This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Geng, H.
Right arrow Articles by Nakano, M. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Geng, H.
Right arrow Articles by Nakano, M. M.

 Previous Article  |  Next Article 

Journal of Bacteriology, March 2007, p. 1745-1755, Vol. 189, No. 5
0021-9193/07/$08.00+0     doi:10.1128/JB.01502-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Characterization of ResDE-Dependent fnr Transcription in Bacillus subtilis{triangledown}

Hao Geng, Yi Zhu, Karl Mullen, Cole S. Zuber, and Michiko M. Nakano*

Department of Environmental and Biomolecular Systems, OGI School of Science and Engineering, Oregon Health and Science University, 20000 NW Walker Road, Beaverton, Oregon 97006

Received 24 September 2006/ Accepted 12 December 2006


arrow
ABSTRACT
 
The ResD-ResE signal transduction system is required for transcription of genes involved in aerobic and anaerobic respiration in Bacillus subtilis. Phosphorylated ResD (ResD~P) interacts with target DNA to activate transcription. A strong sequence similarity was detected in promoter regions of some ResD-controlled genes including fnr and resA. Single-base substitutions in the fnr and resA promoters were performed to determine a ResD-binding sequence. DNase I footprinting analysis indicated that ResD~P itself does not bind to fnr, but interaction of ResD~P with the C-terminal domain of the {alpha} subunit ({alpha}CTD) of RNA polymerase (RNAP) facilitates cooperative binding of ResD~P and RNAP, thereby increasing fnr transcription initiation. Consistent with this result, amino acid substitutions in {alpha}CTD, such as Y263A, K267A, A269I, or N290A, sharply reduced fnr transcription in vivo, and the K267A {alpha}CTD protein, unlike the wild-type protein, did not increase ResD~P binding to the fnr promoter. Amino acid residues of {alpha}CTD required for ResD-dependent fnr transcription, with the exception of N290, which may interact with DNA, constitute a distinct surface, suggesting that these residues likely interact with ResD~P.


arrow
INTRODUCTION
 
The ResD-ResE signal transduction system is required for aerobic and anaerobic respiration in Bacillus subtilis (25, 30). ResE is a membrane-bound sensor histidine kinase that, upon autophosphorylation, donates a high-energy phosphate to its cognate response regulator, ResD. The resD and resE genes constitute an operon with the three upstream genes, resABC (30). ResA is a thiol-disulfide oxidoreductase involved in cytochrome c maturation (5, 6, 8), and ResB and ResC were also shown to play an essential role in cytochrome c synthesis (15). resD and resE are transcribed from a resDE-specific promoter and the resA operon promoter, the latter of which is dependent on ResD and ResE (30).

ResDE-controlled genes that are involved in anaerobic respiration include fnr encoding an anaerobic transcriptional regulator and nasDEF, which constitute an operon encoding subunits of nitrite reductase (21, 25). Fnr is essential for nitrate respiration, as expression of the respiratory nitrate reductase operon, narGHJI, is dependent on Fnr (7). In addition, expression of the flavohemoglobin gene hmp is highly induced by ResDE upon oxygen limitation (14). Transcription of fnr, nasD, and hmp is activated in vitro by ResD when the protein is phosphorylated by incubation with ResE and ATP (10), indicating that phosphorylated ResD (ResD~P) directly interacts with the regulatory region of these genes.

Direct interaction of ResD~P with ResD-controlled promoters, which was confirmed by DNase I or hydroxyl radical footprinting, in ctaA encoding heme A synthase (31), nasD (10, 24), hmp (10, 24), fnr (24), and yclJK encoding a two-component regulatory protein pair (11) has been reported. Binding of ResD~P to fnr was observed either at a higher concentration of ResD~P (24) or was not detected (10). ResD~P bound to three distinct regions of ctaA (A1, positions –209 to –179 relative to the transcription start site; A2, –108 to –55; A3, –2 to +43) (31). The A1 site may be involved in ResD-dependent activation of the divergently transcribed ctaBCDEF operon encoding heme O synthase and the subunits of cytochrome caa3 (17). A consensus ResD-binding site was proposed as TTTGTGAAT (consensus sequence a in Fig. 1A) by a sequence alignment of nasD, hmp, fnr, and ctaA (24, 31). Subsequently, a bioinformatics approach was used to compare ResD-binding sites in the nasD, hmp, fnr, yclJ, and ctaA promoters and TTGTN6TTTNTN2A (consensus sequence b in Fig. 1B) was proposed as a revised consensus ResD-binding sequence (11). The validity of the two proposed consensus sequences has pros and cons (Fig. 1A and B). For example, only a weak similarity (five of nine matches) to consensus sequence a was found in the ResD-binding region (–92 to –68) of yclJ (11), whereas the binding region contains a sequence with a perfect match to consensus sequence b. On the other hand, consensus sequence a, but not consensus sequence b, was identified in the ctaA3 region. We had previously shown that a region including –61 to –58 of the fnr promoter is indispensable for ResD-dependent activation (24); however, consensus sequence b in the fnr promoter does not include the sequence. Finally, although deletion analysis showed that a nasD region between –87 and –76 is critical for ResD-dependent activation, this region is not included either in consensus sequence a or consensus sequence b. Both consensus sequences share 5' TTGT, and the major difference between these sequences is GAA adjacent to TTGT that was proposed only for consensus sequence a.


Figure 1
View larger version (40K):
[in this window]
[in a new window]

  		FIG. 1. Comparison of proposed consensus ResD-binding sequences. (A) Consensus sequence a was proposed by a sequence alignment of experimentally detected ResD-binding regions (24, 31). Numbering is relative to the transcription start site. ctaA has three distinct ResD-binding regions (31). Bases identical to those in the consensus sequence are indicated by bold type, and the numbers in parentheses show the number of bases of the total number of bases that match those in the consensus sequence. An asterisk indicates the sequence on the complementary strand. (B) Consensus sequence b was defined by bioinformatics approach (11). Sequence similarity with consensus sequence b is not detected in ctaA3. (C) A newly proposed consensus ResD-binding sequence (consensus sequence c) for nasD, hmp, and fnr. The full consensus sequence is shown by a box above the sequence. Half-site a is TTGTGAAN2, and half-site b is NTTTN4A. The arrows show the orientation of each half-site. Nucleotides doubly underlined are the nucleotides protected by ResD~P from attack by hydroxyl radicals, and a dotted double line shows protected nucleotides located on the opposite face of DNA helix (10).

Meanwhile, we noticed that there is a strong similarity among the promoter regions of ResD-controlled genes, namely, resA, fnr, and sboA encoding subtilosin (23, 33). As shown in Fig. 2A, a direct repeat (TTCA N7 TTCA) is present in resA and sboA, and a single TTCA sequence is present in fnr. Interestingly, TGAA, which is complementary to TTCA, is the sequence present in consensus sequence a but not in consensus sequence b of ctaA1, ctaA2, ctaA3, and yclJ (Fig. 1A). As shown in Fig. 1A, ctaA1 carries the exact match to the TTCA N7 TTCA sequence on the complementary strand, which is likely used as a ResD-binding sequence of the divergently transcribed ctaBCDEF. The direct repeat is also present in the ResD-binding consensus sequence a of yclJK identified by DNase I footprinting experiments (11) but on the noncoding strand (Fig. 1A), suggesting that the yclIH gene transcribed divergently from yclJK might be regulated by ResDE as well. Expression of lacZ fused to the yclI promoter region was activated upon oxygen limitation, and this activation required ResDE (C. S. Zuber and M. M. Nakano, unpublished results), indicating that yclIH also belongs to the ResDE regulon. These results suggested that TGAA (and TTCA) proposed as a part of consensus sequence a is indeed used as a ResD-binding site. In this study, we generated base substitutions in the fnr and resA promoters to determine whether the TTCA sequence is required for ResD-dependent control. Furthermore, we investigated how binding of ResD~P to the fnr promoter facilitates transcription initiation. DNase I footprinting analysis and in vivo studies using the C-terminal domain of the {alpha} subunit ({alpha}CTD) alanine-scanning mutants indicated that interaction of ResD~P with {alpha}CTD is required for activation of fnr transcription.


Figure 2
View larger version (37K):
[in this window]
[in a new window]

  		FIG. 2. Mutational analysis of the fnr and resA promoters. (A) Comparison of the promoter regions of ResDE-dependent sboA, resA, and fnr. The positions of nucleotides are relative to the transcription start site. The conserved TTCA sequence is underlined. (B) Effects of base substitutions in the fnr promoter region on transcription. (C) Effects of base substitutions in the resA promoter region on transcription. Cells were grown anaerobically in 2x YT supplemented with 1% glucose and 0.2% KNO3, and ß-galactosidase activities were measured as described in Materials and Methods. Experiments were repeated two to six times, and the averages of maximal activities around T1 (1 h after the end of exponential growth) are shown with standard deviations.


arrow
MATERIALS AND METHODS
 
Strains and plasmids. All B. subtilis strains used in this study are derivatives of B. subtilis JH642 and listed in Table 1. Construction of rpoA with alanine substitution at each residue in {alpha}CTD was previously described (32). Plasmids are also listed in Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Bacillus subtilis strains and plasmids

Measurement of ß-galactosidase activity. B. subtilis cells were grown anaerobically in 2x yeast extract-tryptone (2x YT) medium (22) supplemented with 1% glucose and 0.2% KNO3 (the starting optical density at 600 nm was 0.02) with 5 µg/ml of chloramphenicol. Samples were taken at 1-h intervals, and ß-galactosidase activity was measured as described previously (19). The activity reaches its maximum level around T1 (1 h after the end of exponential growth), which was shown in Fig. 2.

Mutational analysis of the fnr and resA promoters. Base substitutions in the fnr promoter were generated by PCR amplification using an upstream mutagenic oligonucleotide (oligonucleotides listed in Table 2) and a downstream oligonucleotide, oMN97-3 with chromosomal DNA isolated from B. subtilis JH642 as the template. Base substitutions in the resA promoter were generated by two-step PCR amplification. Briefly, two PCR products were generated from pMMN649 using one of the complementary mutagenic oligonucleotides (Table 2) together with either oMN06-305 or oMN06-306. The two PCR products were annealed and extended, and the resulting product was used as the template in a second PCR with oMN06-305 and oMN06-306. The resultant PCR products carrying the fnr (positions –62 to +265 relative to the transcription start site) and resA (–134 to +65) promoters were digested with EcoRI and BamHI and inserted into pTKlac (13) which had been digested with the same enzymes, to generate transcriptional lacZ fusions. Each mutation was verified by sequencing. The fnr-lacZ and resA-lacZ constructs were integrated into the SPß locus of JH642 chromosome as described previously (34).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Oligonucleotides

Protein purification. Purification of ResD and ResE proteins were described elsewhere (10). RNA polymerase (RNAP) was purified from B. subtilis MH5646 that produces the RNAP ß' subunit fused to a 10x His tag by Ni2+-nitrilotriacetic acid chromatography as described previously (29), followed by HiQ column chromatography (26). {sigma}A, {alpha}, and {alpha}CTD (between residues 239 and 314) proteins were produced in Escherichia coli carrying pSN64, pSN28, and pSN106, respectively, and purified as described in previous papers (20, 26, 27). {alpha}CTD protein with the K267A mutation was produced using E. coli carrying pZY18 and purified by the procedure used for the wild-type {alpha}CTD purification (26). pZY18 was constructed in the same manner as pSN106 was (26), except chromosomal DNA isolated from B. subtilis ORB5262 (rpoA encoding {alpha}CTD with the K267A mutation [32]) was used as the template for PCR.

DNase I footprinting. A fragment carrying the wild-type fnr promoter (positions –136 to +20) was amplified by PCR using oligonucleotides oHG-2 and oHG-9 with pMMN408 as the template. Mutant promoters (–47T to C and –48T to G) were amplified using the same oligonucleotides with pHG45 and pHG71, respectively, as the template. Two PCR products were generated from pMMN408 using oligonucleotide pairs, oMMN99-24/oHG66 and oMMN99-25/oHG65 where oHG65 and oHG66 are complementary mutagenic oligonucleotides. The second PCR was carried out using the first PCR products with oMN99-24 and oMMN99-25. The second PCR product digested with EcoRI and BamHI was cloned into pUC18 digested with the same enzymes to generate pHG45. pHG71 was generated similarly, except oHG101 and oHG102 were used for mutagenic oligonucleotides. To label the coding or noncoding strand, one of the primers was phosphorylated with T4 polynucleotide kinase and [{gamma}-32P]ATP. The labeled PCR products were separated on a nondenaturing polyacrylamide gel and purified as previously described (24), and DNase I footprinting was carried out as previously described (11). The dideoxynucleotide sequence ladder was obtained by using the Thermo Sequenase cycle sequencing kit (United States Biochemical) using the labeled primer and pMMN408 as the template.


arrow
RESULTS
 
Mutational analysis of the fnr promoter. As described in the introduction, a high similarity was detected in a region between positions –61 and –45 of the resA, sboA, and fnr promoters (Fig. 2A). A direct repeat (TTCA N7 TTCA) is present between –47 and –61 of the resA and sboA promoters. However, the sequence corresponding to the promoter-proximal TTCA repeat present in resA and sboA is GTTA in the fnr promoter. We speculated that a region of the direct repeat accommodates two ResD~P monomers (or one dimer). Our previous deletion analysis of the fnr promoter indicated that the cis-acting sequence required for ResD-dependent fnr expression is located within a region downstream from –61 and a deletion to –58 severely reduced ResD-dependent transcription activation (24), which is in good agreement with the hypothesis that ResD~P binds to the sequence between –60 and –46. In addition, the previous observation that ResD~P weakly binds to the fnr promoter (10) might be explained by the less conserved promoter-proximal half-site (GTTA instead of TTCA).

In order to examine this possibility, we carried out mutational analysis of the fnr promoter. The result showed that any single base substitution in the TTCA sequence between –60 and –57 resulted in sharply reduced transcription, indicating that the TTCA sequence is important for ResD-dependent fnr transcription (Fig. 2B). The substitution of C at –56, at a position adjacent to the TTCA motif, moderately affected transcription. As described above, the resA and sboA promoters carry a downstream TTCA sequence. Therefore, we next determined whether generating a promoter-proximal TTCA sequence at the corresponding site of the fnr promoter increases transcription. fnr expression was slightly increased by the change of T at –47 to C that brings the sequence (GTCA) closer to the TTCA consensus sequence. The substitution of –48T to G (resulting in GGTA) nearly abolished fnr expression, which was also expected if the TTCA sequence is important for ResD binding. However, the change of G at –49 to the consensus T (resulting in TTTA) led to a decrease in transcription, and the substitution of –46A to G (resulting in GTTG) did not impair fnr expression, suggesting that generating the direct repeat does not increase fnr transcription. In fact, simultaneous substitutions of three nucleotides (–49G to T, –47T to C, and –45G to A), which create the consensus TTCA sequence, resulted in transcription slightly lower than the wild-type promoter, although the adverse effect of the –49G to T change appeared to be compensated by the changes in –47T and –45G (see Discussion). ß-Galactosidase activity of the mutant promoters in the resDE mutant is similar (around 30 Miller units), indicating that the mutations do not affect basal level of transcription, which confirmed that the affected bases are important for ResD-dependent activation.

Mutational analysis of the resA promoter. The results of base substitutions in the fnr promoter indicated that the single TTCA site, not TTCA repeats, is essential for ResD-dependent fnr transcription. As shown in Fig. 2A, the resA and sboA promoters contain the TTCA repeats. The next question was whether both TTCA sequences are important for ResD-dependent transcription of resA and sboA or whether the promoter-distal TTCA sequence is sufficient for transcription. The result obtained from mutational analysis of the resA promoter showed that the promoter-distal TTCA site is critical for transcription as shown by the substitution of T at –60 or C at –59, which greatly reduced promoter activity (Fig. 2C). In the downstream TTCA sequence, the second T (–49T), which corresponds to –48T in the fnr promoter, is the most critical residue, as is the case with fnr. The –48C to T change led to 60% reduction of transcription, which is consistent with the result that the –47T to C change in fnr moderately increased transcription. In contrast, the base substitution of T at –50 or A at –47 did not show any significant effect on resA transcription. These results support the conclusion obtained from the mutational analysis of the fnr promoter that TTCA (–60 to –57 of fnr) and –48T are essential for ResD-dependent transcription activation.

DNase I footprinting analysis of the fnr promoter. In order to determine whether ResD~P binds to the TTCA sequence in fnr, we carried out DNase I footprinting analysis of the wild-type and mutant fnr promoters. We used two mutant promoters with T–48G and T–47C, the nucleotide substitutions of which resulted in defective and slightly increased transcription, respectively (Fig. 2B). It was noticed that cleavage patterns of free DNA between positions –46 and –50 reproducibly changed when T at –48 was substituted with G (see Fig. 3 to 5). In the wild-type promoter, sensitivity to DNase I was similar among nucleotides at –47, –48, and –50 of the coding strand, whereas in the mutant promoter the position at –50 was hypersensitive to DNase I attack. On the noncoding strand of the wild-type promoter, cleavage sites were located at –46, –48, and –50. When T at –48 was substituted with G, residues –48 and –50 became resistant to DNase I, and a new hypersensitive site appeared at position –49. In the mutant T–47C promoter, a cleavage pattern around this region was similar to that of the wild-type DNA, except –46 of the coding strand became slightly more sensitive to cleavage.


Figure 3
View larger version (56K):
[in this window]
[in a new window]

  		FIG. 3. DNase I footprinting analysis of the wild-type (wt) and mutant (T–47C and T–48G) promoters in the presence of ResD~P and RNAP. The coding (A) and noncoding (B) strands of each promoter fragment were labeled as described in Materials and Methods. ResD was phosphorylated with 2 µM ResE. Two concentrations of ResD, 3 µM (+) or 6 µM (++), were used (–, none). B. subtilis RNAP (50 nM) was used with 50 nM purified {sigma}A. Regions of strong and weak protection from DNase I digestion are marked by solid and dotted lines, respectively. A box and an arrow show TTCA (positions –60 to –57) and T at –48, respectively, which are critical for fnr transcription. A hypersensitive site detected in the T–48G promoter is shown with an asterisk. Dideoxynucleotide sequencing reactions are also shown, and nucleotide positions are marked relative to the transcription start site.


Figure 5
View larger version (51K):
[in this window]
[in a new window]

  		FIG. 5. DNase I footprinting analysis of the wild-type (wt) and mutant (T–47C and T–48G) promoters in the presence of ResD~P and {alpha}. DNase I footprinting analysis was carried out in a manner similar to that described in the legend to Fig. 3, except that 3 µM {alpha} was used instead of RNAP. A box and an arrow show TTCA (–60 to –57) and T at –48, respectively. A hypersensitive site detected in the T–48G promoter is shown with an asterisk. Dideoxynucleotide sequencing reactions are also shown, and nucleotide positions are marked relative to the transcription start site.

Figure 3 indicated that ResD~P alone did not bind to either the wild-type or mutant promoters (lanes 9, 14, and 19). Increasing the concentration of ResD~P up to 12 µM did not protect the fnr promoter, although 4 µM ResD~P fully protected nasD (data not shown). RNAP alone induced reproducible changes in cleavage patterns of the promoter DNA (lanes 6, 11, and 16), suggesting a distortion of DNA. In the presence of both ResD~P and RNAP, the region between positions –63 and –40 was strongly protected in the wild-type and T–47C promoters, and a region downstream from –32 was also weakly protected (lanes 7, 8, 17, and 18). In contrast, the protection was not detected in the T–48G promoter (lanes 12 and 13), which is consistent with the in vivo result showing that the T–48G mutation impaired transcription.

The region protected in the presence of ResD~P and RNAP included the critical T at –48 (shown by an arrow in Fig. 3) and the TTCA sequence (–60 to –57; marked with a box), which we proposed as the ResD-binding site. This result indicated that ResD~P in the presence of RNAP binds to both the TTCA sequence and the sequence around –48 or that ResD~P binds to TTCA, while a subunit of RNAP makes contact with T at –48. If the latter is the case, the C-terminal domain of the {alpha} subunit is a likely candidate, given the position of –48 with respect to the promoter DNA. Consistent with this notion, a region around –48 was strongly protected by purified {alpha} (Fig. 4A and B). Some weaker protection by {alpha} was observed in upstream and downstream regions of –48, which may be caused by nonspecific binding. When {alpha}CTD (amino acid residues 239 to 314) was used instead of {alpha}, the extended protection disappeared and only the region between –38 and –50 was protected (Fig. 4C and D). This result suggested that {alpha}CTD interacts with the region around –48; however, the substitution of T at –48 to G did not decrease the binding of either {alpha} or {alpha}CTD itself, suggesting that T at –48 is not involved in a direct contact with {alpha}CTD. A possible effect of the T–48G mutation on fnr transcription will be further discussed in the Discussion.


Figure 4
View larger version (87K):
[in this window]
[in a new window]

  		FIG. 4. DNase I footprinting analysis of the wild-type (wt) and mutant (T–47C and T–48G) promoters in the presence of {alpha} or {alpha}CTD. The coding (A and C) and noncoding (B and D) strands of each promoter fragment were labeled as described in Materials and Methods. (A and B) Increased concentrations of {alpha} (0.75, 1.5, 3, and 6 µM) were used. –, none. (C and D) Increased concentrations of {alpha}CTD (12.5, 25, and 50 µM) were used. Regions of strong and weak protection from DNase I digestion are marked by solid and dotted lines, respectively. A box and an arrow show TTCA (positions –60 to –57) and T at –48, respectively. A hypersensitive site detected in the T–48G promoter is shown with an asterisk. Dideoxynucleotide sequencing reactions are also shown, and nucleotide positions are marked relative to the transcription start site.

When ResD~P was present, the protection by {alpha} was further extended to a region including the TTCA sequence in the wild-type promoter and the T–47C promoter (Fig. 5). Consistent with the footprinting results with ResD~P and RNAP (Fig. 3), no cofootprinting with ResD~P and {alpha} in the T–48G promoter was observed. These results indicated that interaction between {alpha} and ResD~P stabilizes ResD~P binding to the fnr promoter and strongly suggested that ResD~P makes contact with TTCA and the change of T at –48 to G adversely affects formation of the RNAP-ResD~P-DNA ternary complex.

Effects of amino acid substitutions in {alpha}CTD on fnr expression. The results of DNase I footprinting analysis described above strongly suggested that interaction of ResD~P with {alpha}, very likely {alpha}CTD, is important in transcriptional activation of fnr. To examine this possibility, we took advantage of a recently constructed {alpha}CTD mutant library (32). This mutant library was constructed by replacing the wild-type rpoA with a mutant allele bearing an alanine codon substitution. B. subtilis strains carrying substitutions E255A, R261A, R268A, R289A, and G292A could not be isolated probably because these residues are essential for transcription of housekeeping genes. Residues 269, 278, and 301 are alanine in the wild-type {alpha}, and we substituted only residue 269 with isoleucine. Within the residues 251 to 314, substitutions Y263A, K267A, A269I, and N290A strongly reduced fnr expression (Fig. 6). I253A, E254A, V260A, L266A, and G309A had a moderate effect on transcription (30 to 50% of the wild-type expression). A previous study showed that substitutions of these residues had no significant effect on expression of rpsD encoding ribosomal S6, except that Y263C and Y263A led to 50% and 40% reduction in rpsD expression, respectively (32; H. Geng and M. M. Nakano, unpublished results), indicating that the effect of these substitutions in transcriptional activation is specific to fnr.


Figure 6
View larger version (29K):
[in this window]
[in a new window]

  		FIG. 6. Effects of single alanine substitutions in {alpha}CTD on fnr-lacZ expression. The wild-type (WT) strain and each {alpha}CTD mutant were grown anaerobically in 2x YT supplemented with 1% glucose and 0.2% KNO3, and ß-galactosidase activities were measured as described in Materials and Methods. Experiments were repeated two or three times, and the average of maximal activities around T1 (1 h after the end of exponential growth) is expressed as percentage of the activity in cells carrying the wild-type rpoA gene. Black bars indicate substitutions that reduced the expression to less than 20% of the wild-type level, and gray bars indicate substitutions that reduced expression to 30 to 50%. Substitutions of residues with no data indicate those substitutions not obtained in B. subtilis (32).

K267 of {alpha}CTD is essential for interaction with ResD~P. One possible explanation of the reduced fnr expression in cells producing certain {alpha}CTD mutants is that the mutated residues are essential for interaction with ResD~P. To examine this possibility, DNase I footprinting analysis of the wild-type fnr promoter was carried out using the {alpha}CTD (K267A) protein. As shown in Fig. 7, the mutant {alpha}CTD binds to the region between –38 and –50 with an affinity similar to that of the wild-type {alpha}CTD (lanes 6 to 8 and 12 to 14). In the presence of ResD~P, the wild-type {alpha}CTD was able to bind to DNA at a concentration lower than the {alpha}CTD alone, and the protected region extended to the upstream region including the TTCA sequence (lanes 9 to 11), indicating that interaction between {alpha}CTD and ResD~P stabilizes binding of both proteins to DNA. In contrast, ResD~P did not significantly increase binding of the mutant {alpha}CTD to DNA, and the mutant {alpha}CTD did not stimulate binding of ResD~P to the TTCA sequence (lanes 15 to 17). We concluded from these results that K267 of {alpha}CTD is essential for interaction with ResD~P at the fnr promoter and that the interaction accelerates transcription initiation of fnr.


Figure 7
View larger version (96K):
[in this window]
[in a new window]

  		FIG. 7. DNase I footprinting analysis of the wild-type fnr promoter in the presence of ResD~P and the wild-type or K267A {alpha}CTD. The coding strand of the fnr promoter was labeled as described in Materials and Methods. ResD (6 µM) was phosphorylated with 2 µM ResE. An increased concentration (25, 50, and 100 µM) of the wild-type (wt) or K267A {alpha}CTD was included when indicated. The dotted line indicates the region protected by {alpha}CTD, and the solid line indicates the region protected in the presence of the wild-type {alpha}CTD and ResD~P. The box and arrow show TTCA (–60 to –57) and T at –48, respectively. Dideoxynucleotide sequencing reactions are also shown, and nucleotide positions are marked relative to the transcription start site. –, none.


arrow
DISCUSSION
 
On the basis of the results described here, we propose that ResD~P binds to the TTCA sequence in the fnr and resA promoters and likely the sequence in the sboA promoter, which supports the hypothesis that TGAA in the proposed consensus sequence a (Fig. 1A) is important for binding of ResD~P to activate transcription of ctaA, ctaB, and yclJ. Given that either TTCA or TGAA is likely used as a ResD-binding site in divergently transcribed ctaA and ctaB as well as yclI and yclJ, we assume that ResD~P could bind to DNA in both orientations. ResD~P dissociation constants for the ctaA genes calculated by Zhang and Hulett were as follows: 8.9 nM for ctaA1, 200 nM for ctaA2, and 47 nM for ctaA3 (31). This may suggest that ResD~P binds to a direct repeat (TTCA N7 TTCA or TGAA N7 TGAA present in ctaA1 and ctaA3) with an affinity higher than that to a single TGAA in ctaA2. TTCA N7 TTCA is similar to a half-site of the PhoP-binding sequence [TT(A/T/C)ACA N4-6 TT(A/T/C)ACA] that accommodates a PhoP dimer (12, 16). Previous work showed that resA expression is positively regulated by the ResDE and PhoPR signal transduction systems and that PhoP~P binds to the resA promoter region that includes the direct repeat sequence shown in Fig. 2 (2), indicating that the same sequence (or overlapping sequence) of resA is used for binding of the two response regulators.

We have shown by hydroxyl radical footprinting that ResD~P binds to five distinct regions of nasD (positions –90 to –45) and hmp (–80 to –40), and we have shown by deletion analysis that cis regions required for activation by ResD~P reside downstream of –87 (nasD) or –67 (hmp). Within these regions, no TGAA sequence is identified, but a single TTCA sequence (–87 to –84 in nasD and –48 to –45 in hmp) is detected, suggesting that ResD~P binds to and activates these promoters in a way different from cta and ycl that contain the direct repeat. In light of consensus sequence a and b, we propose TTGTGAAN3TTTN4A (Fig. 1C) as a consensus ResD box (consensus sequence c) for fnr, nasD, and hmp. Consensus sequence c is composed of two half-sites of 9 bases—site a (TTGTGAANN), which is a part of consensus sequence a, and site b (NTTTN4A), which is similar to the 3' end of consensus sequence b. Figure 1C shows a possible alignment of these sites in the nasD, hmp, and fnr promoters. Each regulatory sequence listed in Fig. 1C was shown to be sufficient for full activation by ResD~P (24). Two consensus sequences (8 of 11 matches) are present in nasD, which likely accommodate two ResD~P dimers. Upstream of the two putative ResD~P dimer-binding sites, half-site a is present in the opposite orientation. The hmp and fnr promoters each contain a single full-site and one half-site a that is oriented oppositely to the full site. The half-site resides upstream (in fnr) or downstream (in hmp) of the full site. Site a, but not site b, is able to serve as a stand-alone half-site, suggesting that ResD~P binds to site b only after site a is occupied by ResD~P. We reexamined the result of base substitutions in the fnr promoter in the context of the newly proposed consensus ResD-binding sequence. Among the bases in fnr-2a (TTGTTAG), the first G (–49), and the third T (–48) in particular, are important for ResD-dependent transcription because the change from TTGTTAG to either TTTTTAG or TTGGTAG led to moderate and severe reduction of fnr transcription (Fig. 2A), respectively. This result is in good agreement with the proposed consensus sequence half-site a (TTGTGAAN2). The change of A at –46 to G (resulting in TTGTTGG) and G at –45 to A (TTGTTAA) did not significantly decrease or increase transcription, indicating that the two A residues in the proposed half-site a are not critical. We showed that the adverse effect of the G–49T mutation was suppressed by the T–47C and G–45A mutations (Fig. 2B). We interpreted this result as meaning that the TTCA direct repeat does not increase fnr transcription; however, we now suggest another interpretation. The G–49T substitution (TTTT) impairs the function of the critical TTGT sequence in consensus sequence c, thereby adversely affecting fnr transcription. Introduction of the T–47C mutation in the G–49T promoter generates TTCA, and the resultant TTCA N7 TTCA sequence now functions as a ResD-binding site like those found in cta and ycl. Further studies of base substitutions, including the first two T residues in the TTGTGAAN2 sequence, are required to properly evaluate the putative consensus sequence c and to understand how differences in the half-site arrangement and orientation in various promoters affect binding and activation by ResD~P.

This study also uncovered the critical role of {alpha}CTD for ResD-dependent fnr transcription. Transcriptional activators bind to specific DNA sequences and often interact with {alpha}CTD. This protein-protein interaction increases the affinity of RNAP to the promoter site to initiate transcription (reviewed in reference 4). Such an activator-dependent transcription is usually observed in promoters lacking the consensus –35 sequence as is the case with ResD-controlled promoters. The results of our footprinting analysis showed that ResD~P alone does not bind to the fnr promoter, unlike the nasD or hmp promoter (10) and the cta (31) or ycl (11) promoter. Hence, a simple recruiting model in which ResD~P binds to the ResD box and recruits RNAP to the promoter seems inapplicable to explain how ResD~P activates fnr transcription. A ternary complex by ResD~P, RNAP, and fnr is formed presumably through direct interaction between ResD~P and {alpha}CTD. Screening of the {alpha}CTD alanine substitution library identified residues that are important for fnr activation by ResD~P. Among these residues, E254, V260, Y263, K267, and A269 form a surface-exposed patch (Fig. 8). Residues D258, K271, A272, E273, and L289 in E. coli {alpha}CTD, which correspond to E254, K267, R268, A269, and M285, respectively, in B. subtilis {alpha}CTD, are proposed to interact with the Fis transcriptional activator (1, 18). Therefore, ResD~P likely interacts with a surface of {alpha}CTD similar to the "273 determinant" that interacts with Fis. N290, which is important for ResD-dependent activation of fnr, is not a part of this surface patch proposed to interact with ResD~P, and might be involved in interaction with DNA. In fact, the corresponding residue (N294) in E. coli {alpha}CTD was shown to interact with DNA (9). Our DNase I footprinting analysis demonstrated that {alpha}CTD carrying the K267A mutation binds to fnr DNA nearly as well as the wild-type {alpha}CTD does; however, the mutant {alpha}CTD, unlike the wild-type protein, does not facilitate binding of ResD~P to the TTCA sequence, confirming our hypothesis that the surface including K267 interacts with ResD~P. The residue K267, as well as Y263, C265, and L266, was shown to be required for optimal ComA-dependent activation of the srf operon encoding proteins that function in the control of competence development and in nonribosomal peptide synthesis (32). The C265A mutation that most severely affected srf expression had no effect on fnr expression. Conversely, A269I conferred severely reduced fnr expression but only moderately affected srf transcription. Therefore, it is likely that ResD~P and ComA~P interact with overlapping but distinct surfaces of {alpha}CTD.


Figure 8
View larger version (85K):
[in this window]
[in a new window]

  		FIG. 8. Structure of B. subtilis {alpha}CTD (28) indicating the residues identified as important for ResD-dependent activation of fnr. Amino acid residues with more than twofold effect are dark gray.

The ResD-{alpha}CTD-DNA ternary complex was not formed with fnr carrying the T–48G mutation, suggesting that the in vivo defect in transcription caused by the mutation could be due to the lack of cooperative binding between ResD~P and {alpha}CTD. Purified {alpha}CTD protects a region around –48, raising the possibility that {alpha}CTD might make direct contact with T at –48. However, the change of –48T to G did not weaken the binding of {alpha}CTD (Fig. 4), instead we observed DNase I-hypersensitive sites unique to T–48G DNA (Fig. 3 to 5). One could imagine that the substitution of T at –48 with G introduces a kink around –50, which results in interference between ResD~P and RNAP, thus disrupting the ResD~P-RNAP-DNA ternary complex.

Although {alpha}CTD alone binds to a region between –50 and –38, it remains unclear whether the protection of the region in the presence of {alpha}CTD and ResD~P is caused by binding of {alpha}CTD, ResD~P, or both because the sequence between –51 and –34 contains two half-sites that show similarity to consensus sequence c (Fig. 1C). One possibility is that ResD~P and {alpha}CTD interact by binding to the same region but to different faces of the DNA helix as seen with BvgA and {alpha}CTD in E. coli (3). DNase I footprinting experiments in Fig. 3 showed that hypersensitive sites that were observed in the presence of RNAP disappeared when both RNAP and ResD~P were present, indicating that the interaction with ResD~P and {alpha}CTD remodels RNAP-promoter interaction to initiate transcription.


arrow
ACKNOWLEDGMENTS
 
We thank Peter Zuber for providing the {alpha}-CTD mutant library and for critical reading of the manuscript. We also thank Shunji Nakano for helpful discussions and for providing purified {sigma}A, {alpha}, and {alpha}CTD proteins and Ying Zhang for constructing pZY18.

This study was supported in part by grant MCB0110513 from the National Science Foundation.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Department of Environmental and Biomolecular Systems, OGI School of Science and Engineering, Oregon Health and Science University, 20000 NW Walker Road, Beaverton, OR 97006. Phone: (503) 748-4078. Fax: (503) 748-1464. E-mail: mnakano{at}ebs.ogi.edu. Back

{triangledown} Published ahead of print on 22 December 2006. Back

Present address: Department of Genetics, Washington University, St. Louis, MO. Back


arrow
REFERENCES
 
    1
  1. Aiyar, S. E., S. M. McLeod, W. Ross, C. A. Hirvonen, M. S. Thomas, R. C. Johnson, and R. L. Gourse. 2002. Architecture of Fis-activated transcription complexes at the Escherichia coli rrnB P1 and rrnE P1 promoters. J. Mol. Biol. 316:501-516.[CrossRef][Medline]
    2. 2
  2. Birkey, S. M., W. Liu, X. Zhang, M. F. Duggan, and F. M. Hulett. 1998. Pho signal transduction network reveals direct transcriptional regulation of one two-component system by another two-component regulator: Bacillus subtilis PhoP directly regulates production of ResD. Mol. Microbiol. 30:943-953.[CrossRef][Medline]
    3. 3
  3. Boucher, P. E., A. E. Maris, M. S. Yang, and S. Stibitz. 2003. The response regulator BvgA and RNA polymerase alpha subunit C-terminal domain bind simultaneously to different faces of the same segment of promoter DNA. Mol. Cell 11:163-173.[CrossRef][Medline]
    4. 4
  4. Busby, S., and R. H. Ebright. 1994. Promoter structure, promoter recognition, and transcription activation in prokaryotes. Cell 79:743-746.[Medline]
    5. 5
  5. Colbert, C. L., Q. Wu, P. J. Erbel, K. H. Gardner, and J. Deisenhofer. 2006. Mechanism of substrate specificity in Bacillus subtilis ResA, a thioredoxin-like protein involved in cytochrome c maturation. Proc. Natl. Acad. Sci. USA 103:4410-4415.[Abstract/Free Full Text]
    6. 6
  6. Crow, A., R. M. Acheson, N. E. Le Brun, and A. Oubrie. 2004. Structural basis of redox-coupled protein substrate selection by the cytochrome c biosynthesis protein ResA. J. Biol. Chem. 279:23654-23660.[Abstract/Free Full Text]
    7. 7
  7. Cruz Ramos, H., L. Boursier, I. Moszer, F. Kunst, A. Danchin, and P. Glaser. 1995. Anaerobic transcription activation in Bacillus subtilis: identification of distinct FNR-dependent and -independent regulatory mechanisms. EMBO J. 14:5984-5994.[Medline]
    8. 8
  8. Erlendsson, L. S., R. M. Acheson, L. Hederstedt, and N. E. Le Brun. 2003. Bacillus subtilis ResA is a thiol-disulfide oxidoreductase involved in cytochrome c synthesis. J. Biol. Chem. 278:17852-17858.[Abstract/Free Full Text]
    9. 9
  9. Gaal, T., W. Ross, E. E. Blatter, H. Tang, X. Jia, V. V. Krishnan, N. Assa-Munt, R. H. Ebright, and R. L. Gourse. 1996. DNA-binding determinants of the alpha subunit of RNA polymerase: novel DNA-binding domain architecture. Genes Dev. 10:16-26.[Abstract/Free Full Text]
    10. 10
  10. Geng, H., S. Nakano, and M. M. Nakano. 2004. Transcriptional activation by Bacillus subtilis ResD: tandem binding to target elements and phosphorylation-dependent and -independent transcriptional activation. J. Bacteriol. 186:2028-2037.[Abstract/Free Full Text]
    11. 11
  11. Härtig, E., H. Geng, A. Hartmann, A. Hubacek, R. Münch, R. W. Ye, D. Jahn, and M. M. Nakano. 2004. Bacillus subtilis ResD induces expression of potential regulatory genes yclJK upon oxygen limitation. J. Bacteriol. 186:6477-6484.[Abstract/Free Full Text]
    12. 12
  12. Hulett, F. M. 2001. The pho regulon, p. 193-201. In A. L. Sonenshein, J. A. Hoch, and R. Losick (ed.), Bacillus subtilis and its closest relatives: from genes to cells. ASM Press, Washington, DC.
    13. 13
  13. Kenney, T. J., and C. P. Moran, Jr. 1991. Genetic evidence for interaction of {sigma}A with two promoters in Bacillus subtilis. J. Bacteriol. 173:3282-3290.[Abstract/Free Full Text]
    14. 14
  14. LaCelle, M., M. Kumano, K. Kurita, K. Yamane, P. Zuber, and M. M. Nakano. 1996. Oxygen-controlled regulation of flavohemoglobin gene in Bacillus subtilis. J. Bacteriol. 178:3803-3808.[Abstract/Free Full Text]
    15. 15
  15. Le Brun, N. E., J. Bengtsson, and L. Hederstedt. 2000. Genes required for cytochrome c synthesis in Bacillus subtilis. Mol. Microbiol. 36:638-650.[CrossRef][Medline]
    16. 16
  16. Liu, W., and F. M. Hulett. 1998. Comparison of PhoP binding to the tuaA promoter with PhoP binding to other Pho-regulon promoters establishes a Bacillus subtilis Pho core binding site. Microbiology 144:1443-1450.[Abstract/Free Full Text]
    17. 17
  17. Liu, X., and H. W. Taber. 1998. Catabolite regulation of the Bacillus subtilis ctaBCDEF gene cluster. J. Bacteriol. 180:6154-6163.[Abstract/Free Full Text]
    18. 18
  18. McLeod, S. M., S. E. Aiyar, R. L. Gourse, and R. C. Johnson. 2002. The C-terminal domains of the RNA polymerase alpha subunits: contact site with Fis and localization during co-activation with CRP at the Escherichia coli proP P2 promoter. J. Mol. Biol. 316:517-529.[CrossRef][Medline]
    19. 19
  19. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
    20. 20
  20. Nakano, M. M., H. Geng, S. Nakano, and K. Kobayashi. 2006. The nitric oxide-responsive regulator NsrR controls ResDE-dependent gene expression. J. Bacteriol. 188:5878-5887.[Abstract/Free Full Text]
    21. 21
  21. Nakano, M. M., T. Hoffmann, Y. Zhu, and D. Jahn. 1998. Nitrogen and oxygen regulation of Bacillus subtilis nasDEF encoding NADH-dependent nitrite reductase by TnrA and ResDE. J. Bacteriol. 180:5344-5350.[Abstract/Free Full Text]
    22. 22
  22. Nakano, M. M., M. A. Marahiel, and P. Zuber. 1988. Identification of a genetic locus required for biosynthesis of the lipopeptide antibiotic surfactin in Bacillus subtilis. J. Bacteriol. 170:5662-5668.[Abstract/Free Full Text]
    23. 23
  23. Nakano, M. M., G. Zheng, and P. Zuber. 2000. Dual control of sbo-alb operon expression by the Spo0 and ResDE systems of signal transduction under anaerobic conditions in Bacillus subtilis. J. Bacteriol. 182:3274-3277.[Abstract/Free Full Text]
    24. 24
  24. Nakano, M. M., Y. Zhu, M. LaCelle, X. Zhang, and F. M. Hulett. 2000. Interaction of ResD with regulatory regions of anaerobically induced genes in Bacillus subtilis. Mol. Microbiol. 37:1198-1207.[CrossRef][Medline]
    25. 25
  25. Nakano, M. M., P. Zuber, P. Glaser, A. Danchin, and F. M. Hulett. 1996. Two-component regulatory proteins ResD-ResE are required for transcriptional activation of fnr upon oxygen limitation in Bacillus subtilis. J. Bacteriol. 178:3796-3802.[Abstract/Free Full Text]
    26. 26
  26. Nakano, S., K. N. Erwin, M. Ralle, and P. Zuber. 2005. Redox-sensitive transcriptional control by thiol/disulphide switch in the global regulator, Spx. Mol. Microbiol. 55:498-510.[CrossRef][Medline]
    27. 27
  27. Nakano, S., M. M. Nakano, Y. Zhang, M. Leelakriangsak, and P. Zuber. 2003. A regulatory protein that interferes with activator-stimulated transcription in bacteria. Proc. Natl. Acad. Sci. USA 100:4233-4238.[Abstract/Free Full Text]
    28. 28
  28. Newberry, K. J., S. Nakano, P. Zuber, and R. G. Brennan. 2005. Crystal structure of the Bacillus subtilis anti-alpha, global transcriptional regulator, Spx, in complex with the alpha C-terminal domain of RNA polymerase. Proc. Natl. Acad. Sci. USA 102:15839-15844.[Abstract/Free Full Text]
    29. 29
  29. Qi, Y., and F. M. Hulett. 1998. PhoP-P and RNA polymerase {sigma}A holoenzyme are sufficient for transcription of Pho regulon promoters in Bacillus subtilis: Pho-P activator sites within the coding region stimulate transcription in vitro. Mol. Microbiol. 28:1187-1197.[CrossRef][Medline]
    30. 30
  30. Sun, G., E. Sharkova, R. Chesnut, S. Birkey, M. F. Duggan, A. Sorokin, P. Pujic, S. D. Ehrlich, and F. M. Hulett. 1996. Regulators of aerobic and anaerobic respiration in Bacillus subtilis. J. Bacteriol. 178:1374-1385.[Abstract/Free Full Text]
    31. 31
  31. Zhang, X., and F. M. Hulett. 2000. ResD signal transduction regulator of aerobic respiration in Bacillus subtilis: cta promoter regulation. Mol. Microbiol. 37:1208-1219.[CrossRef][Medline]
    32. 32
  32. Zhang, Y., S. Nakano, S. Y. Choi, and P. Zuber. 2006. Mutational analysis of the Bacillus subtilis RNA polymerase alpha C-terminal domain supports the interference model of Spx-dependent repression. J. Bacteriol. 188:4300-4311.[Abstract/Free Full Text]
    33. 33
  33. Zheng, G., L. Z. Yan, J. C. Vederas, and P. Zuber. 1999. Genes of the sbo-alb locus of Bacillus subtilis are required for production of the antilisterial bacteriocin subtilosin. J. Bacteriol. 181:7346-7355.[Abstract/Free Full Text]
    34. 34
  34. Zuber, P., and R. Losick. 1987. Role of AbrB in Spo0A- and Spo0B-dependent utilization of a sporulation promoter in Bacillus subtilis. J. Bacteriol. 169:2223-2230.[Abstract/Free Full Text]

Journal of Bacteriology, March 2007, p. 1745-1755, Vol. 189, No. 5
0021-9193/07/$08.00+0     doi:10.1128/JB.01502-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Esbelin, J., Armengaud, J., Zigha, A., Duport, C. (2009). ResDE-Dependent Regulation of Enterotoxin Gene Expression in Bacillus cereus: Evidence for Multiple Modes of Binding for ResD and Interaction with Fnr. J. Bacteriol. 191: 4419-4426 [Abstract] [Full Text]  
  • Seino, Y., Takahashi, T., Hihara, Y. (2009). The Response Regulator RpaB Binds to the Upstream Element of Photosystem I Genes To Work for Positive Regulation under Low-Light Conditions in Synechocystis sp. Strain PCC 6803. J. Bacteriol. 191: 1581-1586 [Abstract] [Full Text]  
  • Puri-Taneja, A., Schau, M., Chen, Y., Hulett, F. M. (2007). Regulators of the Bacillus subtilis cydABCD Operon: Identification of a Negative Regulator, CcpA, and a Positive Regulator, ResD. J. Bacteriol. 189: 3348-3358 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Geng, H.
Right arrow Articles by Nakano, M. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Geng, H.
Right arrow Articles by Nakano, M. M.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»