JOURNAL OF BACTERIOLOGY, Aug. 2006, p. 5712–5721 0021-9193/06/$08.00⫹0 doi:10.1128/JB.00405-06 Copyright © 2006, American Society for Microbiology. All Rights Reserved. Vol. 188, No. 16 Activity of Rhodobacter sphaeroides RpoHII, a Second Member of the Heat Shock Sigma Factor Family Heather A. Green and Timothy J. Donohue* Department of Bacteriology, University of Wisconsin—Madison, Madison, Wisconsin 53706 Received 22 March 2006/Accepted 31 May 2006 The ability to regulate gene expression through the use of alternate sigma factors enables bacteria to respond quickly to changes in the environment. These secondary sigma factors guide RNA polymerase to promoter elements whose target sequences differ from those of the primary sigma factors, thereby directing the transcription of a specific set of genes (44). One such rapid response to environmental perturbations, known as the heat shock response, uses an alternate sigma factor to increase the expression of proteins needed to maintain an intracellular milieu conducive for protein folding. While these heat-inducible gene products have traditionally been referred to as the heat shock proteins, increased heat shock protein expression is also induced by other stress conditions (11, 33). We are interested in alternate sigma factors that play a role in the heat shock or other stress responses of the ␣-proteobacterium Rhodobacter sphaeroides. In the ␥-proteobacterium Escherichia coli, ␴32 (encoded by rpoH) is the alternate sigma factor responsible for recognizing heat shock gene promoters (12). As a member of the ␴70 family of eubacterial sigma factors, ␴32 recognizes unique promoter sequences centered at positions ⫺10 and ⫺35 relative to the transcriptional start site (44). All ␴32-like proteins are defined by a conserved region known as the “RpoH box”; they also contain conserved sequences in regions 2.4 and 4.2 that recognize the ⫺10 and ⫺35 elements, respectively, of heat shock promoters (29). While E. coli, with its single rpoH gene, sets the stage for much of what we know about bacterial heat shock regulation, several ␣-proteobacteria are known to encode multiple RpoH homologs. Three rpoH-like genes in Bradyrhizobium japonicum have been reported (31, 32), while Sinorhizobium meliloti contains two rpoH genes (36, 37). Individual RpoH family members from these bacteria can completely, or partially, complement the temperature sensitivity of an E. coli rpoH mutant (31, 36, 37), indicating that they are functionally similar to ␴32. In both Rhizobium species, however, each of the RpoH homologs appears to have different but overlapping roles in the organism’s response to stress (31, 32, 36, 37). Past work established that R. sphaeroides RpoHI (previously called ␴37) was a member of the ␴32 family of alternate sigma factors, since the rpoHI gene complemented the inability of an E. coli ␴32 mutant to support phage growth (19). In addition, a ⬃37-kDa protein isolated from RNA polymerase preparations transcribed several E. coli heat shock promoters in vitro when reconstituted with core RNA polymerase. The R. sphaeroides RpoHI-null mutant (⌬RpoHI), however, mounted a typical heat shock response, implying the existence of a second system by which this bacterium could activate heat shock promoters (19, 24). Further evidence supporting this idea came from in vitro transcription assays that demonstrated that a ⬃38-kDa protein (previously called ␴38) purified from ⌬RpoHI cells recognized both a known E. coli ␴32 promoter (19) and the R. sphaeroides cycA P1 promoter, which contains sequence elements related to the E. coli ␴32 promoter consensus (24). The identity of this ⬃38-kDa protein, however, and its possible similarity to other alternate sigma factors were unknown at the time. In this report, we illustrate that the rpoHII gene encodes a second alternate sigma factor of the ␴32 family in R. sphaeroides. We also show that recombinant RpoHI and RpoHII can each transcribe several heat shock promoters when reconstituted with core RNA polymerase from either E. coli or R. sphaeroides. While both RpoHI and RpoHII recognize R. sphaeroides promoters that resemble the E. coli ␴32 consensus, there are differences in the ability of each protein to recognize individual promoters in vitro. We discuss the possibility that each R. sphaeroides RpoH homolog may regulate different, yet * Corresponding author. Mailing address: Department of Bacteriology, University of Wisconsin—Madison, Madison, WI 53706. Phone: (608) 262-4663. Fax: (608) 262-9865. E-mail: tdonohue@bact.wisc.edu. 5712 Downloaded from http://jb.asm.org/ on August 15, 2015 by guest We have identified a second RpoH homolog, RpoHII, in the ␣-proteobacterium Rhodobacter sphaeroides. Primary amino acid sequence comparisons demonstrate that R. sphaeroides RpoHII belongs to a phylogenetically distinct group with RpoH orthologs from ␣-proteobacteria that contain two rpoH genes. Like its previously identified paralog, RpoHI, RpoHII is able to complement the temperature-sensitive phenotype of an Escherichia coli ␴32 (rpoH) mutant. In addition, we show that recombinant RpoHI and RpoHII each transcribe two E. coli ␴32-dependent promoters (rpoD PHS and dnaK P1) when reconstituted with E. coli core RNA polymerase. We observed differences, however, in the ability of each sigma factor to recognize six R. sphaeroides promoters (cycA P1, groESL1, rpoD PHS, dnaK P1, hslO, and ecfE), all of which resemble the E. coli ␴32 promoter consensus. While RpoHI reconstituted with R. sphaeroides core RNA polymerase transcribed all six promoters, RpoHII produced detectable transcripts from only four promoters (cycA P1, groESL1, hslO, and ecfE). These results, in combination with previous work demonstrating that an RpoHI mutant mounts a typical heat shock response, suggest that while RpoHI and RpoHII have redundant roles in response to heat, they may also have roles in response to other environmental stresses. VOL. 188, 2006 ACTIVITY OF R. SPHAEROIDES RpoHII overlapping, regulons in response to one or more environmental stress signals. MATERIALS AND METHODS in seven 1-ml fractions with Tris-EDTA buffer (10 mM Tris-HCl [pH 8.0] and 0.1 mM EDTA) supplemented with 0.75 M NaCl and 40% propylene glycol. Fractions containing core RNA polymerase subunits, as visualized by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), were combined and dialyzed against 1 liter of TGED medium (10 mM Tris-HCl [pH 7.9], 10% glycerol, 0.1 mM EDTA, and fresh 0.1 mM dithiothreitol) containing 47.5% glycerol and 100 mM NaCl. E. coli core RNA polymerase was purchased from Epicenter Technologies, Inc. (Madison, WI). All protein concentrations were determined using the Bio-Rad protein assay (Bio-Rad Laboratories, Hercules, CA). The proteins were stored at ⫺20°C prior to use. In vitro transcription assays. Core RNA polymerase, whether from E. coli or R. sphaeroides, was reconstituted with either His6-RpoHI or His6-RpoHII by incubating the proteins for 1 h at 32°C in transcription buffer (40 mM Tris-HCl [pH 7.9], 2 mM EDTA, 5 mM MgCl, 100 mM KCl, 1 mM dithiothreitol, 100 ␮g/ml acetylated bovine serum albumin). The mixture was added to a 20-␮l reaction mixture containing 20 nM supercoiled plasmid template in transcription buffer and incubated for 25 min at 32°C. Transcription assays were initiated with the addition of a mixture of ribonucleotide triphosphates at final concentrations of 250 ␮M GTP, CTP, and ATP; 25 ␮M UTP; and 1 ␮Ci [␣-32P]UTP (3,000 Ci/mmol). Reaction mixtures were incubated at 32°C for 25 min and terminated with 6 ␮l of 95% (wt/vol) formamide loading buffer (6). Samples were heated to 95°C and resolved on 6% polyacrylamide–7 M urea gels (19). The transcripts were visualized and quantified using the Molecular Dynamics (Sunnyvale, CA) PhosphorImaging system and ImageQuant software. Identification of candidate RpoH-dependent promoters from R. sphaeroides. The R. sphaeroides genome (GenBank accession numbers CP000143 to CP000147) was scanned using the program PromScan (http://www.promscan .uklinux.net) to search for candidate ␴32-like promoter sequences. This query used DNA sequences upstream of four R. sphaeroides promoters whose in vivo transcript levels increased upon an increase in temperature (cycA P1, groESL1, rpoD, PHS, and rrnB) (19). We chose to analyze candidate promoters positioned upstream of genes encoding known proteins in other bacteria. RESULTS Primary sequence similarity of R. sphaeroides RpoH homologs. The R. sphaeroides genome sequence predicts that this ␣-proteobacterium encodes two members of the heat shock sigma factor family, RpoHI (19) and RpoHII (see below). R. sphaeroides RpoHII shares 46% amino acid identity (64% similarity) with its paralog, R. sphaeroides RpoHI (19), and 36% amino acid identity with E. coli ␴32 (23) (Fig. 1A). RpoHII shares the greatest degree of amino acid identity with RpoH proteins from ␣-proteobacteria known to contain two RpoH factors. It displays 50% amino acid identity to Mesorhizobium loti RpoH-like sigma factor C, 47% identity to Bartonella quintana RpoH2, 46% identity to a Brucella melitensis ␴32 factor, and 42% identity to Sinorhizobium meliloti RpoH2 (36, 37). Together, R. sphaeroides RpoHII and RpoHI are most similar, displaying ⬃81% and ⬃84% amino acid identities, respectively, to the cognate RpoH proteins of two marine heterotrophs of the Roseobacter clade, Jannaschia helgolandensis (42) and Silicibacter pomeroyi (27). These values are consistent with a phylogenetic tree of ␴32 homologs (Fig. 2), which predicts that six of these proteins form a distinct cluster with R. sphaeroides RpoHII. In contrast, R. sphaeroides RpoHI falls into a larger cluster with proteins from ␣-proteobacterial species that contain either one or more rpoH genes. Figure 1B and C shows amino acid alignments of RpoH proteins from ␣-proteobacteria known to contain multiple rpoH genes and E. coli ␴32. All proteins in the heat shock family of alternate sigma factors contain the RpoH box, a conserved stretch of nine amino acids in region C (29) that has been implicated in the regulation of E. coli ␴32 activity (3, 18, 28). The second amino acid residue in the RpoH box is characteristically a lysine (K) in proteins from the ␣-proteobacteria Downloaded from http://jb.asm.org/ on August 15, 2015 by guest Bacterial strains and growth conditions. E. coli DH5␣ was used as a plasmid host. Cells were grown in Luria-Bertani (LB) medium (38) supplemented with ampicillin (100 ␮g/ml) as necessary. For the complementation assays, we grew E. coli CAG9333, an rpoH mutant capable of growth at 40°C (22), in LB medium supplemented with ampicillin (25 ␮g/ml) when required. All E. coli cells were grown at 37°C unless otherwise indicated. Comparison of amino acid sequences of RpoH-like proteins. The amino acid alignments and the phylogenetic relationships of RpoHI and RpoHII were determined by comparing their amino acid sequences with those of other RpoH homologs (DDBJ/EMBL/GenBank accession numbers are in parentheses): Agrobacterium tumefaciens (accession number D50828), Alcaligenes xylosoxydans (accession number AB009990), Bartonella quintana (accession numbers BQ12120 [rpoH2] and BQ02820 [rpoH1]), Bradyrhizobium japonicum (accession numbers U55047 [rpoH1], Y09502 [rpoH2], and Y09666 [rpoH3]), Brucella melitensis (accession numbers AB3287 [gene BMEI0280] and AD3299 [gene BMEI0378]), Buchnera aphidicola (accession number BAU35400), Caulobacter crescentus (accession number U39791), Citrobacter freundii (accession number X14960), Coxiella burnetii (accession number AF120928), Enterobacter cloacae (accession number D50829), Escherichia coli K-12 (accession number U00096), Haemophilus influenzae (accession number U32713), Hydrogenophilus thermoluteolus (accession number AB009991), Jannaschia sp. strain CCS1 (genome draft in progress for gene 3800 and gene 400 [JGI]), Mesorhizobium loti (accession numbers MLR3862 [RpoH-like sigma factor C] and MLR3741 [sigma factor]), Methylovorus sp. strain SS1 (accession number AF177466), Myxococcus xanthus (accession numbers X55500 [sigB], L12992 [sigC], and AF023661 [sigE]), Proteus mirabilis (accession number D50830), Pseudomonas aeruginosa (accession number S77322), Pseudomonas putida (accession number AB025418), Ralstonia sp. strain CH34 (accession number J05278), Rhodobacter capsulatus (accession number AF017436), Rhodobacter sphaeroides (accession numbers U82397 [rpoHI] and CP000143 [rpoHII]), Rickettsia prowazekii strain Madrid E (accession number AJ235271), Serratia marcescens (accession number D50831), Silicibacter pomeroyi (accession numbers SPO0406 [rpoH-1] and SPO1409 [rpoH-2]), Sinorhizobium meliloti (accession numbers AF128845 [rpoH1] and AF149031 [rpoH2]), Stigmatella aurantiaca (accession numbers U27311 [sigC] and Z14970 [sigB]), Vibrio cholerae (accession number U44432), Xanthomonas campestris pv. campestris (accession number AF042156), and Zymomonas mobilis (accession number D50832). Amino acid alignments were generated using ClustalW and BoxShade 3.21. The phylogenetic tree was created using PAUP* version 4.0beta (Sinauer Associates, Sunderland, Mass.). Plasmid constructions. The rpoHI and rpoHII genes were amplified from an R. sphaeroides 2.4.1 cosmid (pU18106) and genomic DNA, respectively, with 2.5 units of Pfu polymerase (Stratagene, La Jolla, CA) using primers specific for each gene. The PCR products were cloned into pUC18 and pUC19 in each orientation relative to the lac promoter. N-terminal hexahistidine-tagged RpoHI and RpoHII were obtained by cloning either rpoHI or rpoHII into pET-15b at the NdeI and NdeI-BamHI sites, respectively (Novagen, Madison, WI). The cloned portions of the resulting plasmids, pHAG7 and pHAG16, were sequenced to ensure that they encoded the desired His6-tagged versions of RpoHI and RpoHII, respectively. Plasmids used as in vitro transcription templates were derived from either pRKK96 (34) or pRKK137 (24), which both carry the spot 42 transcription terminator (1). Transcription templates containing R. sphaeroides cycA P1 as well as the E. coli dnaK P1 and rpoD PHS promoters have been described previously (19, 24). Candidate R. sphaeroides promoters (groESL1, rpoD PHS, dnaK, hslO, and ecfE) were PCR amplified from 2.4.1 genomic DNA as described above and sequenced with vector-specific primers to guarantee that the desired region was cloned in the proper orientation. All potential promoter regions lie within 150 bp of the predicted start of translation, except rpoD PHS, which is positioned ⬃400 bp upstream of the open reading frame start. Sequences of primers used in this study are available upon request. Proteins used for in vitro transcription assays. Conditions for expression and purification of His6-RpoHI, His6-RpoHII, and R. sphaeroides core RNA polymerase were described previously (1), with the following exceptions. Core RNA polymerase was obtained from an ⌬RpoHI null strain (19) by affinity chromatography (⬃40 g of cells) on a ⬃2-ml resin bed containing the 4RA2 monoclonal antibody against the ␣ subunit of E. coli RNA polymerase (Richard Burgess, University of Wisconsin—Madison). Proteins bound to the column were eluted 5713 5714 GREEN AND DONOHUE J. BACTERIOL. and an arginine (R) in those from the ␥-proteobacteria (19). Not surprisingly, both RpoH paralogs from R. sphaeroides contain a lysine at this position. While the third amino acid residue in the RpoH box from ␥-proteobacteria has invariably been a lysine, this residue is either an arginine or a lysine in proteins related to RpoHI from the ␣ subdivision (45). In contrast, proteins that cluster phylogenetically with R. sphaeroides RpoHII contain an uncharged amino acid, either alanine (A), serine (S), or valine (V), in the analogous position. This alignment also illustrates that R. sphaeroides RpoHII and its most closely related homologs have considerably less amino acid conservation than the RpoHI orthologs (Fig. 1B) in regions 2.1 and 2.2, which comprise a domain implicated in the binding of core RNA polymerase (7, 13). Function of R. sphaeroides rpoH genes in an E. coli rpoH mutant. Since R. sphaeroides RpoHII has significant amino acid conservation in regions 2.4 (⫺10 promoter recognition) and 4.2 (⫺35 promoter recognition) (Fig. 1B and C), we predicted that it should recognize heat shock promoters. To test this hypothesis, we asked whether the R. sphaeroides rpoHII gene could complement the temperature-sensitive phenotype of the E. coli ␴32-null strain CAG9333. While this tester strain lacks a functional copy of rpoH, a compensatory R40 mutation that results in enhanced GroES and GroEL synthesis allows for Downloaded from http://jb.asm.org/ on August 15, 2015 by guest FIG. 1. Alignments of proteobacterial RpoH proteins. (A) Alignment of R. sphaeroides RpoHI and RpoHII and E. coli ␴32. Identical/similar amino acid residues are shaded. Numbers on the left indicate the amino acid position relative to the start of each protein. Regions that are conserved among eubacterial sigma factors are denoted by the numbers below the sequences. Region C and the RpoH box are also shown. Gaps introduced to maximize the alignment are indicated by dashes. (B) Partial alignment of RpoH homologs from eight ␣-proteobacteria containing multiple rpoH genes and E. coli ␴32 (␥ subgroup) showing the amino acid sequence from region 2.1 to the end of region C. (C) Region 4.2 of the proteins listed in B showing the predicted helix-turn-helix (H-T-H) DNA-binding motif. For simplicity, the “. . .” indicates an intervening sequence between region C and region 4.2. Abbreviations: Bqui, Bartonella quintana; Bjap, Bradyrhizobium japonicum; BmelAB3287, Brucella melitensis RpoH gene BMEI0280; BmelAD3299, B. melitensis RpoH gene BMEI0378; Ecoli, Escherichia coli; Jann3800, Jannaschia sp. strain CCS1-3800/ CDS; Jann0400, Jannaschia sp. strain CCS1-0400/CDS; MlotRpoHC, Mesorhizobium loti RpoH-like sigma factor C; MlotMLR3741, M. loti sigma factor; RsphRpoH1, R. sphaeroides RpoHI; RsphRpoH2, R. sphaeroides RpoHII; Spom, Silicibacter pomeroyi; Smel, Sinorhizobium meliloti. See Materials and Methods for accession numbers. VOL. 188, 2006 ACTIVITY OF R. SPHAEROIDES RpoHII 5715 growth at up to 40°C (22). When we introduced a plasmid containing rpoHII downstream of the lac promoter into CAG9333 and tested for growth at 37°C (permissive temperature) and 44°C (restrictive temperature), we found that rpoHII was sufficient to restore growth at 44°C. In contrast, CAG9333, containing a control plasmid (pUC18), did not grow at 44°C (Table 1). We had previously shown that R. sphaeroides rpoHI TABLE 1. Complementation of an E. coli ⌬RpoH mutant (CAG 9333)a Complementing gene 37°C 44°C rpoD PHS::lacZ expressionb rpoHI rpoHII Nonec ⫹ ⫹ ⫹ ⫹ ⫹ ⫺ ⫹ ⫹ ⫺ a All phenotypes were scored on solid medium. Scored by monitoring a LacZ phenotype (see the text); “⫹,” blue; “⫺,” white. c Cells containing pUC18 control plasmid. b supported phage growth of an E. coli ␴32 mutant (19), so we expected a copy of this gene to allow for growth at 44°C. CAG9333 also carries a chromosomal rpoD PHS::lacZ fusion that can be used to score for transcription from this known heat shock promoter (22). We demonstrated in previous work that RNA polymerase holoenzymes purified from either R. sphaeroides wild-type or ⌬RpoHI cells were able to transcribe the E. coli rpoD PHS promoter in vitro (19). As expected, strains carrying either rpoHI or rpoHII formed blue colonies on plates containing X-Gal (5-bromo-4-chloro-3-indolyl-␤-D-galactopyranoside) and IPTG (isopropyl-␤-D-thiogalactopyranoside) (Table 1). These results indicate that the rpoHII gene product as well as the rpoHI gene product can functionally replace ␴32, presumably because they are capable of binding E. coli core RNA polymerase and transcribing one or more heat shock promoters. Recombinant RpoHI and RpoHII can direct transcription from E. coli heat shock promoters in vitro. To test the hypothesis that recombinant RpoHI and RpoHII can direct transcription from E. coli heat shock promoters in vitro, we reconsti- Downloaded from http://jb.asm.org/ on August 15, 2015 by guest FIG. 2. Phylogenetic tree of ␴32 homologs from 31 different proteobacterial species analyzed by the neighbor-joining method using PAUP* 4.0beta (Sinauer Associates, Sunderland, Mass.). Sigma factor names other than RpoH are shown in parentheses. The subdivisions (␣, ␤, ␥, and ␦) of the proteobacteria are shown to the right of the tree. As indicated by the numbers to the right of each branch point, 100 pseudoreplicates were used in this bootstrap analysis. Only values greater than 70% are shown. See Materials and Methods for accession numbers. 5716 GREEN AND DONOHUE J. BACTERIOL. TABLE 2. Comparison of promoters from R. sphaeroides and E. coli recognized by RpoHI and RpoHII Promoter ⫺35 regiona Spacer length (bp) ⫺10 regiona R. sphaeroides cycA P1 groESL1 rpoD PHS dnaK hslO ecfE GTATTGATT CCGTTGACA CCCTTGAAA CGCTTGCAG CCCTTGGTC CCCTTGAGG 13 14 14 14 13 14 ACCTATAT TCATATCT ACCCATGT CCCTATAT GCCTATAT GACCATCT E. coli rpoD PHS dnaK P1 CCCTTGAAA CCCTTGATG 14 13 GACGATAT CCCCATTT ␴32 consensus CNCTTGAAA 13/14 CCCCATNT 32 Those bases matching the E. coli ␴ consensus sequence (11) are shown in boldface type. The N indicates no base preference. tuted E. coli core RNA polymerase with histidine-tagged versions of either RpoHI or RpoHII and assayed transcription from two known E. coli heat shock promoters, rpoD PHS and dnaK P1 (Table 2). In order to determine the amount of sigma factor required to produce optimal transcription from each promoter, we performed multiple-round transcription using a fixed amount of E. coli core RNA polymerase (37.5 nM) and increasing amounts of either RpoHI (5.75 nM to 115 nM) or RpoHII (2 nM to 40 nM). As expected, transcript abundance from rpoD PHS and dnaK P1 increased with increasing concentrations of either RpoHI or RpoHII (titration curves represent- FIG. 3. In vitro transcription of E. coli rpoD PHS and dnaK P1 promoters by E. coli core RNA polymerase reconstituted with either purified R. sphaeroides RpoHI (A) or RpoHII (B). Shown are products from multiple-round transcription assays performed in the presence of increasing amounts of either RpoHI (A) (5.75, 11.5, 34.5, 57.5, and 115 nM) or RpoHII (B) (2, 4, 12, 20, and 40 nM) with a constant amount of core RNA polymerase (37.5 nM). Lanes marked with an asterisk contain products from assays using a control template in the presence of either a 3:1 (A) or a 1:1 (B) molar ratio of the indicated RpoH sigma factor to core RNA polymerase. Panels C and D represent the quantification and summary of transcription results from A and B as well as additional assays. The relative transcript abundance (pixel intensity) from each promoter was measured by pixel intensity on a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and adjusted for background. Downloaded from http://jb.asm.org/ on August 15, 2015 by guest a ing an average of several assays are shown in Fig. 3C and D). In the presence of RpoHI, we reproducibly obtained optimum transcript levels from both promoters when the sigma factor was in an approximately twofold molar excess over core RNA polymerase (Fig. 3C). In comparison, adequate transcription from reaction mixtures containing RpoHII required an approximately equimolar ratio of sigma to core RNA polymerase (Fig. 3D) over the course of several different assays. The transcripts derived from each of these promoters (Fig. 3A and B) are identical in length to those synthesized by E. coli E␴32 (20), indicating that pure RpoHI and RpoHII each recognize the established heat shock promoters. The results of these in vitro transcription assays also revealed that different levels of transcript were produced from rpoD PHS compared to dnaK P1 when the E. coli core was reconstituted with either RpoHI or RpoHII. At the concentration of purified RpoHI required for optimal transcription, the RNA product was twofold more abundant from the dnaK P1 promoter than from the rpoD PHS promoter (Fig. 3C). Similarly, greater than twofold more transcript resulted from the dnaK P1 promoter than from the rpoD PHS promoter in the presence of a 1:1 molar ratio of RpoHII to core (Fig. 3D). Recombinant RpoHI and RpoHII can direct transcription from R. sphaeroides heat shock promoters in vitro. To provide further evidence that RpoHI and RpoHII are members of the ␴32 family, we tested their ability to transcribe the R. sphaeroides cycA P1 and groESL1 promoters (Table 2). Primer extension assays indicate that the abundance of cycA P1 and groESL1-specific transcripts increases when either R. sphaeroides wild-type or ⌬RpoHI cells mount a heat shock response VOL. 188, 2006 ACTIVITY OF R. SPHAEROIDES RpoHII 5717 (19); hence, we predicted that these promoters would be recognized by RNA polymerase containing either RpoHI or RpoHII in vitro. We also tested the more physiologically relevant holo-RNA polymerase in these reactions by reconstituting the individual RpoH proteins with core enzyme isolated from R. sphaeroides. To eliminate RpoHI activity from our core RNA polymerase preparations, we purified this enzyme from an RpoHI mutant (19). The in vivo levels of RpoHI and RpoHII differ vastly. Protein gels of purified holo-RNA polymerase (24), as well as Western blot analyses of RpoHI and RpoHII levels in crude cell extracts (data not shown), indicate that RpoHI is considerably more abundant than RpoHII under aerobic growth conditions. This R. sphaeroides core RNA polymerase does not contain any detectable RpoHII activity with these promoters by multiple-round transcription (data not shown). When we performed multiple-round transcription using a fixed amount of R. sphaeroides core RNA polymerase (160 nM) and increasing amounts of RpoHI (20 nM to 800 nM), we found that increasing the concentration of RpoHI stimulated transcription from both cycA P1 and groESL1 (results from a representative assay are shown in Fig. 4A). While maximal levels of the groESL1 transcript appeared to require a sigmato-core ratio of only 0.5 in the presence of RpoHI, we chose to focus on a ratio of sigma to core that was optimal for both promoters tested. Thus, with this reconstituted RNA polymerase holoenzyme, transcript abundance from cycA P1 was ap- proximately twofold greater than that from groESL1 when the molar ratio of sigma to core was 1:1 (Fig. 4C). Similarly, when we increased the molar concentration of pure RpoHII (5 nM to 160 nM) in the presence of a constant concentration of core RNA polymerase (160 nM), the transcript abundance from both promoters increased, reaching a maximum at 80 to 160 nM RpoHII (Fig. 4B). At all levels of RpoHII tested, however, we detected only low levels of transcript produced from groESL1. Consequently, RpoHII-dependent transcription from cycA P1 was approximately 16-fold greater than that from groESL1 when the sigma-to-core RNA polymerase molar ratio was approximately 1:1 (Fig. 4D). If transcript abundance is taken as an estimate of promoter function, cycA P1 is transcribed with greater efficiency than groESL1 in vitro when core RNA polymerase is reconstituted with either RpoHI or RpoHII. On the other hand, if one analyzes the ⫺10 and ⫺35 regions of cycA P1 and groESL1, it is the R. sphaeroides groESL1 promoter that is most similar in sequence to the E. coli ␴32 consensus (Table 2), intimating that sequence comparisons alone may not be good indicators of promoter recognition by either RpoHI or RpoHII (see Discussion). The molar concentrations of purified RpoH proteins essential to observe optimal transcription were similar to that seen when we used E. coli core RNA polymerase in reactions instead. However, as indicated by control experiments, each of the E. coli heat shock promoters that we tested are stronger than the R. sphaeroides promoters analyzed (data not shown). Downloaded from http://jb.asm.org/ on August 15, 2015 by guest FIG. 4. In vitro transcription of R. sphaeroides cycA P1 and groESL1 promoters after adding either RpoHI (A) or RpoHII (B) to R. sphaeroides core RNA polymerase. Shown are products from multiple-round transcription assays performed in the presence of increasing amounts of either RpoHI (A) (20, 40, 80, 160, 480, and 800 nM) or RpoHII (B) (5, 10, 20, 40, 80, and 160 nM) with a constant amount of core RNA polymerase (160 nM). Panels C and D represent the quantification and summary of transcription results from A and B and depict typical titration curves for these two assays. The products marked with an asterisk in B are of unknown origin, but control experiments indicate that they are derived from an RpoHII-dependent promoter on the vector (pRKK96) used to generate these transcription templates (data not shown). The relative transcript abundance (pixel intensity) from each promoter was measured by pixel intensity on a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and adjusted for background. 5718 GREEN AND DONOHUE FIG. 5. In vitro transcription of R. sphaeroides cycA P1, groESL1, rpoD PHS, dnaK P1, hslO, and ecfE promoters by either RpoHI (230 nM) or RpoHII (160 nM) reconstituted with either R. sphaeroides (160 nM) (A) or E. coli (37.5 nM) core RNA polymerase (B). Open arrows denote transcription products of the expected length from each promoter. The product marked with an asterisk in the 4th, 6th, and 12th lanes is of unknown origin, but control experiments indicate that it is derived from an RpoHII-dependent promoter on the vector (pRKK96) used to generate these transcription templates (data not shown). transcript in assays where RpoHII is added to either R. sphaeroides (Fig. 5A, lane 8) or E. coli (Fig. 5B, lane 8) core RNA polymerase suggests that this gene either lacks an RpoHIIdependent promoter or contains one whose activity is below the detection level of this assay (see Discussion). Control experiments showed that transcript production from these six promoters is dependent on either RpoHI or RpoHII, because core RNA polymerase alone, from either E. coli or R. sphaeroides, does not produce a detectable product from any of the six heat shock promoters (data not shown). These in vitro transcription assays also revealed that an additional transcript is generated by one of our transcription templates (pRKK96) that is dependent on the presence of RpoHII (see transcription products marked with asterisks in Fig. 4B and 5A and B). The only difference between pRKK96 and pRKK137, the other template used in these assays, is the presence of a ⬃2-kbp spectinomycin resistance cartridge cloned upstream of the promoter. Thus, it would appear that this RpoHII-dependent transcript is derived from either an uncharacterized promoter within the spectinomycin resistance cartridge or one created by the insertion of this element into Downloaded from http://jb.asm.org/ on August 15, 2015 by guest Recognition of additional R. sphaeroides promoters by RpoHI and RpoHII. To further investigate promoters recognized by RpoHI and RpoHII, we examined the ability of reconstituted RNA polymerase holoenzymes to transcribe other R. sphaeroides candidate target genes. Having established the sigma-to-core ratios that are required for efficient transcription from the cycA P1 and groESL1 promoters, we chose four additional promoters to assay under these in vitro transcription conditions: rpoD PHS, dnaK P1, hslO (10, 16), and ecfE (9). Each of these promoters contains sequences related to the E. coli ␴32 consensus promoter sequence (Table 2). We knew from previous results that rpoD PHS-specific transcripts increase in both wild-type and ⌬RpoH1 cells upon a temperature increase (19). The remaining three promoters (dnaK P1, hslO, and ecfE) were selected after querying the R. sphaeroides genome with a matrix-driven program, PromScan (40), that utilized the ⫺10 and ⫺35 regions of known R. sphaeroides heat shock promoters. RpoHI reconstituted with R. sphaeroides core RNA polymerase stimulated transcription from the two control promoters, cycA P1 and groESL1, as well as from each of the four other test promoters, rpoD PHS, dnaK P1, hslO, and ecfE (Fig. 5A). When we reconstituted RpoHII with R. sphaeroides core enzyme, we detected transcription products from each of the promoters except rpoD PHS and dnaK P1 (Fig. 5A, lanes 6 and 8). The transcript lengths from each set of promoters transcribed by the two different holoenzymes were identical except for hslO (Fig. 5A, lanes 9 and 10), which reproducibly yielded a transcript that was 1 to 2 nucleotides shorter in the presence of RpoHII. One possible explanation for this feature of the hslO promoter is that RpoHII-containing RNA polymerase either initiates or terminates transcription at a different position than core enzyme reconstituted with RpoHI. The lack of a detectable RpoHII-dependent transcript from the dnaK promoter region (Fig. 5A, lane 8) suggests that the dnaK operon may not be controlled by this sigma factor (see Discussion). The absence of a detectable transcript from rpoD PHS (Fig. 5A, lane 6) using holoenzyme containing RpoHII, however, was surprising, since cells lacking RpoHI still experience an increase in rpoD-specific mRNA from this promoter after a heat shock (19). One feasible explanation for the failure to detect transcription from either rpoD PHS or dnaK P1 is that a protein(s) present in our R. sphaeroides core preparation reduces transcription in the presence of RpoHII (see Discussion). To test for the presence of such an activity in our R. sphaeroides core RNA polymerase, we repeated the above-described transcription reactions using a highly purified preparation of E. coli core RNA polymerase. As predicted by the above-described results (Fig. 5A), E. coli core RNA polymerase reconstituted with RpoHI transcribed all six promoters (Fig. 5B). However RNA polymerase holoenzyme formed by adding E. coli core enzyme to RpoHII generated detectable transcripts from only five R. sphaeroides promoters: cycA P1, groESL1, rpoD PHS, hslO, and ecfE (Fig. 5B). The presence of an rpoD PHS transcript from assays using RpoHII and E. coli core RNA polymerase is consistent with the hypothesis that our R. sphaeroides core RNA polymerase may contain a factor(s) capable of inhibiting transcription from this promoter (see Discussion). The inability to detect a dnaK P1-specific J. BACTERIOL. VOL. 188, 2006 ACTIVITY OF R. SPHAEROIDES RpoHII pRKK137. Simple inspection of the relevant sequences did not identify a candidate promoter, so additional experiments are under way to define this proposed RpoHII-dependent promoter. DISCUSSION constitute the three most highly conserved RpoH box residues among RpoH homologs (Fig. 1B). Alternatively, the third residue in the RpoH box of proteins that form a distinct cluster with RpoHII is either an alanine, serine, or valine, deviating from the positively charged arginine or lysine otherwise found in this position. To our knowledge, no one has studied the effects that altering this position has on ␴32’s, or another RpoH factor’s, ability to bind core RNA polymerase. If the RpoH box indeed has a regulatory role, as many lines of evidence suggest, the presence of an uncharged residue at this position may influence the activity of proteins sharing the clade with RpoHII. Regions 2.1 and 2.2 are conserved among all members of the ␴70 superfamily. Not surprisingly, amino acid substitutions in these regions alter core RNA polymerase binding (7, 13). In the case of E. coli ␴32, amino acid substitutions in region 2.2 reduced its affinity for core RNA polymerase (17), while amino acid substitutions in region 2.1 suggested that this region may bind DnaK (15, 35). These observations concur with the view that DnaK and core RNA polymerase compete for binding to specific regions of ␴32 (45, 46). The proteins most closely related to R. sphaeroides RpoHII, however, show less primary amino acid conservation, as a group, in regions 2.1 and 2.2 than their respective paralogs (Fig. 1B). The relative plasticity in the primary amino acid sequences of regions 2.1 and 2.2 among the RpoHII clade may modify their affinities for core RNA polymerase or, possibly, the binding of chaperones that could negatively regulate their activity (14, 41). Thus, the differences in promoter activity that we have observed for RpoHI and RpoHII may in part be due to their relative efficiencies in binding core RNA polymerase, as reflected by the amino acid sequence variations in regions 2.1 and 2.2 and the RpoH box. While domains 2 and 4 of the ␴70-type sigma factors are structurally conserved, it is the primary amino acid sequence variation in regions 2.4 and 4.2 that accounts for differences in promoter recognition among the diverse members of this superfamily (13). In light of the fact that RpoHI and RpoHII each recognize several heat shock promoters, it is not surprising that regions 2.4 and 4.2 of these two proteins show little sequence variation from other ␴32 family members. In addition, both RpoHI and RpoHII share several amino acid residues with regions 2.4 and 4.2 of E. coli ␴32 that are believed to contact the ⫺10 and ⫺35 elements of the groE promoter (21). Promoter recognition by RpoHI and RpoHII. Of the six R. sphaeroides promoters that we examined, all were transcribed by either R. sphaeroides or E. coli core RNA polymerase reconstituted with RpoHI. When recombinant RpoHII was combined with R. sphaeroides core enzyme, however, we were unable to detect transcripts from two promoters, rpoD PHS and dnaK P1. This was surprising, since rpoD PHS is one of three promoters we chose to analyze based on the knowledge that their transcript abundance increases in ⌬RpoHI cells after a heat shock (19). On the other hand, when RpoHII was added to E. coli core RNA polymerase, we detected a product from rpoD PHS. The lack of a product from rpoD PHS in the presence of RpoHII and the R. sphaeroides core enzyme may have been due to a protein, or proteins, in the core RNA polymerase preparation that reduces in vitro transcription from this particular promoter. Another feasible explanation is that promoter escape from R. sphaeroides rpoD PHS may be more Downloaded from http://jb.asm.org/ on August 15, 2015 by guest Previous studies identified R. sphaeroides RpoHI as a member of the ␴32 family of alternate sigma factors (19). Several lines of evidence, however, intimated the existence of a second sigma factor that recognized heat-inducible promoters (19, 24). In this work, we provide evidence that R. sphaeroides RpoHII is this second sigma factor. Amino acid sequence alignments show that RpoHII is a member of the ␴32 family. Like its paralog, RpoHI, expression of RpoHII complements the growth defect of a temperature-sensitive E. coli ␴32 mutant and initiates transcription from the E. coli ␴32-dependent promoter rpoD PHS both in vivo and in vitro. Through the use of purified components, we demonstrate that RpoHI and RpoHII recognize a number of known and presumed heat shock promoters from R. sphaeroides. We also identify differences in the abilities of RpoHI and RpoHII to recognize individual promoters in vitro. RpoHII is a second sigma factor that recognizes R. sphaeroides heat shock promoters. Previously, two proteins copurifying with R. sphaeroides RNA polymerase were eluted from SDS-PAGE gels, individually reconstituted with core enzyme, and shown to transcribe the E. coli dnaK P1 and R. sphaeroides cycA P1 promoters (19, 24). One of these proteins, ␴37, was tentatively identified as RpoHI, but the identity of the other protein, known as ␴38, was not known. Several lines of evidence link RpoHII to the ␴38 activity described by previous studies. Both recombinant RpoHII (this work) and gel-purified ␴38 (19) recognized the heat-inducible promoters dnaK P1 and cycA P1, and both proteins have similar apparent molecular weights when analyzed by SDS-PAGE (data not shown). Accordingly, antibody raised to His-tagged RpoHII reacts with a protein of similar molecular weight in R. sphaeroides RNA polymerase preparations (data not shown). In addition to identifying RpoHII as a second member of the ␴32 family in R. sphaeroides, our experiments provide the first in vitro analysis of promoter recognition by a protein belonging to the clade containing RpoHII. Comparison of RpoH primary amino acid sequences. There are now seven organisms, all members of the ␣-proteobacteria, predicted to contain two rpoH genes. R. sphaeroides RpoHII falls into a distinct phylogenetic clade of alternate sigma factors whose primary amino acid sequences differ somewhat from their respective paralogs, most noticeably in conserved regions 2.1 and 2.2 and the RpoH box. The RpoH box lies in region C, a stretch of 23 amino acids that initially was implicated in the DnaK-mediated degradation of ␴32 (25, 28). Subsequent studies indicated that instead, region C directly interacts with core RNA polymerase and that the RpoH box may give ␴32 a competitive advantage over other sigma factors in binding to this enzyme (3, 18). For example, amino acid substitutions in the fifth and sixth residues (F136L and F137E) of the E. coli ␴32 RpoH box (132QRKLFFNLR140) each decreased the binding of core RNA polymerase (3, 18). These two phenylalanines, together with L135, the fourth residue, 5719 5720 GREEN AND DONOHUE or RpoHII, dnaK P1 was a stronger promoter than rpoD PHS. Among the E. coli ␴32-dependent promoters that have been studied, those with higher activities tend to more closely mimic the ␴32 consensus sequence (43). Indeed, the ⫺10 element of E. coli dnaK P1 is a perfect match to the ␴32 consensus. In addition, a recent study demonstrated that a tryptophan (W108) in the boundary between regions 2.3 and 2.4 of ␴32 contacts a conserved ⫺13 C-G base pair (21). The presence of a tryptophan residue at this position in both RpoHI and RpoHII may explain why these sigma factors transcribe E. coli dnaK P1, which contains this conserved base pair, with greater efficiency than rpoD PHS. The number of R. sphaeroides promoters tested, along with the variation in their sequences, makes it difficult to propose a consensus promoter for either RpoHI or RpoHII. The fact that these two sigma factors do not recognize an identical repertoire of promoters in vitro makes constructing such a consensus for R. sphaeroides even more challenging. A future interest is to determine whether R. sphaeroides has separate RpoHIand RpoHII-specific target genes that allow this ␣-proteobacterium to respond to different stress, or other environmental, signals. For example, recent studies have shown that RpoHII is part of the RpoE regulon (2). RpoE is required for R. sphaeroides to mount a transcriptional response to singlet oxygen (2). Consequently, RpoHII may directly regulate the expression of genes whose products are required to mitigate the damaging effects of this reactive oxygen species as well as play a role with RpoHI in responding to stress caused by elevated temperatures. ACKNOWLEDGMENTS This study was supported by grants GM37509 and GM75273 (National Institute of General Medical Sciences) and DE-FG0205ER15653 (Department of Energy) to T.J.D. We are grateful to Matias Cafaro for constructing the phylogenetic tree and to Yann Dufour for his computer expertise. We also thank Rachelle Stenzel and Archna Bhasin for their discussions and thoughtful reading of the manuscript. REFERENCES 1. Anthony, J., H. A. Green, and T. J. Donohue. 2003. Rhodobacter sphaeroides RNA polymerase and its sigma factors. Methods Enzymol. 370:54–65. 2. Anthony, J. R., K. L. Warczak, and T. J. Donohue. 2005. A transcriptional response to singlet oxygen, a toxic byproduct of photosynthesis. Proc. Natl. Acad. Sci. USA 102:6502–6507. 3. Arsene, F., T. Tomoyasu, A. Mogk, C. Schirra, A. Schulze-Specking, and B. Bukau. 1999. Role of region C in regulation of the heat shock gene-specific sigma factor of Escherichia coli, ␴32. J. Bacteriol. 181:3552–3561. 4. Artsimovitch, I., V. Svetlov, L. Anthony, R. R. Burgess, and R. Landick. 2000. RNA polymerases from Bacillus subtilis and Escherichia coli differ in recognition of regulatory signals in vitro. J. Bacteriol. 182:6027–6035. 5. Bardwell, J. C. A., and E. A. Craig. 1984. Major heat shock gene of Drosophila and the Escherichia coli heat-inducible dnaK gene are homologous. Proc. Natl. Acad. Sci. USA 81:848–852. 6. Barker, M. M., T. Gaal, C. A. Josaitis, and R. L. Gourse. 2001. Mechanism of regulation of transcription initiation by ppGpp. I. Effects of ppGpp on transcription initiation in vivo and in vitro. J. Mol. Biol. 305:673–688. 7. Burgess, R. R., and L. Anthony. 2001. How sigma docks to RNA polymerase and what sigma does. Curr. Opin. Microbiol. 4:126–131. 8. Cowing, D. W., J. C. A. Bardwell, E. A. Craig, C. Woolford, R. W. Hendrix, and C. A. Gross. 1985. Consensus sequence for Escherichia coli heat shock gene promoters. Proc. Natl. Acad. Sci. USA 82:2679–2683. 9. Dartigalongue, C., H. Loferer, and S. Raina. 2001. EcfE, a new essential inner membrane protease: its role in the regulation of heat shock response in Escherichia coli. EMBO J. 20:5908–5918. 10. Graf, P. C., and U. Jakob. 2002. Redox-regulated molecular chaperones. Cell. Mol. Life Sci. 59:1624–1631. 11. Gross, C. A. 1996. Function and regulation of the heat shock proteins, p. 1382–1399. In F. C. Neidhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, Downloaded from http://jb.asm.org/ on August 15, 2015 by guest efficient when RpoHII is added to E. coli core RNA polymerase. For example, Artsimovitch and coworkers reconstituted E. coli ␴32 with either Bacillus subtilis or E. coli core RNA polymerase and demonstrated that the two holoenzymes differed vastly in their promoter selectivities (4). Their results suggested that while promoter recognition resides mainly in the sigma subunit, promoter utilization depends primarily on core RNA polymerase and that contacts made between the core subunits and promoter DNA may either facilitate or inhibit promoter escape (4). Among the R. sphaeroides promoters that we examined, rpoD PHS is most similar to the E. coli ␴32 consensus in both the ⫺10 and ⫺35 regions, suggesting that specific contacts made between E. coli core RNA polymerase and this promoter may result in more efficient transcription than those made with the R. sphaeroides core enzyme. Why the R. sphaeroides dnaK P1 promoter was not transcribed by either the E. coli or R. sphaeroides core enzymes in the presence of RpoHII is unknown at this time. The gene product DnaK is an Hsp70 protein (5, 8), and we saw a diminished increase in the synthesis rate of a ⬃75-kDa protein in ⌬RpoHI cells after a heat shock relative to wild-type cells (19). If this ⬃75-kDa protein is DnaK, a second mechanism, independent of RpoHII, may be responsible for the heat induction of this promoter in cells lacking RpoHI. In contrast, the heatinduced synthesis of a 76-kDa protein in S. meliloti appears to be totally dependent on RpoH1 (37). While it is conceivable that dnaK P1 is simply not recognized by RpoHII, the ⫺10 region of dnaK P1 is almost identical to that of the strongest R. sphaeroides promoter for RpoHII that we tested, cycA P1. The ⫺35 regions of these two promoters, however, share only the TTG motif, and previous studies have shown that a point mutation in the ⫺35 region of cycA P1 (from ⫺34TTGA⫺31 to ⫺34 TTGC⫺31) reduced activity by ⬃60% (24). Since dnaK P1 has a C in this position (TTGC), RpoHII may have a greater requirement for an A in the TTGA motif of the ⫺35 element than RpoHI. If we use transcript abundance as an estimate of promoter strength, the strongest R. sphaeroides promoter that we tested with either RpoHI or RpoHII was cycA P1. Despite this, the ⫺10 and ⫺35 elements of cycA P1 have fewer matches to the E. coli ␴32 consensus sequence than the other five promoters examined (Table 2). Thus, the canonical idea of what makes a strong heat shock promoter may not apply to R. sphaeroides RpoHI and RpoHII. Less-than-optimal heat shock promoter sequences for other ␣-proteobacteria have been noted (26, 30, 39). Based on the upstream regions of nine dnaKJ and groESL operons from eight ␣-proteobacteria, Segal and Ron constructed a putative consensus sequence (CTTG[17 to 18 bp]C YTAT-T--G) that differs from the E. coli ␴32 consensus at several positions (39). Only the R. sphaeroides dnaK P1 and hslO promoter regions, however, abide by this proposed consensus sequence. The existence of relatively weak promoters that are regulated by two or more sigma factors may represent a mechanism that evolved to more subtly regulate gene expression in the absence or presence of different stress conditions. Based on transcript abundance alone, both of the E. coli promoters that we examined, dnaK P1 and rpoD PHS, were considerably stronger than any of the six R. sphaeroides promoters analyzed. Furthermore, whether a holoenzyme was generated by mixing E. coli core RNA polymerase with RpoHI J. BACTERIOL. VOL. 188, 2006 12. 13. 14. 15. 16. 17. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 5721 of the ␴32 polypeptide is involved in DnaK-mediated negative control of the heat shock response in Escherichia coli. Proc. Natl. Acad. Sci. USA 91: 10280–10284. Nakahigashi, K., H. Yanagi, and T. Yura. 1995. Isolation and sequence analysis of rpoH genes encoding ␴32 homologs from gram negative bacteria: conserved mRNA and protein segments for heat shock regulation. Nucleic Acids Res. 23:4383–4390. Narberhaus, F., M. Kowarik, C. Beck, and H. Hennecke. 1998. Promoter selectivity of the Bradyrhizobium japonicum RpoH transcription factors in vivo and in vitro. J. Bacteriol. 180:2395–2401. Narberhaus, F., P. Krummenacher, H. Fischer, and H. Hennecke. 1997. Three disparately regulated genes for ␴32-like transcription factors in Bradyrhizobium japonicum. Mol. Microbiol. 24:93–104. Narberhaus, F., W. Weiglhofer, H. M. Fischer, and H. Hennecke. 1996. The Bradyrhizobium japonicum rpoH1 gene encoding a ␴32-like protein is part of a unique heat shock gene cluster together with groESL1 and three small heat shock genes. J. Bacteriol. 178:5337–5346. Neidhardt, F. C., and R. A. VanBogelen. 1987. Heat shock response, p. 1334– 1346. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology, vol. 2. American Society of Microbiology, Washington, D.C. Newman, J. D., M. J. Falkowski, B. A. Schilke, L. C. Anthony, and T. J. Donohue. 1999. The Rhodobacter sphaeroides ECF sigma factor, ␴E, and the target promoters cycA P3 and rpoE P1. J. Mol. Biol. 294:307–320. Obrist, M., and F. Narberhaus. 2005. Identification of a turnover element in region 2.1 of Escherichia coli ␴32 by a bacterial one-hybrid approach. J. Bacteriol. 187:3807–3813. Oke, V., B. G. Rushing, E. J. Fisher, M. Moghadam-Tabrizi, and S. R. Long. 2001. Identification of the heat-shock sigma factor RpoH and a second RpoH-like protein in Sinorhizobium meliloti. Microbiology 147:2399–2408. Ono, Y., H. Mitsui, T. Sato, and K. Minamisawa. 2001. Two RpoH homologs responsible for the expression of heat shock protein genes in Sinorhizobium meliloti. Mol. Gen. Genet. 264:902–912. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Segal, G., and E. Z. Ron. 1995. The dnaKJ operon of Agrobacterium tumefaciens: transcriptional analysis and evidence for a new heat shock promoter. J. Bacteriol. 177:5952–5958. Studholme, D. J., M. Buck, and T. Nixon. 2000. Identification of potential ␴N-dependent promoters in bacterial genomes. Microbiology 146:3021–3023. Tomoyasu, T., T. Ogura, T. Tatsuta, and B. Bukau. 1998. Levels of DnaK and DnaJ provide tight control of heat shock gene expression and protein repair in Escherichia coli. Mol. Microbiol. 30:567–581. Wagner-Dobler, I., H. Rheims, A. Felske, R. Pukall, and B. J. Tindall. 2003. Jannaschia helgolandensis gen. nov., sp. nov., a novel abundant member of the marine Roseobacter clade from the North Sea. Int. J. Syst. Evol. Microbiol. 53:731–738. Wang, Y., and P. L. deHaseth. 2003. Sigma 32-dependent promoter activity in vivo: sequence determinants of the groE promoter. J. Bacteriol. 185:5800– 5806. Wosten, M. M. 1998. Eubacterial sigma-factors. FEMS Microbiol. Rev. 22: 127–150. Yura, T., M. Kanemori, and M. T. Morita. 2000. The heat shock response: regulation and function, p. 3–18. In G. Storz and R. Hengge-Aronis (ed.), Bacterial stress responses. ASM Press, Washington, D.C. Zhao, K., M. Liu, and R. R. Burgess. 2005. The global transcriptional response of Escherichia coli to induced ␴32 protein involves ␴32 regulon activation followed by inactivation and degradation of ␴32 in vivo. J. Biol. Chem. 280:17758–17768. Downloaded from http://jb.asm.org/ on August 15, 2015 by guest 18. K. B. Low, B. Magasanik, W. S. Reznikoff, M. Rily, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1. American Society for Microbiology, Washington, D.C. Grossman, A. D., J. W. Erickson, and C. A. Gross. 1984. The htpR gene product of E. coli is a sigma factor for heat-shock promoters. Cell 38:383– 390. Gruber, T. M., and C. A. Gross. 2003. Multiple sigma subunits and the partitioning of bacterial transcription space. Annu. Rev. Microbiol. 57:441– 466. Guisbert, E., C. Herman, C. Z. Lu, and C. A. Gross. 2004. A chaperone network controls the heat shock response in E. coli. Genes Dev. 18:2812– 2821. Horikoshi, M., T. Yura, S. Tsuchimoto, Y. Fukumori, and M. Kanemori. 2004. Conserved region 2.1 of Escherichia coli heat shock transcription factor ␴32 is required for modulating both metabolic stability and transcriptional activity. J. Bacteriol. 186:7474–7480. Jakob, U., W. Muse, M. Eser, and J. C. A. Bardwell. 1999. Chaperone activity with a redox switch. Cell 96:341–352. Joo, D. M., N. Ng, and R. Calendar. 1997. A ␴32 mutant with a single amino acid change in the highly conserved region 2.2 exhibits reduced core RNA polymerase affinity. Proc. Natl. Acad. Sci. USA 94:4907–4912. Joo, D. M., A. Nolte, R. Calendar, Y. N. Zhou, and D. J. Jin. 1998. Multiple regions on the Escherichia coli heat shock transcription factor ␴32 determine core RNA polymerase binding specificity. J. Bacteriol. 180:1095–1102. Karls, R. K., J. Brooks, P. Rossmeissl, J. Luedke, and T. J. Donohue. 1998. Metabolic roles of a Rhodobacter sphaeroides member of the ␴32 family. J. Bacteriol. 180:10–19. Karls, R. K., D. J. Jin, and T. J. Donohue. 1993. Transcription properties of RNA polymerase holoenzymes isolated from the purple nonsulfur bacterium Rhodobacter sphaeroides. J. Bacteriol. 175:7629–7638. Kourennaia, O. V., L. Tsujikawa, and P. L. deHaseth. 2005. Mutational analysis of Escherichia coli heat shock transcription factor sigma 32 reveals similarities with sigma 70 in recognition of the ⫺35 promoter element and differences in promoter DNA melting and ⫺10 recognition. J. Bacteriol. 187:6762–6769. Kusukawa, N., and T. Yura. 1988. Heat shock protein GroE of Escherichia coli: key protective roles against thermal stress. Genes Dev. 2:874–882. Landick, R., V. Vaughn, E. T. Lau, R. A. VanBogelen, J. W. Erickson, and F. C. Neidhardt. 1984. Nucleotide sequence of the heat shock regulatory gene of E. coli suggests its protein product may be a transcription factor. Cell 38:175. MacGregor, B. J., R. K. Karls, and T. J. Donohue. 1998. Transcription of the Rhodobacter sphaeroides cycA P1 promoter by alternate RNA polymerase holoenzymes. J. Bacteriol. 180:1–9. McCarty, J. S., S. Rudiger, H.-J. Schonfeld, J. Schneider-Mergener, K. Nakahigashi, T. Yura, and B. Bukau. 1996. Regulatory region C of the E. coli heat shock transcription factor, ␴32, constitutes a DnaK binding site and is conserved among eubacteria. J. Mol. Biol. 256:829–837. Minder, A. C., F. Narberhaus, M. Babst, H. Hennecke, and H. M. Fischer. 1997. The dnaKJ operon belongs to the ␴32-dependent class of heat shock genes in Bradyrhizobium japonicum. Mol. Gen. Genet. 254:195–206. Moran, M. A., A. Buchan, J. M. Gonzalez, J. F. Heidelberg, W. B. Whitman, R. P. Kiene, J. R. Henriksen, G. M. King, R. Belas, C. Fuqua, L. Brinkac, M. Lewis, S. Johri, B. Weaver, G. Pai, J. A. Eisen, E. Rahe, W. M. Sheldon, W. Ye, T. R. Miller, J. Carlton, D. A. Rasko, I. T. Paulsen, Q. Ren, S. C. Daugherty, R. T. Deboy, R. J. Dodson, A. S. Durkin, R. Madupu, W. C. Nelson, S. A. Sullivan, M. J. Rosovitz, D. H. Haft, J. Selengut, and N. Ward. 2004. Genome sequence of Silicibacter pomeroyi reveals adaptations to the marine environment. Nature 432:910–913. Nagai, H., H. Yuzawa, M. Kanemori, and T. Yura. 1994. A distinct segment ACTIVITY OF R. SPHAEROIDES RpoHII