and ygiH double deletion cannot be because of inactivation of a minor pathway (using acyl-PO4) for 1-acyl-G3P synthesis.
Interestingly, PlsX activity becomes indispensable for growth of E. coli cells in which PlsB activity is supplied from a plasmid copy of plsB26. Growth, in minimal medium, of E. coli SJ22 cells harboring the plsB mutation on the chromosome is dependent on a supply of exogenous G3P. Depletion of G3P in the growth medium has been found to result in accumulation of acyl-ACP due to the blockage of 1-acyl-G3P synthesis, leading to a severe reduction in fatty acid biosynthesis because accumulated acyl-ACP causes feedback inhibition of enzymes in the biosynthetic pathway [27, 28]. G3P-acyltransferase activity expressed from a plasmid-borne copy of plsB26 may be less than that in wild type cells, and acyl-ACP may therefore accumulate in such cells. PlsX may be able to suppress the deleterious accumulation of acyl-ACP by converting acyl-ACP to acyl-PO4. PlsY may also be involved in control of the intracellular acyl-ACP concentration. E. coli PlsB has been shown to utilize both acyl-ACP and acyl-CoA as acyl donors in the synthesis of 1-acyl-G3P , but does not use these donors to make acyl-PO4 . On the other hand, S. pneumoniae PlsY has been shown to synthesize 1-acyl-G3P using both acyl-ACP and acyl-PO4. These results suggest that E. coli PlsY, in the absence of PlsX, may synthesize 1-acyl-G3P using acyl-ACP, and this activity would then contribute to consumption of acyl-ACP. It is therefore possible that both PlsX and PlsY contribute independently to the maintenance of an appropriate intracellular acyl-ACP concentration, and simultaneous inactivation of both proteins therefore results in deleterious accumulation of acyl-ACP. Overexpression of PlsB increases 1-acyl-G3P synthesis activity and may suppress the accumulation of acyl-ACP resulting from the plsX and ygiH double deletion. Suppression of growth defects due to the plsX and ygiH double deletion by overexpression of thioesterase I strongly supports our hypothesis.
Acyl-ACP has been shown to be not only a central cofactor for fatty acid and phospholipid synthesis, but also a regulatory molecule, coordinating the synthesis of these lipids with cell growth. Our results strongly suggest that PlsX, PlsY and PlsB form a complex network functioning to supply appropriate levels of lipid biosynthetic precursors, especially acyl-ACP. In turn, this controls the synthesis of 1-acyl-G3P in E. coli cells, resulting in appropriate levels of 1-acyl-G3P under various growth conditions. Further studies are needed to reveal the precise roles of PlsX, PlsY and PlsB.
Both plsX and yneS are essential for 1-acyl-G3P synthesis in B. subtilis cells lacking plsB, in agreement with recent reports on the biochemical functions of these genes. Both genes will be essential for bacterial growth lacking plsB. In E. coli, PlsB plays a principal role in 1-acyl-G3P synthesis and is also essential for bacterial growth. PlsX and YgiH also, however, play important roles in E. coli growth, probably by regulating the intracellular concentrations of acyl-ACP at appropriate levels. These proteins are therefore important targets for the development of new antibacterial reagents.
Materials and strains
The B. subtilis and E. coli strains used in this study are listed in Tables S1 [see Additional file 1] and S2 [see Additional file 2], respectively. Primers and plasmids are specified in Tables S3 [see Additional file 3] and S4 [see Additional file 4], respectively. B. subtilis cells were transformed as described previously . E. coli strains DH5α (Takara) and JM105 (Takara), were used throughout as cloning hosts.
Full-length sequences of E. coli plsB, plsX (EcplsX), and ygiH, and B. subtilis plsX (BsplsX) and yneS, were amplified and cloned into plasmids pSTV28 or pSTV29 (Takara, 12–15 copies per cell) as summarized in Table S3 [see Additional file 3]. A DNA fragment containing a mutated plsB sequence, plsB26, was PCR-amplified from genomic DNA of E. coli strain TL84 , and cloned into pSTV29. Inserts of pSTV29ygiH and pSTV29yneS were transferred into pMW118 (Nippon Gene, ca. 5 copies per cell), because the replication of pMW118 is compatible with that of pSTV28 and pSTV29.
The IAA (3β-indoleacrylic acid, Sigma)-inducible trp promoter fragment was PCR amplified from W3110 chromosomal DNA. The PCR product was digested with the restriction enzymes Mun I and Eco RI, and cloned into the Eco RI site of pMC1403 , generating pMC1403Pw. Full-length sequences of E. coli plsB, plsX, and ygiH were amplified and cloned into the Eco RI/Bam HI, Bam HI, or Eco RI/Bam HI sites of pMC1403-Pw, respectively, to obtain pMC1403Pw-plsB, pMC1403Pw-plsX and pMC1403Pw-ygiH.
The full-length sequence of E. coli tesA was amplified and cloned into plasmid pUC18S. Plasmid pUC18S was created by replacing the Aat II/Eam 1105I fragment (containing the ampicillin-resistance gene) of pUC18 (Takara) with an Aat II/Eam 1105I fragment (containing the spectinomycin-resistance gene) amplified, by PCR, from pAPNC213 .
B. subtilis strains were grown in Luria-Bertani (LB) medium, or DSM, supplemented with chloramphenicol (5 μg/ml), kanamycin (5 μg/ml), or erythromycin (0.5 μg/ml), as appropriate.E. coli strains were grown in LB or M9 minimal media, supplemented with ampicillin (50 μg/ml), chloramphenicol (5 μg/ml), spectinomycin (50 μg/ml), or kanamycin (25 μg/ml), as appropriate.
Isolation of the B. subtilis yneS-ts mutant
The yneS-ts mutant was isolated according to the procedure of F. Kawamura (personal communication). The B. subtilis strain MY101, in which the sequence 45–103 bp downstream of the termination codon of yneS (in the coding sequence of yneR, which is downstream of yneS) was replaced with a cat gene, was generated as follows. The regions upstream and downstream of the insertion site were amplified by PCR using a yneS-F1 and yneS-R1 primer pair, and a yneS-F2 and yneS-R2 primer pair, respectively. The cat gene of pCBB31  was amplified using primers cat-F and cat-R. The primer pairs yneS-R1 and cat-R, and cat-F and yneS-F2, contain overlapping sequences, and the three fragments obtained were ligated by recombinant PCR using yneS-F1 and yneS-R2 primers. B. subtilis strain 168 was transformed with the resulting fragment, and chloramphenicol-resistant transformants were selected (MY101, ΔyneR::cat+). Next, the pCH11 plasmid containing an inactive cat gene (a nonsense mutation is located in the gene), a kanamycin-resistance gene, and a ts origin of replication, was introduced into MY101. We selected for kanamycin resistance and chloramphenicol sensitivity in order to isolate cells with the inactive cat gene on the chromosome, which occurred when gene conversion from the plasmid copy was successful . The introduced plasmid was removed by cultivation at the temperature restrictive for pCH11 replication, to generate strain MY102 (ΔyneR::cat-). PCR mutagenesis of the yneS-cat fragment was performed using rTaq polymerase (Takara) with MY101 chromosomal DNA as template. The mutated yneS-cat fragment was used to transform strain MY102. Chloramphenicol-resistant transformants were selected at 30°C, and colonies that were unable to grow above 42°C were screened to isolate strain MY103 (yneS-ts).
Construction of the B. subtilis plsC-ts mutant
As the growth of cells in which gfp (encoding green fluorescent protein) was translationally fused to the C-terminus of PlsC became temperature-sensitive (data not shown), we predicted that the required plsC-ts mutant could be obtained by modification of the C-terminal region of the plsC gene. We therefore created a series of mutants with deletions of different numbers of amino acids from the C-terminus of PlsC and found that deletion of the 7 end residues resulted in a ts phenotype (strain MY112). To generate strain MY112, the 3' end of the plsC coding region (with the 7 codons deleted) was amplified, and a termination codon was added by PCR using the primers plsC-F1 and plsC-CΔ7-R1. The region downstream of plsC was amplified with the oligonucleotides plsC-F2 and plsC-R2. The two fragments were ligated to the 5' and 3' ends of the cat gene (amplified from pCBB31 using plsC-cat-F and plsC-cat-R primers) by recombinant PCR with overlapping sequences contained within the plsC-CΔ7-R1 and plsC-cat-F primer pair, and the plsC-F2 and plsC-cat-R primer pair. The resulting fragment was used to transform B. subtilis strain 168 with selection for chloramphenicol resistance.
Construction of an IPTG-dependent mutant of B. subtilis plsX
The plsX gene is located in the fatty acid biosynthesis operon composed of fapR, plsX, fabD, fabG, and acpA. To generate cells in which plsX expression was under the control of the IPTG-inducible Pspac promoter, we initially integrated Pspac-plsX into the aprE locus of the bacterial chromosome. The plsX gene was amplified by PCR using the pAP-plsX-F and pAP-plsX-R primers, and inserted between the Pspac promoter and the kanamycin-resistance gene (kan) on pAPNCK, that contains sequences upstream and downstream of aprE, flanking Pspac and kan. Plasmid pAPNCK was created by replacing the Hind III fragment (containing the spectinomycin-resistance gene) of pAPNC213  with the Hind III fragment (containing the kan gene) of pDG780 . The resulting plasmid was transformed into B. subtilis cells and selection of a double crossover using kanamycin resistance yielded strain MY110 (ΔaprE::Pspac-plsX-kan). Next, the native plsX gene was replaced with the cat gene, without the promoter and terminator sequences, to ensure the correct expression of downstream essential genes. To achieve this, upstream and downstream regions of plsX were amplified using the plsX-F1and plsX-R1 primer pair, and the plsX-F2 and plsX-R2 primer pair, respectively. Fragments were ligated to the 5' and 3' ends of the cat gene (amplified from pCBB31 using primers cat-F-p and cat-R-t) by recombinant PCR using overlapping sequences in the plsX-R1 and cat-F-p primer pair, and the cat-R-t and plsX-F2 primer pair. The resulting fragment was used to transform strain MY110, and chloramphenicol-resistant transformants were selected to generate strain MY111 (ΔaprE:: Pspac-plsX-kan ΔplsX::cat-p-t).
Analysis of phospholipid composition
Cells of B. subtilis strains 168, yneS-ts (MY103), and plsC-ts (MY112) were cultured in LB medium at 30°C until an OD600 of 0.4 was attained, and the temperature was then shifted to 42°C. Cells were pulse-labeled with 32 [Pi] (50 μCi/ml, Amersham Biosciences) for 5 min after 1 h of cultivation at 42°C. Incorporation of the label was terminated by the addition of 0.8 ml of culture to 3.0 ml of chloroform:methanol (1:2, v/v). Lipids were extracted using the method of Bligh and Dyer . The phospholipids produced were examined by two-dimensional thin-layer chromatography on Silica Gel 60 plates (Merck) developed with chloroform:methanol:water (65:25:4) in the first dimension and chloroform:methanol:acetic acid (65:25:10) in the second dimension . The incorporation of label into lysophosphatidic acid and other lipids was determined and quantitated using the BAS2500 image analyzer (Fuji).
Pspac-plsX (MY111) cells were cultured in LB medium containing 50 μM IPTG at 37°C until an OD600 value of 0.4 was attained. Cells were harvested, washed twice with LB medium, and suspended in 5 mL LB medium with or without 1 mM IPTG at final densities (OD600 values) of 0.06 or 0.1, respectively. Next, cells were pulse-labeled with 32 [Pi] (50 μCi/ml) for 5 min after 1 h of cultivation at 37°C. Lipid extraction and analysis were performed as described above.
Gene disruption in E. coli
Deletions of the plsB, plsX, and ygiH genes were achieved using a one-step chromosomal gene inactivation method involving the phage λ Red recombination system, as developed by Datsenko and Wanner [22, 34]. The E. coli strain, BW25113, was used. The primers used for target gene disruptions are listed in Table S4 [see Additional file 4]. The 5' end of each primer contained 60–70 bp of sequences upstream or downstream of the target gene, while the 3' end contained specific 20-nucleotide sequences from the template, pKD13. The primer pairs plsBdelup and plsBdeldown, plsXdelup and plsXdeldown, and ygiHdelup and ygiHdeldown, were used to generate ΔplsB::kan (in strain MEC001), ΔplsX::kan (in strain MEC002), and ΔygiH::kan (in strain MEC003) mutations, respectively. Because E. coli plsB was essential for growth, inactivation of plsB was performed in the presence of a plasmid harboring the gene (pSTV28plsB-p). Deletion mutations were confirmed using colony PCR with a kan-specific primer (k1 or k2)  and the locus-specific primers listed in Table S4 [see Additional file 4]. The deletion mutations, ΔplsX::kan and ΔygiH::kan, were transferred into prototrophic strain W3110 by P1vir transduction to obtain strains MEC005 and MEC006, respectively. The kan cassettes in strains MEC005 and MEC006 were removed with FLP recombinase, expressed by plasmid pCP20, as described previously , to obtain strains MEC102 and MEC103, respectively.
Construction of E. coli strains harboring IAA-inducible genes
To introduce plsB,plsX and ygiH genes, under the control of IAA-inducible trp promoters, into the E. coli chromosome, cloned gene fragments in pMC1403-based plasmids were transferred into the λRZ5 phage by recA-mediated homologous recombination as previously described . The technique involves double crossovers between common upstream ampicillin-resistance genes and downstream lac operon sequences. The resulting λPw-plsB, λPw-plsX, and λPw-ygiH phages were lysogenized into the W3110 chromosome, to obtain strains MEC199, MEC308, and MEC309, respectively. The ΔplsB::kan mutation was transferred into strain MEC199 by P1vir transduction, and the kan cassette was then removed with FLP recombinase, expressed from plasmid pCP20, to obtain strain MEC201. The λPw-plsX and λPw-ygiH phages were also lysogenized into the strain MEC102 chromosome, to generate strains MEC303 and MEC304, respectively. The ΔygiH::kan mutation was then transferred into strains MEC303 and MEC304 by P1vir transduction.
Cronan JEJ, Rock CO: Biosynthesis of membrane lipids. Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. Edited by: Neidhardt FC. 1987, American Society for Microbiology, 474-497.
de Chateau M, Bjorck L: Protein PAB, a mosaic albumin-binding bacterial protein representing the first contemporary example of module shuffling. J Biol Chem. 1994, 269 (16): 12147-12151.
Larson TJ, Ludtke DN, Bell RM: sn-Glycerol-3-phosphate auxotrophy of plsB strains of Escherichia coli: evidence that a second mutation, plsX, is required. J Bacteriol. 1984, 160 (2): 711-717.
Heath RJ, Rock CO: A conserved histidine is essential for glycerolipid acyltransferase catalysis. J Bacteriol. 1998, 180 (6): 1425-1430.
Coleman J: Characterization of Escherichia coli cells deficient in 1-acyl-sn-glycerol-3- phosphate acyltransferase activity. J Biol Chem. 1990, 265 (28): 17215-17221.
Heath RJ, Goldfine H, Rock CO: A gene (plsD) from Clostridium butyricum that functionally substitutes for the sn-glycerol-3-phosphate acyltransferase gene (plsB) of Escherichia coli. J Bacteriol. 1997, 179 (23): 7257-7263.
Ray TK, Cronan JE: Acylation of sn-glycerol 3-phosphate in Escherichia coli. Study of reaction with native palmitoyl-acyl carrier protein. J Biol Chem. 1975, 250 (21): 8422-8427.
Lu YJ, Zhang YM, Grimes KD, Qi J, Lee RE, Rock CO: Acyl-phosphates initiate membrane phospholipid synthesis in Gram-positive pathogens. Mol Cell. 2006, 23 (5): 765-772. 10.1016/j.molcel.2006.06.030.
Lu YJ, Zhang F, Grimes KD, Lee RE, Rock CO: Topology and Active Site of PlsY: THE BACTERIAL ACYLPHOSPHATE:GLYCEROL- 3-PHOSPHATE ACYLTRANSFERASE. J Biol Chem. 2007, 282 (15): 11339-11346. 10.1074/jbc.M700374200.
Kobayashi K, Ehrlich SD, Albertini A, Amati G, Andersen KK, Arnaud M, Asai K, Ashikaga S, Aymerich S, Bessieres P: Essential Bacillus subtilis genes. Proc Natl Acad Sci USA. 2003, 100 (8): 4678-4683. 10.1073/pnas.0730515100.
Schujman GE, Paoletti L, Grossman AD, de Mendoza D: FapR, a bacterial transcription factor involved in global regulation of membrane lipid biosynthesis. Dev Cell. 2003, 4 (5): 663-672. 10.1016/S1534-5807(03)00123-0.
Seki T, Miyachi H, Yoshikawa H, Kawamura F, Saito H: An improved methd of prophage transformation in Bacillus subtilis . J Gen Appl Microbiol. 1986, 32: 73-79.
Murakami T, Haga K, Takeuchi M, Sato T: Analysis of the Bacillus subtilis spoIIIJ gene and its Paralogue gene, yqjG. J Bacteriol. 2002, 184 (7): 1998-2004. 10.1128/JB.184.7.1998-2004.2002.
Beijer L, Nilsson RP, Holmberg C, Rutberg L: The glpP and glpF genes of the glycerol regulon in Bacillus subtilis. J Gen Microbiol. 1993, 139 (2): 349-359.
Glatz E, Persson M, Rutberg B: Antiterminator protein GlpP of Bacillus subtilis binds to glpD leader mRNA. Microbiology. 1998, 144 (Pt 2): 449-456.
OH YK, Freese EB, Freese E: Abnormal septation and inhibition of sporulation by accumulation of L- -glycerophosphate in Bacillus subtilis mutants. J Bacteriol. 1973, 113 (2): 1034-1045.
Morimoto T, Loh PC, Hirai T, Asai K, Kobayashi K, Moriya S, Ogasawara N: Six GTP-binding proteins of the Era/Obg family are essential for cell growth in Bacillus subtilis . Microbiology. 2002, 148 (Pt 11): 3539-3552.
Freese EB, Oh YK: Adenosine 5'-triphosphate release and membrane collapse in glycerol-requiring mutants of Bacillus subtilis . J Bacteriol. 1974, 120 (1): 507-515.
Oh YK, Freese EB, Freese E: Abnormal septation and inhibition of sporulation by accumulation of L- -glycerophosphate in Bacillus subtilis mutants. J Bacteriol. 1973, 113 (2): 1034-1045.
Bligh EG, Dyer WJ: A rapid method of total lipid extraction and purification. Can J Biochem Physiol. 1959, 37 (8): 911-917.
Gerdes SY, Scholle MD, Campbell JW, Balazsi G, Ravasz E, Daugherty MD, Somera AL, Kyrpides NC, Anderson I, Gelfand MS: Experimental determination and system level analysis of essential genes in Escherichia coli MG1655. J Bacteriol. 2003, 185 (19): 5673-5684. 10.1128/JB.185.19.5673-5684.2003.
Datsenko KA, Wanner BL: One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A. 2000, 97 (12): 6640-6645. 10.1073/pnas.120163297.
Heath RJ, Rock CO: A missense mutation accounts for the defect in the glycerol-3-phosphate acyltransferase expressed in the plsB26 mutant. J Bacteriol. 1999, 181 (6): 1944-1946.
Jiang P, Cronan JEJ: Inhibition of fatty acid synthesis in Escherichia coli in the absence of phospholipid synthesis and release of inhibition by thioesterase action. J Bacteriol. 1994, 176 (10): 2814-2821.
Cho H, Cronan JEJ: Escherichia coli thioesterase I, molecular cloning and sequencing of the structural gene and identification as a periplasmic enzyme. J Biol Chem. 1993, 268 (13): 9238-9245.
Jacobs MA, Alwood A, Thaipisuttikul I, Spencer D, Haugen E, Ernst S, Will O, Kaul R, Raymond C, Levy R: Comprehensive transposon mutant library of Pseudomonas aeruginosa . Proc Natl Acad Sci U S A. 2003, 100 (24): 14339-14344. 10.1073/pnas.2036282100.
Rock CO, Jackowski S: Regulation of phospholipid synthesis in Escherichia coli. Composition of the acyl-acyl carrier protein pool in vivo. J Biol Chem. 1982, 257 (18): 10759-10765.
Heath RJ, Rock CO: Regulation of malonyl-CoA metabolism by acyl-acyl carrier protein and beta-ketoacyl-acyl carrier protein synthases in Escherichia coli . J Biol Chem. 1995, 270 (26): 15531-15538. 10.1074/jbc.270.26.15531.
Dubnau D, Davidoff-Abelson R: Fate of transforming DNA following uptake by competent Bacillus subtilis. I. Formation and properties of the donor-recipient complex. J Mol Biol. 1971, 56 (2): 209-221. 10.1016/0022-2836(71)90460-8.
Casadaban MJ, Chou J, Cohen SN: In vitro gene fusions that join an enzymatically active β-galactosidase segment to amino-terminal fragments of exogenous proteins: Escherichia coli plasmid vectors for the detection and cloning of translational initiation signals. J Bacteriol. 1980, 143 (2): 971-980.
Yamada K: Characterization of spoII genes of Bacillus subtilis. 1989, PhD Thesis Hiroshima University
Kawamura F, Wang LF, Doi RH: Catabolite-resistant sporulation (crsA) mutations in the Bacillus subtilis RNA polymerase sigma 43 gene (rpoD) can suppress and be suppressed by mutations in spo0 genes. Proc Natl Acad Sci U S A. 1985, 82 (23): 8124-8128. 10.1073/pnas.82.23.8124.
Guerout-Fleury AM, Shazand K, Frandsen N, Stragier P: Antibiotic-resistance cassettes for Bacillus subtilis . Gene. 1995, 167 (1–2): 335-336. 10.1016/0378-1119(95)00652-4.
Oshima T, Aiba H, Masuda Y, Kanaya S, Sugiura M, Wanner BL, Mori H, Mizuno T: Transcriptome analysis of all two-component regulatory system mutants of Escherichia coli K-12. Mol Microbiol. 2002, 46 (1): 281-291. 10.1046/j.1365-2958.2002.03170.x.
Ito K, Kawakami K, Nakamura Y: Multiple control of Escherichia coli lysyl-tRNA synthetase expression involves a transcriptional repressor and a translational enhancer element. Proc Natl Acad Sci U S A. 1993, 90 (1): 302-306. 10.1073/pnas.90.1.302.
We are grateful to Fujio Kawamura (Rikkyo University) for providing plasmid pCH11. TL84 (ME8400) was supplied by the National BioResource Project for E. coli (National Institute of Genetics). We thank Toshifumi Inada (Nagoya University) for helpful discussions. This work was supported by KAKENHI (Grant-in-Aid for Scientific Research) on Priority Areas "Systems Genomics" from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and a Grant-in-Aid for JSPS Research Fellowships for Young Scientists to MY.
All experimental work was carried out by MY under the supervision of TO and NO. TO provided the E. coli transduction system. MY and TO wrote the draft manuscript. NO finalized the analysis and completed the manuscript. All authors read and approved the final manuscript.