Repression of the activities of two extracytoplasmic function σ factors, σ^M and σ^V, of Bacillus subtilis by glucolipids in Escherichia coli cells

ABSTRACT

Extracytoplasmic function (ECF) σ factors respond to environmental stresses and regulate numerous genes required for adaptation. Under normal growth conditions, the ECF σ factors are sequestered by transmembrane anti-σ factor proteins, from which they are released under stress conditions. In Bacillus subtilis ugtP null mutant cells, which lack glucolipids, three of the seven ECF σ factors, σ^M, σ^V and σ^X, are activated. The Escherichia coli cell membrane does not contain glucolipids. When the genes for these three ECF σ and anti-σ factors were introduced into E. coli cells, expression of lacZ fused to the ECF σ factor-regulated promoters indicated ECF σ factor activity. Additional expression of the ugtP gene in these E. coli cells led to the synthesis of small amounts of glucolipids, and the activities of σ^M and σ^V were repressed, but the activity of σ^X was unaffected. It is likely that glucolipids directly influence anti-σ^M and anti-σ^V factors by stabilizing conformations that sequester the respective ECF σ factors.

MAIN

The membrane lipid composition of Bacillus subtilis is complex. It includes acidic phospholipids (phosphatidylglycerol [PG] and cardiolipin [CL]), zwitterionic phospholipids (phosphatidylethanolamine [PE] and lysylphosphatidylglycerol [LysPG]) and glucolipids (monoglucosyldiacylglycerol [MGDG], diglucosyldiacylglycerol [DGDG] and triglucosyldiacylglycerol [TGDG]) (de Mendoza et al., 2002; Matsumoto et al., 2012). In contrast, the membrane composition of Escherichia coli is much simpler: it comprises only the three major phospholipids, PG, CL and PE, except for the outer leaflet of the outer membrane which is composed of lipopolysaccharide (Shibuya, 1992). B. subtilis glucolipids are synthesized by the gene product of ugtP, which processively transfers glucose from UDP-glucose to diacylglycerol (Jorasch et al., 1998). MGDG, DGDG and TGDG amount to 1.2%, 9.8% and 0.3%, respectively, of total membrane lipids (Kawai et al., 2006). A ugtP null mutant lacking membrane glucolipids shows abnormal cell morphology (Price et al., 1997; Matsuoka et al., 2011a) and constitutive activation of three extracytoplasmic function (ECF) σ factors, σ^M, σ^V and σ^X (Salzberg and Helmann, 2008; Matsuoka et al., 2011a; Hashimoto et al., 2013).

ECF σ factors respond to environmental stresses and direct the transcription of genes involved in tasks such as the maintenance of cell envelope integrity. B. subtilis has seven ECF σ factors: σ^M, σ^V, σ^W, σ^X, σ^Y, σ^Z and σ^YlaC. Except for σ^Z, these are regulated directly by their respective cognate transmembrane anti-σ factors, which sequester the σ factors. The genes for the σ factors and the anti-σ factors form operons whose transcription is directed by the cognate σ factors; the activity of the promoter of an ECF σ factor gene is thus indicative of the activity of the respective ECF σ factor (Helmann, 2002; Yoshimura et al., 2004; Asai et al., 2005).

In this study, we introduced σ factor–anti-σ factor operons for σ^M, σ^V and σ^X into E. coli cells and monitored the σ factor activity via lacZ fusion to the promoters of the operons. Glucolipids were synthesized in this heterologous system, and their influence on the activity of the ECF σ factors was analyzed. A heterologous system offers the advantage of allowing the introduced components to be analyzed independently of other effects that might disturb their interactions in the native system.

The bacterial strains and plasmids used in this study are listed in Table 1. Luria-Bertani (LB) medium (Miller, 1992) was used with addition of antibiotics, where appropriate, at the following concentrations (μg ml^–1): ampicillin, 50; chloramphenicol, 5; kanamycin, 5; tetracycline, 10; spectinomycin, 100. Cell growth was monitored with a mini photo 518R photometer (TAITEC) equipped with a 530-nm interference filter and a 14-mm light path adaptor or a Klett-Summerson colorimeter equipped with a no. 54 filter.

Table 1. Bacterial strains and plasmids used for this study

Strain or plasmid	Relevant genotype or description	Construction, source or referencea
E. coli
DH5α	recA1 gyrA relA1 thi-1 glnV44 hsdR17 endA1 Δ(lacIZYA-argF)U169deoR (ϕ80dlac Δ(lacZ)M15)	Laboratory collection
B. subtilis
168	trpC2	Laboratory collection
YTB019	168 ugtP::kan	(Matsuoka et al., 2011b)
MBS001	YTB019 aprE::(Pspac-ugtP spc)	pAPNC-ugtP → YTB019
SK71	trpC2 lys-1 aprEΔ3 nprE18 nprR2 amyE::(PsigM′-lacZ cat)	(Kosono et al., 2004)
SK72	trpC2 lys-1 aprEΔ3 nprE18 nprR2 amyE::(PsigV-lacZ cat)	(Kosono et al., 2004)
SK74	trpC2 lys-1 aprEΔ3 nprE18 nprR2 amyE::(PsigX′-lacZ cat)	(Kosono et al., 2004)
YTB004	168 amyE::(PsigM′-lacZ cat)	SK71 → 168
YTB041	168 amyE::(PsigV-lacZ cat)	SK72 → 168
YTB033	168 amyE::(PsigX′-lacZ cat)	SK74 → 168
YTB002	YTB019 amyE::(PsigM′-lacZ cat)	SK71 → YTB019
YTB040	YTB019 amyE::(PsigV-lacZ cat)	SK72 → YTB019
YTB030	YTB019 amyE::(PsigX′-lacZ cat)	SK74 → YTB019
YTB020	MBS001 amyE::(PsigM′-lacZ cat)	SK71 → MBS001
RMB001	MBS001 amyE::(PsigV-lacZ cat)	SK72 → MBS001
YTB027	MBS001 amyE::(PsigX′-lacZ cat)	SK74 → MBS001
Plasmids
pAPNC213	aprE ′specr lacI Pspac ′aprE ampr	(Morimoto et al., 2002)
pAPNC-ugtP	pAPNC213 Pspac-ugtP	This study
pFZY1	F′lac replicon, lacZY ampr	(Koop et al., 1987)
pFZY1-PsigM′	pFZY1 PsigM′-lacZY	This study
pFZY1-PsigV	pFZY1 PsigV-lacZY	This study
pFZY1-PsigX′	pFZY1 PsigX′-lacZY	This study
pBR322	ampr tetr	(Balbás et al., 1986)
pBR-sigM-antisigM	pBR322 PsigM′-sigM-yhdL-yhdK tetr	This study
pBR-sigV-antisigV	pBR322 PsigV-sigV-rsiV tetr	This study
pBR-sigX-antisigX	pBR322 PsigX′-sigX-rsiX tetr	This study
pHR718	pSC101 replicon, lacIq Ptrc specr	(Shiba et al., 2004)
pHR718-ugtP-his	pHR718 Ptrc-ugtP-his6	This study

^a  →, transformation with chromosomal or plasmid DNA.

First we confirmed the activation of σ^M, σ^V and σ^X in the B. subtilis ugtP null mutant. Activities of σ^M, σ^V and σ^X factors in B. subtilis were monitored by measuring β-galactosidase activities expressed from transcriptional lacZ fusions to the promoters of the operons of the genes for the ECF σ factors and the anti-σ factors, which are autoregulated by the respective σ factors. The lacZ fusions were integrated into the chromosomal amyE region. The β-galactosidase assay method using o-nitrophenyl-β-d-galactoside as substrate and the unit definition were as described by Wang and Doi (1984). As we previously reported (Hashimoto et al., 2013), σ^M, σ^V and σ^X factors were activated in the ugtP mutant cells: 6.9-fold, 3.4-fold and 1.5-fold, respectively (Fig. 1). When the ugtP gene was placed under the control of the isopropyl-thio-β-d-galactoside (IPTG)-inducible promoter P_spac and integrated into the aprE region of the ugtP null mutant, IPTG induction of this ectopic ugtP gene decreased the activities of the ECF σ factors (Fig. 1). Thus the activation of σ^M, σ^V and σ^X in the ugtP null mutant was complemented by an ectopic copy of ugtP, indicating that it is the defect of ugtP itself, not a polar effect of the ugtP disruption, that is responsible for the activation of the σ factors.

Fig. 1.

Activation of σ^M, σ^V and σ^X in the B. subtilis ugtP null mutant and its suppression by ectopic expression of ugtP. The strains used were YTB004 (WT), YTB002 (ugtP) and YTB020 (P_spac-ugtP) in A, YTB041 (WT), YTB040 (ugtP) and RMB001 (P_spac-ugtP) in B, and YTB033 (WT), YTB030 (ugtP) and YTB027 (P_spac-ugtP) in C. Overnight cultures were diluted 100-fold in LB medium containing either no IPTG or the indicated concentration of IPTG. After 2 h cultivation with shaking, β-galactosidase activity was measured. The means and standard errors of three measurements are shown.

To monitor the activities of σ^M, σ^V and σ^X in E. coli, the promoters of the operons of genes for the ECF σ factors and their anti-σ factors were cloned into a promoter-probe mini-F vector, pFZY1 (Koop et al., 1987), to construct transcriptional lacZ fusions. For construction of plasmids, see Supplementary Information. The plasmids were introduced into the lacZ-defective E. coli strain DH5α. Empty vectors, pBR322 (Balbás et al., 1986) and pHR718 (Shiba et al., 2004), for the plasmids used in the subsequent experiments were also introduced. In these transformants a small amount of β-galactosidase activity (background) was detected (Fig. 2). This activity may be due to transcription by σ⁷⁰-containing RNA polymerase of E. coli.

Fig. 2.

Activity of σ^M, σ^V and σ^X in E. coli cells and the effect of expression of ugtP. The strain used for background in each panel was DH5α harboring pFZY1-P_sigM′ and empty vectors pBR322 and pHR718 (A), pFZY1-P_sigV, pBR322 and pHR718 (B), or pFZY1-P_sigX′, pBR322 and pHR718 (C). The strain used for the second bar of each panel was DH5α harboring pFZY1-P_sigM′, pBR-sigM-antisigM and the empty vector pHR718 (A), pFZY1- P_sigV, pBR-sigV-antisigV and pHR718 (B), or pFZY1-P_sigX′, pBR-sigX-antisigX and pHR718 (C). The strain used with increasing concentrations of IPTG was DH5α harboring pFZY1-P_sigM′, pBR-sigM-antisigM and pHR718-ugtP-his (A), pFZY1-P_sigV, pBR-sigV-antisigV and pHR718-ugtP-his (B), or pFZY1-P_sigX′, pBR-sigX-antisigX and pHR718-ugtP-his (C). Overnight cultures were diluted 100-fold in LB medium containing either no IPTG or the indicated concentration of IPTG and cultivated with shaking. At 0.25 on the mini photo 518R photometer, β-galactosidase activity was measured. The means and standard errors of three measurements are shown.

Next, pBR322-derived multicopy plasmids carrying the ECF σ factor–anti-σ factor operons and the empty vector pHR718 were introduced. Introduction of a plasmid carrying the genes for σ^M and anti-σ^M factors (YhdL and YhdK; σ^M has two anti-σ factors) led to a more than 10-fold increase in β-galactosidase activity compared with the background activity (Fig. 2A). This increment corresponds to the σ^M activity in the absence of glucolipid since the E. coli membrane does not contain glucolipid. E. coli RNA polymerase core enzyme seems to form holoenzyme with B. subtilis σ^M. Further introduction of the ugtP gene under the control of the IPTG-inducible promoter P_trc on pHR718 reduced the activity to ca. 75% even in the absence of IPTG. Addition of IPTG at concentrations of >30 μM reduced the activity to ca. 60% of that in the absence of the ugtP gene. Reduction in the activity in the presence of pHR718-ugtP-his can most likely be ascribed to anti-σ^M factors that sequester σ^M in the presence of glucolipids.

Introduction of a plasmid carrying the genes for σ^V and anti-σ^V factor (RsiV) led to a 10-fold increase in β-galactosidase activity compared with the background activity (Fig. 2B). E. coli RNA polymerase core enzyme seems to form holoenzyme with B. subtilis σ^V. Further introduction of the plasmid carrying the ugtP gene under the control of P_trc reduced the activity to ca. 50% even in the absence of IPTG. Addition of IPTG slightly reduced the activity. These results indicate that anti-σ^V factor sequesters σ^V in the presence of glucolipids and that the function is weakened in the absence of glucolipids.

Introduction of a plasmid carrying the genes for σ^X and anti-σ^X factor (RsiX) led to an eight-fold increase in β-galactosidase activity compared with the background activity (Fig. 2C). E. coli RNA polymerase core enzyme seems to form holoenzyme with B. subtilis σ^X. However, additional introduction of a plasmid carrying the ugtP gene did not reduce the activity even after addition of IPTG; instead, it slightly increased the activity. The presence of glucolipids did not strengthen the function of anti-σ^X factor in E. coli.

Repression of the activity of σ^M and σ^V by ugtP expression in E. coli was also confirmed by real-time RT-PCR (see Supplementary Information and Supplementary Fig. S1).

We tried to clone the genes for the ECF σ factors without the genes for the cognate anti-σ factors. For cloning procedures, see Supplementary Information. Repeated attempts to do so, however, were unsuccessful. The anti-σ factors are not likely to completely lose their ability to sequester the cognate σ factors even in the absence of glucolipids. In the absence of anti-σ factors large amounts of the ECF σ factors, whose genes are positively autoregulated on multicopy plasmids, would freely interact with RNA polymerase core enzyme and cause a shortage of σ⁷⁰-containing holoenzyme. We suppose that this was the reason why we could not clone the genes for the ECF σ factors alone.

Lipid compositions of the strains used for the assay of the ECF σ activities were analyzed by continuous labeling with [¹⁴C]acetic acid followed by two-dimensional thin layer chromatography of the extracted lipids as described (Hashimoto et al., 2013) (Fig. 3, Table 2). In these experiments, the ugtP gene was expressed in E. coli under the control of an IPTG-inducible P_trc promoter using pHR718-ugtP-his, as in Fig. 2. At IPTG concentrations above 100 μM, glucolipid content ceased to increase, and even at 1 mM it was only about 1% in total. This is much lower than that in wild-type B. subtilis cells (ca. 10%). The ratio of MGDG/DGDG/TGDG is also markedly different from that in wild-type B. subtilis cells (1.2%/9.8%/0.3%) (Kawai et al., 2006). MGDG content was below a detectable level even in the presence of 1 mM IPTG, which is consistent with the report by Jorasch et al. (1998), who expressed ugtP (formerly called ypfP) in E. coli.

Fig. 3.

Two-dimensional thin layer chromatography of lipids of E. coli cells expressing ugtP. Cells were labeled with 3.75 × 10⁴ Bq ml^–1 of [1-¹⁴C]acetic acid (Amersham; 2.12 × 10⁹ Bq mmol^–1) in LB medium at 37 ℃, continuously from initial inoculation for preculture to cell harvesting. At about 100 Klett units, cells were harvested, and lipids were extracted and analyzed by two-dimensional thin layer chromatography as described (Hashimoto et al., 2013). Shown here as examples are the results of DH5α harboring pFZY1-P_sigM′, pBR-sigM-antisigM and pHR718 (A), and pFZY1-P_sigM′, pBR-sigM-antisigM and pHR718-ugtP-his grown in the presence of 1 mM IPTG (B). CDP-DG is CDP-diacylglycerol.

Table 2. Lipid composition of E. coli cells used for the assay of ECF σ factor activities

DH5α harboring:	IPTG (μM)	mol%a
		PE	PG	CL	CDP-DG	MGDG	DGDG	TGDG
pFZY1-PsigM′, pBR-sigM-antisigM and pHR718	0	80.21	15.99	3.58	0.23	<0.05	<0.05	<0.05
pFZY1-PsigM′, pBR-sigM-antisigM and pHR718-ugtP-his	0	80.81	15.65	3.30	0.24	<0.05	<0.05	<0.05
	10	81.28	15.71	2.77	0.19	<0.05	0.05	<0.05
	30	79.87	16.43	3.25	0.17	<0.05	0.22	0.06
	100	79.77	15.80	3.49	0.13	<0.05	0.52	0.30
	300	82.23	13.71	2.87	0.15	<0.05	0.59	0.45
	1000	82.57	12.99	2.97	0.26	<0.05	0.62	0.59
pFZY1-PsigV, pBR-sigV-antisigV and pHR718	0	79.95	16.39	3.44	0.22	<0.05	<0.05	<0.05
pFZY1-PsigV, pBR-sigV-antisigV and pHR718-ugtP-his	0	80.75	16.09	2.89	0.26	<0.05	<0.05	<0.05
	10	81.56	15.50	2.65	0.25	<0.05	<0.05	<0.05
	30	80.77	15.74	2.99	0.22	<0.05	0.05	<0.05
	100	81.43	14.62	2.72	0.21	<0.05	0.58	0.44
	300	81.49	14.18	3.11	0.22	<0.05	0.58	0.42
	1000	81.41	14.47	2.88	0.20	<0.05	0.58	0.46
pFZY1-PsigX′, pBR-sigX-antisigX and pHR718	0	79.39	16.21	4.15	0.25	<0.05	<0.05	<0.05
pFZY1-PsigX′, pBR-sigX-antisigX and pHR718-ugtP-his	0	80.53	14.82	4.32	0.33	<0.05	<0.05	<0.05
	10	80.05	15.62	4.02	0.25	<0.05	0.05	<0.05
	30	82.27	13.19	4.09	0.29	<0.05	0.11	0.05
	100	81.54	13.73	3.72	0.28	<0.05	0.40	0.33
	300	82.83	12.22	3.99	0.28	<0.05	0.32	0.37
	1000	82.14	13.04	3.75	0.26	<0.05	0.37	0.44

^a  Cells were labeled and lipids were analyzed as described in the legend to Fig. 3. Molar percentage of each lipid was calculated from PSL units obtained with an FLA-7000 bioimaging analyzer (Fujifilm). <0.05, below a detectable level.

If the glucolipid content could be further increased in E. coli, a higher level of repression of σ^M and σ^V activities might be observed. However, as shown in Fig. 2, a considerable level of repression was observed even without addition of IPTG in E. coli. In the absence of IPTG there would be leaky expression from the P_trc promoter, but the glucolipid content was below a detectable level. With increasing concentrations of IPTG to 100 μM, glucolipid content also increased, and the repression levels of σ^M and σ^V activities were slightly higher. We surmise that a low glucolipid content suffices to potentiate anti-σ^M and anti-σ^V factors to a considerable level.

In E. coli cells, heterologous σ^M and σ^V from B. subtilis showed higher activity in the absence of glucolipids than in their presence. This parallels the activation of these σ factors in the ugtP mutant of B. subtilis (Matsuoka et al., 2011a; Hashimoto et al., 2013). Glucolipids are most likely incorporated into the membrane in E. coli cells, and the anti-σ factors are canonical membrane proteins with hydrophobic transmembrane segments and presumably behave as membrane proteins in E. coli cells. We suggest that transmembrane anti-σ factors are potentiated to sequester the σ factors by the presence of glucolipids. Since glucolipids in the heterologous E. coli cells exhibited the same effect as in the B. subtilis cells, we infer that glucolipids probably directly influence anti-σ^M and anti-σ^V factors. Glucolipids probably help the anti-σ factors to maintain conformations suitable for sequestering the respective ECF σ factors. In the absence of glucolipids, anti-σ factors cannot stabilize the conformations, and the ECF σ factors are released and allowed to associate with RNA polymerase core enzyme. In this sense, glucolipids may have a chaperone-like function. It has been reported that PE acts as a chaperone in the assembly of the lactose permease LacY in E. coli (Bogdanov et al., 1996).

In contrast, the activity of σ^X was not repressed by the presence of glucolipids in E. coli. In fact, its activity was slightly increased in their presence; the reason for this is not known. In B. subtilis cells some additional factor(s), which are absent in E. coli, may be involved in the regulation of anti-σ^X factor function by glucolipids.

It has been reported that the activation of σ^V of B. subtilis involves regulated intramembrane proteolysis (RIP) of the anti-σ^V factor (Hastie et al., 2013). When B. subtilis cells are challenged by lysozyme, lysozyme-bound anti-σ^V factor is cleaved in its extracytoplasmic domain by a signal peptidase (Hastie et al., 2014), and intramembrane proteolysis by the site 2 protease RasP follows (Hastie et al., 2013). The activation of σ^V in the absence of glucolipids in E. coli may involve RIP by E. coli proteases. Alternatively, a conformational change of anti-σ^V may activate σ^V weakly: σ^V activation in the absence of glucolipids is only several-fold, whereas lysozyme treatment causes 65-fold activation (Hastie et al., 2013).

ACKNOWLEDGMENTS

This study was supported in part by a Grant-in-Aid for Scientific Research (C) from the Japan Society for the Promotion of Science.

REFERENCES

•  Asai,  K.,  Matsumoto,  T., and  Sadaie,  Y. (2005) ECF (extracytoplasmic function) sigma factors of Bacillus subtilis. In: Survival and Death in Bacteria (ed.: Yamada, M.), pp. 143–153. Research Signpost, Kerala, India.
•  Balbás,  P.,  Soberon,  X.,  Merino,  E.,  Zurita,  M.,  Lomeli,  H.,  Valle,  F.,  Flores,  N., and  Bolivar,  F. (1986) Plasmid vector pBR322 and its special-purpose derivatives — a review. Gene 50, 3–40.
•  Bogdanov,  M.,  Sun,  J.,  Kaback,  H. R., and  Dowhan,  W. (1996) A phospholipid acts as a chaperone in assembly of a membrane transport protein. J. Biol. Chem. 271, 11615–11618.
•  de Mendoza,  D.,  Schujman,  G. E., and  Aguilar,  P. S. (2002) Biosynthesis and function of membrane lipids. In: Bacillus subtilis and Its Closest Relatives (eds.: Sonenshein, A. L., Hoch, J. A., and Losick, R.), pp. 43–55. ASM Press, Washington, D. C.
•  Hashimoto,  M.,  Seki,  T.,  Matsuoka,  S.,  Hara,  H.,  Asai,  K.,  Sadaie,  Y., and  Matsumoto,  K. (2013) Induction of extracytoplasmic function sigma factors in Bacillus subtilis cells with defects in lipoteichoic acid synthesis. Microbiology 159, 23–35.
•  Hastie,  J. L.,  Williams,  K. B., and  Ellermeier,  C. D. (2013) The activity of σ^V, an extracytoplasmic function σ factor of Bacillus subtilis, is controlled by regulated proteolysis of the anti-σ factor RsiV. J. Bacteriol. 195, 3135–3144.
•  Hastie,  J. L.,  Williams,  K. B.,  Sepulveda,  C.,  Houtman,  J. C.,  Forest,  K. T., and  Ellermeier,  C. D. (2014) Evidence of a bacterial receptor for lysozyme: binding of lysozyme to the anti-sigma factor RsiV controls activation of the ECF σ factor σ^V. PLoS Genet. 10, e1004643.
•  Helmann,  J. D. (2002) The extracytoplasmic function (ECF) sigma factors. Adv. Microb. Physiol. 46, 47–110.
•  Jorasch,  P.,  Wolter,  F. P.,  Zahringer,  U., and  Heinz,  E. (1998) A UDP glucosyltransferase from Bacillus subtilis successively transfers up to four glucose residues to 1,2-diacylglycerol: expression of ypfP in Escherichia coli and structural analysis of its reaction products. Mol. Microbiol. 29, 419–430.
•  Kawai,  F.,  Hara,  H.,  Takamatsu,  H.,  Watabe,  K., and  Matsumoto,  K. (2006) Cardiolipin enrichment in spore membranes and its involvement in germination of Bacillus subtilis Marburg. Genes Genet. Syst. 81, 69–76.
•  Koop,  A. H.,  Hartley,  M. E., and  Bourgeois,  S. (1987) A low-copy-number vector utilizing β-galactosidase for the analysis of gene control elements. Gene 52, 245–256.
•  Kosono,  S.,  Asai,  K.,  Sadaie,  Y., and  Kudo,  T. (2004) Altered gene expression in the transition phase by disruption of a Na⁺/H⁺ antiporter gene (shaA) in Bacillus subtilis. FEMS Microbiol. Lett. 232, 93–99.
•  Matsumoto,  K.,  Matsuoka,  S., and  Hara,  H. (2012) Membranes and lipids. In: Escherichia coli and Bacillus subtilis: The Frontier of Molecular Microbiology Revisited (eds.: Sadaie, Y., and Matsumoto, K.), pp. 61–91. Research Signpost, Kerala, India.
•  Matsuoka,  S.,  Chiba,  M.,  Tanimura,  Y.,  Hashimoto,  M.,  Hara,  H., and  Matsumoto,  K. (2011a) Abnormal morphology of Bacillus subtilis ugtP mutant cells lacking glucolipids. Genes Genet. Syst. 86, 295–304.
•  Matsuoka,  S.,  Hashimoto,  M.,  Kamiya,  Y.,  Miyazawa,  T.,  Ishikawa,  K.,  Hara,  H., and  Matsumoto,  K. (2011b) The Bacillus subtilis essential gene dgkB is dispensable in mutants with defective lipoteichoic acid synthesis. Genes Genet. Syst. 86, 365–376.
•  Miller,  J. H. (1992) A Short Course in Bacterial Genetics: a Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
•  Morimoto,  T.,  Loh,  P. C.,  Hirai,  T.,  Asai,  K.,  Kobayashi,  K.,  Moriya,  S., and  Ogasawara,  N. (2002) Six GTP-binding proteins of the Era/Obg family are essential for cell growth in Bacillus subtilis. Microbiology 148, 3539–3552.
•  Price,  K. D.,  Roels,  S., and  Losick,  R. (1997) A Bacillus subtilis gene encoding a protein similar to nucleotide sugar transferases influences cell shape and viability. J. Bacteriol. 179, 4959–4961.
•  Salzberg,  L. I., and  Helmann,  J. D. (2008) Phenotypic and transcriptomic characterization of Bacillus subtilis mutants with grossly altered membrane composition. J. Bacteriol. 190, 7797–7807.
•  Shiba,  Y.,  Yokoyama,  Y.,  Aono,  Y.,  Kiuchi,  T.,  Kusaka,  J.,  Matsumoto,  K., and  Hara,  H. (2004) Activation of the Rcs signal transduction system is responsible for the thermosensitive growth defect of an Escherichia coli mutant lacking phosphatidylglycerol and cardiolipin. J. Bacteriol. 186, 6526–6535.
•  Shibuya,  I. (1992) Metabolic regulation and biological functions of phospholipids in Escherichia coli. Prog. Lipid Res. 31, 245–299.
•  Wang,  P. Z., and  Doi,  R. H. (1984) Overlapping promoters transcribed by Bacillus subtilis σ⁵⁵ and σ³⁷ RNA polymerase holoenzymes during growth and stationary phases. J. Biol. Chem. 259, 8619–8625.
•  Yoshimura,  M.,  Asai,  K.,  Sadaie,  Y., and  Yoshikawa,  H. (2004) Interaction of Bacillus subtilis extracytoplasmic function (ECF) sigma factors with the N-terminal regions of their potential anti-sigma factors. Microbiology 150, 591–599.