A Vitamin B12-Based System for Conditional Expression Reveals ...

JOURNAL OF BACTERIOLOGY, May 2009, p. 3108–3119 0021-9193/09/$08.00⫹0 doi:10.1128/JB.01737-08 Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Vol. 191, No. 9

A Vitamin B12-Based System for Conditional Expression Reveals dksA To Be an Essential Gene in Myxococcus xanthus䌤 Diana García-Moreno,1 María Carmen Polanco,1 Gloria Navarro-Avile´s,1† Francisco J. Murillo,1 S. Padmanabhan,2 and Montserrat Elías-Arnanz1* ´ rea de Gene´tica (Unidad Asociada al IQFR-CSIC), Facultad de Biología, Departamento de Gene´tica y Microbiología, A Universidad de Murcia, Murcia 30100, Spain,1 and Instituto de Química Física Rocasolano, Consejo Superior de Investigaciones Científicas, Serrano 119, 28006 Madrid, Spain2 Received 11 December 2008/Accepted 17 February 2009

Myxococcus xanthus is a prokaryotic model system for the study of multicellular development and the response to blue light. The previous analyses of these processes and the characterization of new genes would benefit from a robust system for controlled gene expression, which has been elusive so far for this bacterium. Here, we describe a system for conditional expression of genes in M. xanthus based on our recent finding that vitamin B12 and CarH, a MerR-type transcriptional repressor, together downregulate a photoinducible promoter. Using this system, we confirmed that M. xanthus rpoN, encoding ␴54, is an essential gene, as reported earlier. We then tested it with ftsZ and dksA. In most bacteria, ftsZ is vital due to its role in cell division, whereas null mutants of dksA, whose product regulates the stringent response via transcriptional control of rRNA and amino acid biosynthesis promoters, are viable but cause pleiotropic effects. As with rpoN, it was impossible to delete endogenous ftsZ or dksA in M. xanthus except in a merodiploid background carrying another functional copy, which indicates that these are essential genes. B12-based conditional expression of ftsZ was insufficient to provide the high intracellular FtsZ levels required. With dksA, as with rpoN, cells were viable under permissive but not restrictive conditions, and depletion of DksA or ␴54 produced filamentous, aberrantly dividing cells. dksA thus joins rpoN in a growing list of genes dispensable in many bacteria but essential in M. xanthus.

extracytoplasmic function ␴ factor CarQ, has been used in the controlled expression of genes such as csgA and socE, implicated in developmental cell-cell interactions, or relA, required for (p)ppGpp synthesis in M. xanthus (7, 17, 27, 28, 50). The applicability of this system to the study of essential genes in M. xanthus, however, remains to be demonstrated. A second system designed for conditional gene expression in M. xanthus makes use of the PpilA-lacO-RBS promoter (where RBS is ribosome binding site), an isopropyl-␤-D-thiogalactopyranoside (IPTG)-inducible version of the native pilA promoter, whose transcription initiation depends on the alternative sigma factor ␴54 (19). In this system, the target gene is cloned in a plasmid under the control of PpilA-lacO-RBS, which integrates by homologous recombination at the pilA chromosomal locus. The gene encoding the LacI repressor is supplied in a separate plasmid that integrates in the chromosome by recombination at the Mx8 attB site. A limitation of this system for characterizing essential genes by studying the conditional mutant phenotype produced on turning off their expression is the relatively high level of basal expression observed from PpilA-lacO-RBS (19). Moreover, the system as designed requires two plasmids and uses up both of the available selectable markers typically used in M. xanthus (kanamycin or tetracycline resistance), which is an impediment in the study of essential genes. An alternative design for conditional gene expression in M. xanthus is reported here. It is based on our recent finding that vitamin B12 and a MerR-type transcriptional repressor, CarH, collaborate to downregulate the light-inducible M. xanthus PB promoter in the dark, whereas light derepresses it (Fig. 1) (45). PB, a ␴A-dependent promoter, drives the expression of the

The gram-negative Myxococcus xanthus and other myxobacteria are striking in the prokaryotic world for their ability to undergo a complex developmental program that culminates in the production of multicellular fruiting bodies. This process, triggered by starvation, has served as a prokaryotic model for the study of the molecular mechanisms underlying coordinated cell movements and intercellular interactions, as well as for the cellular differentiation process that occurs during multicellular development (22, 23, 59). M. xanthus is also undoubtedly an important model system for the genetic-molecular analysis of the reception and transduction of the blue light signal and of the consequent transcriptional regulation of structural genes involved in carotenoid biosynthesis (11). Genetic and molecular biological tools such as transposon insertions of transcriptional reporters and nonpolar in-frame gene deletions have enabled a functional analysis and characterization of the genes involved in these processes. However, many studies in M. xanthus have been hindered by the lack of autonomously replicating plasmids (one has only recently been reported [58]) and of a tightly regulated gene expression system that allows functional analysis of essential genes. One system that employs light activation of the carQRS promoter, which depends on the * Corresponding author. Mailing address: Departamento de Gene´tica ´ rea de Gene´tica (Unidad Asociada al IQFRy Microbiología, A CSIC), Facultad de Biología, Universidad de Murcia, Murcia 30100, Spain. Phone: 34 968 367134. Fax: 34 968 363963. E-mail: [email protected]. † Present address: Department of Environmental Protection, Estacio ń Experimental del Zaidín, Consejo Superior de Investigaciones Científicas, 18008 Granada, Spain. 䌤 Published ahead of print on 27 February 2009. 3108

VOL. 191, 2009

B12-BASED CONDITIONAL GENE EXPRESSION IN M. XANTHUS

3109

MATERIALS AND METHODS

FIG. 1. Schematic of the M. xanthus PB promoter region and its regulation. Promoter PB drives expression of the gene clusters represented by the squat arrows (constituent genes are labeled); the arrowhead indicates the direction of transcription. Ovals depict protein factors involved in PB regulation, as labeled. Arrows indicate positive regulation, and blunt-ended lines indicate negative regulation. Other regulatory proteins or light-dependent genes or operons implicated in carotenogenesis are omitted for clarity (see reference 11 for details).

carB-carA gene cluster, which groups together all but one of the structural genes encoding the enzymes involved in carotenoid synthesis (reviewed in reference 11). In addition to repression by B12 and CarH, expression of PB is also repressed in the dark by CarA, another MerR-type transcriptional factor remarkably similar to CarH in domain organization but whose action does not depend on B12. CarH (in the presence of B12) and CarA bind to the same bipartite operator to impede promoter access to RNA polymerase (RNAP) and thereby repress PB (29, 30, 44, 45). RNAP holoenzyme containing the extracytoplasmic function ␴ factor CarQ, together with a host of transcriptional factors (11), drives expression of the carQRS operon to produce CarS, an antirepressor that physically binds to CarA and CarH to derepress PB in the light. Mutually exclusive binding of CarA or CarH to operator DNA and to CarS is the molecular basis for this mechanism of repression and antirepression (38, 45). We have exploited the repression of PB by CarH and B12 to design a system for the regulatable expression of genes in M. xanthus. In this single-plasmid-based system, the gene of interest is placed under the control of PB and CarH so as to tightly repress its expression in the dark when B12 is present. We employed the system to confirm a previous finding that rpoN, which encodes the alternative sigma factor ␴54, is an essential gene in M. xanthus, unlike in other bacteria, where it appears to be required for growth only under specific nutrient-limiting conditions (25). We then tested our system with other candidate genes. One of these was the M. xanthus dksA gene. In Escherichia coli, DksA has been shown to regulate the stringent response via transcriptional control of rRNA and amino acid biosynthesis promoters (18, 40, 41, 43, 51). Gene disruption of dksA indicated that it is not essential in E. coli, but a strain lacking DksA grew poorly under nutrient-limiting conditions (24). Moreover, DksA has been implicated in the regulation of ␴54-dependent transcription in vivo, thus establishing a functional link between rpoN and dksA (3, 52). The B12-based system for conditional expression described in this study shows that dksA, like rpoN, is an essential gene in M. xanthus.

Bacterial strains and growth conditions. Table 1 lists the strains and plasmids used in this study. M. xanthus was grown at 33°C in rich Casitone-Tris ([CTT] (⬍1 nM B12, according to the manufacturer estimates) medium (4) supplemented, when indicated (see below and see the figure legends), with antibiotics (kanamycin at 40 ␮g/ml or tetracycline at 10 ␮g/ml) and 0.75 ␮M B12. The concentration of B12 in the medium was based on the observation that PB repression levels are maintained for a 1 nM to 1 ␮M concentration range of B12 (cyanocobalamin) or derivatives such as methylcobalamin, adenosylcobalamin, or hydroxycobalamin (45). E. coli strain DH5␣ was used for plasmid constructions, and BL21(DE3) was used for overexpressing His-tagged proteins. E. coli growth was carried out at 37°C in Luria broth supplied with the appropriate antibiotic as needed, while protein overexpression was induced overnight at 25°C. Plasmid and strain constructions. Standard protocols and commercial kits were used in the preparation and manipulation of chromosomal and plasmid DNA. All constructs were verified by DNA sequencing. For protein overexpression, the M. xanthus dksA gene was PCR amplified from genomic DNA and cloned into the NdeI site of the pET15b vector (Novagen) to generate pMR3042. Plasmids pMR2915 and pMR2979, described next, were used in the construction of pMR2991. Plasmid pMR2915 was generated by replacing the Kmr selectable marker in pMR2700, a plasmid that contains a 1.38-kb fragment of M. xanthus DNA for chromosomal integration by homologous recombination (44), by a Tcr selectable marker. For this, pMR2700 was digested with MluI to release a 1.5-kb DNA fragment containing the Kmr marker and then ligated to a 1.3-kb PCRamplified DNA fragment containing the Tcr marker from pACYC184 (47). Plasmid pMR2979 was generated by ligating into PstI/KpnI-digested pMR2915 a 121-bp PstI/XbaI-digested DNA fragment containing the PB promoter/operator region and an XbaI/KpnI-digested 3.26-kb DNA fragment containing the lacZ gene with its ribosomal binding site. To construct pMR2991, the ⬃3.4-kb PstI/ KpnI fragment from pMR2979 containing the PB-lacZ fusion and a ⬃0.9-kb KpnI/EcoRI fragment corresponding to carH were first ligated into PstI/EcoRIdigested pMR2915 to generate pMR2981. This plasmid was then cut with KpnI in order to insert a 92-bp DNA fragment containing Pc, a ␴A-dependent promoter that has been previously used in M. xanthus for constitutive expression of genes at a heterologous site (44). The resulting plasmid, pMR2991, which integrates into the M. xanthus chromosome by homologous recombination via the 1.38-kb DNA fragment, therefore carries carH under the control of a constitutive promoter and a reporter lacZ gene under PB control. Plasmids pMR3120, pMR3168, and pMR3085 are derivatives of pMR2991 that were obtained by cloning the M. xanthus rpoN, ftsZ, or dksA (each with its own RBS) genes, respectively, into the XbaI site of pMR2991. Plasmids pMR3121, pMR3165, and pMR3254 are pBJ113/pBJ114 derivatives (21) that were generated to create complete in-frame deletions of the M. xanthus rpoN, ftsZ, or dksA genes, respectively, as described next. All three plasmids contain a Kmr gene for positive selection and a galK gene that confers sensitivity to galactose (Gals) for negative selection. To construct pMR3121, two chromosomal fragments were PCR amplified and ligated into pBJ114 to replace the complete rpoN gene by an XbaI site: (i) a 1.01-kb fragment corresponding to the DNA immediately upstream of rpoN, with a HindIII site added to its 5⬘ end and an XbaI site added to its 3⬘ end; (ii) a 0.99-kb fragment corresponding to the DNA immediately downstream of rpoN, with an XbaI site added to its 5⬘ end and a KpnI site added to its 3⬘ end. Plasmid pMR3165, with a replacement of the ftsZ gene by an EcoRI site, was constructed by ligating into pBJ113 a 0.99-kb fragment of chromosomal DNA upstream of ftsZ (flanked by KpnI and EcoRI sites) and a 1.02-kb fragment of chromosomal DNA downstream of ftsZ (flanked by EcoRI and HindIII sites). Plasmid pMR3254 was constructed by PCR amplification of 0.85 kb of DNA upstream of dksA (flanked by SalI and XbaI sites) and 1.02 kb of DNA downstream of dksA (flanked by XbaI and EcoRI sites). After the two DNA fragments were ligated into pBJ114, an XbaI site replaced the dksA gene. Plasmid pMR3202 was constructed by ligating into PstI/KpnI-digested pMR2915 a 1.54-kb PstI/XbaI DNA fragment including ftsZ and 330 bp of upstream DNA and an XbaI/KpnI 3.26-kb DNA fragment containing the lacZ gene with its ribosomal binding site. Plasmids were introduced into M. xanthus cells by electroporation, and integration of the plasmids by homologous recombination was selected for on CTT plates containing the appropriate antibiotic. Plasmid integration at the correct chromosomal site in M. xanthus was verified by PCR in all cases. Integration of plasmids pMR3121, pMR3165, and pMR3254 into the chromosome yielded Kmr merodiploids containing both the wild-type and deletion alleles of rpoN, ftsZ, and dksA, respectively. Independent merodiploids were grown for several generations in CTT medium in the absence of kanamycin and then plated on CTT

3110

GARCIÁ-MORENO ET AL.

J. BACTERIOL. TABLE 1. Bacterial strains and plasmids used in this work

Strain or plasmid

Strains M. xanthus DK1050 MR1461 MR1716 MR1859 MR1860 MR1862 MR1866 MR1867 MR1870 MR1871 MR1925 MR1926 MR1929 MR1930 MR1933 E. coli BL21(DE3) DH5␣ Plasmids pET15b pACYC184 pBJ113 pBJ114 pMR2700 pMR2915 pMR2979 pMR2981 pMR2991 pMR3042 pMR3085 pMR3120 pMR3121 pMR3165 pMR3168 pMR3202 pMR3254

Relevant genotype or description

Wild type ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH ⌬carA ⌬carH dksA/⌬dksA

PB::lacZ Pc::carH rpoN/⌬rpoN ftsZ/⌬ftsZ ftsZ/⌬ftsZ PB::ftsZ-lacZ Pc::carH ftsZ/⌬ftsZ P330::ftsZ-lacZ rpoN/⌬rpoN PB::rpoN-lacZ Pc::carH ⌬rpoN PB::rpoN-lacZ Pc::carH rpoN PB::rpoN-lacZ Pc::carH dksA PB::dksA-lacZ Pc::carH ⌬dksA PB::dksA-lacZ Pc::carH dksA/⌬dksA dksA/⌬dksA PB::dksA-lacZ Pc::carH

Source or reference

48 This 45 This This This This This This This This This This This This

work work work work work work work work work work work work work

F⫺ lon ompT rB⫺ mB⫺ ␭DE3 F⫺ supE44 ⌬lacU169 (␾80d lacZ⌬M15) hsdR17 recA1 endA1 gyrA96 thi-1 relA1

Novagen Gibco-BRL

Vector for protein overexpression; Ampr Cloning vector; Tcr Cmr Cloning vector; Gals Kmr Cloning vector; Gals Kmr; same as pBJ113, but with polylinker in opposite orientation Contains 1.38-kb M. xanthus DNA fragment for chromosomal integration; Kmr Tcr derivative of pMR2700 pMR2915 plus PstI/XbaI fragment containing PB and XbaI/KpnI fragment containing lacZ pMR2915 plus PstI/KpnI fragment from pMR2979 containing PB::lacZ and KpnI/EcoRI fragment containing carH pMR2981 plus KpnI/KpnI promoter fragment Vector for protein overexpression of dksA; Ampr pMR2991 plus XbaI/XbaI fragment containing dksA pMR2991 plus XbaI/XbaI fragment containing rpoN pBJ114 plus 1.01-kb HindIII/XbaI fragment upstream of rpoN and 0.99-kb XbaI/KpnI fragment downstream of rpoN pBJ113 plus 0.99-kb KpnI/EcoRI fragment of chromosomal DNA upstream of ftsZ and 1.02-kb EcoRI/HindIII fragment of chromosomal DNA downstream of ftsZ pMR2991 plus XbaI/XbaI fragment containing ftsZ pMR2915 plus PstI/XbaI fragment containing P330::ftsZ and XbaI/KpnI fragment containing lacZ pBJ114 plus 0.85-kb SalI/XbaI fragment of chromosomal DNA upstream of dksA and 1-kb XbaI/EcoRI fragment of chromosomal DNA downstream of dksA

Novagen 47 21 21 44 This work This work This work

plates supplemented with 10 mg/ml galactose to select for the loss of the vector Gals marker. This evicts the vector DNA together with either the wild-type or the deletion allele by intramolecular recombination to yield haploid Gal⫹ Kms cells. The presence of the wild-type or the deletion allele was diagnosed by PCR. Purification and characterization of DksA. E. coli BL21(DE3) freshly transformed with pMR3042 was grown in 10 ml of Luria broth-ampicillin medium to an A600 of 0.6 to 1.0 at 37°C and then inoculated into 1 liter of fresh medium. After growth to an A600 of 0.6 to 1.0 at 37°C, 0.5 mM IPTG was added to induce dksA expression overnight at 25°C. Analysis of cell extracts from 1 ml of culture showed that His6-DksA is overexpressed as a soluble protein. The protein was purified using Talon metal affinity resin (Clontech) and the accompanying native protein purification protocol. The imidazole was eliminated by extensive dialysis in buffer A (50 mM phosphate buffer, pH 7.5, 5 mM ␤-mercaptoethanol) plus 150 mM NaCl, and the sample was further purified by passage through a phosphocellulose column equilibrated with the same buffer or, when required, treated with thrombin (GE Health Sciences), passed through phosphocellulose, and dialyzed extensively against buffer A plus 150 mM NaCl. Analysis of DksA for bound zinc was carried out using the colorimetric 4-(2-pyridylazo)resorcinol (PAR) assay (reference 42 and references therein). Briefly, DksA without the His6 tag was preincubated with p-chloromercuribenzoate to modify sulfhydryl (-SH) groups and release Zn. Freshly prepared PAR (100 mM) was then added to complex the freed Zn, and the absorbance at 500 nm was used to estimate Zn.

This This This This This

work work work work work

This work This work This work This work

In the pull-down assay to confirm interaction between DksA and M. xanthus RNA polymerase (obtained as described previously [30]), 200 ␮l of Talon in two 2-ml SigmaPrep (Sigma-Aldrich) columns was washed five times with 600 ␮l of low-salt buffer (buffer A plus 50 mM NaCl). A total of 400 ␮l of His6-DksA (⬃25 nmol) was applied to one column but not to the other, which served as a control. After incubation for 1 h, the resin in each column was washed with 30 volumes of low-salt buffer. To each column 20 pmol of RNAP (200 ␮l) was added and incubated for 1 to 2 h with gentle rotary agitation. Flowthrough from each column was collected separately. The resin was then extensively washed with low-salt buffer and subsequently eluted in five 200-␮l fractions with high-salt buffer (buffer A plus 500 mM NaCl). These and the low-salt flowthrough (after RNAP incubation) were analyzed using 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis with silver staining. ␤-Galactosidase activity assays. ␤-Galactosidase activity was qualitatively assessed on CTT plates with 40 mg ml⫺1 of X-Gal (5-bromo-4-chloro-3-indolyl-␤D-galactopyranoside) and measured (in nmol of o-nitrophenyl ␤-D-galactoside hydrolyzed min⫺1 mg⫺1 of protein) as described elsewhere for vegetatively growing liquid cultures of M. xanthus (1). Comparison of reporter lacZ expression from pMR2991 in the dark and in the light and in the presence or absence of B12 was carried out as follows. A starter culture of the corresponding M. xanthus strain was grown in the dark in 10 ml of CTT medium supplemented with the required antibiotic. The culture (0.25 ml) was added to each of two separate


VOL. 191, 2009

3111

culture flasks containing 12 ml of fresh CTT medium and antibiotic, with 0.75 ␮M B12 included in one of the two flasks. After 16 h of growth in the dark at 33°C, each 12-ml culture was split into two; one of these remained in the dark, and the other was exposed to light while growth continued for another 8 h. Samples were then harvested for quantitative analysis of ␤-galactosidase activity. Cell growth and microscopy. Cells were grown at 33°C and 300 rpm under permissive conditions (in the light) to an optical density at 550 nm (OD550) of ⬇0.5. Ten-microliter aliquots (⬃106 cells) were transferred to each of three separate culture flasks containing 10 ml of fresh CTT medium. One of the cultures was incubated in the light (permissive conditions), a second was incubated in the dark, and a third also was incubated in the dark but after the addition of B12 to 0.75 ␮⌴ (restrictive conditions). All three cultures were grown at 33°C and 300 rpm, and the OD550 was monitored periodically. Samples of 100 ␮l of each culture were taken at different time points during the incubation and mixed with the fluorescent dye 4⬘-6-diamino-2-phenylindole (DAPI; excitation maximum, 350 nm; emission maximum, 461 nm) to a final concentration of 1 ng/␮l. A 4-␮l drop of this cell suspension was placed on a glass plate coated with 3-aminopropyl-triethoxysilane to facilitate cell immobilization and then examined using a Nikon Eclipse 80i microscope equipped with a Nikon Plan Apo VC 100⫻ oil immersion objective (numerical aperture, 1.40) and a Hamamatsu ORCA-AG charge-coupled-device camera. Images were processed with Metamorph, version 4.5 (Universal Imaging Group).

RESULTS Design of a vitamin B12-based system for regulatable expression in M. xanthus. In the dark, PB is repressed by CarA or CarH-B12 binding to a bipartite operator that consists of a high-affinity site pI (a perfect interrupted palindrome located upstream of PB) and a low-affinity site pII (a pI-like pseudopalindrome that straddles the ⫺35 region of PB) (30). CarS, produced on illumination, relieves repression of PB by physically interacting with CarA or CarH to dismantle repressoroperator DNA complexes (Fig. 1) (29, 30, 44, 45). To design the vitamin B12-based system for conditional gene expression in M. xanthus, we first constructed a plasmid (pMR2991) with a reporter lacZ probe under the transcriptional control of a 121-bp DNA fragment that includes PB and its bipartite operator, a Tcr marker for selection, and a 1.38-kb DNA fragment for plasmid integration into the M. xanthus chromosome by homologous recombination. In addition, pMR2991 bears a cassette for constitutive expression of the CarH repressor (Fig. 2A) (see Materials and Methods for details); in its natural context, expression of carA and carH appears to be driven by a light-independent ␴54-dependent promoter as well as by transcriptional readthrough from PB in the light (11, 49). Plasmid pMR2991 was then introduced by electroporation into the M. xanthus MR1716 strain, which lacks both carA (whose product represses PB in a B12-independent manner) as well as the adjacent carH, to generate strain MR1461. The levels of reporter lacZ expression for MR1461 cells grown in the dark and in the light in the absence or presence of B12 were then assessed (Fig. 2B) (see also Materials and Methods). In the dark and with B12 present, ␤-galactosidase activity was very low, consistent with the role of B12 in mediating PB repression by CarH. Accordingly, ␤-galactosidase activity in the dark was about sixfold higher when B12 was absent. An additional 1.4fold increase in PB expression was observed for samples grown in the light, which induces production of the antirepressor CarS. The maximum level of reporter lacZ expression from PB was observed in the light with B12 also present. This stimulatory effect of B12 on the light activation of PB occurs even in strains lacking CarA and CarH (45). Although the reasons for this effect of B12 in the light remain to be deciphered, it pro-

FIG. 2. Regulated gene expression mediated by CarH and B12. (A) Schematic of plasmid pMR2991. The plasmid is introduced into the ⌬carA ⌬carH double-deleted M. xanthus strain MR1716, where it integrates into the M. xanthus chromosome by homologous recombination via the 1.38-kb DNA fragment to yield strain MR1461. (B) Expression of the reporter lacZ in MR1461 in the dark or in the light with or without 0.75 ␮M B12. Samples were obtained as described in Materials and Methods and then harvested for quantitative analysis of ␤-galactosidase activity (in nmol of o-nitrophenyl ␤-D-galactoside hydrolyzed/min/mg of protein). Mean values and errors are from two or more independent measurements.

vides an added level of modulating gene expression in this B12-regulated system. In sum, the results demonstrate the following for a gene (e.g., lacZ) under PB control: (i) that there is essentially no expression for growth in the dark with B12 present; (ii) that the expression is significantly greater for growth in the dark with no B12 added; and (iii) that expression levels can be further enhanced by light and even more so by including B12. The system thus provides a graduated ability to fine-tune expression of a candidate gene. Use of the B12-based system for conditional expression of the essential rpoN gene in M. xanthus. Keseler and Kaiser (25) showed that rpoN, which encodes ␴54, is an essential gene in M. xanthus based on the impossibility of obtaining rpoN null mutants using various strategies for gene knockout (direct disruption by fragment insertions, resolution of tandem gene duplications, or transduction of the null allele). The finding was also noteworthy in that no other member of the ␴54 family had been reported to be essential for growth until then. Moreover, it has been reported that many developmentally regulated genes in M. xanthus are expressed from ␴54-dependent promoters (20). Since conditional expression of rpoN has not been investigated, we reasoned that this would be a suitable test case for the applicability of the B12-based system that we have designed. To confirm that rpoN is essential for cell viability in M. xanthus, strain MR1716 was first electroporated with plasmid pMR3121, which bears a complete in-frame deletion of rpoN

3112


J. BACTERIOL.

FIG. 3. Schematic showing construction of a M. xanthus strain conditionally expressing rpoN. Plasmid pMR3120, which expresses the rpoN gene under the control of the PB promoter, is introduced into the merodiploid strain MR1859 (containing wild-type and deleted copies of rpoN in the ⌬carA ⌬carH double-deleted genetic background). Rectangles in gray and black indicate genomic regions upstream and downstream of the deleted gene, respectively. Integration of pMR3120 into the chromosome by homologous recombination via the 1.38-kb DNA fragment yields strain MR1867. From this, subsequent loss of the ⌬rpoN allele at the native site yields strain MR1871 (with two copies of rpoN, of which one is expressed from PB at an unlinked site), while loss of the wild-type rpoN allele at the native site yields strain MR1870 (with the ⌬rpoN allele at the native site and rpoN expressed from PB at the unlinked site).

(⌬rpoN allele) flanked by about 1 kb of the genomic DNA on either side of the gene for homologous recombination. Plasmid integration at the correct chromosomal locus to produce a merodiploid (strain MR1859) with the wild type as well as the ⌬rpoN allele was verified by PCR. The resulting Gals Kmr merodiploid was then electroporated with pMR3120 (Fig. 3), a plasmid that was constructed by cloning the rpoN gene and its RBS immediately downstream of the PB promoter in pMR2991, to generate strain MR1867 (with pMR3120 integrated by homologous recombination via the 1.38-kb DNA fragment, as verified by PCR). Haploids were generated by growing MR1867 for several generations in the absence of kanamycin and, in order to optimize rpoN expression from PB, in the light. Haploid Galr Kms cells arising from recombination between the two rpoN alleles were selected on plates containing galactose and grown in the light. A total of 133 recombinants were picked, and the presence of the wild-type or the ⌬rpoN allele at the native chromosomal locus was diagnosed by PCR. Out of these, 24 recombinants had the ⌬rpoN allele while the rest retained the wild-type allele. One of each type was

then selected for comparing the ability to grow on CTT plates in the light or in the dark and in the absence or the presence of B12. As shown in Fig. 4A, the recombinant bearing the wild-type rpoN allele (strain MR1871) grew equally well under the four conditions tested, and the red colony color in the light was the result of light-induced carotenogenesis. In marked contrast, the recombinant with the ⌬rpoN allele and an rpoN copy under PB control (strain MR1870) grew well in the light with or without B12, less so in the absence of B12 in the dark, and hardly at all when B12 was also present. Thus, if supplied with rpoN under the control of PB, a strain with the ⌬rpoN allele is viable when grown in the light but not when grown in the dark with B12 present. Cultures of MR1870 and MR1871 were grown under permissive conditions (light) and then shifted for subsequent growth in the light, in the dark with no B12, and in the dark with B12. The growth curve for MR1870 in the light (Fig. 4B) resembled curves obtained for MR1871 under the three growth conditions (data not shown). Consistent with the apparent lower growth rate of MR1870 on plates in the dark with

VOL. 191, 2009


3113

FIG. 4. PB-dependent expression of rpoN confirms it to be an essential gene. (A) Growth under the conditions indicated for M. xanthus strains with the wild-type rpoN allele (MR1871) or the ⌬rpoN allele (MR1870) at the native site and with an rpoN copy under PB control at an unlinked site. Cells grown on CTT plates under permissive conditions (light) were streaked on CTT plates with or without B12 and incubated at 33°C under the indicated conditions for 2 days. The cells are red in the light due to carotenoid synthesis. In the dark, cells are yellow when B12 is present, but yellow gradually turns orange when B12 is absent (due to expression of the carB operon). Note that there is essentially no growth in the dark when B12 is present for strain MR1870. (B) Growth curves for MR1870 in the light (filled circles) and in the dark with no B12 (open circles) or with B12 (filled triangles), obtained as described in Materials and Methods. The abscissa indicates time (in hours) after the shift in growth conditions. (C) Cellular morphology of M. xanthus strains MR1870 and MR1871, under the conditions indicated (permissive, light; restrictive, dark and B12). Cells were examined at an OD550 of ⬃0.8 by DIC and fluorescence microscopy after DAPI staining to visualize the nucleoid, as described in Materials and Methods. Note the aberrant cell division phenotype of MR1870 on shifting to restrictive conditions. Bar, 5 ␮m.

no B12 (Fig. 4A), liquid cultures were also observed to grow more slowly in the dark than in the light (Fig. 4B). The presence of B12 not only caused MR1870 to grow even more slowly in the dark but also arrested growth at a significantly lower OD550 relative to that in the light or in the dark without B12 (Fig. 4B). We examined the cellular morphology of MR1870 under each of these conditions by differential interference contrast (DIC) and fluorescence microscopy of cells stained with DAPI to visualize the nucleoid and compared it to that of MR1871. As can be seen in Fig. 4C, both strains exhibited

normal cell lengths (⬃4 to 8 ␮m) under permissive conditions (light), with cells containing one or two nucleoids at the midcell section. A similar, normal cellular morphology characterized MR1870 grown in the dark with no B12 present (data not shown), despite the slower-growth phenotype observed. By contrast, on shifting to restrictive conditions (dark with B12), MR1870 produced cells that were ⬎10 ␮m (about 48% of the total examined), with some cells reaching lengths of up to ⬃35 ␮m (Fig. 4C, inset). These abnormally long, filamentous cells appeared to contain several DAPI-stained foci distributed

3114


J. BACTERIOL.

FIG. 5. Schematic showing construction of a M. xanthus strain with ftsZ expressed from promoters PB or P330. (A) Plasmid pMR3168, with the ftsZ gene under the control of PB, is electroporated into the merodiploid strain MR1860 (containing wild-type and deleted copies of ftsZ in the ⌬carA ⌬carH genetic background). Rectangles in gray and black indicate genomic regions upstream and downstream of the deleted gene, respectively. Integration of pMR3168 into the chromosome at a heterologous site via recombination through the 1.38-kb DNA fragment generates strain MR1862, which was grown in the light and in the absence of kanamycin to generate Galr Kms recombinants. These latter were analyzed for the presence of the wild-type or deleted ftsZ allele. (B) Plasmid pMR3202 is similar to pMR3168 in panel A except that P330 (see text) replaces PB, and the carH expression cassette is absent. As in panel A, it was electroporated into strain MR1860 and the Tcr transformants were grown in the absence of kanamycin to generate Galr Kms recombinants, which were then analyzed for the presence of the wild-type or deleted ftsZ allele.

along the cell that were less distinct and well defined than those observed for MR1870 under permissive conditions or for MR1871 under restrictive conditions. This aberrant cell division phenotype that we observe on limiting rpoN expression in M. xanthus is reminiscent of that reported in Caulobacter crescentus when rpoN (not essential in this bacterium) is disrupted (5). In this latter case, it appears to be a consequence of the rpoN disruption mutant’s being impaired in a global regulatory mechanism that links ␴54-dependent flagellar biogenesis to cell division. In sum, our results show that the B12-based system engineered here is suitable for conditionally expressing rpoN, whose limited expression is detrimental to M. xanthus growth, as would be expected for an essential gene. This prompted us to test the applicability of this system for some other M. xanthus genes, as discussed next.

Analysis of the M. xanthus ftsZ gene. The highly conserved cell division protein FtsZ is vital in most bacteria including E. coli, Bacillus subtilis, and C. crescentus (2, 8, 56), with Streptomyces coelicolor being one known exception (36). We tested if M. xanthus ftsZ is also essential and whether it could be conditionally expressed using our B12-based system. For this, we used the same strategy as described above for rpoN and depicted in Fig. 3. Plasmid pMR3165, with a complete in-frame deletion of ftsZ (the ⌬ftsZ allele), was used to generate strain MR1860, a Gals Kmr merodiploid resulting from integration of the plasmid at the correct chromosomal locus in MR1716, as verified by PCR. MR1860 was then electroporated with pMR3168, which expresses ftsZ under the control of PB (Fig. 5A). Plasmid integration at the heterologous site (via recombination at the 1.38-kb M. xanthus DNA fragment) generated

VOL. 191, 2009


strain MR1862. After MR1862 was grown for several generations in the absence of kanamycin under permissive conditions, 200 Galr Kms recombinants obtained were picked. All of these grew well under restrictive conditions, and PCR analysis of 48 of them did not reveal any recombinant that carried the ⌬ftsZ allele. In bacteria where it has been studied, FtsZ is typically present in high amounts in vivo, and such elevated intracellular levels are presumably required for viability of actively dividing cells (10, 31, 32, 35). The inability to obtain recombinants with the ⌬ftsZ allele in M. xanthus could therefore be a consequence of the levels of FtsZ produced, when PB drives its expression, being lower than those required for viability. This possibility was supported by the following experiment. We first constructed pMR3202 (Fig. 5B), which contains ftsZ together with 0.33 kb of intergenic DNA (P330) immediately preceding ftsZ in the genome. The latter has promoter activity, as inferred from expression of the reporter lacZ gene in pMR3202, which was about fivefold higher than the maximum levels observed when expressed from PB (data not shown). Strain MR1860 was then electroporated with pMR3202, whose chromosomal integration at a heterologous site (via recombination at the 1.38-kb M. xanthus DNA fragment) generated strain MR1866. The latter was grown for several generations in the absence of kanamycin, and the resulting Galr Kms recombinants were analyzed by PCR for the presence of the wild-type or ⌬ftsZ allele at the natural site. In this case, out of the 15 haploid colonies that we checked by PCR, 11 were found to contain the ⌬ftsZ allele. Thus, the natural ftsZ allele can be deleted in a merodiploid background in which a second copy of ftsZ is expressed from P330 but not PB, consistent with the fivefold higher expression from P330. Thus, we found that the B12-based system for conditional expression of ftsZ is likely insufficient to provide the higher intracellular levels of FtsZ required and that, as with its counterparts in most other bacteria, ftsZ appears to be an essential gene in M. xanthus. The M. xanthus dksA gene. The finding that rpoN is required for growth only under specific nutrient-limiting conditions in most bacteria but is essential in M. xanthus and the implication of DksA in the regulation of ␴54-dependent transcription in vivo prompted us to examine dksA, whose counterparts in other bacteria have a role in response to starvation conditions. As noted before, E. coli dksA is not an essential gene, but a strain lacking it grows poorly on nutrient limitation (24). In E. coli, DksA regulates the stringent response by enhancing the inhibitory effects of (p)ppGpp (GDP 3⬘-diphosphate and its pentaphosphate derivative) on rRNA transcription and by activating promoters for amino acid biosynthesis and transport (18, 40, 41, 43). Accumulation of (p)ppGpp in response to amino acid starvation has been shown to initiate the early gene expression leading to fruiting body formation in M. xanthus (17). The role of DksA in the M. xanthus life cycle, however, remains to be examined. A protein BLAST (http://www.ncbi.nlm.nih.gov/BLAST) search of the M. xanthus genome (13) with E. coli DksA as query reveals four proteins with predicted zinc-binding motifs (made up of Cys) of the type found in the DksA/TraR family. These four proteins (annotated as MXAN_3200, MXAN_5718, MXAN_3006, and MXAN_7086 with NCBI accession codes YP_631401, YP_633856, YP_631215, and YP_635199, respectively) are 120, 130, 130, and 101 residues long, respectively, compared to the

3115

151-residue E. coli DksA (Fig. 6A). We assigned the 120-residue protein as DksA based on the observations that its sequence shares the maximum identity/similarity (37%/58%) to E. coli DksA and that it conserves all four Cys and the two Asp residues implicated in function (Fig. 6A) (40, 43). A putative promoter with a consensus ⫺35 TTGACA hexamer and a less-conserved TGAAGC at the ⫺10 position can be identified 60 bp upstream of the start codon of dksA, which appears to be expressed as an independent unit with both genes flanking it transcribed in the opposite sense. The one immediately upstream of dksA encodes a putative uncharacterized protein while the gene downstream expresses a protein with a forkhead-associated domain (a phosphoprotein recognition module with diverse roles in cellular physiology) (13, 20, 39). Further downstream of dksA are genes encoding a serine/threonine protein kinase and, more notably given the link between DksA function and the stringent response, a pppGpp synthetase. DksA is an acidic protein (for example, the theoretical pI is 4.9 for E. coli DksA and 5.04 for Pseudomonas aeruginosa DksA) that has been shown to contain zinc and to interact directly with native RNAP in vitro (40, 46). The protein that we assigned as DksA is also acidic (theoretical pI of 5.22). Moreover, after removal of the His6 tag, purified DksA was confirmed to contain zinc using the colorimetric PAR assay, as described in Materials and Methods (data not shown). RNAP retention at low salt and elution at high salt by His6-DksA immobilized on a metal affinity resin has been employed to demonstrate direct binding of E. coli DksA to RNAP in vitro (40). Using this pull-down assay, we also observed that His6DksA retained RNAP (Fig. 6B). Thus, the protein that we assigned as DksA binds to zinc and RNAP, as has been reported for its counterparts in other bacteria. dksA is an essential gene in M. xanthus. To assess the importance of dksA function in M. xanthus, we set out to generate a strain with a complete deletion of the gene. For this, we generated plasmid pMR3254, which bears a complete in-frame deletion of the dksA gene (⌬dksA allele) flanked by genomic DNA on either side of the gene for homologous recombination (0.85 kb and 1 kb of DNA upstream and downstream of dksA, respectively). Plasmid pMR3254 was introduced by electroporation into the wild-type strain DK1050. A Gals Kmr merodiploid, strain MR1933, resulting from integration of the plasmid at the correct chromosomal locus (as verified by PCR) was picked. Strain MR1933 was then grown for several generations in the absence of kanamycin, and haploid cells arising from recombination between the DNA segments flanking the wildtype and ⌬dksA alleles were selected on plates containing galactose. Out of 86 recombinants diagnosed by PCR, none was found to have the ⌬dksA allele, suggesting that dksA is an essential gene in M. xanthus. Conditional expression using the B12-dependent system described here was used to further confirm if dksA is indeed an essential gene. Strain MR1716 was first electroporated with plasmid pMR3254 to generate a merodiploid with the wildtype and ⌬dksA alleles in the ⌬carA ⌬carH genetic background. Such a merodiploid was then electroporated with plasmid pMR3085, a derivative of pMR2991 (Fig. 2) that was constructed by cloning the dksA gene and its RBS immediately downstream of the PB promoter. Haploids were generated by growing the resulting strain for several generations in the ab-

3116


J. BACTERIOL.

FIG. 6. Analysis of M. xanthus DksA. (A) Sequence alignment of E. coli DksA (NCBI accession code NP_414687) and its sequence homologs in M. xanthus (MXAN_3200, NCBI accession code YP_631401; MXAN_5718, NCBI accession code YP_633856; MXAN_3006, NCBI accession code YP_631215; MXAN_7086, NCBI accession code YP_635199). In the alignment, identical residues are shaded black (with an asterisk in the consensus line), and similar residues are in gray. Black arrows indicate the four conserved zinc-binding Cys residues and the unfilled arrows are the two Asp residues implicated in function. (B) Analysis of RNAP retention by His6-DksA immobilized on TALON metal affinity resin. Shown is a silver-stained 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel of low-salt (50 mM) and high-salt (500 mM) eluates on passing M. xanthus core RNAP through a column of Talon resin alone (Control) or that bound to His6-DksA. The ␣, ␤, and ␤⬘ RNAP subunits are indicated. (C) Growth on CTT plates, under the conditions indicated, for M. xanthus strains with the wild-type dksA allele (MR1925) or the ⌬dksA allele (MR1926) at the native site and a dksA copy under PB control at an unlinked site. Cells grown on CTT plates under permissive conditions (light) were streaked on CTT plates with or without B12 and incubated at 33°C under the indicated conditions for 2 days. Both strains appear red in the light due to carotenoid synthesis. In the dark, MR1925 is yellow if B12 is present, but yellow gradually turns to orange when B12 is absent (due to expression of the carB operon). MR1926 in the dark appears to have lost the yellow, suggesting a link between DksA depletion and impaired yellow pigment synthesis. Note that there is virtually no growth in the dark with B12 present for strain MR1926.

sence of kanamycin and in the light for optimal expression of dksA from PB. Haploid cells arising from recombination between the two dksA alleles were selected on plates containing galactose and grown in the light. A total of 87 recombinants were picked, and the presence of the wild-type or the ⌬dksA allele at the native chromosomal locus was diagnosed by PCR. Out of these, 17 had the ⌬dksA allele while the rest retained

the wild-type allele. One of each type was then selected, and the ability to grow on CTT plates in the light or in the dark, with and without B12, was compared. As shown in Fig. 6C, the recombinant bearing the wild-type dksA allele (strain MR1925) grew equally well under all four growth conditions tested. In marked contrast, the recombinant with the ⌬dksA allele (strain MR1926) grew well in the light with or without


VOL. 191, 2009

3117

FIG. 7. Cellular morphology of M. xanthus strains MR1925 and MR1926 under permissive (light) and restrictive (dark with B12) conditions. Cells were examined at an OD550 of ⬃0.8 by DIC and fluorescence microscopy after DAPI staining to visualize the nucleoid, as described in Materials and Methods. Arrows point to regions depleted in DAPI fluorescence in abnormally long MR1926 cells observed on shifting to restrictive growth conditions (dark and B12). Scale bar, 5 ␮m.

B12, less in the absence of B12 in the dark, and very poorly, if at all, in the dark with B12 present. Thus, if provided with dksA under PB control, a strain with the ⌬dksA allele is viable when grown in the light but not when grown in the dark with B12 present. Moreover, these data also indicate a lack of functional redundancy between the gene that we have assigned to be dksA and any of the other three that code for the similar proteins shown in Fig. 6A. Examination of cellular morphology by DIC and fluorescence microscopy after DAPI staining of the cells showed that both MR1925 and MR1926 grew normally under permissive conditions to produce similar, rod-shaped cells of normal lengths (⬃4 to 8 ␮m), with one or two nucleoids discernible at the mid-cell section (Fig. 7). However, on shifting to restrictive conditions, MR1926 displayed an aberrantly dividing cellular morphology. In this case, about 56% of the cells examined had cell lengths of ⬎10 ␮m, with about half of these being ⬎15 ␮m. Moreover, a skewed distribution of the DAPI-stained foci to one portion of the cell was frequently observed in these longer cells. Interestingly, a filamentous phenotype has also been reported with a ⌬dksA E. coli strain (34) even though dksA is not essential in this bacterium, as noted earlier (24). Thus, consistent with the finding that dksA is indispensable in M. xanthus, limiting its expression is deleterious to M. xanthus growth, and this directly or indirectly affects cell division. DISCUSSION B12-dependent conditional expression in M. xanthus. The ability to change the expression levels of a given gene using a tightly regulated promoter cassette allows the study of a con-

ditional mutant phenotype for essential genes simply by turning off their expression. Controlled gene expression systems that have been developed for use in bacteria include those based on effectors such as short aliphatic amides like acetamide (10) or drugs like the antibiotic anhydrotetracycline (33), but among the regulatory switches most widely used are promoters that control expression of genes and operons involved in the utilization of carbon sources. In these, promoter activity is modulated by repressors or activators in response to the availability of sugars such as lactose, arabinose, or xylose; the IPTG-inducible promoter for the lac operon is possibly the best known such system for regulating the expression of a gene of interest (9, 15, 26, 33, 37). There is a dearth of conditional expression systems available for use in the myxobacteria, and the one lac-based system described for M. xanthus has the drawback of a relatively high level of basal expression under noninduced conditions, thereby limiting its use for the study of essential genes (19). This report describes a system for conditional gene expression in M. xanthus whose modulation depends on a novel effector: vitamin B12. In this system, the candidate gene is placed under the control of the PB promoter which is repressed in the dark by the CarH repressor when vitamin B12 is present. Incremental levels of gene expression can be achieved by the absence of B12 in the dark, by illumination with no B12 present, or by exposure to light in the presence of B12, in that order. An important prerequisite for studying essential genes and their cellular roles is that their expression can be completely turned off, as required, by the tightly regulated promoter under whose control they are placed. In this respect, the B12-based system does repress the expression of a candidate gene to

3118


marginal levels under restrictive conditions. Thus, the system allowed us to confirm that M. xanthus rpoN is essential, as was previously reported based on different experimental criteria (25). Additionally, our system has revealed that the gene we assigned as dksA in M. xanthus is also essential for viability. This system has also revealed that a gene coding for a protein of unknown function, whose details will be reported elsewhere, is an essential gene in M. xanthus (García-Moreno et al., unpublished data). A fourth gene in M. xanthus that we examined, ftsZ, was concluded to be essential based on the observation that the natural ftsZ allele could be disrupted only in a merodiploid background carrying another functional ftsZ copy. However, expression of ftsZ from PB in the B12-based system, in the absence of the gene at its native site, did not allow cell survival. Our data suggest that this may be because PB-driven expression of ftsZ produces lower than the normal amounts required for viability of M. xanthus. Thus, for the B12-based system to be broadly applicable, it will be necessary to engineer PB promoter strength for greater expression levels of target genes on activation without compromising low expression under conditions of repression. This remains an objective for future studies and would certainly benefit from our ongoing work aimed at understanding the molecular basis of how B12 and CarH act jointly in repressing PB. The applicability of the B12-based system to conditionally express genes involved in M. xanthus development remains to be examined. With this system, achieving maximal levels of expression requires illumination, but light has been reported to inhibit sporulation though not fruiting body formation (28). Since a candidate gene can also be expressed in the dark with no B12 present, the need for light could be dispensed with, especially once a PB of optimal promoter strength is engineered. PB and the promoter from which CarH is expressed in the B12-based system are both ␴A dependent. Since ␴A levels have been shown to persist at a high level until 12 h into development (16), our B12-based system could be applicable at least for the genes expressed early in development. Why might dksA be essential in M. xanthus? The choice of dksA as a candidate gene for study was motivated by its functional link to ␴54-dependent transcription and the involvement of both in the response to nutrient-limiting conditions that is observed in bacteria such as E. coli. In contrast to other bacteria, rpoN coding for ␴54 is a vital gene in M. xanthus (25), and it is not entirely surprising that we found dksA also to be essential, given the functional link between the two genes. A third example of a gene known to be dispensable in many other bacteria but not in M. xanthus is lonV, which encodes an ATPdependent protease (53). This is striking, given that a second gene, lonD, that encodes a very similar product also exists in M. xanthus (12, 54). Both lonV and lonD are expressed during vegetative growth and development, but while lonV is essential under all conditions of growth, lonD is required only for multicellular development (12, 53, 54). The functions underpinning the essential nature of rpoN and lonV in M. xanthus remain to be elucidated, and an in-depth experimental analysis of this for dksA is also outside the scope of the present study. A common feature of rpoN, lonV, and dksA is that they are implicated in many functions, and their absence causes pronounced pleiotropic phenotypes (5, 6, 14, 18, 24, 51, 57). Moreover, even in bacteria where these genes can be deleted with-

J. BACTERIOL.

out impairing cell viability, growth in rich medium is required, and they are indispensable under specific, nutrient-limiting conditions. Given that dksA, like rpoN, is implicated in regulation of transcription, the essential nature of dksA and rpoN may stem from controlling expression of a gene essential for M. xanthus viability. For instance, ␴54-dependent promoters require, in addition to ␴54-RNAP holoenzyme, a specialized transcription factor called an enhancer-binding activator protein (EBP) that usually binds to regulatory DNA sequences upstream of the promoter (6). Analysis of the M. xanthus genome indicates that it has an unusually high number of EBP-encoding genes (52 such putative EBP genes have been identified [20]). A number of these EBPs have been implicated in development, heat shock response, and motility, but the functions of many others remain to be discovered. The high abundance of these EBPs (possibly the largest number found so far in bacteria) suggests a mode of regulation that is very likely critical in the life style of M. xanthus. It also implies that ␴54 is at the nexus of a large regulatory network, of which at least some parts may be essential for viability. Likewise, dksA may be a vital part of a chain of transcriptional control. Studies in numerous other bacteria have also implicated DksA in chaperone activity, DNA replication, recombination and repair, cell division, quorum sensing, responses to envelope stress, and a number of other critical cellular processes in concert with or independent of (p)ppGpp (18, 34, 51, 55), and our data indicate that the progressive depletion of DksA (or ␴54) in M. xanthus results in aberrant cell division. Thus, the direct or indirect role in any of the essential processes attributed to dksA may underlie its vital nature in M. xanthus. ACKNOWLEDGMENTS We thank Rick Gourse (University of Wisconsin-Madison) for useful discussions on DksA and for a short stay in his laboratory by one of us (G.N.A.). We also thank J. A. Madrid for technical assistance and C. Flores for DNA sequencing. This work was supported by MEC (Spain) grants BFU2006-14524 (to M.E.A.) and BFU2005-01040 (to S.P.) and FPI fellowships (D.G.M. and G.N.A.). REFERENCES 1. Balsalobre, J. M., R. M. Ruiz-Va ´zquez, and F. M. Murillo. 1987. Light induction of gene expression in Myxococcus xanthus. Proc. Natl. Acad. Sci. USA 84:2359–2362. 2. Beall, B., and J. Lutkenhaus. 1991. FtsZ in Bacillus subtilis is required for vegetative septation and for asymmetric septation during sporulation. Genes Dev. 5:447–455. 3. Bernardo, L. M., L. U. Johansson, D. Solera, E. Skarfstad, and V. Shingler. 2006. The guanosine tetraphosphate (ppGpp) alarmone, DksA and promoter affinity for RNA polymerase in regulation of ␴54-dependent transcription. Mol. Microbiol. 60:749–764. 4. Bretscher, A. P., and D. Kaiser. 1978. Nutrition of Myxococcus xanthus, a fruiting myxobacterium. J. Bacteriol. 133:763–768. 5. Brun, Y. V., and L. Shapiro. 1992. A temporally controlled sigma-factor is required for polar morphogenesis and normal cell division in Caulobacter. Genes Dev. 6:2395–2408. 6. Buck, M., M. T. Gallegos, D. J. Studholme, Y. Guo, and J. D. Gralla. 2000. The bacterial enhancer-dependent ␴54 (␴N) transcription factor. J. Bacteriol. 182:4129–4136. 7. Crawford, E. W., Jr., and L. J. Shimkets. 2000. The stringent response in Myxococcus xanthus is regulated by SocE and the CsgA C-signaling protein. Genes Dev. 14:483–492. 8. Dai, K., and J. Lutkenhaus. 1991. ftsZ is an essential cell division gene in Escherichia coli. J. Bacteriol. 173:3500–3506. 9. de Boer, H. A., L. J. Comstock, and M. Vasser. 1983. The tac promoter: a functional hybrid derived from the trp and lac promoters. Proc. Natl. Acad. Sci. USA 80:21–25.

VOL. 191, 2009


10. Dziadek, J., S. A. Rutherford, M. V. Madiraju, M. A. Atkinson, and M. Rajagopalan. 2003. Conditional expression of Mycobacterium smegmatis ftsZ, an essential cell division gene. Microbiology 149:1593–1603. 11. Elías-Arnanz, M., M. Fontes, and S. Padmanabhan. 2008. Carotenogenesis in Myxococcus xanthus: a complex regulatory network, p. 211–225. In D. E. Whitworth (ed.), Myxobacteria: multicellularity and differentiation. ASM Press, Washington, DC. 12. Gill, R. E., M. Karlok, and D. Benton. 1993. Myxococcus xanthus encodes an ATP-dependent protease which is required for developmental gene transcription and intercellular signaling. J. Bacteriol. 175:4538–4544. 13. Goldman, B. S., W. C. Nierman, D. Kaiser, S. C. Slater, A. S. Durkin, J. A. Eisen, C. M. Ronning, W. B. Barbazuk, M. Blanchard, C. Field, C. Halling, G. Hinkle, O. Iartchuk, H. S. Kim, C. Mackenzie, R. Madupu, N. Miller, A. Shvartsbeyn, S. A. Sullivan, M. Vaudin, R. Wiegand, and H. B. Kaplan. 2006. Evolution of sensory complexity recorded in a myxobacterial genome. Proc. Natl. Acad. Sci. USA 103:15200–15205. 14. Gottesman, S. 1996. Proteases and their targets in Escherichia coli. Annu. Rev. Genet. 30:465–506. 15. Guzman, L. M., D. Belin, M. J. Carson, and J. Beckwith. 1995. Tight regulation, modulation, and high-level expression by vectors containing the arabinose PBAD promoter. J. Bacteriol. 177:4121–4130. 16. Hao, T., D. Biran, G. J. Velicer, and L. Kroos. 2002. Identification of the ⍀4514 regulatory region, a developmental promoter of Myxococcus xanthus that is transcribed in vitro by the major vegetative RNA polymerase. J. Bacteriol. 184:3348–3359. 17. Harris, B. Z., D. Kaiser, and M. Singer. 1998. The guanosine nucleotide (p)ppGpp initiates development and A-factor production in M. xanthus. Genes Dev. 12:1022–1035. 18. Haugen, S. P., W. Ross, and R. L. Gourse. 2008. Advances in bacterial promoter recognition and its control by factors that do not bind DNA. Nat. Rev. Microbiol. 6:507–519. 19. Jelsbak, L., and D. Kaiser. 2005. Regulating pilin expression reveals a threshold for S motility in Myxococcus xanthus. J. Bacteriol. 187:2105–2112. 20. Jelsbak, L., M. Givskov, and D. Kaiser. 2005. Enhancer-binding proteins with a forkhead-associated domain and the ␴54 regulon in Myxococcus xanthus fruiting body development. Proc. Natl. Acad. Sci. USA 102:3010–3015. 21. Julien, B., D. Kaiser, and A. Garza. 2000. Spatial control of cell differentiation in Myxococcus xanthus. Proc. Natl. Acad. Sci. USA 97:9098–9103. 22. Kaiser, D. 2003. Coupling cell movement to multicellular development in myxobacteria. Nat. Rev. Microbiol. 1:45–54. 23. Kaiser, D. 2004. Signaling in myxobacteria. Annu. Rev. Microbiol. 58:75–98. 24. Kang, P. J., and E. Craig. 1990. Identification and characterization of a new Escherichia coli gene that is a dosage-dependent suppressor of a dnaK deletion mutation. J. Bacteriol. 172:2055–2064. 25. Keseler, I. M., and D. Kaiser. 1997. ␴54, a vital protein for Myxococcus xanthus. Proc. Natl. Acad. Sci. USA 94:1979–1984. 26. Kim, L., A. Mogk, and W. Schumann. 1996. A xylose-inducible Bacillus subtilis integration vector and its application. Gene 181:71–76. 27. Letouvet-Pawlak, B., C. Monnier, S. Barray, D. A. Hodgson, and J. F. Guespin-Michel. 1990. Comparison of ␤-galactosidase production by two inducible promoters in Myxococcus xanthus. Res. Microbiol. 141:425–435. 28. Li, S., B. U. Lee, and L. Shimkets. 1992. csgA expression entrains Myxococcus xanthus development. Genes Dev. 6:401–410. 29. Lo ´pez-Rubio, J. J., M. Elías-Arnanz, S. Padmanabhan, and F. J. Murillo. 2002. A repressor-antirepressor pair links two loci controlling light-induced carotenogenesis in Myxococcus xanthus. J. Biol. Chem. 277:7262–7270. 30. Lo ´pez-Rubio, J. J., S. Padmanabhan, J. M. La ´zaro, M. Salas, F. J. Murillo, and M. Elías-Arnanz. 2004. Operator design and mechanism for CarA repressor-mediated down-regulation of the photoinducible carB operon in Myxococcus xanthus. J. Biol. Chem. 279:28945–28953. 31. Lu, C., J. Stricker, and H. P. Erickson. 1998. FtsZ from Escherichia coli, Azotobacter vinelandii, and Thermotoga maritime—quantitation, GTP hydrolysis, and assembly. Cell Motil. Cytoskel. 40:71–86. 32. Lutkenhaus, J., and S. G. Addinall. 1997. Bacterial cell division and the Z ring. Annu. Rev. Biochem. 66:93–116. 33. Lutz, R., and H. Bujard. 1997. Independent and tight regulation of transcriptional units in Escherichia coli via the LacR/O, the TetR/O and AraC/ I1–I2 regulatory elements. Nucleic Acids Res. 25:1203–1210. 34. Magnusson, L. U., B. Gummesson, P. Joksimovic´, A. Farewell, and T. Nystro ¨m. 2007. Identical, independent, and opposing roles of ppGpp and DksA in Escherichia coli. J. Bacteriol. 189:5193–5202. 35. Margolin, W. 2005. FtsZ and the division of prokaryotic cells and organelles. Nat. Rev. Mol. Cell Biol. 6:862–871. 36. McCormick, J. R., E. P. Su, A. Driks, and R. Losick. 1994. Growth and viability of Streptomyces coelicolor mutant for the cell division gene ftsZ. Mol. Microbiol. 14:243–254. 37. Meisenzahl, A. C., L. Shapiro, and U. Jenal. 1997. Isolation and character-

38.

39.

40.

41.

42.

43.

44.

45.

46.

47. 48.

49.

50.

51. 52.

53.

54.

55.

56. 57.

58.

59.

3119

ization of a xylose-dependent promoter from Caulobacter crescentus. J. Bacteriol. 179:592–600. Navarro-Avile´s, G., M. A. Jimeńez, M. C. Pe´rez-Marín, C. Gonza ´lez, M. Rico, F. J. Murillo, M. Elías-Arnanz, and S. Padmanabhan. 2007. Structural basis for operator and antirepressor recognition by Myxococcus xanthus CarA repressor. Mol. Microbiol. 63:980–994. Pallen, M., R. Chaudhuri, and A. Khan. 2002. Bacterial FHA domains: neglected players in the phospho-threonine signalling game? Trends Microbiol. 10:556–563. Paul, B. J., M. M. Barker, W. Ross, D. A. Schneider, C. Webb, J. W. Foster, and R. L. Gourse. 2004. DksA: a critical component of the transcription initiation machinery that potentiates the regulation of rRNA promoters by ppGpp and the initiating NTP. Cell 118:311–322. Paul, B. J., M. B. Berkmen, and R. L. Gourse. 2005. DksA potentiates direct activation of amino acid promoters by ppGpp. Proc. Natl. Acad. Sci. USA 102:7823–7828. Pen ˜ alver-Mellado, M., F. García-Heras, S. Padmanabhan, D. GarcíaMoreno, F. J. Murillo, and M. Elías-Arnanz. 2006. Recruitment of a novel zinc-bound transcriptional factor by a bacterial HMGA-type protein is required for regulating multiple processes in Myxococcus xanthus. Mol. Microbiol. 61:910–926. Perederina, A., V. Svetlov, M. N. Vassylyeva, T. H. Tahirov, S. Yokoyama, I. Artsimovitch, and D. G. Vassylyev. 2004. Regulation through the secondary channel-structural framework for ppGpp-DksA synergism during transcription. Cell 118:297–309. Pe´rez-Marín, M. C., J. J. Lo ´pez-Rubio, F. J. Murillo, M. Elías-Arnanz, and S. Padmanabhan. 2004. The N-terminus of Myxococcus xanthus CarA repressor is an autonomously folding domain that mediates physical and functional interactions with both operator DNA and antirepressor protein. J. Biol. Chem. 279:33093–33103. Pe´rez-Marín, M. C., S. Padmanabhan, M. C. Polanco, F. J. Murillo, and M. Elías-Arnanz. 2008. Vitamin B12 partners the CarH repressor to downregulate a photoinducible promoter in Myxococcus xanthus. Mol. Microbiol. 67: 804–819. Perron, K., R. Comte, and C. van Delden. 2005. DksA represses ribosomal gene transcription in Pseudomonas aeruginosa by interacting with RNA polymerase on ribosomal promoters. Mol. Microbiol. 56:1087–1102. Rose, R. E. 1988. The nucleotide sequence of pACYC184. Nucleic Acids Res. 16:355. Ruiz-Va ´zquez, R. M., and F. J. Murillo. 1984. Abnormal motility and fruiting behavior of Myxococcus xanthus bacteriophage-resistant strains induced by a clear-plaque mutant of bacteriophage Mx8. J. Bacteriol. 160:818–821. Ruiz-Va ´zquez, R. M., M. Fontes, and F. J. Murillo. 1993. Clustering and co-ordinated activation of carotenoid genes in Myxococcus xanthus by blue light. Mol. Microbiol. 10:25–34. Singer, M., and D. Kaiser. 1995. Ectopic production of guanosine penta- and tetraphosphate can initiate early developmental gene expression in Myxococcus xanthus. Genes Dev. 9:1633–1644. Srivatsan, A., and J. D. Wang. 2008. Control of bacterial transcription, translation and replication by (p)ppGpp. Curr. Opin. Microbiol. 11:100–105. Szalewska-Palasz, A., L. U. Johansson, L. M. Bernardo, E. Ska ¨rfstad, E. Stec, K. Bra ¨nnstro ¨m, and V. Shingler. 2007. Properties of RNA polymerase bypass mutants: implications for the role of ppGpp and its co-factor DksA in controlling transcription dependent on ␴54. J. Biol. Chem. 282:8046–18056. Tojo, N., S. Inouye, and T. Komano. 1993. Cloning and nucleotide sequence of the Myxococcus xanthus lon gene: indispensability of lon for vegetative growth. J. Bacteriol. 175:2271–2277. Tojo, N., S. Inouye, and T. Komano. 1993. The lonD gene is homologous to the lon gene encoding an ATP-dependent protease and is essential for the development of Myxococcus xanthus. J. Bacteriol. 175:4545–4549. Trautinger, B. W., R. P. Jaktaji, E. Rusakova, and R. G. Lloyd. 2005. RNA polymerase modulators and DNA repair activities resolve conflicts between DNA replication and transcription. Mol. Cell 19:247–258. Wang, Y., B. D. Jones, and Y. V. Brun. 2001. A set of ftsZ mutants blocked at different stages of cell division in Caulobacter. Mol. Microbiol. 40:347–360. Wright, R., C. Stephens, G. Zweiger, L. Shapiro, and M. R. Alley. 1996. Caulobacter Lon protease has a critical role in cell-cycle control of DNA methylation. Genes Dev. 10:1532–1542. Zhao, J. Y., L. Zhong, M. J. Shen, Z. J. Xia, Q. X. Cheng, X. Sun, G. P. Zhao, Y. Z. Li, and Z. J. Qin. 2008. Discovery of an autonomously replicating plasmid pMF1 from Myxococcus fulvus and the development of a gene cloning system in Myxococcus xanthus. Appl. Environ. Microbiol. 74:1980– 1987. Zusman, D. R., A. E. Scott, Z. Yang, and J. R. Kirby. 2007. Chemosensory pathways, motility and development in Myxococcus xanthus. Nat. Rev. Microbiol. 5:862–872.