Spo0A-Dependent Activation of an Extended –10 Region Promoter in Bacillus subtilis
http://www.100md.com
《细菌学杂志》
Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia 30322
ABSTRACT
At the onset of endospore formation in Bacillus subtilis the DNA-binding protein Spo0A directly activates transcription from promoters of about 40 genes. One of these promoters, Pskf, controls expression of an operon encoding a killing factor that acts on sibling cells. AbrB-mediated repression of Pskf provides one level of security ensuring that this promoter is not activated prematurely. However, Spo0A also appears to activate the promoter directly, since Spo0A is required for Pskf activity in a abrB strain. Here we investigate the mechanism of Pskf activation. DNase I footprinting was used to determine the locations at which Spo0A bound to the promoter, and mutations in these sites were found to significantly reduce promoter activity. The sequence near the –10 region of the promoter was found to be similar to those of extended –10 region promoters, which contain a TRTGn motif. Mutational analysis showed that this extended –10 region, as well as other base pairs in the –10 region, is required for Spo0A-dependent activation of the promoter. We found that a substitution of the consensus base pair for the nonconsensus base pair at position –9 of Pskf produced a promoter that was active constitutively in both abrB and spo0A abrB strains. Therefore, the base pair at position –9 of Pskf makes its activity dependent on Spo0A binding, and the extended –10 region motif of the promoter contributes to its high level of activity.
INTRODUCTION
Bacillus subtilis initiates a complex developmental program that leads to the differentiation of the cell into a dormant endospore as the final step in a series of responses to nutrient depletion (for a review, see reference 18). Initiation of this differentiation is dependent upon phosphorylation of a response regulator, Spo0A (6). A complex signal transduction network comprising multiple kinases and phosphatases controls the level of phosphorylated Spo0A (Spo0AP) in the cells (6, 16). Spo0AP directly regulates expression of about 121 genes through either repression or activation (14) by binding specifically to a 7-bp DNA element, referred to as an 0A box (5'-TGNCGAA-3') (24, 31). For example, Spo0AP binding directly represses the promoters for abrB, fruR, flgB, ftsE, sdp, and rocD (14, 17, 20). Other genes are directly activated by Spo0AP (14, 17, 20), and still others are activated by Spo0AP indirectly (14, 24). For example, the spo0H gene is repressed by AbrB until Spo0AP produced at the end of the exponential-growth phase represses abrB, relieving spo0H from AbrB-mediated repression (28). Moreover, Spo0A-activated promoters include those used by RNA polymerase containing the primary sigma factor, A (e.g., spoIIG and spoIIE promoters) (20, 30) and those used by RNA polymerase containing the secondary sigma factor, H (e.g., spoIIA and racA promoters) (2, 3, 29), which introduces additional combinations of control.
In a strain mutant with respect to both kinA and kinB, two of the kinases that control the level of Spo0AP, a low level of Spo0AP accumulates that is sufficient to repress abrB transcription but insufficient for the activation of other Spo0AP-dependent genes required for sporulation (25). Therefore, the responses of some Spo0A-regulated promoters to Spo0AP appear to be more sensitive to Spo0AP levels than are the responses of other Spo0A-regulated promoters. A progressive accumulation of Spo0AP at the end of the exponential-growth phase produces a temporal pattern of responses by Spo0A-regulated promoters that reflects the specific response of each promoter to low- or high-threshold levels of Spo0AP (9). Thus, in addition to the combinations of controls described above, Spo0A-regulated promoters can be classified as either high- or low-threshold activated or repressed promoters (9). High-threshold promoters require a high level of Spo0AP to be activated or repressed, probably in part because the regulatory regions for these genes have relatively weak affinities for Spo0AP, whereas low-threshold promoters respond to a low level of Spo0AP because their regulatory regions have relatively high-affinity binding sites for Spo0AP (9).
The progressive accumulation of Spo0AP allows cells to try less-drastic responses to nutrient depletion before commitment to differentiation into a dormant cell type. Moreover, Spo0A phosphorylation occurs in only a fraction of the population (7, 9, 10). Activation of Spo0A in this fraction of the population enables these cells to delay progression into sporulation by activating two operons, the sporulation killing factor (skf) and the sporulation delaying protein (sdp) operons (10). The products of these two operons prevent non-Spo0AP-expressing sibling cells from sporulating (sdp) and cause them to lyse (skf). The cells that have activated Spo0A early are then able to feed off the nutrients released, allowing them to continue growing and thus to delay their differentiation into dormant endospores. The skf promoter must be highly active to drive production of the secreted toxin, and it must be activated early in response to low levels of phosphorylated Spo0A (9, 10). However, its expression must be tightly regulated to prevent premature expression of the killing factor. Here we investigate how this promoter can be both highly active in response to low levels of Spo0AP and tightly controlled to prevent premature expression of the toxin.
The sequence of the skf promoter, which is activated by low levels of Spo0AP (9), appears to indicate that Spo0AP may directly activate this promoter by a mechanism that is different from that of the spoIIG promoter, which has been studied more extensively (4, 12, 19-21, 23). Comparison of the relative affinities of Spo0A for the high- and low-threshold promoters shows that spoIIG has a Kd of 1,700 nM whereas the skf promoter has an apparent Kd value of 26 nM (9). A bioinformatics search identified two Spo0A binding sites (14); however, the search did not reveal other features of the skf promoter that would contribute to its positive regulation by low levels of Spo0AP.
Here we report that skf is a A-dependent promoter with multiple Spo0A binding sites, two of which span the –35 region. Moreover, upstream of the mapped transcriptional start site is a sequence that is very similar to the TRTGn motif found in extended –10 promoters in Bacillus subtilis (13, 26, 27), another feature that is not present in the spoIIG promoter. We show that this sequence is necessary for the high level of Spo0A-dependent promoter activity. We also describe other features of this promoter that are necessary for its tight regulation by Spo0A.
MATERIALS AND METHODS
Bacterial strains and culture media. Routine microbiological manipulations and procedures were carried out by standard techniques as described in reference 8. The concentrations of antibiotics used for selection on Luria broth (LB) or Difco sporulation medium (DSM) were 5 μg/ml for chloramphenicol, 100 μg/ml for spectinomycin, 10 μg/ml for kanamycin, and 100 μg/ml for ampicillin. Cultures were grown in LB, and sporulation was induced by nutrient exhaustion in DSM. Competent cells were prepared and transformed by the two-step method as described in reference 8.
To create the abrB deletion strain (CMBS001) (Table 1), the 5'- and 3'-flanking DNA of abrB was PCR amplified from chromosomal DNA. The 5'-flanking DNA PCR product was cloned into pDG784 digested with SphI and PstI. The 3'-flanking DNA PCR fragment was then cloned into the resulting plasmid digested with BamHI and EcoRI to generate plasmid pDG784abrB. Plasmid pDG784abrB was sequenced to ensure the correct DNA sequences of the cloned fragments and was used to transform competent JH642. Kanamycin-resistant colonies were selected and analyzed further by PCR to determine whether substitution of the kanamycin resistance cassette for abrB occurred, as described in reference 12.
To clone the skf promoter, the DNA region from the stop codon (TAA) of ybcM (the upstream gene adjacent to the skf operon) to the start codon (ATG) of skfA (the first structural gene of the skf operon) was PCR amplified with oligonucleotides GNC1 and GNC2 (Table 2) and inserted into EcoRI- and HindIII-digested pBG1PLK to create pBG1PLK-SKF. The plasmid was checked by sequencing and used to transform competent JH642 to chloramphenicol resistance. Integration of pBG1PLK-SKF into the wild-type B. subtilis chromosomal amyE locus by homologous recombination was confirmed by testing for loss of amylase production by iodine staining and by PCR and DNA sequencing. The resulting strain was later transformed with abrB and/or spo0A knockout plasmids (12), producing corresponding B. subtilis strains (Table 1).
Overlapping PCR was used to introduce base pair substitutions in the wild-type skf promoter. First, two QuikChange oligonucleotides were designed that were complementary to each other and annealed to the skf promoter but carried the desired nucleotide change. Each was used in combination with an external oligonucleotide (GNC1 or GNC2) to perform PCR using the plasmid pBG1PLK-SKF as the template. The two amplified products, which correspond to the 5' and 3' portions of the skf promoter and partially overlap the mutated region, were gel purified. The products were then mixed, annealed, and used as the template for the second step of PCR performed with the two external primers, GNC1 and GNC2. The product obtained, which should carry the skf promoter sequence with the desired substitution, was sequenced and then inserted into EcoRI- and HindIII-digested pBG1PLK. The resulting plasmids (Table 1) were then used to transform competent JH642.
RNA preparations. Cultures of B. subtilis strains THWB2, THWB18, THWB24 and THWB9 (Table 1) were incubated in DSM. Two hours after the end of the exponential-growth phase, the bacteria were harvested by centrifugation and stored at –80°C. RNA was prepared as described in reference 8.
Primer extension reactions. A total of 100 ng of primer GNC12 (Table 2) was end labeled with [-32P]ATP by use of T4 polynucleotide kinase from New England Biolabs. The labeled primer was purified using an Amersham Biosciences MicroSpin G-25 column (Amersham Pharmacia Biotech, Piscataway, N.J.). A 15-ng volume of this purified labeled primer was added to 50 μg of total RNA along with hybridization buffer (150 mM KCl, 10 mM Tris [pH 8.3], and 1 mM EDTA) in a final volume of 30 μl. The RNA and labeled primer were incubated in 90 μl of elongation buffer (20 mM Tris [pH 8.3], 10 mM MgCl2, 7 mM dithiothreitol) at 30°C for 16 h. A total of 20 units of avian myeloblastosis virus reverse transcriptase (Promega) was added, and the reaction mixture was incubated at 42°C for 1 h. The extension was terminated by adding 205 μl diethyl pyrocarbonate water, and the extension products were digested with 4.2 μg RNase A at 37°C for 15 min. The products were extracted with 300 μl phenol:chloroform:isoamyl alcohol (25:24:1) and precipitated by adding 30 μl of 3 M sodium acetate and 2.5 volumes of ethanol. The dried pellet was then suspended in 4 μl Tris-EDTA and 2 μl formamide sequencing loading buffer and subjected to electrophoresis in a 6% polyacrylamide gel containing 7 M urea, alongside a sequencing reaction ladder generated with labeled primer GNC12.
Preparation of end-labeled DNA and DNase I footprinting. The oligonucleotide SKF3For (50 pmol) was labeled with [-32P]ATP by use of T4 polynucleotide kinase (NEB) per the manufacturer's instructions. The probe was separated from unincorporated nucleotides with a G-25 MicroSpin column (Amersham Pharmacia Biotech, Piscataway, N.J.). The purified labeled probe was used in a PCR containing 36 pmol of unlabeled SKF4Rev primer and Herculase DNA polymerase (Stratagene). The PCR product was purified by elution from a G-50 MicroSpin column (Amersham Pharmacia Biotech, Piscataway, N.J.).
For DNase I protection assays, end-labeled DNA fragments were preincubated with or without the purified C-terminal domain of Spo0A (C-Spo0A), purified as described in reference 22, for 15 min in assay buffer [20 mM TrisCl (pH 7.4), 50 mM KCl, 1 mM dithiothreitol, poly(dI-dC) (2 ng/μl), 400 μl MgCl2, 200 μM CaCl2, bovine serum albumin (100 μg/ml)]. DNase I prepared from lyophilized enzyme (Sigma) was added at 100 ng/ml for 1 min at 37°C. The digestion was quenched by the addition of 10 volumes of ice-cold precipitation buffer (570 mM NH4OAc, 50 μg tRNA/ml, 80% ethanol). Electrophoretic analysis was carried out by resuspending the dried pellet from the protection reaction in 80% formamide-50 mM Tris-HCl-50 mM borate-1.4 mM Na2-EDTA-0.1% (wt/vol) xylene cyanol-0.1% (wt/vol) bromophenol blue. After denaturation by incubation at 90°C followed by chilling in ice, the samples were electrophoresed through 8% (wt/vol) acrylamide-8.3 M urea gels alongside a sequencing reaction generated with the labeled primer. The electrophoresis buffer was 90 mM Tris-HCl-89 mM borate-2.5 mM Na2-EDTA. After electrophoresis, the gels were exposed to KODAK XAR-2 film at –79°C with an intensifying screen.
-Galactosidase activity. Duplicate cultures were incubated in DSM with chloramphenicol (5 μg/ml). After 2 h of growth, two 300-μl aliquots of each culture were collected every half hour for 6 h. The first of these aliquots was used to measure the optical density, while bacteria from the second aliquot were harvested by centrifugation and stored at –80°C until assayed for -galactosidase (8).
RESULTS
Anatomy of the skf promoter. We used primer extension analysis to determine the start point of skf transcription. RNA was isolated about 2.5 h after the end of the exponential-growth phase from cultures of four strains (wild type, spo0A, abrB, and spo0A abrB) that contained an skfA'-'lacZ translational fusion integrated at the amyE locus of the chromosome. A primer complementary to part of the lacZ sequence was used in primer extension reactions, and the products were analyzed as described in Materials and Methods. The major transcription product found in RNA isolated from an otherwise wild-type strain mapped to a position 53 bp upstream from the start codon of skfA, AUG (Fig. 1 and 2). This product was absent in RNA isolated from an spo0A null strain (Fig. 2, lane h). In results consistent with measurements of the -galactosidase accumulated by the strains (see below), the RNA transcript was most abundant in RNA isolated from an abrB mutant strain (Fig. 2, lane a) and was present at a low level in a spo0A and abrB double-mutant strain (Fig. 2, lane b). The start point of transcription seen is consistent with the effects of mutations on promoter activity described below. A transcript that mapped to the same start point was observed in experiments using a primer that was complementary to part of the wild-type allele of skfA at its normal chromosomal location (data not shown).
We used the C-terminal domain of Spo0A (C-Spo0A), which binds and activates transcription independently of phosphorylation (11), in DNase I footprinting to precisely determine the site(s) at which Spo0A binds within the skf promoter. C-Spo0A protected a region extending approximately from bases –14 to –72 on the nontemplate strand of the skf promoter (Fig. 3). Within this protected region it was apparent that there were sites with different affinities. When the lowest amount of C-Spo0A was added (12 nM), we observed protection from bases –20 to –53 (Fig. 3, lane b), with base –19 showing hypersensitivity to cleavage that gradually diminished with increasing amounts of C-Spo0A. On the other hand, base –27 shows increased hypersensitivity with increasing concentrations of C-Spo0A (Fig. 3, lanes c to f). Examination of the sequence within this region for Spo0A binding sites identified a region centered at –28 that is a perfect match to the consensus Spo0A binding site (5'-TGNCGAA-3') (box 1 in Fig. 1) and other sites partially matching the consensus Spo0A binding site (boxes 2 to 4 in Fig. 1). Another DNase I-protected site was observed between bases –50 and –72 when we used higher concentrations of C-Spo0A (Fig. 3, lanes d to f). Spo0A is thought to bind as a head-to-tail dimer (31), and therefore we denote 0A boxes 1 and 2 as site 1 and boxes 3 and 4 as site 2. Site 1 is a higher-affinity site than site 2, which was expected because it contained the perfect consensus 0A binding site. In addition to these sites, titration with C-Spo0A also showed protection and hypersensitive bands downstream from the start point of transcription. The results determined by Fujita et al. (9) suggest that high levels of Spo0A act as a repressor for this promoter, perhaps by binding to these downstream sites. Since binding to these sites would mean that the promoter is repressed at higher levels of Spo0A, and with our interest being in how the promoter is activated, we did not try to resolve all the binding sites beyond the start point of transcription. Examination of C-Spo0A binding by use of the same DNA fragment that had been end labeled on the template strand showed a similar pattern of protection where site 1 had greater affinity than site 2 and sites beyond the transcriptional start site had the lowest affinity (data not shown). Examination of site 1 and site 2 in relationship to other conserved elements of the promoter shows a poorly conserved –35-like element centered at positions –32 and –33, which lies between the two 0A boxes of site 1 (Fig. 1). This places the highest affinity 0A binding site downstream of this –35-like element.
Spo0A binding sites are required for skf promoter activity. To test whether the sequences bound by Spo0A are required for skf promoter activity, we examined the effects of base pair substitutions at the consensus Spo0A binding site. The wild-type or mutant promoters were fused to a promoterless lacZ and integrated at the amyE chromosomal locus of a wild-type strain and in isogenic derivatives containing deletions of the spo0A and/or abrB locus. In a wild-type background, lacZ production under the control of the skf promoter was minimal until the end of the exponential-growth phase (T0) (Fig. 4A). After this point, the activity increased steadily until about T2, declining slightly as the culture approached T4 (Fig. 4A). Production of lacZ in a abrB strain showed a higher basal level of transcription as well as a slightly earlier activation (T–0.5). The activity in this strain was significantly higher than that seen in the wild-type background but reached a peak at a similar time (T2.5). Activity in the spo0A strain remained constant throughout the time course at a level similar to the basal level seen in the wild-type strain. We also observed the activity of the fusion in a spo0A abrB strain (Fig. 4A). As in the spo0A background, the activity in this strain was low; however, we did note that early activity (before T0) was greater than late activity (after T0.) All of these results were consistent with those previously reported by Fujita et al. (9).
Having confirmed the findings of Fujita et al. (9), we proceeded to investigate the consensus Spo0A binding site centered at position –28. We made base substitutions at positions –25, –27, and –29. The effects of these base-substituted promoter fusions were then examined in the spo0A, abrB, and spo0A abrB strains (Fig. 4B). All the base substitutions had a negative effect in the abrB strain, the most dramatic resulting from the –29 substitution, which reduced activity to basal level. There was no noticeable activity in the spo0A or spo0A abrB strains (data not shown). This confirmed that binding to the high-affinity site 1 was important for regulation of the skf promoter by Spo0A. We also created a truncated version of the promoter in which we deleted the site 2 binding sites (0A boxes 3 and 4) (e.g., strain AKB407) and tested the activation of the promoter in spo0A, abrB, and spo0A abrB strains. No activity from the promoter was detected in any of the strains (data not shown), suggesting that both sites (site 1 and site 2) are needed for optimal activation of the promoter by Spo0A.
Spo0A-dependent activation of the skf promoter requires an extended –10 region. The sequence centered 10 bp upstream from the putative start point of transcription identified by primer extension is identical to the consensus –10 region at five of six positions (5'-TATTAT-3') found at promoters used by RNA polymerase containing the primary sigma factor A (15) (Fig. 1). We examined the effects of several single-base-pair substitutions to test whether this sequence is important for promoter activity. When examined in a wild-type strain, base substitutions at positions –10 and –11 resulted in a decrease in promoter activity (Fig. 4C and data not shown). However, substitution of A for the T at position –9, which made the –10 sequence a perfect match to the consensus sequence, resulted in a significant increase in promoter activity (Fig. 1 and 4C). These results are consistent with the model that RNA polymerase containing A uses the skf promoter.
The sequence at the –10 region of the promoter is also similar to those of extended –10 region promoters, which contain a TRTGn motif (1, 5, 13, 26, 27). To determine whether this motif played a role in promoter activity we examined the effect of a 2-bp substitution at positions –15 and –14 (T to C and G to T, respectively). The activity for this promoter was significantly reduced in both wild-type and abrB mutant strains (Fig. 4D). These results show that the extended –10 motif, as well as other base pairs in the –10 region, is required for Spo0A-dependent activation of this promoter.
We also noted that there was a second sequence, centered between –6 and –7, that was similar to a consensus –10 sequence (nontemplate strand 5'-TATCGT-3'). The short distance between this region and the start of transcription led us to believe that it was unlikely to serve as the –10 site for the promoter. However, a double-base-pair substitution of AA for the CG within this sequence, which makes it more consensus-like, resulted in slightly elevated promoter activity (Fig. 1 and data not shown). Therefore, to further test the role of this potential –10-like sequence we examined the effect of a single-base-pair substitution at position –4, which alters the most highly conserved position in the –10 consensus sequence (15). If this sequence were acting as a –10 element we would expect that this change would reduce promoter activity. However, this substitution had no effect on promoter activity (data not shown).
A mutation in the skf promoter that results in Spo0A-independent activity. Extended –10 region promoters do not require extensive similarity to consensus at the –35 region for their activity (1, 5, 13, 26, 27). Therefore, it was unclear why the extended –10 region of the skf promoter was insufficient for promoter activity in the absence of Spo0A and AbrB. The –10 region of the skf promoter contains only a single nonconsensus base pair, located at position –9. As mentioned above, a base substitution conferring the consensus base to this position resulted in an overall increase in activity in the wild-type strain. We examined the effect of this base substitution in abrB and spo0A abrB strains and found that the substitution conferred Spo0A-independent activity to the promoter (Fig. 4E). Moreover, primer extension analyses showed that the start point of transcription from this promoter was identical to that of the wild-type promoter (data not shown). These results indicate that, at least for the skf promoter, the base pair at position –9 is critical for transcription activity and that a nonconsensus base pair at this position results in a promoter that requires the assistance of Spo0A despite the presence of an extended –10 motif.
DISCUSSION
Expression of the skf operon must be highly activated in response to low levels of phosphorylated Spo0A (9); however, its expression must be tightly regulated to prevent premature expression of the killing factor. This tight regulation of a highly active promoter involves two processes. The skf promoter is repressed by AbrB (9). Synthesis of AbrB is repressed by low levels of phosphorylated Spo0A; therefore, the production of low levels of Spo0A leads to reduced synthesis of AbrB and therefore derepression of the skf promoter activity. However, even in an abrB mutant strain, skf promoter activity also requires direct activation by Spo0A. The skf promoter contains high-affinity Spo0A binding sites, two of which span the –35 region of the promoter. We found that these sites, in addition to the upstream sites, are required for the Spo0A-dependent promoter activation.
An unusual feature of the skf promoter is its extended –10 region motif. Many promoters in Escherichia coli and B. subtilis contain a conserved sequence motif at the upstream end of the –10 region, the so-called extended –10 motif, or the TRTGn motif in B. subtilis (1, 5, 13, 26, 27). In E. coli this sequence can compensate for a weakly conserved –35 region. We found that the base substitutions in the TG motif located at positions –15 and –14 greatly reduce the Spo0A-dependent promoter activity. Evidently, RNA polymerase requires interaction with both Spo0A and the extended –10 region sequence to utilize the skf promoter efficiently. Because the extended –10 region plays a role in skf promoter activity, and because promoters containing extended –10 region sequences are thought to not require sequences at their –35 regions that are highly similar to the consensus –35 sequence, it seemed surprising that Spo0A was required for skf promoter activity even in the absence of AbrB repression. We found that a single-base-pair substitution at position –9 that produced a perfect match to a consensus –10 region yielded a promoter that no longer required Spo0A for its activation. Therefore, the specific base pair at position –9 of the promoter makes skf promoter activity directly dependent on Spo0A binding, and the extended –10 region motif of the promoter contributes to its high level of activity.
The best-understood mechanism of promoter activation by Spo0A is the activation of the high-threshold, A-dependent spoIIG promoter (4, 12, 19-21, 23). In this promoter, sequences similar to the –10 and –35 hexamers, which signal recognition of promoters by A-RNA polymerase, are separated by 22 bp, a distance that is ordinarily considered too great for productive interaction of these sequences with A (15). At this promoter, the consensus –35-like promoter element is centered between positions –37 and –36 and is overlapped by a Spo0A binding site (20). Based on molecular modeling and genetic assays, it has been proposed that binding of Spo0AP occludes binding of A region 4 to the consensus –35-like sequence and interaction between Spo0A and A positions in region 4 of A 18 bp upstream from the –10 region of the spoIIG promoter (12). This positioning of A allows region 2 of A to interact productively with the –10 region of the spoIIG promoter, thus stimulating its activity. It is likely that Spo0A activates some other A-dependent promoters (e.g., spoIIE and yneE) by a similar mechanism, because they are similar to spoIIG in that their –35-like sequences are separated from the –10 region by 21 bp (reference 30 and unpublished data). Moreover, their –35-like sequences overlap a Spo0A binding site (reference 30 and unpublished data). However, in contrast to these promoters the skf promoter does not have a consensus –35 sequence at or upstream from its –35 region, and the positions of Spo0A binding sites relative to the start point of transcription for the skf promoter differ from the spoIIG-like promoter sites. Therefore, activation of the skf promoter by Spo0A may involve novel interactions between Spo0A and RNA polymerase.
ACKNOWLEDGMENTS
We gratefully acknowledge Rich Losick and Gordon Churchward for their suggestions on the work.
The work was supported by Public Health Services grant GM54395 from the National Institute of General Medical Sciences.
G.C., A.K., and T.H.W. contributed equally to this work.
REFERENCES
Barne, K. A., J. A. Bown, S. J. Busby, and S. D. Minchin. 1997. Region 2.5 of the Escherichia coli RNA polymerase sigma70 subunit is responsible for the recognition of the "extended –10" motif at promoters. EMBO J. 16:4034-4040.
Ben-Yehuda, S., D. Z. Rudner, and R. Losick. 2003. RacA, a bacterial protein that anchors chromosomes to the cell poles. Science 299:532-536.
Bird, T., D. Burbulys, J. J. Wu, M. A. Strauch, J. A. Hoch, and G. B. Spiegelman. 1992. The effect of supercoiling on the in vitro transcription of the spoIIA operon from Bacillus subtilis. Biochimie 74:627-634.
Bird, T. H., J. K. Grimsley, J. A. Hoch, and G. B. Spiegelman. 1993. Phosphorylation of Spo0A activates its stimulation of in vitro transcription from the Bacillus subtilis spoIIG operon. Mol. Microbiol. 9:741-749.
Browning, D. F., and S. J. Busby. 2004. The regulation of bacterial transcription initiation. Nat. Rev. Microbiol. 2:57-65.
Burbulys, D., K. A. Trach, and J. A. Hoch. 1991. Initiation of sporulation in B. subtilis is controlled by a multicomponent phosphorelay. Cell 64:545-552.
Chung, J. D., G. Stephanopoulos, K. Ireton, and A. D. Grossman. 1994. Gene expression in single cells of Bacillus subtilis: evidence that a threshold mechanism controls the initiation of sporulation. J. Bacteriol. 176:1977-1984.
Cutting, S. M., and C. Harwood. 1990. Molecular biological methods for Bacillus. Wiley, Chichester, United Kingdom.
Fujita, M., J. E. Gonzalez-Pastor, and R. Losick. 2005. High- and low-threshold genes in the Spo0A regulon of Bacillus subtilis. J. Bacteriol. 187:1357-1368.
Gonzalez-Pastor, J. E., E. C. Hobbs, and R. Losick. 2003. Cannibalism by sporulating bacteria. Science 301:510-513.
Grimsley, J. K., R. B. Tjalkens, M. A. Strauch, T. H. Bird, G. B. Spiegelman, Z. Hostomsky, J. M. Whiteley, and J. A. Hoch. 1994. Subunit composition and domain structure of the Spo0A sporulation transcription factor of Bacillus subtilis. J. Biol. Chem. 269:16977-16982.
Kumar, A., C. Buckner Starke, M. DeZalia, and C. P. Moran, Jr. 2004. Surfaces of Spo0A and RNA polymerase sigma factor A that interact at the spoIIG promoter in Bacillus subtilis. J. Bacteriol. 186:200-206.
Mitchell, J. E., D. Zheng, S. J. Busby, and S. D. Minchin. 2003. Identification and analysis of "extended –10" promoters in Escherichia coli. Nucleic Acids Res. 31:4689-4695.
Molle, V., M. Fujita, S. T. Jensen, P. Eichenberger, J. E. Gonzalez-Pastor, J. S. Liu, and R. Losick. 2003. The Spo0A regulon of Bacillus subtilis. Mol. Microbiol. 50:1683-1701.
Moran, C. P., Jr., N. Lang, S. F. LeGrice, G. Lee, M. Stephens, A. L. Sonenshein, J. Pero, and R. Losick. 1982. Nucleotide sequences that signal the initiation of transcription and translation in Bacillus subtilis. Mol. Gen. Genet. 186:339-346.
Perego, M., P. Glaser, and J. A. Hoch. 1996. Aspartyl-phosphate phosphatases deactivate the response regulator components of the sporulation signal transduction system in Bacillus subtilis. Mol. Microbiol. 19:1151-1157.
Perego, M., J. J. Wu, G. B. Spiegelman, and J. A. Hoch. 1991. Mutational dissociation of the positive and negative regulatory properties of the Spo0A sporulation transcription factor of Bacillus subtilis. Gene 100:207-212.
Piggot, P. J., and D. W. Hilbert. 2004. Sporulation of Bacillus subtilis. Curr. Opin. Microbiol. 7:579-586.
Rowe-Magnus, D. A., and G. B. Spiegelman. 1998. DNA strand separation during activation of a developmental promoter by the Bacillus subtilis response regulator Spo0A. Proc. Natl. Acad. Sci. USA 95:5305-5310.
Satola, S., P. A. Kirchman, and C. P. Moran, Jr. 1991. Spo0A binds to a promoter used by sigma A RNA polymerase during sporulation in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 88:4533-4537.
Satola, S. W., J. M. Baldus, and C. P. Moran, Jr. 1992. Binding of Spo0A stimulates spoIIG promoter activity in Bacillus subtilis. J. Bacteriol. 174:1448-1453.
Schyns, G., C. M. Buckner, and C. P. Moran, Jr. 1997. Activation of the Bacillus subtilis spoIIG promoter requires interaction of Spo0A and the sigma subunit of RNA polymerase. J. Bacteriol. 179:5605-5608.
Seredick, S. D., B. M. Turner, and G. B. Spiegelman. 2003. Assay of transcription modulation by SpoOA of Bacillus subtilis. Methods Enzymol. 370:312-323.
Strauch, M., V. Webb, G. Spiegelman, and J. A. Hoch. 1990. The SpoOA protein of Bacillus subtilis is a repressor of the abrB gene. Proc. Natl. Acad. Sci. USA 87:1801-1805.
Trach, K. A., and J. A. Hoch. 1993. Multisensory activation of the phosphorelay initiating sporulation in Bacillus subtilis: identification and sequence of the protein kinase of the alternate pathway. Mol. Microbiol. 8:69-79.
Voskuil, M. I., and G. H. Chambliss. 1998. The –16 region of Bacillus subtilis and other gram-positive bacterial promoters. Nucleic Acids Res. 26:3584-3590.
Voskuil, M. I., K. Voepel, and G. H. Chambliss. 1995. The –16 region, a vital sequence for the utilization of a promoter in Bacillus subtilis and Escherichia coli. Mol. Microbiol. 17:271-279.
Weir, J., M. Predich, E. Dubnau, G. Nair, and I. Smith. 1991. Regulation of spo0H, a gene coding for the Bacillus subtilis H factor. J. Bacteriol. 173:521-529.
Wu, J. J., P. J. Piggot, K. M. Tatti, and C. P. Moran, Jr. 1991. Transcription of the Bacillus subtilis spoIIA locus. Gene 101:113-116.
York, K., T. J. Kenney, S. Satola, C. P. Moran, Jr., H. Poth, and P. Youngman. 1992. Spo0A controls the A-dependent activation of Bacillus subtilis sporulation-specific transcription unit spoIIE. J. Bacteriol. 174:2648-2658.
Zhao, H., T. Msadek, J. Zapf, Madhusudan, J. A. Hoch, and K. I. Varughese. 2002. DNA complexed structure of the key transcription factor initiating development in sporulating bacteria. Structure (Cambridge) 10:1041-1050.(Guangnan Chen, Amrita Kum)
ABSTRACT
At the onset of endospore formation in Bacillus subtilis the DNA-binding protein Spo0A directly activates transcription from promoters of about 40 genes. One of these promoters, Pskf, controls expression of an operon encoding a killing factor that acts on sibling cells. AbrB-mediated repression of Pskf provides one level of security ensuring that this promoter is not activated prematurely. However, Spo0A also appears to activate the promoter directly, since Spo0A is required for Pskf activity in a abrB strain. Here we investigate the mechanism of Pskf activation. DNase I footprinting was used to determine the locations at which Spo0A bound to the promoter, and mutations in these sites were found to significantly reduce promoter activity. The sequence near the –10 region of the promoter was found to be similar to those of extended –10 region promoters, which contain a TRTGn motif. Mutational analysis showed that this extended –10 region, as well as other base pairs in the –10 region, is required for Spo0A-dependent activation of the promoter. We found that a substitution of the consensus base pair for the nonconsensus base pair at position –9 of Pskf produced a promoter that was active constitutively in both abrB and spo0A abrB strains. Therefore, the base pair at position –9 of Pskf makes its activity dependent on Spo0A binding, and the extended –10 region motif of the promoter contributes to its high level of activity.
INTRODUCTION
Bacillus subtilis initiates a complex developmental program that leads to the differentiation of the cell into a dormant endospore as the final step in a series of responses to nutrient depletion (for a review, see reference 18). Initiation of this differentiation is dependent upon phosphorylation of a response regulator, Spo0A (6). A complex signal transduction network comprising multiple kinases and phosphatases controls the level of phosphorylated Spo0A (Spo0AP) in the cells (6, 16). Spo0AP directly regulates expression of about 121 genes through either repression or activation (14) by binding specifically to a 7-bp DNA element, referred to as an 0A box (5'-TGNCGAA-3') (24, 31). For example, Spo0AP binding directly represses the promoters for abrB, fruR, flgB, ftsE, sdp, and rocD (14, 17, 20). Other genes are directly activated by Spo0AP (14, 17, 20), and still others are activated by Spo0AP indirectly (14, 24). For example, the spo0H gene is repressed by AbrB until Spo0AP produced at the end of the exponential-growth phase represses abrB, relieving spo0H from AbrB-mediated repression (28). Moreover, Spo0A-activated promoters include those used by RNA polymerase containing the primary sigma factor, A (e.g., spoIIG and spoIIE promoters) (20, 30) and those used by RNA polymerase containing the secondary sigma factor, H (e.g., spoIIA and racA promoters) (2, 3, 29), which introduces additional combinations of control.
In a strain mutant with respect to both kinA and kinB, two of the kinases that control the level of Spo0AP, a low level of Spo0AP accumulates that is sufficient to repress abrB transcription but insufficient for the activation of other Spo0AP-dependent genes required for sporulation (25). Therefore, the responses of some Spo0A-regulated promoters to Spo0AP appear to be more sensitive to Spo0AP levels than are the responses of other Spo0A-regulated promoters. A progressive accumulation of Spo0AP at the end of the exponential-growth phase produces a temporal pattern of responses by Spo0A-regulated promoters that reflects the specific response of each promoter to low- or high-threshold levels of Spo0AP (9). Thus, in addition to the combinations of controls described above, Spo0A-regulated promoters can be classified as either high- or low-threshold activated or repressed promoters (9). High-threshold promoters require a high level of Spo0AP to be activated or repressed, probably in part because the regulatory regions for these genes have relatively weak affinities for Spo0AP, whereas low-threshold promoters respond to a low level of Spo0AP because their regulatory regions have relatively high-affinity binding sites for Spo0AP (9).
The progressive accumulation of Spo0AP allows cells to try less-drastic responses to nutrient depletion before commitment to differentiation into a dormant cell type. Moreover, Spo0A phosphorylation occurs in only a fraction of the population (7, 9, 10). Activation of Spo0A in this fraction of the population enables these cells to delay progression into sporulation by activating two operons, the sporulation killing factor (skf) and the sporulation delaying protein (sdp) operons (10). The products of these two operons prevent non-Spo0AP-expressing sibling cells from sporulating (sdp) and cause them to lyse (skf). The cells that have activated Spo0A early are then able to feed off the nutrients released, allowing them to continue growing and thus to delay their differentiation into dormant endospores. The skf promoter must be highly active to drive production of the secreted toxin, and it must be activated early in response to low levels of phosphorylated Spo0A (9, 10). However, its expression must be tightly regulated to prevent premature expression of the killing factor. Here we investigate how this promoter can be both highly active in response to low levels of Spo0AP and tightly controlled to prevent premature expression of the toxin.
The sequence of the skf promoter, which is activated by low levels of Spo0AP (9), appears to indicate that Spo0AP may directly activate this promoter by a mechanism that is different from that of the spoIIG promoter, which has been studied more extensively (4, 12, 19-21, 23). Comparison of the relative affinities of Spo0A for the high- and low-threshold promoters shows that spoIIG has a Kd of 1,700 nM whereas the skf promoter has an apparent Kd value of 26 nM (9). A bioinformatics search identified two Spo0A binding sites (14); however, the search did not reveal other features of the skf promoter that would contribute to its positive regulation by low levels of Spo0AP.
Here we report that skf is a A-dependent promoter with multiple Spo0A binding sites, two of which span the –35 region. Moreover, upstream of the mapped transcriptional start site is a sequence that is very similar to the TRTGn motif found in extended –10 promoters in Bacillus subtilis (13, 26, 27), another feature that is not present in the spoIIG promoter. We show that this sequence is necessary for the high level of Spo0A-dependent promoter activity. We also describe other features of this promoter that are necessary for its tight regulation by Spo0A.
MATERIALS AND METHODS
Bacterial strains and culture media. Routine microbiological manipulations and procedures were carried out by standard techniques as described in reference 8. The concentrations of antibiotics used for selection on Luria broth (LB) or Difco sporulation medium (DSM) were 5 μg/ml for chloramphenicol, 100 μg/ml for spectinomycin, 10 μg/ml for kanamycin, and 100 μg/ml for ampicillin. Cultures were grown in LB, and sporulation was induced by nutrient exhaustion in DSM. Competent cells were prepared and transformed by the two-step method as described in reference 8.
To create the abrB deletion strain (CMBS001) (Table 1), the 5'- and 3'-flanking DNA of abrB was PCR amplified from chromosomal DNA. The 5'-flanking DNA PCR product was cloned into pDG784 digested with SphI and PstI. The 3'-flanking DNA PCR fragment was then cloned into the resulting plasmid digested with BamHI and EcoRI to generate plasmid pDG784abrB. Plasmid pDG784abrB was sequenced to ensure the correct DNA sequences of the cloned fragments and was used to transform competent JH642. Kanamycin-resistant colonies were selected and analyzed further by PCR to determine whether substitution of the kanamycin resistance cassette for abrB occurred, as described in reference 12.
To clone the skf promoter, the DNA region from the stop codon (TAA) of ybcM (the upstream gene adjacent to the skf operon) to the start codon (ATG) of skfA (the first structural gene of the skf operon) was PCR amplified with oligonucleotides GNC1 and GNC2 (Table 2) and inserted into EcoRI- and HindIII-digested pBG1PLK to create pBG1PLK-SKF. The plasmid was checked by sequencing and used to transform competent JH642 to chloramphenicol resistance. Integration of pBG1PLK-SKF into the wild-type B. subtilis chromosomal amyE locus by homologous recombination was confirmed by testing for loss of amylase production by iodine staining and by PCR and DNA sequencing. The resulting strain was later transformed with abrB and/or spo0A knockout plasmids (12), producing corresponding B. subtilis strains (Table 1).
Overlapping PCR was used to introduce base pair substitutions in the wild-type skf promoter. First, two QuikChange oligonucleotides were designed that were complementary to each other and annealed to the skf promoter but carried the desired nucleotide change. Each was used in combination with an external oligonucleotide (GNC1 or GNC2) to perform PCR using the plasmid pBG1PLK-SKF as the template. The two amplified products, which correspond to the 5' and 3' portions of the skf promoter and partially overlap the mutated region, were gel purified. The products were then mixed, annealed, and used as the template for the second step of PCR performed with the two external primers, GNC1 and GNC2. The product obtained, which should carry the skf promoter sequence with the desired substitution, was sequenced and then inserted into EcoRI- and HindIII-digested pBG1PLK. The resulting plasmids (Table 1) were then used to transform competent JH642.
RNA preparations. Cultures of B. subtilis strains THWB2, THWB18, THWB24 and THWB9 (Table 1) were incubated in DSM. Two hours after the end of the exponential-growth phase, the bacteria were harvested by centrifugation and stored at –80°C. RNA was prepared as described in reference 8.
Primer extension reactions. A total of 100 ng of primer GNC12 (Table 2) was end labeled with [-32P]ATP by use of T4 polynucleotide kinase from New England Biolabs. The labeled primer was purified using an Amersham Biosciences MicroSpin G-25 column (Amersham Pharmacia Biotech, Piscataway, N.J.). A 15-ng volume of this purified labeled primer was added to 50 μg of total RNA along with hybridization buffer (150 mM KCl, 10 mM Tris [pH 8.3], and 1 mM EDTA) in a final volume of 30 μl. The RNA and labeled primer were incubated in 90 μl of elongation buffer (20 mM Tris [pH 8.3], 10 mM MgCl2, 7 mM dithiothreitol) at 30°C for 16 h. A total of 20 units of avian myeloblastosis virus reverse transcriptase (Promega) was added, and the reaction mixture was incubated at 42°C for 1 h. The extension was terminated by adding 205 μl diethyl pyrocarbonate water, and the extension products were digested with 4.2 μg RNase A at 37°C for 15 min. The products were extracted with 300 μl phenol:chloroform:isoamyl alcohol (25:24:1) and precipitated by adding 30 μl of 3 M sodium acetate and 2.5 volumes of ethanol. The dried pellet was then suspended in 4 μl Tris-EDTA and 2 μl formamide sequencing loading buffer and subjected to electrophoresis in a 6% polyacrylamide gel containing 7 M urea, alongside a sequencing reaction ladder generated with labeled primer GNC12.
Preparation of end-labeled DNA and DNase I footprinting. The oligonucleotide SKF3For (50 pmol) was labeled with [-32P]ATP by use of T4 polynucleotide kinase (NEB) per the manufacturer's instructions. The probe was separated from unincorporated nucleotides with a G-25 MicroSpin column (Amersham Pharmacia Biotech, Piscataway, N.J.). The purified labeled probe was used in a PCR containing 36 pmol of unlabeled SKF4Rev primer and Herculase DNA polymerase (Stratagene). The PCR product was purified by elution from a G-50 MicroSpin column (Amersham Pharmacia Biotech, Piscataway, N.J.).
For DNase I protection assays, end-labeled DNA fragments were preincubated with or without the purified C-terminal domain of Spo0A (C-Spo0A), purified as described in reference 22, for 15 min in assay buffer [20 mM TrisCl (pH 7.4), 50 mM KCl, 1 mM dithiothreitol, poly(dI-dC) (2 ng/μl), 400 μl MgCl2, 200 μM CaCl2, bovine serum albumin (100 μg/ml)]. DNase I prepared from lyophilized enzyme (Sigma) was added at 100 ng/ml for 1 min at 37°C. The digestion was quenched by the addition of 10 volumes of ice-cold precipitation buffer (570 mM NH4OAc, 50 μg tRNA/ml, 80% ethanol). Electrophoretic analysis was carried out by resuspending the dried pellet from the protection reaction in 80% formamide-50 mM Tris-HCl-50 mM borate-1.4 mM Na2-EDTA-0.1% (wt/vol) xylene cyanol-0.1% (wt/vol) bromophenol blue. After denaturation by incubation at 90°C followed by chilling in ice, the samples were electrophoresed through 8% (wt/vol) acrylamide-8.3 M urea gels alongside a sequencing reaction generated with the labeled primer. The electrophoresis buffer was 90 mM Tris-HCl-89 mM borate-2.5 mM Na2-EDTA. After electrophoresis, the gels were exposed to KODAK XAR-2 film at –79°C with an intensifying screen.
-Galactosidase activity. Duplicate cultures were incubated in DSM with chloramphenicol (5 μg/ml). After 2 h of growth, two 300-μl aliquots of each culture were collected every half hour for 6 h. The first of these aliquots was used to measure the optical density, while bacteria from the second aliquot were harvested by centrifugation and stored at –80°C until assayed for -galactosidase (8).
RESULTS
Anatomy of the skf promoter. We used primer extension analysis to determine the start point of skf transcription. RNA was isolated about 2.5 h after the end of the exponential-growth phase from cultures of four strains (wild type, spo0A, abrB, and spo0A abrB) that contained an skfA'-'lacZ translational fusion integrated at the amyE locus of the chromosome. A primer complementary to part of the lacZ sequence was used in primer extension reactions, and the products were analyzed as described in Materials and Methods. The major transcription product found in RNA isolated from an otherwise wild-type strain mapped to a position 53 bp upstream from the start codon of skfA, AUG (Fig. 1 and 2). This product was absent in RNA isolated from an spo0A null strain (Fig. 2, lane h). In results consistent with measurements of the -galactosidase accumulated by the strains (see below), the RNA transcript was most abundant in RNA isolated from an abrB mutant strain (Fig. 2, lane a) and was present at a low level in a spo0A and abrB double-mutant strain (Fig. 2, lane b). The start point of transcription seen is consistent with the effects of mutations on promoter activity described below. A transcript that mapped to the same start point was observed in experiments using a primer that was complementary to part of the wild-type allele of skfA at its normal chromosomal location (data not shown).
We used the C-terminal domain of Spo0A (C-Spo0A), which binds and activates transcription independently of phosphorylation (11), in DNase I footprinting to precisely determine the site(s) at which Spo0A binds within the skf promoter. C-Spo0A protected a region extending approximately from bases –14 to –72 on the nontemplate strand of the skf promoter (Fig. 3). Within this protected region it was apparent that there were sites with different affinities. When the lowest amount of C-Spo0A was added (12 nM), we observed protection from bases –20 to –53 (Fig. 3, lane b), with base –19 showing hypersensitivity to cleavage that gradually diminished with increasing amounts of C-Spo0A. On the other hand, base –27 shows increased hypersensitivity with increasing concentrations of C-Spo0A (Fig. 3, lanes c to f). Examination of the sequence within this region for Spo0A binding sites identified a region centered at –28 that is a perfect match to the consensus Spo0A binding site (5'-TGNCGAA-3') (box 1 in Fig. 1) and other sites partially matching the consensus Spo0A binding site (boxes 2 to 4 in Fig. 1). Another DNase I-protected site was observed between bases –50 and –72 when we used higher concentrations of C-Spo0A (Fig. 3, lanes d to f). Spo0A is thought to bind as a head-to-tail dimer (31), and therefore we denote 0A boxes 1 and 2 as site 1 and boxes 3 and 4 as site 2. Site 1 is a higher-affinity site than site 2, which was expected because it contained the perfect consensus 0A binding site. In addition to these sites, titration with C-Spo0A also showed protection and hypersensitive bands downstream from the start point of transcription. The results determined by Fujita et al. (9) suggest that high levels of Spo0A act as a repressor for this promoter, perhaps by binding to these downstream sites. Since binding to these sites would mean that the promoter is repressed at higher levels of Spo0A, and with our interest being in how the promoter is activated, we did not try to resolve all the binding sites beyond the start point of transcription. Examination of C-Spo0A binding by use of the same DNA fragment that had been end labeled on the template strand showed a similar pattern of protection where site 1 had greater affinity than site 2 and sites beyond the transcriptional start site had the lowest affinity (data not shown). Examination of site 1 and site 2 in relationship to other conserved elements of the promoter shows a poorly conserved –35-like element centered at positions –32 and –33, which lies between the two 0A boxes of site 1 (Fig. 1). This places the highest affinity 0A binding site downstream of this –35-like element.
Spo0A binding sites are required for skf promoter activity. To test whether the sequences bound by Spo0A are required for skf promoter activity, we examined the effects of base pair substitutions at the consensus Spo0A binding site. The wild-type or mutant promoters were fused to a promoterless lacZ and integrated at the amyE chromosomal locus of a wild-type strain and in isogenic derivatives containing deletions of the spo0A and/or abrB locus. In a wild-type background, lacZ production under the control of the skf promoter was minimal until the end of the exponential-growth phase (T0) (Fig. 4A). After this point, the activity increased steadily until about T2, declining slightly as the culture approached T4 (Fig. 4A). Production of lacZ in a abrB strain showed a higher basal level of transcription as well as a slightly earlier activation (T–0.5). The activity in this strain was significantly higher than that seen in the wild-type background but reached a peak at a similar time (T2.5). Activity in the spo0A strain remained constant throughout the time course at a level similar to the basal level seen in the wild-type strain. We also observed the activity of the fusion in a spo0A abrB strain (Fig. 4A). As in the spo0A background, the activity in this strain was low; however, we did note that early activity (before T0) was greater than late activity (after T0.) All of these results were consistent with those previously reported by Fujita et al. (9).
Having confirmed the findings of Fujita et al. (9), we proceeded to investigate the consensus Spo0A binding site centered at position –28. We made base substitutions at positions –25, –27, and –29. The effects of these base-substituted promoter fusions were then examined in the spo0A, abrB, and spo0A abrB strains (Fig. 4B). All the base substitutions had a negative effect in the abrB strain, the most dramatic resulting from the –29 substitution, which reduced activity to basal level. There was no noticeable activity in the spo0A or spo0A abrB strains (data not shown). This confirmed that binding to the high-affinity site 1 was important for regulation of the skf promoter by Spo0A. We also created a truncated version of the promoter in which we deleted the site 2 binding sites (0A boxes 3 and 4) (e.g., strain AKB407) and tested the activation of the promoter in spo0A, abrB, and spo0A abrB strains. No activity from the promoter was detected in any of the strains (data not shown), suggesting that both sites (site 1 and site 2) are needed for optimal activation of the promoter by Spo0A.
Spo0A-dependent activation of the skf promoter requires an extended –10 region. The sequence centered 10 bp upstream from the putative start point of transcription identified by primer extension is identical to the consensus –10 region at five of six positions (5'-TATTAT-3') found at promoters used by RNA polymerase containing the primary sigma factor A (15) (Fig. 1). We examined the effects of several single-base-pair substitutions to test whether this sequence is important for promoter activity. When examined in a wild-type strain, base substitutions at positions –10 and –11 resulted in a decrease in promoter activity (Fig. 4C and data not shown). However, substitution of A for the T at position –9, which made the –10 sequence a perfect match to the consensus sequence, resulted in a significant increase in promoter activity (Fig. 1 and 4C). These results are consistent with the model that RNA polymerase containing A uses the skf promoter.
The sequence at the –10 region of the promoter is also similar to those of extended –10 region promoters, which contain a TRTGn motif (1, 5, 13, 26, 27). To determine whether this motif played a role in promoter activity we examined the effect of a 2-bp substitution at positions –15 and –14 (T to C and G to T, respectively). The activity for this promoter was significantly reduced in both wild-type and abrB mutant strains (Fig. 4D). These results show that the extended –10 motif, as well as other base pairs in the –10 region, is required for Spo0A-dependent activation of this promoter.
We also noted that there was a second sequence, centered between –6 and –7, that was similar to a consensus –10 sequence (nontemplate strand 5'-TATCGT-3'). The short distance between this region and the start of transcription led us to believe that it was unlikely to serve as the –10 site for the promoter. However, a double-base-pair substitution of AA for the CG within this sequence, which makes it more consensus-like, resulted in slightly elevated promoter activity (Fig. 1 and data not shown). Therefore, to further test the role of this potential –10-like sequence we examined the effect of a single-base-pair substitution at position –4, which alters the most highly conserved position in the –10 consensus sequence (15). If this sequence were acting as a –10 element we would expect that this change would reduce promoter activity. However, this substitution had no effect on promoter activity (data not shown).
A mutation in the skf promoter that results in Spo0A-independent activity. Extended –10 region promoters do not require extensive similarity to consensus at the –35 region for their activity (1, 5, 13, 26, 27). Therefore, it was unclear why the extended –10 region of the skf promoter was insufficient for promoter activity in the absence of Spo0A and AbrB. The –10 region of the skf promoter contains only a single nonconsensus base pair, located at position –9. As mentioned above, a base substitution conferring the consensus base to this position resulted in an overall increase in activity in the wild-type strain. We examined the effect of this base substitution in abrB and spo0A abrB strains and found that the substitution conferred Spo0A-independent activity to the promoter (Fig. 4E). Moreover, primer extension analyses showed that the start point of transcription from this promoter was identical to that of the wild-type promoter (data not shown). These results indicate that, at least for the skf promoter, the base pair at position –9 is critical for transcription activity and that a nonconsensus base pair at this position results in a promoter that requires the assistance of Spo0A despite the presence of an extended –10 motif.
DISCUSSION
Expression of the skf operon must be highly activated in response to low levels of phosphorylated Spo0A (9); however, its expression must be tightly regulated to prevent premature expression of the killing factor. This tight regulation of a highly active promoter involves two processes. The skf promoter is repressed by AbrB (9). Synthesis of AbrB is repressed by low levels of phosphorylated Spo0A; therefore, the production of low levels of Spo0A leads to reduced synthesis of AbrB and therefore derepression of the skf promoter activity. However, even in an abrB mutant strain, skf promoter activity also requires direct activation by Spo0A. The skf promoter contains high-affinity Spo0A binding sites, two of which span the –35 region of the promoter. We found that these sites, in addition to the upstream sites, are required for the Spo0A-dependent promoter activation.
An unusual feature of the skf promoter is its extended –10 region motif. Many promoters in Escherichia coli and B. subtilis contain a conserved sequence motif at the upstream end of the –10 region, the so-called extended –10 motif, or the TRTGn motif in B. subtilis (1, 5, 13, 26, 27). In E. coli this sequence can compensate for a weakly conserved –35 region. We found that the base substitutions in the TG motif located at positions –15 and –14 greatly reduce the Spo0A-dependent promoter activity. Evidently, RNA polymerase requires interaction with both Spo0A and the extended –10 region sequence to utilize the skf promoter efficiently. Because the extended –10 region plays a role in skf promoter activity, and because promoters containing extended –10 region sequences are thought to not require sequences at their –35 regions that are highly similar to the consensus –35 sequence, it seemed surprising that Spo0A was required for skf promoter activity even in the absence of AbrB repression. We found that a single-base-pair substitution at position –9 that produced a perfect match to a consensus –10 region yielded a promoter that no longer required Spo0A for its activation. Therefore, the specific base pair at position –9 of the promoter makes skf promoter activity directly dependent on Spo0A binding, and the extended –10 region motif of the promoter contributes to its high level of activity.
The best-understood mechanism of promoter activation by Spo0A is the activation of the high-threshold, A-dependent spoIIG promoter (4, 12, 19-21, 23). In this promoter, sequences similar to the –10 and –35 hexamers, which signal recognition of promoters by A-RNA polymerase, are separated by 22 bp, a distance that is ordinarily considered too great for productive interaction of these sequences with A (15). At this promoter, the consensus –35-like promoter element is centered between positions –37 and –36 and is overlapped by a Spo0A binding site (20). Based on molecular modeling and genetic assays, it has been proposed that binding of Spo0AP occludes binding of A region 4 to the consensus –35-like sequence and interaction between Spo0A and A positions in region 4 of A 18 bp upstream from the –10 region of the spoIIG promoter (12). This positioning of A allows region 2 of A to interact productively with the –10 region of the spoIIG promoter, thus stimulating its activity. It is likely that Spo0A activates some other A-dependent promoters (e.g., spoIIE and yneE) by a similar mechanism, because they are similar to spoIIG in that their –35-like sequences are separated from the –10 region by 21 bp (reference 30 and unpublished data). Moreover, their –35-like sequences overlap a Spo0A binding site (reference 30 and unpublished data). However, in contrast to these promoters the skf promoter does not have a consensus –35 sequence at or upstream from its –35 region, and the positions of Spo0A binding sites relative to the start point of transcription for the skf promoter differ from the spoIIG-like promoter sites. Therefore, activation of the skf promoter by Spo0A may involve novel interactions between Spo0A and RNA polymerase.
ACKNOWLEDGMENTS
We gratefully acknowledge Rich Losick and Gordon Churchward for their suggestions on the work.
The work was supported by Public Health Services grant GM54395 from the National Institute of General Medical Sciences.
G.C., A.K., and T.H.W. contributed equally to this work.
REFERENCES
Barne, K. A., J. A. Bown, S. J. Busby, and S. D. Minchin. 1997. Region 2.5 of the Escherichia coli RNA polymerase sigma70 subunit is responsible for the recognition of the "extended –10" motif at promoters. EMBO J. 16:4034-4040.
Ben-Yehuda, S., D. Z. Rudner, and R. Losick. 2003. RacA, a bacterial protein that anchors chromosomes to the cell poles. Science 299:532-536.
Bird, T., D. Burbulys, J. J. Wu, M. A. Strauch, J. A. Hoch, and G. B. Spiegelman. 1992. The effect of supercoiling on the in vitro transcription of the spoIIA operon from Bacillus subtilis. Biochimie 74:627-634.
Bird, T. H., J. K. Grimsley, J. A. Hoch, and G. B. Spiegelman. 1993. Phosphorylation of Spo0A activates its stimulation of in vitro transcription from the Bacillus subtilis spoIIG operon. Mol. Microbiol. 9:741-749.
Browning, D. F., and S. J. Busby. 2004. The regulation of bacterial transcription initiation. Nat. Rev. Microbiol. 2:57-65.
Burbulys, D., K. A. Trach, and J. A. Hoch. 1991. Initiation of sporulation in B. subtilis is controlled by a multicomponent phosphorelay. Cell 64:545-552.
Chung, J. D., G. Stephanopoulos, K. Ireton, and A. D. Grossman. 1994. Gene expression in single cells of Bacillus subtilis: evidence that a threshold mechanism controls the initiation of sporulation. J. Bacteriol. 176:1977-1984.
Cutting, S. M., and C. Harwood. 1990. Molecular biological methods for Bacillus. Wiley, Chichester, United Kingdom.
Fujita, M., J. E. Gonzalez-Pastor, and R. Losick. 2005. High- and low-threshold genes in the Spo0A regulon of Bacillus subtilis. J. Bacteriol. 187:1357-1368.
Gonzalez-Pastor, J. E., E. C. Hobbs, and R. Losick. 2003. Cannibalism by sporulating bacteria. Science 301:510-513.
Grimsley, J. K., R. B. Tjalkens, M. A. Strauch, T. H. Bird, G. B. Spiegelman, Z. Hostomsky, J. M. Whiteley, and J. A. Hoch. 1994. Subunit composition and domain structure of the Spo0A sporulation transcription factor of Bacillus subtilis. J. Biol. Chem. 269:16977-16982.
Kumar, A., C. Buckner Starke, M. DeZalia, and C. P. Moran, Jr. 2004. Surfaces of Spo0A and RNA polymerase sigma factor A that interact at the spoIIG promoter in Bacillus subtilis. J. Bacteriol. 186:200-206.
Mitchell, J. E., D. Zheng, S. J. Busby, and S. D. Minchin. 2003. Identification and analysis of "extended –10" promoters in Escherichia coli. Nucleic Acids Res. 31:4689-4695.
Molle, V., M. Fujita, S. T. Jensen, P. Eichenberger, J. E. Gonzalez-Pastor, J. S. Liu, and R. Losick. 2003. The Spo0A regulon of Bacillus subtilis. Mol. Microbiol. 50:1683-1701.
Moran, C. P., Jr., N. Lang, S. F. LeGrice, G. Lee, M. Stephens, A. L. Sonenshein, J. Pero, and R. Losick. 1982. Nucleotide sequences that signal the initiation of transcription and translation in Bacillus subtilis. Mol. Gen. Genet. 186:339-346.
Perego, M., P. Glaser, and J. A. Hoch. 1996. Aspartyl-phosphate phosphatases deactivate the response regulator components of the sporulation signal transduction system in Bacillus subtilis. Mol. Microbiol. 19:1151-1157.
Perego, M., J. J. Wu, G. B. Spiegelman, and J. A. Hoch. 1991. Mutational dissociation of the positive and negative regulatory properties of the Spo0A sporulation transcription factor of Bacillus subtilis. Gene 100:207-212.
Piggot, P. J., and D. W. Hilbert. 2004. Sporulation of Bacillus subtilis. Curr. Opin. Microbiol. 7:579-586.
Rowe-Magnus, D. A., and G. B. Spiegelman. 1998. DNA strand separation during activation of a developmental promoter by the Bacillus subtilis response regulator Spo0A. Proc. Natl. Acad. Sci. USA 95:5305-5310.
Satola, S., P. A. Kirchman, and C. P. Moran, Jr. 1991. Spo0A binds to a promoter used by sigma A RNA polymerase during sporulation in Bacillus subtilis. Proc. Natl. Acad. Sci. USA 88:4533-4537.
Satola, S. W., J. M. Baldus, and C. P. Moran, Jr. 1992. Binding of Spo0A stimulates spoIIG promoter activity in Bacillus subtilis. J. Bacteriol. 174:1448-1453.
Schyns, G., C. M. Buckner, and C. P. Moran, Jr. 1997. Activation of the Bacillus subtilis spoIIG promoter requires interaction of Spo0A and the sigma subunit of RNA polymerase. J. Bacteriol. 179:5605-5608.
Seredick, S. D., B. M. Turner, and G. B. Spiegelman. 2003. Assay of transcription modulation by SpoOA of Bacillus subtilis. Methods Enzymol. 370:312-323.
Strauch, M., V. Webb, G. Spiegelman, and J. A. Hoch. 1990. The SpoOA protein of Bacillus subtilis is a repressor of the abrB gene. Proc. Natl. Acad. Sci. USA 87:1801-1805.
Trach, K. A., and J. A. Hoch. 1993. Multisensory activation of the phosphorelay initiating sporulation in Bacillus subtilis: identification and sequence of the protein kinase of the alternate pathway. Mol. Microbiol. 8:69-79.
Voskuil, M. I., and G. H. Chambliss. 1998. The –16 region of Bacillus subtilis and other gram-positive bacterial promoters. Nucleic Acids Res. 26:3584-3590.
Voskuil, M. I., K. Voepel, and G. H. Chambliss. 1995. The –16 region, a vital sequence for the utilization of a promoter in Bacillus subtilis and Escherichia coli. Mol. Microbiol. 17:271-279.
Weir, J., M. Predich, E. Dubnau, G. Nair, and I. Smith. 1991. Regulation of spo0H, a gene coding for the Bacillus subtilis H factor. J. Bacteriol. 173:521-529.
Wu, J. J., P. J. Piggot, K. M. Tatti, and C. P. Moran, Jr. 1991. Transcription of the Bacillus subtilis spoIIA locus. Gene 101:113-116.
York, K., T. J. Kenney, S. Satola, C. P. Moran, Jr., H. Poth, and P. Youngman. 1992. Spo0A controls the A-dependent activation of Bacillus subtilis sporulation-specific transcription unit spoIIE. J. Bacteriol. 174:2648-2658.
Zhao, H., T. Msadek, J. Zapf, Madhusudan, J. A. Hoch, and K. I. Varughese. 2002. DNA complexed structure of the key transcription factor initiating development in sporulating bacteria. Structure (Cambridge) 10:1041-1050.(Guangnan Chen, Amrita Kum)