ABSTRACT
In the bacterium Myxococcus xanthus, starvation triggers the formation of multicellular fruiting bodies containing thousands of stress-resistant spores. Recent work showed that fruiting body development is regulated by a cascade of transcriptional activators called enhancer binding proteins (EBPs). The EBP Nla6 is a key component of this cascade; it regulates the promoters of other EBP genes, including a downstream-functioning EBP gene that is crucial for sporulation. In recent expression studies, hundreds of Nla6-dependent genes were identified, suggesting that the EBP gene targets of Nla6 may be part of a much larger regulon. The goal of this study was to identify and characterize genes that belong to the Nla6 regulon. Accordingly, a direct repeat [consensus, C(C/A)ACGNNGNC] binding site for Nla6 was identified using in vitro and in vivo mutational analyses, and the sequence was subsequently used to find 40 potential developmental promoter (88 gene) targets. We showed that Nla6 binds to the promoter region of four new targets (asgE, exo, MXAN2688, and MXAN3259) in vitro and that Nla6 is important for their normal expression in vivo. Phenotypic studies indicate that all of the experimentally confirmed targets of Nla6 are primarily involved in sporulation. These targets include genes involved in transcriptional regulation, cell-cell signal production, and spore differentiation and maturation. Although sporulation occurs late in development, all of the developmental loci analyzed here show an Nla6-dependent burst in expression soon after starvation is induced. This finding suggests that Nla6 starts preparing cells for sporulation very early in the developmental process.
IMPORTANCE Bacterial development yields a remarkable array of complex multicellular forms. One such form, which is commonly found in nature, is a surface-associated aggregate of cells known as a biofilm. Mature biofilms are structurally complex and contain cells that are highly resistant to antibacterial agents. When starving, the model bacterium Myxococcus xanthus forms a biofilm containing a thin mat of cells and multicellular structures that house a highly resistant cell type called a myxospore. Here, we identify the promoter binding site of the transcriptional activator Nla6, identify genes in the Nla6 regulon, and show that several of the genes in the Nla6 regulon are important for production of stress-resistant spores in starvation-induced M. xanthus biofilms.
INTRODUCTION
Bacterial development yields a remarkable array of complex multicellular forms. Arguably, one of the most interesting forms of bacterial multicellularity is a surface-associated aggregate of cells known as a biofilm (1). Mature biofilms are complex and often contain highly ordered structural features, such as towers of cells. Particular sets of genes must be expressed at each stage in the development of biofilms; however, the regulation and function of such genes are not well understood in the vast majority of bacteria. In this study, we identified a set of developmental promoter/gene targets of Nla6, a key component in a cascade of transcriptional activators that regulates the formation of Myxococcus xanthus starvation-induced biofilms.
M. xanthus is a Gram-negative bacterium that is capable of forming two structurally distinct biofilms depending on nutrient availability (2). To obtain nutrients, M. xanthus cells form swarming biofilms that collectively hunt and feed on prey bacteria using hydrolytic enzymes (3). When their supply of food begins to run out, cells in the swarming biofilm reorganize to form a second type of biofilm that contains a mat of rod-shaped cells, known as peripheral rods (4), and multicellular fruiting bodies that house about 100,000 dormant and stress-resistant spores.
Most studies of M. xanthus starvation-induced biofilms have focused on the elaborate fruiting body structures, and much is known about the steps that lead to their formation: cells cluster into an aggregation center, the cell aggregate becomes larger and eventually develops the dome-shaped appearance of a fruiting body, the rod-shaped cells in the newly formed fruiting body differentiate into spherical cells, and the spherical cells mature into stress-resistant spores. A number of signals have been implicated in the development of spore-filled fruiting bodies. Among these are two early developmental signals, the intracellular starvation signal (p)ppGpp (5–7) and an extracellular signal known as A-signal (8, 9), which is a mixture of amino acids and peptides that is thought to function as a reporter of the local cell density (10–12).
The accumulation of these early signals and many subsequent developmental events are regulated by enhancer binding proteins (EBPs) (13). EBPs work with the σ54 protein to activate transcription. Specifically, σ54 directs RNA polymerase to conserved sequences in the −24 and −12 regions of target promoters (14), and EBPs, which bind to tandem repeat sequences or enhancers located upstream of the −24 and −12 regions (15–17), use the energy derived from ATP hydrolysis to help σ54 RNA polymerase form an open promoter complex and initiate transcription (18–20). Most EBPs, including those linked to fruiting body development, are predicted to be components in signal transduction circuits; they activate transcription in response to input from a signal transduction partner that detects a particular environmental signal (21–23).
Six of the signal-responsive EBPs that are expressed in the early to middle stages of M. xanthus fruiting body development form a regulatory cascade, which is reminiscent of the sigma factor cascade that controls the sequential stages of spore development in Bacillus subtilis (13, 24). Namely, pairs of EBPs functioning at one stage of development directly activate transcription of an EBP gene important for the next developmental stage. Because of this design, expression of the EBP genes themselves is responsive to signal input; however, the activating signals have yet to be identified. To date, EBPs in the cascade have been linked to the regulation of (p)ppGpp and A-signal accumulation and to sporulation (13, 25–30).
The EBP Nla6, which is a key component in the regulatory cascade, begins to influence developmental gene expression prior to the onset of aggregation (13). Despite its link to changes in gene expression in this early, preaggregation stage of development, the results of phenotypic studies suggest that Nla6's primary function may be to regulate sporulation (13, 27), which is a late developmental event. Indeed, a mutation in nla6 causes a slight aggregation delay and a strong sporulation defect (27). DNA microarray studies showed that Nla6 is important for the normal expression of hundreds of developmental genes (13); however, it is unclear which of these genes are direct targets of Nla6.
The primary goals of this work were to determine how Nla6 identifies its target promoters/genes and to begin probing for the developmental functions of these genes. A previous study revealed that multiple EBPs, including Nla6, regulate the developmental promoters of the EBP gene operons actB, nla6, and nla28 (13). Here, a direct repeat [consensus, C(C/A)ACGNNGNC] binding site for Nla6 was identified using in vitro and in vivo mutational analyses and the sequence was subsequently used to find 40 potential developmental promoter (88 gene) targets. We showed that Nla6 binds to the promoter region of four of the new targets (asgE, exo, MXAN2688, and MXAN3259) in vitro and that Nla6 is important for their normal expression in vivo. Phenotypic studies indicate that all of the experimentally confirmed targets of Nla6 are primarily involved in sporulation (25–27, 31–33), suggesting that Nla6's primary developmental function is indeed to regulate M. xanthus sporulation genes. The confirmed targets of Nla6 include genes involved in transcriptional regulation, cell-cell signal production, and spore differentiation and maturation. The predicted functions of the putative Nla6 targets include transcriptional regulation/signal transduction, cell wall/membrane biogenesis, and solute transport. Although sporulation occurs late in development, all of the developmental loci analyzed here show an Nla6-dependent burst in expression soon after starvation is induced. This finding suggests that Nla6 starts preparing cells for sporulation very early in the developmental process.
MATERIALS AND METHODS
Bacterial strains and plasmids.The strains, plasmids and primers used in this study are shown in Tables S1 and S2 in the supplemental material.
Growth and development.M. xanthus strains were grown at 32°C in CTTYE (1% Casitone, 10mM Tris-HCl [pH 7.6], 1 mM KH2PO4, 8 mM MgSO4, 0.5% yeast extract) broth or on plates containing CTTYE broth and 1.5% agar. CTTYE broth and CTTYE agar plates were supplemented with 50 μg/ml of kanamycin as needed. Fruiting body development was induced by placing M. xanthus cells on plates containing TPM (10 mM Tris-HCl [pH 7.6], 1 mM KH2PO4, 8 mM MgSO4) and 1.5% agar or in 6-well microtiter plates containing MC7 buffer (10 mM morpholinepropanesulfonic acid [MOPS; pH 7.0], 1 mM CaCl2) and incubating the plates at 32°C. Fruiting body development was monitored as previously described (27). CTT soft agar (1.0% Casitone, 10.0 mM Tris-HCl [pH 8.0], 1.0 mM KH2PO4, 8.0 mM MgSO4, 0.7% agar) was used for assaying sporulation efficiency. CTTYE, TPM, MC7, and CTT soft agar were made as previously described (27, 34).
Escherichia coli strain BL21(DE3) (35) was grown in Luria-Bertani (LB) broth or plates containing LB broth and 1.5% agar. LB broth and LB agar plates were supplemented with 100 μg/ml of ampicillin or 50 μg/ml of kanamycin as needed. For protein expression and purification, E. coli BL21(DE3) cells were grown in 2× YT medium (YT is 1.6% tryptone, 1% yeast extract, and 0.5% NaCl [pH 7.0]) supplemented with 100 μg/ml of ampicillin or 50 μg/ml of kanamycin as needed. LB and 2× YT media were made as previously described (34).
Preparation of the Nla6 DBD.A PCR-generated DNA fragment that codes for the Nla6 DNA binding domain (DBD) was cloned into the pMBP-parallel 1 vector (36), and the resulting plasmid (pKG14) was transformed into E. coli strain BL21(DE3). The Nla6 DBD was purified on amylose resin (New England BioLabs) and concentrated to 1 μg/μl using Amicon Ultra 30K centrifugal filters (Millipore) as previously described (13). Protein concentrations were estimated using Quick Start Bradford dye reagent (Bio-Rad).
Electrophoretic mobility shift assays (EMSAs).The PCR-generated fragments of the putative Nla6 target promoters contained approximately 150 to 200 bp of DNA upstream of the σ54 RNA polymerase binding sites, which were identified experimentally (13, 37–40) or using a bioinformatics tool (PromScan) that was specifically developed to find such sites in the sequences of bacterial DNA (41). The DNA fragments were end labeled using [γ-32P]ATP (MP Biomedicals) and T4 polynucleotide kinase (New England BioLabs) and purified using the QIAquick nucleotide removal kit (Qiagen). Binding reactions were carried out as previously described using a 1 μM concentration of the Nla6 DBD and 1 ng of the labeled promoter fragment (13, 42). The reaction mixtures were electrophoresed through native PAGE Novex (Invitrogen) gels, visualized using a storage phosphor screen (Amersham Biosciences) and the Typhoon-9410 imager (GE Healthcare), and analyzed using ImageQuant software (Molecular Dynamics).
Identifying potential targets of Nla6 using bioinformatics.The five promoter fragments (actB P1, asgE P1, asgE P2, nla6 P1a, and nla28 P1a) to which the Nla6 DBD bound in our initial studies were scanned using the Consensus bioinformatics tool (43); similar 10-bp direct repeat sequences were identified in all of the fragments, and the direct repeat sequences were used to generate a consensus Nla6 binding site or enhancer element [C(C/A)ACGNNGNC-(N1–N60)-C(C/A)ACGNNGNC]. Subsequently, a genome-scale DNA pattern bioinformatics search tool (http://rsat.bigre.ulb.ac.be/rsat//RSAT_home.cgi) (43) was used to scan the M. xanthus genome sequence (23) for close matches to this consensus Nla6 enhancer site. In particular, we looked for putative Nla6 enhancers with half-sites that contained the conserved C at position 1 (conserved in all of the enhancer half-sites used to generate the consensus) and no more than two mismatches to the consensus. In addition, the half-sites had to be separated by 1 to 60 nucleotides. Two hundred eighty-seven putative Nla6 enhancer sites were identified by using this strategy. Since Nla6-mediated regulation of developmental genes was the focus of this study, we determined whether the 287 putative enhancers are located in the promoter region of a developmentally regulated single gene or operon. For the gene or operon to be considered developmentally regulated, expression had to increase at least 2-fold during development (our DNA microarray data on Gene Expression Omnibus [accession number GSE13523] [13]). The developmentally regulated single genes and operons that had a potential Nla6 enhancer in their promoter regions were also required to have a known σ54 promoter or a putative σ54 promoter identified using the PromScan bioinformatics tool (41). In addition, the potential Nla6 enhancer had to be located in the 1 kb of DNA immediately upstream of the σ54 or putative σ54 promoter. Nineteen developmentally regulated operons (containing 67 genes) and 21 single genes met these two criteria (see Table S3 in the supplemental material). Three of the promoters that contained a putative Nla6 enhancer element were tested for the Nla6 DBD binding and were positive, leading to the revised consensus Nla6 enhancer CCACG(C/G)(C/G)GNC-(N1–N60)-CCACG(C/G)(C/G)GNC.
Mutational analysis of the putative Νla6 enhancer.A purified DNA fragment of the nla28 promoter region and the nla6 promoter region were cloned into the pCR2.1 TOPO vector (Invitrogen), and mutations in the putative Nla6 enhancer were made as previously described (13). Portions of the plasmid-borne wild-type and mutant nla promoter fragments were PCR amplified and used in the EMSAs described above. In addition, the full-length wild-type and mutant nla28 promoter fragments were cloned into the promoterless lacZ expression vector pREG1727 (44), the plasmids were introduced into wild-type strain DK1622, and cells carrying a plasmid integrated at the Mx8 phage attachment site in the chromosome were identified via PCR. The in vivo activities of wild-type and nla28 mutant promoters were determined by measuring the specific activity of β-galactosidase in cells that developed on TPM starvation agar for various amounts of time (45). All promoter fragments used in the in vitro and in vivo assays were analyzed via DNA sequencing.
qPCR.To examine expression of the asgE gene during vegetative growth, wild-type and nla6 mutant cells were harvested at different times after inoculation into CTTYE broth. To examine expression of the remaining Nla6 target genes under study, wild-type cells and nla6 mutant cells were harvested during vegetative growth (0 h) and at various times after starvation-initiated development. Total cellular RNA was isolated and cDNA was generated from vegetatively growing cells as described by Sarwar and Garza (34) and Ossa et al. (30), respectively. Total cellular RNA was isolated from developmental cells using the RNAprotect bacterial reagent (Qiagen) and the RNeasy minikit (Qiagen) as described in the manufacturer's protocols. To help lyse developmental cells, 0.1-mm-diameter glass beads were added after the lysis buffer, and the cell suspension was shaken vigorously using a VWR DVX-2500 multitube vortex mixer. For each developmental time point, total RNA was isolated from seven independent wild-type samples and from seven independent nla6 mutant samples. The seven wild-type RNA samples from each time point were pooled, the seven nla6 mutant RNA samples from each time point were pooled, and the pooled samples were subsequently used to generate cDNA. cDNA was generated from 100 ng of pooled RNA using the iScript cDNA synthesis kit (Bio-Rad). A 10-fold dilution series of cDNA for each wild-type and nla6 mutant time point was added to quantitative PCR (qPCR) mixtures containing 150 nmol of gene-specific primers and 1× SYBR green master mix (Bio-Rad). The CFX Connect real-time PCR detection system (Bio-Rad) was used to perform the qPCR analysis, which was done in triplicate. The reaction cycles were as follows: 1 cycle of 95°C for 2 min, 40 cycles of 95°C for 15 s, 55°C for 30 s, and 72°C for 30 s, followed by 1 cycle of 95°C for 1 min, 68°C for 10 s, and 95°C for 1 min. Standard-curve R2 values and amplification efficiency values ranged from 0.990 to 1.0 and 93% to 100%, respectively. Relative changes (n-fold) in asgE mRNA levels were calculated by the ΔΔCT method using the reference gene rpoD as previously described (30, 34). Because the levels of rpoD mRNA were significantly different in wild-type cells and nla6 mutant cells late in development, the reference used for the developmental mRNA expression profiles was 16S rRNA; developmental expression of 16S rRNA appears to be constitutive, and the levels are similar in the wild-type and nla6 mutant cells throughout development. The relative changes in developmental mRNA levels were calculated using this reference and the ΔΔCT method.
RESULTS
The Nla6 DBD binds to the vegetative promoter of asgE, a gene that is important for A-signal accumulation and sporulation.Previous work showed that the Nla6 DBD binds to the promoter region of the nla6, nla28, and actB operons (13). The nla and actB EBP genes are all part of the regulatory cascade that modulates changes in gene expression during various stages of M. xanthus fruiting body development. To date, no Nla6 targets outside the EBP regulatory cascade have been identified. To probe for additional Nla6 targets, we chose the σ54-like promoters identified upstream of the asgE, mbhA, and spi genes (37, 38, 40). The promoters of mbhA and spi were selected because they become active prior to the onset of aggregation (preaggregation), which is the stage of development that Nla6 first influences gene expression (13). The asgE vegetative promoter (asgEveg) was selected because it appears to be important for accumulation of normal levels of A-signal (31) and because Nla6 was previously linked to A-signal accumulation in extracellular complementation studies (27). Furthermore, mutations in asgE and nla6 produce similar developmental phenotypes: relatively mild defects in aggregation and strong sporulation defects (27, 31).
EMSAs were performed with end-labeled fragments of the asgEveg promoter region, the mbhA promoter region, and the spi promoter region and purified Nla6 DBD. Two fragments of each promoter region were used in these assays. Fragment 1 contained 150 to 200 bp of DNA located immediately upstream of the putative σ54 RNA polymerase binding site, and fragment 2 contained 150 to 200 bp of DNA located upstream of fragment 1. An end-labeled fragment of the dev promoter, which is a non-σ54 promoter element (46), was used as a control for nonspecific Nla6 DBD binding. As shown in Fig. 1, we did not detect binding to the negative-control dev promoter fragment. We also failed to detect binding to either of the spi or mbhA promoter fragments (data not shown). We did, however, detect Nla6 DBD binding to both fragments of the asgEveg promoter (Fig. 1). This finding suggests that Nla6 might directly regulate the asgE gene, which is consistent with previous studies linking asgE and nla6 to A-signal accumulation and sporulation (27, 31).
EMSAs performed with the Nla6 DBD and fragments of the asgEveg promoter. The binding reactions were performed with (+) or without (−) a 1 μM concentration of purified Nla6 DBD and an end-labeled fragment of the dev promoter region (negative control) or the end-labeled P1 or P2 fragment of the asgEveg promoter region.
Expression of the asgE, actB, and nla28 genes is altered in an nla6 mutant.The results of the Nla6 DBD binding assays suggested that actB, asgE, nla6, and nla28 promoters might be in vivo targets of Nla6 (13). Consequently, we wanted to test whether expression of asgE and the actB and nla operons is altered in an nla6 mutant. Previous DNA microarray and qPCR analyses showed that expression of MXAN4043, which is a gene in the nla6 operon, is reduced about 2.5-fold in an nla6 mutant relative to that in wild-type cells (13). To test whether expression of the actB, asgE, and nla28 genes is altered in an nla6 mutant, we performed qPCR with primer pairs that are specific to each gene.
We determined the relative levels of asgE mRNA in wild-type and nla6 mutant cells grown in CTTYE broth, since the Nla6 DBD binds to the asgE vegetative promoter. In particular, cells were isolated during the transition from lag phase to exponential growth (40 Klett units), exponential growth (150 Klett units), and the transition from exponential growth to stationary phase (250 Klett units) (Fig. 2A), and the relative levels of asgE mRNA were determined (Fig. 2B). The levels of asgE mRNA in wild-type cells were about 2- to 3-fold higher during exponential growth and the transition into stationary phase than during the transition into exponential growth. The level of asgE mRNA in nla6 mutant cells was similar to that in wild-type cells during the transition into exponential growth; however, asgE mRNA levels were reduced about 5-fold and 3-fold compared to those in wild-type cells during exponential growth and the transition into stationary phase, respectively. These findings indicate that Nla6 is important for the observed increase in asgE mRNA during vegetative growth.
Relative expression of asgE, actB, and nla28 in wild-type and nla6 mutant cells. Expression of asgE mRNA, actB mRNA, and nla28 mRNA in wild-type and nla6 mutant cells was determined using qPCR. (A) Growth of wild-type cells (squares) and nla6 mutant cells (circles). Cells were grown at 32°C in CTTYE broth and harvested at densities of 40, 150, and 250 Klett units. (B) asgE mRNA levels in wild-type (gray bars) and nla6 mutant (black bars) cells harvested during growth. Total cellular RNA was isolated from three independent biological replicates, cDNA was generated from each RNA sample, qPCRs were performed in triplicate, and the relative levels of asgE mRNA were calculated using the reference gene rpoD and the ΔΔCT method. Values are means. (C and D) actB (C) and nla28 (D) mRNA levels in cells harvested at 0, 1, 8, 12, and 24 h of development. Total cellular RNA from seven independent biological replicates was pooled, cDNA was generated from the pooled RNA, qPCRs were performed in triplicate, and the relative levels of actB mRNA and nla28 mRNA, respectively, were calculated by using the ΔΔCT method, with 16S rRNA as the reference. Values (C and D) are means and standard deviations of the means for wild-type cells (squares) and nla6 mutant cells (circles). The error bars in the actB and nla28 mRNA expression profiles do not extend past the symbols.
Since the actB and nla28 genes are developmentally regulated (13, 47), we determined the relative levels of actB mRNA and nla28 mRNA in wild-type cells and nla6 mutant cells at various developmental time points (Fig. 2C and D). Using the vegetative-growth time point (0 h) as the baseline, we detected two relatively large (about 2- to 10-fold) increases in actB mRNA and nla28 mRNA in wild-type cells. The initial increases in actB and nla28 mRNAs occurred soon after starvation initiated development (0 h to 1 h of development). The second increase in actB mRNA and nla28 mRNA occurred in the middle to late stages of development: actB mRNA levels increased during the aggregation stage (8 to 12 h of development) and nla28 mRNA levels increased in the time period that spans the formation of the fruiting body structures and the onset of spore differentiation (12 to 24 h of development).
In nla6 mutant cells, the levels of actB mRNA were about 1.4- to 6.0-fold lower than in wild-type cells at each time point after starvation-initiated development. The levels of nla28 mRNA in nla6 mutant cells were 3- to 6-fold lower than in wild-type cells at the early developmental time points, similar to those in wild-type cells at 12 h of development (aggregation) and about 2-fold higher than those in wild-type cells at the late developmental time point (24 h), which corresponds to the onset of spore differentiation.
Together, the results of the qPCR analysis suggest that Nla6 is an important regulator of actB and nla28 mRNA expression in developing cells. Furthermore, the data are consistent with the idea that Nla6 functions as an activator of actB expression and that Nla6 functions as an activator and inhibitor of nla28 expression in developing cells. In the latter case, Nla6 seems to function as an activator in early development and an inhibitor late in development. It should be noted that the nla6 mutation alters expression of both actB and nla28 mRNA in growing cells (0 h), a point that is addressed in Discussion.
Mutations in the putative Nla6 enhancers in the nla6 and nla28 promoters inhibit Nla6 DBD binding in vitro.EBP dimers bind to enhancer elements, which are repeat sequences that are separated by a variable number of bases (22). As shown in Table 1, all the promoter fragments (actB P1, asgEveg P1, asgEveg P2, nla6 P1a, and nla28 P1a) to which the Nla6 DBD binds in vitro contain similar 10-bp direct-repeat sequences, which are good candidates for Nla6 enhancer elements. To examine whether we identified bona fide Nla6 binding sites, we subjected the putative enhancer element, nla28 enhancer element 1 (EE1), in the nla28 promoter to mutational analysis. We chose the nla28 promoter for this initial set of experiments because nla28 EE1 is separated from other regulatory sequences in the promoter and the Nla6 DBD specifically binds to the nla28 P1a fragment of the nla28 promoter, which contains nla28 EE1 (13). The 151-bp nla28 P1a fragment was used as a template to generate the nla28 P1b and nla28 P1c fragments, which contain a CC-to-TT substitution in the first nla28 EE1 half-site and a CA-to-TT substitution in the second nla28 EE1 half-site, respectively (Fig. 3A). Subsequently, nla28 P1a, nla28 P1b, and nla28 P1c promoter fragments were end labeled, and EMSAs were performed using purified Nla6 DBD. As shown in Fig. 3B, we detected Nla6 DBD binding to the wild-type nla28 P1a promoter fragment, which is consistent with previous results (13). In contrast, the Nla6 DBD failed to bind to the nla28 P1b and the nla28 P1c fragments, which contain mutations in the first and second nla28 EE1 half-sites, respectively.
Tandem repeat sequences in Nla6 target promoters
EMSAs performed with the Nla6 DBD and nla promoter fragments containing Nla6 enhancer element mutations. (A) nla28 promoter region. The 151-bp nla28 P1a fragment contains the wild-type Nla6 enhancer element nla28 EE1. The nla28 P1b and nla28 P1c fragments are derivatives of nla28 P1a that contain a CC-to-TT substitution in the first nla28 EE1 half-site and a CA-to-TT substitution in the second nla28 EE1 half-site, respectively. The mutated nucleotides in nla28 EE1 are in bold and underlined. (B) EMSAs performed with (+) or without (−) the Nla6 DBD and the nla28 P1a, nla28 P1b, or nla28 P1c promoter fragment. (C) nla6 promoter region. The nla6 P1a fragment contains the wild-type Nla6 enhancer element nla6 EE1. The nla6 P1b and nla6 P1c fragments are derivatives of nla6 P1a that contain a CAA-to-TTT substitution in the first nla6 EE1 half-site and a CCA-to-TTT substitution in the second nla6 EE1 half-site, respectively. The mutated nucleotides in nla6 EE1 are in bold and underlined. (D) EMSAs performed with (+) or without (−) the Nla6 DBD and the nla6 P1a, nla6 P1b, or nla6 P1c promoter fragment.
To further examine whether the 10-bp direct repeat sequence that we identified is an Nla6 DBD binding site, we generated mutations in the putative enhancer element (nla6 EE1) in the 201-bp nla6 P1a promoter fragment: the nla6 P1b and nla6 P1c fragments are derivatives of nla6 P1a that contain a CAA-to-TTT substitution in the first nla6 EE1 half-site and a CCA-to-TTT substitution in the second nla6 EE1 half-site, respectively (Fig. 3C). The nla6 P1a, nla6 P1b, and nla6 P1c promoter fragments were end labeled, and EMSAs were performed using purified Nla6 DBD. We detected Nla6 DBD binding to the wild-type nla6 P1a promoter fragment as previously reported (13). In contrast, we detected little or no binding of the Nla6 DBD to the nla6 P1b and the nla6 P1c promoter fragments, which contain mutations in the first and second nla6 EE1 half-sites, respectively (Fig. 3D). The fact that mutations in the 5′ end of either of the nla6 EE1 half-sites or the nla28 EE1 half-sites inhibit in vitro Nla6 DBD promoter binding support our proposal that nla6 EE1 and nla28 EE1 function as Nla6 enhancer elements.
Mutations in the putative Nla6 enhancer in the nla28 promoter reduce in vivo activity.The experiments described above indicate that mutations in either of the putative Nla6 enhancer (nla28 EE1) half-sites abolish Nla6 DBD binding to the nla28 promoter in vitro. To test whether these mutations inhibit the in vivo activity of the nla28 promoter, the nla28 P3a fragment of the nla28 promoter, which contains a wild-type nla28 EE1 and the putative site of σ54 RNA polymerase binding, was used as a template to generate mutations in each nla28 EE1 half-site individually. Figure 4A shows the nucleotide positions that were mutated. As described above for the in vitro assays, we generated T substitutions at nucleotide positions 1 and 2 of the first nla28 EE1 half-site (nla28 P3b promoter fragment) or at nucleotide positions 1 and 2 of the second nla28 EE1 half-site (nla28 P3c promoter fragment). Subsequently, nla28 P3a, nla28 P3b, and nla28 P3c were cloned into a plasmid to create lacZ transcriptional fusions (44). M. xanthus strains with a wild-type or mutant nla28 promoter/lacZ fusion plasmid integrated at the Mx8 phage attachment site in the chromosome were generated, and developmental expression of lacZ was monitored to determine the activity of each promoter. Figure 4B shows the relative levels of expression of lacZ fused to the wild-type nla28 P3a promoter fragment, the nla28 P3b mutant promoter fragment or the nla28 P3c mutant promoter fragment at 1 h of development (the time of peak nla28 expression) (Fig. 2D). Expression of lacZ from the nla28 P3b mutant promoter fragment was reduced about 1.7-fold relative to the wild-type nla28 P3a promoter fragment and expression of lacZ from the nla28 P3c mutant promoter fragment was reduced about 1.5-fold relative to the wild-type nla28 P3a promoter fragment. For comparison, nla28 expression in cells that lack Nla6 is about 4-fold less than in wild-type cells after 1 h of development (Fig. 2D). Thus, a mutation in the first or second half-site of the proposed Nla6 enhancer in the nla28 promoter reduces its in vivo activity, but not as dramatically as nla6 inactivation. Perhaps these differences in nla28 promoter activity are due to the levels of Nla28, which is known to positively regulate the nla28 promoter (13). Namely, in the nla6 mutant, the levels of nla28 mRNA and presumably the Nla28 protein are relatively low. In contrast, the strains carrying the Nla6 enhancer mutations have wild-type nla6 and nla28 loci and are predicted to produce wild-type levels of nla28 mRNA and Nla28.
In vivo activity of the nla28 promoter after the introduction of Nla6 enhancer element mutations. (A) The 418-bp nla28 P3a fragment of the nla28 promoter region carries a wild-type σ54 promoter and an intact Nla6 enhancer element (nla28 EE1). nla28 P3b and nla28 P3c are derivatives of nla28 P3a that carry a mutation in the first and second nla28 EE1 half-site, respectively. The mutated nucleotides in nla28 EE1 are in bold and underlined. (B) nla28 P3a, nla28 P3b, and nla28 P3c were cloned into a lacZ expression vector and transferred to the wild-type M. xanthus strain DK1622. At various times during development, β-galactosidase-specific activities (defined as nanomoles of ONP produced per minute per milligram of protein) in cells carrying a wild-type or a mutant promoter fragment were determined. The values are mean β-galactosidase specific activities from 3 samples taken at 1 h of development, which is the time at which nla28 expression peaks. Error bars represent standard deviations of the means.
The results of the in vivo mutational analysis described above and those of the in vitro mutational analysis suggest that the direct repeats that we identified in the nla28 promoter, as well as the actB, asgE, and nla6 promoters, are likely to be Nla6 enhancers.
Identification of additional Nla6 targets using the Nla6 enhancer element.To identify potential members of the Nla6 regulon in the M. xanthus genome sequence, we generated a consensus Nla6 enhancer [C(C/A)ACGNNGNC-(N1–N60)-C(C/A)ACGNNGNC] by aligning the putative half-sites in the actB, asgEveg (P1 and P2), and nla promoter regions. As described in Materials and Methods, we then searched for close matches to the consensus Nla6 enhancer in the promoter region of each operon and single gene in the M. xanthus genome sequence (23, 43) and for a known σ54 promoter or putative σ54 promoter (identified using the PromScan bioinformatics tool [41]) downstream of the matching sequence. In addition, we identified operons and single genes that are developmentally regulated based on DNA microarray analysis (13). Nineteen operons containing 67 genes and 21 single genes met all of our criteria (see Table S3 in the supplemental material). The majority of these 88 genes have unknown functions or have been assigned only general functions. Most of the remaining targets were classified as regulatory genes (transcription/signal transduction), as genes with cell envelope-associated functions such as cell wall/membrane biogenesis, or as genes encoding solute transporters (Table 2). One of the putative Nla6 targets that was previously characterized and has a notable developmental function is the exoA-I operon (32, 33). In particular, an in-frame deletion of the exoC gene affects spore coat assembly, presumably because it inhibits the export of spore coat polysaccharide, and leads to defects in spore differentiation and maturation (33). Similarly, a mutation in the nla6 gene itself produces defects in spore differentiation and maturation (Table 3).
Functional categories of putative Nla6 target genes
Developmental phenotypes of strains carrying mutations in Nla6 target genesa
The Nla6 DBD binds to the promoter regions of exo, MXAN2688, and MXAN3259 in vitro.To examine whether our bioinformatics approach uncovered developmental promoters that are bound by the Nla6 DBD, the promoter regions of three of the putative Nla6 targets (exo, MXAN2688, and MXAN3259) listed in Table S3 in the supplemental material were selected for EMSAs. All of the promoter fragments that we used in the EMSAs contain at least one direct repeat that is a close match to the consensus Nla6 enhancer element (Table 1). These promoter fragments were end labeled and tested for Nla6 DBD binding, and the results are shown in Fig. 5. The Nla6 DBD failed to bind to the negative-control dev promoter fragment, which lacks a putative Nla6 enhancer. In contrast, the Nla6 DBD bound to all of the promoter fragments that contain putative Nla6 enhancers: the P1 and P2 fragments of the exo promoter region, the P2 fragment of the MXAN2688 promoter and the P1 fragment of the MXAN3259 promoter. Thus, all four of the promoter fragments that we tested contain direct repeats similar to the consensus Nla6 enhancer (Table 1), and all four are positive for Nla6 DBD binding in vitro (Fig. 5).
EMSAs performed with the Nla6 DBD and fragments of putative Nla6 target promoters identified using the consensus Nla6 enhancer. The binding reactions were performed with (+) or without (−) a 1 μM concentration of purified Nla6 DBD and the end-labeled negative-control dev promoter fragment, the exo P1 promoter fragment, the exo P2 promoter fragment, the MXAN2688 P2 promoter fragment, or the MXAN3259 P1 promoter fragment.
Developmental expression of exo, MXAN2688, and MXAN3259 is altered in the nla6 mutant.To examine whether Nla6 is important for exo, MXAN2688, and MXAN3259 expression in vivo, we monitored the mRNA level of each gene in an nla6 mutant cells and in wild-type cells during development using qPCR. Figure 6A to C show the relative expression levels of exo mRNA, MXAN3259 mRNA, and MXAN2688 mRNA, respectively, in nla6 mutant cells and in wild-type cells during fruiting body development.
Developmental expression of exoA, MXAN3259, and MXAN2688 in wild-type and nla6 mutant cells. Expression of exoA mRNA, MXAN3259 mRNA, and MXAN2688 mRNA in wild-type and nla6 mutant cells was determined using qPCR. Mean exo mRNA levels (A), MXAN3259 mRNA levels (B) and MXAN2688 mRNA levels (C) in cells harvested at 0, 1, 2, 8, 12 and 24 h of development are shown. Total cellular RNA from seven independent biological replicates was pooled, cDNA was generated from the pooled RNA, qPCRs were performed in triplicate, and the relative levels of exo mRNA, MXAN3259 mRNA, and MXAN2688 mRNA, respectively, were calculated by the ΔΔCT method with 16S rRNA as the reference. Values are means for wild-type cells (squares) and nla6 mutant cells (circles). Error bars show standard deviations of the means. With one exception, the error bars in the exo, MXAN3259, and MXAN2688 mRNA expression profiles do not extend past the symbols.
As observed in our analysis actB and nla28 mRNAs, a relatively large increase (about 3- to 5-fold) in exo mRNA and MXAN3259 mRNA was detected in wild-type cells soon after starvation initiated development (0 h to 1 h of development). After 1 h of development, we observed a general downward trend in the levels of exo mRNA and MXAN3259 mRNA, until they reached their vegetative-growth baseline levels at 12 h (aggregation stage) of development. Between 12 and 24 h of development, when fruiting body formation is completed and spore differentiation begins, it seemed that the levels of MXAN3259 mRNA and perhaps exo mRNA started to rise again.
In nla6 mutant cells, the levels of exo and MXAN3259 mRNAs were about 3- to 20-fold lower than the levels in wild-type cells at the early developmental time points, similar to the levels in wild-type cells at 12 h (aggregation), and about 2-fold higher than the levels in wild-type cells at 24 h (spore differentiation). As previously observed in our actB and nla28 mRNA expression studies, the nla6 mutation alters the levels of exo and MXAN3259 mRNAs in vegetatively growing cells (see Discussion for implications).
Like the other mRNAs analyzed here, the level of MXAN 2688 mRNA in wild-type cells increased (about 3-fold) between 0 h and 1 h of development. At 2 h of development, the level of MXAN2688 mRNA returned to its vegetative-growth baseline and subsequently fluctuated between slightly lower and slightly higher than the baseline level for the remainder of the developmental time course.
In nla6 mutant cells, the MXAN2688 mRNA level was about 8-fold lower than in wild-type cells at the earliest (1-h) developmental time point, similar to or slightly higher than in wild-type cells at the 2- and 8-h time points, and about 4- to 10-fold higher than in wild-type cells at the last two developmental time points, 12 h and 24 h.
The results of these qPCR studies suggest that Nla6 is an important in vivo regulator of exo, MXAN3259, and MXAN2688 mRNA expression in developing cells. Furthermore, these results, coupled with the results of the in vitro Nla6 DBD promoter binding assays, suggest that exo, MXAN2688, MXAN3259, and presumably the developmentally regulated single genes and operons listed in Table S3 in the supplemental material are good candidates for direct regulation by Nla6.
It is worth noting that expression of all of the mRNAs examined here is activated soon after starvation initiates development, and yet, as described below, they are predicted to function late in development. In addition, our data support the idea that Nla6 functions as an activator and inhibitor of exo, MXAN3259, and MXAN2688 mRNA expression. For all three mRNAs, as well as that of nla28, Nla6 seems to function as an activator early in development (1 h) and an inhibitor late in development (24 h).
Strains carrying a mutation in the Nla6 target gene MXAN2688 or MXAN3259 have strong defects in spore development.Our in vitro studies showed that the Nla6 DBD binds to the promoter region of six operons and one single gene. Five of the operons (actB, asgE, exo, nla6, and nla28) have been characterized in detail and have been assigned developmental functions (25–27, 31–33). The five genes that are predicted to be in the remaining operon (MXAN3259 to MXAN3263) and the single gene MXAN2688 are uncharacterized (23). To examine whether these loci are important for development, we generated an insertion in MXAN2688 and one in MXAN3259. Subsequently, we determined whether the insertions affect the formation of multicellular fruiting bodies and/or sporulation. The results of these phenotypic studies are summarized in Table 3. Cells carrying an insertion in MXAN2688 or MXAN3259 started to form aggregates at about the same time as wild-type cells, and the mature fruiting bodies of MXAN2688 and MXAN3259 mutant cells looked similar to those of the wild type. In sporulation assays, however, the MXAN2688 and MXAN3259 mutants showed strong defects. In particular, the number of spherical cells produced by the MXAN2688 mutant was about 70% of the wild-type strain, indicating that this mutant has a minor defect in spore differentiation. The MXAN2688 insertion mutant does, however, have a strong defect in spore maturation: the number of heat- and sonication-resistant spores produced by this mutant was 0.1% that of the wild-type strain. The MXAN3259 insertion mutant produced strong defects in spore differentiation and spore maturation, as indicated by the facts that no spherical cells were visually identified among the 200 developmental cells examined and that the number of heat- and sonication-resistant spores produced by the MXAN3259 mutant was 0.1% that of the wild-type strain. The sporulation phenotype of the MXAN3259 mutant and the sporulation phenotype of the MXAN2688 mutant are reminiscent of the nla6 mutant itself, which is defective for spore differentiation and spore maturation (Table 3). These findings indicate that MXAN2688 and a gene or genes in the MXAN3259 operon are important for sporulation, which is consistent with the previous proposal that Nla6 is a key regulator of sporulation genes in M. xanthus (27).
DISCUSSION
M. xanthus uses a cascade of EBPs to modulate changes in gene expression during sequential stages of M. xanthus fruiting body development. Previous in vitro assays showed that the Nla6 DBD binds to the promoter region of three EBP gene operons that participate in the cascade: actB, nla6, and nla28 (13). In this study, we identified targets of Nla6 that function outside the EBP regulatory cascade. We found that the vegetative promoter of the asgE operon, the promoter regions of two developmentally regulated operons (exo and MXAN3259), and the promoter region of a developmentally regulated single gene (MXAN2688) are positive for Nla6 DBD binding in vitro. All of the promoter fragments to which the Nla6 DBD bound in this study and in previous work (13) contain similar 10-bp direct repeat sequences [consensus, C(C/A)ACGNNGNC-(N1–N60)-C(C/A)ACGNNGNC]. We tested whether these direct repeats are Nla6 enhancer elements using mutational analysis. Specifically, we generated mutations in each half-site of the putative Nla6 enhancer in the nla6 promoter and the nla28 promoter and showed that they abolish or reduce in vitro Nla6 DBD binding. We also showed that the mutations in the putative Nla6 enhancer in the nla28 promoter reduce its in vivo activity during development. It is noteworthy that previous experimental analyses indicate that the actB, asgE, nla6, and nla28 operons use σ54 promoter elements (13, 40, 47), which are the targets of EBP-mediated regulation. In addition, the remaining promoters that were positive for Nla6 DBD binding contain sequences that are good matches to the σ54 promoter consensus (14). Moreover, the locations of all of the putative Nla6 enhancer elements that we identified are consistent with those found in other σ54 promoters (see Fig. S1 in the supplemental material) (48). These results, together with the data showing that a mutation in nla6 affects the in vivo activity of the 7 promoters to which the Nla6 DBD binds (as determined by the developmental or vegetative expression profile of a gene located downstream of the promoter), provide strong evidence that these promoters are targets of Nla6-mediated regulation.
Nla6 may be a general regulator of stress-induced genes.The work described here and much of the early work on Nla6 focused on its role in M. xanthus fruiting body development (13, 27). However, the study by Pan et al. (49) suggested that Nla6 is not simply a regulator of development; Nla6 plays an important role in the regulation of salt tolerance in M. xanthus. Since relatively high levels of salt induce osmotic stress and development is triggered by starvation-induced stress, those authors proposed that Nla6 is a general regulator of stress-associated genes. Indeed, our data indicate that Nla6 directly regulates a set of genes that are activated soon after starvation triggers development and a gene (asgE) that is expressed at relatively high levels when vegetatively growing cells begin to starve and transition into stationary phase. It is noteworthy that we identified putative Nla6 enhancers in the promoter regions of 287 single genes or operons and that only 40 of those 287 are developmentally regulated. Perhaps some or many of these putative Nla6 targets are involved in the response to osmotic stress or some other type of stress.
For many of its targets, Nla6 functions as an activator early and inhibitor late in development.The results of our in vivo expression studies suggest that Nla6 mediates the activation of 4/5 target promoters early in development and the partial inhibition of these promoters late in development, which is consistent with the recent finding that the MXAN4899 EBP mediates the activation and inhibition of some M. xanthus promoters (50). Since nla6 mRNA levels rise early in development and decrease rapidly around the time that cells begin to aggregate (13, 51), we propose that Nla6 directly activates these promoters early in development and indirectly inhibits them late in development. In the case of actB, which requires Nla6 for full activation during the early and middle stages of development, it seems likely that the early activation is directly mediated by Nla6 and the subsequent activation is indirectly mediated by Nla6.
How might Nla6 activate transcription at its target promoters? Nla6 is predicted to be a response regulator in a two-component signal transduction system (23). In two-component systems that contain an EBP such as Nla6, phosphorylation of the EBP by a histidine kinase sensor protein promotes EBP oligomerization (52), which stimulates the ATPase activity of the EBP and the formation of an open promoter complex (20, 53, 54). Presumably, the phosphorylation of enhancer-bound Nla6 by its histidine kinase partner would enable Nla6 to oligomerize and to activate transcription. Although a histidine kinase protein that promotes the phosphorylation Nla6 has yet to be identified, it has been suggested that Nla6S regulates the dephosphorylation of Nla6 in developing cells and, hence, its ability to activate transcription of developmental promoters (55).
Nla6 targets are activated early but are predicted to be important for late development.When we examined the mRNA levels of Nla6 target loci, we detected an increase of at least 3-fold 1 h after starvation initiated fruiting body development. This early increase in mRNA levels occurs well before the predicted time that the loci function; mutational analyses have linked all of these loci to sporulation (25–27, 31–33), which occurs late in development (after the formation of the fruiting body structure). We propose that Nla6 activates these loci as part of an early, starvation-induced stress response and that these loci and their corresponding protein products start to prepare cells for differentiation into spores, which occurs later in development. Perhaps the late induction of some of these loci simply serves to replenish these proteins in cells that are ready to initiate spore differentiation. Of course, there are other explanations for the early induction of these sporulation loci, and additional work is needed to sort out the possibilities. With this in mind, only a fraction of the cells that enter development become spores, and it would be interesting to determine whether the Nla6 target loci are expressed in only a subpopulation of developing cells, i.e., the cells that eventually become spores.
Nla6 regulates asgE as cells enter stationary phase.The data presented here indicate that Nla6 mediates the activation of the asgE vegetative promoter, which is important for asgE expression during exponential growth and the transition into stationary phase (40). The results of expression studies and primer extension analyses suggest that asgE also has a developmental promoter (40); however, the activity of the asgE developmental promoter appears to be dispensable for sporulation during fruiting body development (31, 40). This finding implies that the asgE vegetative promoter, which is regulated by Nla6, provides enough asgE mRNA and presumably AsgE protein for development. Perhaps AsgE is present and poised to act when development begins and the function of the asgE developmental promoter is simply to fine-tune asgE expression (and ultimately AsgE) levels in developing cells.
Nla6 regulates other targets during vegetative growth.In nla6 mutant cells, expression of most of the developmental targets of Nla6 was altered at the 0-h time point, which corresponds to early- to mid-exponential-phase growth. This finding indicates that Nla6 regulates a number of loci during growth, which is consistent with previous data suggesting that nla6 is expressed at a relatively high level in growing cells (49). This finding also suggests that Nla6 is involved in maintaining a relatively low level or baseline expression of these loci in growing cells, or that it is involved in the growth phase regulation of these loci, which has already been shown for asgE (Fig. 2B).
Functions/potential functions of Nla6 targets.Two of the Nla6 targets, nla28 and actB, code for EBPs. These EBPs participate in a regulatory cascade that manages many important developmental events (13). Nla28 and ActB start modulating gene expression in the early and middle stages of development, respectively (13, 26, 56, 57). Based on DNA microarray analysis, Nla28 directly or indirectly regulates hundreds of M. xanthus developmental genes (13). However, direct Nla28 targets that reside outside the EBP regulatory cascade have yet to be identified. Similarly, no direct targets of ActB outside the EBP regulatory cascade have been identified. Sporulation genes are likely to be among the targets of both EBPs, since mutations in nla28 or actB primarily affect sporulation (25–27).
We found that Nla6 regulates a number of loci that are not involved in the EBP regulatory cascade; asgE is one such locus. A mutation in the asgE gene has little effect on aggregation, but it reduces the number of mature spores produced during fruiting body development (31). A detailed characterization of the asgE mutant revealed a partial reduction in the level of the early developmental signal known as A-signal (31). A-signal is believed to be a mixture of amino acids and small peptides generated by extracellular proteolysis (10–12). Because A-signal was produced in rough proportion to the cell density (11) and mutations that abolished A-signal production caused aggregation and sporulation defects under standard assay conditions (58), it was proposed that aggregation proceeds normally only after A-signal specifies that a sufficient number of cells are present for the formation of spore-filled fruiting bodies (11, 58). The more recent work of Berleman and Kirby (59) suggests a caveat to this proposal: under certain environmental conditions, A-signal is important for sporulation but not aggregation. Since the asgE mutant has a partial defect in A-signal production and A-signal is important for sporulation under a variety of assay conditions, it seems likely that the sporulation defect of the asgE mutant is due to its failure to produce normal levels of A-signal (31).
Based on the phenotypes of the asgE mutant and the data presented here, we propose that Nla6's growth-phase regulation of asgE helps ensure that enough A-signal is present to complete the sporulation stage of fruiting body development. Nla6 may provide additional A-signal, which would help promote sporulation, through its developmental regulation of nla28; nla28 was previously linked to A-signal accumulation using extracellular complementation assays (27). Thus, Nla6 may regulate A-signal accumulation via two different loci. The results of extracellular complementation studies, which suggest that A-signal production is reduced in nla6 cells (27), are consistent with the idea that Nla6 is an important regulator of A-signal accumulation.
MXAN2688, MXAN3259, and exo are the three remaining Nla6 targets located outside the EBP regulatory cascade. The Nla6 DBD binds upstream of exoA (MXAN3225), which is the first gene in the exoA-I operon (32, 33). In a previous study, Ueki and Inouye (60) used primer extension analysis to identify three putative transcriptional start sites (PD1, PD2, and PV) upstream of exoA. PD1 and PD2 exo mRNAs were detected only in developing cells; however, PV exo mRNA was detected in growing cells and in cells isolated during early development. Based on the following results, we suggest that Nla6 and its putative σ54 promoter target may be responsible for expression of PV exo mRNA in growing and developing cells. First, the presumed transcriptional start site of PV exo mRNA is located 40 bp downstream of the putative σ54 promoter. Second, Nla6 is important for expression of exo during growth and early development. Third, expression of PV mRNA in growing and developing cells is not dependent on FruA, which is a known regulator of the exo operon (60).
Mutations in the exo operon yield cells that are capable of aggregating but are defective for spore differentiation and spore maturation (32, 33), suggesting that exo functions primarily as sporulation locus during fruiting body development. Presumably, the exo mutant sporulation phenotype is due to a defect in the export of spore coat polysaccharide; several lines of evidence indicate that the M. xanthus spore coat helps maintain osmotic stability and the integrity of the cell envelope during spore morphogenesis (33, 61, 62).
MXAN3259 appears to be the first gene in an operon containing a total of 5 genes, and MXAN2688 appears to be expressed as a single gene (23). An insertion in MXAN3259 or MXAN2688 yields cells with no obvious aggregation defect; however, these cells are defective for spore differentiation and spore maturation. Thus, it seems that MXAN2688, MXAN3259, and exo are all important for spore differentiation and spore maturation.
What are the potential functions of genes in the MXAN3259 and MXAN2688 loci? The protein encoded by the MXAN2688 gene has not been assigned a putative molecular function. However, the product of the MXAN3259 gene is predicted to be a member of the polysaccharide deacetylase family of proteins (23), which include the NodB chitooligosaccharide deacetylases, chitin deacetylases, and peptidoglycan N-acetylglucosamine deacetylase (63–65). In Bacillus subtilis, the polysaccharide deacetylase PdaA is known to be important for spore germination (66). Moreover, the Streptococcus pneumoniae polysaccharide deacetylase PgdA has been implicated in lysozyme resistance (67), a common property of bacterial spores, including those of M. xanthus. MXAN3260, which is the second gene in the MXAN3259 operon, codes for a putative cell membrane protein that has similarity to members of the polysaccharide biosynthesis family (23). A notable member of this family is B. subtilis SpoVB, a protein that is important for spore cortex formation and spore heat resistance (68), suggesting that the product of MXAN3260 may be involved in the formation of the M. xanthus spore cortex.
Model for Nla6-mediated regulation of sporulation.Since the Nla6 targets identified to date are primarily sporulation loci, we propose that the primary developmental function of Nla6 is to regulate sporulation. However, as previously noted, Nla6 and some of its targets may have other functions. We suggest that Nla6 promotes sporulation by regulating at least 3 different classes of genes (Fig. 7). First, Nla6 modulates asgE expression to ensure that a sufficient level of A-signal is present to complete sporulation. Presumably, AsgE does not participate in A-signal accumulation during vegetative growth but is poised to do so when development begins (31). Second, Nla6 modulates developmental expression of the actB and nla28 genes. Previous studies suggest that actB and nla28 code for transcriptional activators (EBPs) that directly or indirectly regulate sporulation loci (25–27). In the case of Nla28, it is possible that a gene or genes involved in A-signal production are one such target, which would reinforce the activity of AsgE. Third, Nla6 modulates developmental expression of the exo, MXAN2688, and MXAN3259 loci, which are important for spore differentiation and spore maturation. The sporulation functions of the MXAN2688 and MXAN3259 are unknown. However, several lines of evidence indicate that the exo locus plays a role in the export of spore coat polysaccharide (32, 33).
Nla6-mediated regulation of sporulation in M. xanthus. The Nla6 targets analyzed here are primarily involved in sporulation. Each of these Nla6 targets can be placed into one of three functional categories: A-signal accumulation (asgE), transcriptional regulation (actB and nla28), or spore differentiation and maturation (exo, MXAN2688, and MXAN3259). We suggest that Nla6 directly activates (solid arrows) expression of the actB, exo, MXAN2688, MXAN3259, and nla28 loci early in development and indirectly (dashed arrows) inhibits or activates their expression late in development. We also suggest that Nla6 directly (solid arrow) activates asgE expression around the time that cells begin the transition into stationary-phase growth. In addition to confirmed Nla6 targets, 17 operons and 20 single genes were classified as potential targets. Many of the putative targets of Nla6 were classified as having regulatory, cell wall/membrane biogenesis, or solute transport functions. For a more detailed description of Nla6 and its targets and putative targets, see Discussion.
Additional candidates for Nla6-mediated regulation.Our bioinformatics data suggest that the genes analyzed here are but a small part of the Nla6 developmental regulon. As shown in Table S3 in the supplemental material, we uncovered 19 operons containing a total of 67 genes and 21 single genes that have a putative Nla6 binding site in their promoter regions. One of these genes (MXAN5181) codes for a putative penicillin binding protein (PBP) (23). PBPs are important for peptidoglycan synthesis, and some B. subtilis PBPs, such as SpoVD, have been linked to the morphogenesis of the spore cortex (69), which is a modified form of peptidoglycan. Thus, MXAN5181 is a candidate for an M. xanthus sporulation gene. Interestingly, the MXAN5181 operon contains the crdS gene, which codes for a histidine kinase that regulates the activity of the CrdA EBP and as a consequence regulates the timing of developmental events (70). Another potential target of Nla6 that is involved in transcriptional regulation is mrpC; this gene codes for an important regulator of many middle-late developmental genes (71–75). The goal of future work will be to determine whether Nla6 directly regulates these genes or its other potential targets, to determine the functions of the uncharacterized targets of Nla6, and ultimately to piece together the functions of the Nla6 regulon.
ACKNOWLEDGMENTS
We thank Roy Welch and Sonia Johns for their help with the manuscript.
This work was supported by National Science Foundation grant IOS-0950976 to A. G. Garza.
FOOTNOTES
- Received 25 October 2014.
- Accepted 21 January 2015.
- Accepted manuscript posted online 2 February 2015.
Supplemental material for this article may be found at http://dx.doi.org/10.1128/JB.02408-14.
- Copyright © 2015, American Society for Microbiology. All Rights Reserved.
