Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ranson-Olson, B. Articles by Zeilstra-Ryalls, J. H. Search for Related Content PubMed PubMed Citation Articles by Ranson-Olson, B. Articles by Zeilstra-Ryalls, J. H.

 Previous Article  |  Next Article 

Journal of Bacteriology, October 2008, p. 6769-6778, Vol. 190, No. 20
0021-9193/08/$08.00+0     doi:10.1128/JB.00828-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Regulation of the Rhodobacter sphaeroides 2.4.1 hemA Gene by PrrA and FnrL

Britton Ranson-Olson^1, and Jill H. Zeilstra-Ryalls^2*

Department of Biological Sciences, Oakland University, Rochester, Michigan 48309,¹ Department of Biological Sciences, Bowling Green State University, Bowling Green, Ohio 43403²

Received 13 June 2008/ Accepted 30 July 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Part of the oxygen responsiveness of Rhodobacter sphaeroides 2.4.1 tetrapyrrole production involves changes in transcription of the hemA gene, which codes for one of two isoenzymes catalyzing 5-aminolevulinic acid synthesis. Regulation of hemA transcription from its two promoters is mediated by the DNA binding proteins FnrL and PrrA. The two PrrA binding sites, binding sites I and II, which are located upstream of the more-5' hemA promoter (P1), are equally important to transcription under aerobic conditions, while binding site II is more important under anaerobic conditions. By using phosphoprotein affinity chromatography and immunoblot analyses, we showed that the phosphorylated PrrA levels in the cell increase with decreasing oxygen tensions. Then, using both in vivo and in vitro methods, we demonstrated that the relative affinities of phosphorylated and unphosphorylated PrrA for the two binding sites differ and that phosphorylated PrrA has greater affinity for site II. We also showed that PrrA regulation is directed toward the P1 promoter. We propose that the PrrA component of anaerobic induction of P1 transcription is attributable to higher affinity of phosphorylated PrrA than of unphosphorylated PrrA for binding site II. Anaerobic activation of the more-3' hemA promoter (P2) is thought to involve FnrL binding to an FNR consensuslike sequence located upstream of the P2 promoter, but the contribution of FnrL to P1 induction may be indirect since the P1 transcription start is within the putative FnrL binding site. We present evidence suggesting that the indirect action of FnrL works through PrrA and discuss possible mechanisms.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The hemA gene of the alphaproteobacterium Rhodobacter sphaeroides 2.4.1 codes for one of two 5-aminolevulinic acid (ALA) synthase isoenzymes responsible for ALA production. ALA is the precursor of all tetrapyrroles, and in keeping with the ability of this bacterium to generate energy from respiratory chains and from anoxygenic photosynthesis, R. sphaeroides synthesizes both heme and bacteriochlorophyll (Bchl), as well as vitamin B₁₂, a necessary cofactor in Bchl formation (14) and also in methionine and cysteine biosynthesis (3). Both the absolute and relative amounts of these tetrapyrroles change dramatically in this bacterium as part of its metabolic response to changes in its environment, and oxygen is a key player in controlling those events. Certainly one of the most notable responses is the >100-fold increase in Bchl production when oxygen tensions are reduced (23), which is necessary in preparation for photosynthesis. At the same time, appropriate levels of heme and vitamin B₁₂ must also be maintained. This radical change in tetrapyrrole biosynthesis is enabled in part by an increase in ALA formation (23), and with respect to oxygen control we have found that among the ALA synthase genes present in this organism, alterations in hemA expression are responsible (38, 40). Therefore, investigations of hemA regulation should provide insights into the mechanisms of oxygen control in the cell. Our previous studies have shown that hemA is transcribed from two promoters and that two global regulatory proteins, PrrA and FnrL, regulate hemA transcription in response to changes in oxygen tension (11, 30).

FnrL is a homolog of the Escherichia coli anaerobic regulatory protein Fnr, an oxygen-labile DNA binding protein that is active anaerobically (20). The hemA upstream sequences contain a perfect FNR consensus sequence (35). FnrL has been shown to activate transcription from the downstream P2 promoter in response to decreasing oxygen tensions (11), and this is consistent with the –45 position of the FNR consensus sequence relative to the transcription start site of P2. However, as the transcription start site of the upstream P1 promoter is centrally located within the FNR consensus sequence, activation of P1 involving FnrL binding to the consensus would not be expected. Nevertheless, we have demonstrated that an intact fnrL gene is also required for increased expression from the P1 promoter in response to decreasing oxygen tensions (11, 38). Therefore, we have proposed that FnrL has an indirect role in regulation of transcription from P1, which predicts the existence of another factor that acts directly.

PrrA is the DNA binding regulator protein of the PrrBA two-component redox-responsive regulatory system in R. sphaeroides. A combination of microarray and other investigations have shown that the PrrA regulon includes tetrapyrrole biosynthesis genes (10, 11, 27, 30, 41), as well as genes with roles in many other biological processes, including photosynthesis, carbon and nitrogen fixation, denitrification, formaldehyde dehydrogenase activity, and aerotaxis (2, 8, 13, 18, 31). For hemA, two PrrA binding sites that are centered at bp –163 and –67 relative to the P1 transcription start site have been identified. In vitro studies indicated that PrrA activates P1 transcription, and while the in vivo data are consistent with the hypothesis that PrrA has an activator role, they also revealed that although the two PrrA binding sites are equally important in transcription under aerobic conditions, one binding site predominates under anaerobic conditions (30).

From these studies, two central questions arise. How is the differential effect of PrrA on the two binding sites achieved, and how do PrrA and FnrL work together to achieve the correct levels of hemA expression? Our investigations of these questions provided new insights into hemA gene expression, and our findings may contribute to understanding the actions of these DNA binding proteins with other members of both regulons.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Bacterial strains, plasmids, and growth conditions. The bacterial strains and plasmids used in this study are listed in Table 1. The E. coli strains were grown at 37°C in Luria-Bertani media (32). Antibiotics were used for strain selection or plasmid maintenance at the following final concentrations: 100 µg/ml of ampicillin, 15 µg/ml of tetracycline (Tc), and 50 µg/ml of kanamycin (Kn) or 50 µg/ml of spectinomycin and 50 µg/ml of streptomycin. R. sphaeroides strains were cultured at 28°C in Sistrom's succinate minimal medium A (5, 7, 34). The following growth conditions were used. Highly aerobic conditions were achieved by sparging liquid cultures with a mixture of 30% oxygen, 2% carbon dioxide, and 68% nitrogen; semiaerobic conditions were achieved by sparging liquid cultures with a mixture of 2% oxygen, 2% carbon dioxide, and 96% nitrogen; and anaerobic conditions were achieved either by sparging liquid cultures with a mixture of 2% carbon dioxide and 98% nitrogen or by growing cells in completely filled screw-cap tubes. When solid media were used for anaerobic incubation, the plates were placed in jars and an anaerobic environment was generated using BD BBL GasPak anaerobic system envelopes (Becton, Dickinson and Company, Sparks, MD). For anaerobic growth in the dark, liquid and solid media were supplemented with yeast extract (final concentration, 1% [wt/vol]) and dimethyl sulfoxide (DMSO) (final concentration, 0.06 M) as an alternate electron acceptor, and for photosynthetic growth anaerobic cultures were placed in front of banks of incandescent lights (ca. 10 W/m²). When appropriate, antibiotics were used at the following final concentrations: 0.5 µg/ml of Tc, 50 µg/ml of spectinomycin and streptomycin, or 50 µg/ml Kn. Reagent-grade antibiotics and chemicals were purchased from Sigma Chemical Co. (St. Louis, MO).

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

TABLE 1. Bacterial strains and plasmids

Conjugation and transformation. Plasmids were mobilized into R. sphaeroides by triparental mating as described previously by Davis et al. (5), using the HB101(pRK2013) helper strain (6, 12). Transformation of E. coli was carried out using CaCl₂-treated cells (32).

DNA manipulation and DNA sequence analysis. Standard protocols (32) were used or the manufacturers' instructions were followed for general DNA and plasmid manipulation, including purification, isolation, restriction endonuclease treatment, and other enzymatic treatments. Plasmid DNA was purified using a FastPlasmid mini kit (Qiagen, Valencia, CA), and a Zymoclean purification kit (Zymo Research Co., Orange, CA) was used for isolation of DNA from agarose. Restriction enzymes were purchased from New England BioLabs, Inc. (Beverly, MA), Gibco-BRL/Life Technologies, Inc. (Gaithersburg, MD), and Promega (Madison, WI). DNA sequencing was performed with an ABI Prism 310 genetic analyzer with an ABI Prism BigDye Terminator cycle sequencing Ready Reaction kit (Applied Biosystems, Inc., Foster City, CA), and the sequencing reaction mixtures were prepared according to the manufacturers' instructions. To improve primer extension reactions, DMSO was added at a final concentration of 5% due to the high G+C content of R. sphaeroides templates. Oligonucleotides were purchased from Integrated DNA Technologies (Coralville, IA).

Construction of β-galactosidase reporter plasmids. A QuikChange site-directed mutagenesis kit (Stratagene, La Jolla, CA) was used to carry out oligonucleotide-directed mutagenesis. In all cases, the integrity of the relevant sequences was confirmed by DNA sequencing. The mutations introduced into the hemA sequences are indicated in Fig. 1. Plasmid templates with 316 bp of hemA sequences upstream of the translation initiation site that either were not altered or were altered at the PrrA binding site I, at the binding site II, or at both binding sites (30) were used as templates in oligonucleotide-directed mutagenesis reactions to create the P2 promoter deletions with primer HemAP2DelUP and the complementary primer, HemAP2DelDOWN (11). Each of the altered hemA upstream sequences (Fig. 1) was excised from the vector using the unique PstI and XbaI sites flanking the hemA sequences and ligated with the promoterless lacZ vector pCF1010 that had been restricted with PstI and XbaI (Table 1).

View larger version (16K): [in this window] [in a new window]

FIG. 1. Description of the relevant hemA wild-type and mutant upstream sequences used in this study. The +1 sites of transcription from hemA promoters P1 and P2 (11, 26) are labeled (P1) and (P2). The FNR consensuslike sequence is also labeled. DNA sequences protected from DNase I cleavage by PrrA, binding sites I and II (30), are indicated by shading, and the identical 9-bp motifs in these protected regions are highlighted. The P2 sequences that were deleted using the oligonucleotides described previously (11) are in parentheses. DNA oligonucleotides used to create mutations within PrrA binding sites I and II were described previously (30). Sequences corresponding to the primers used to generate DNA that was investigated by the EMSAs are indicated by arrows; labeled binding site I DNA was generated using primer "UP" and biotin-labeled primer "I," labeled binding site II DNA was generated using primer "II" and biotin-labeled primer "DOWN," and unlabeled competitor DNA was generated using primers "UP" and "DOWN." For further details concerning the oligonucleotides and their use, see Materials and Methods.

Construction of plasmids with wild-type or mutant prrA genes. The wild-type prrA gene was amplified from purified genomic DNA (1) using primers PrrA5'UP (5'-GGCAAAGCCCTCGCTCGTC-3') and PrrA3'DOWN (5'-CTTGATCGCAGCCTCGAACCAG-3') and ligated into the EcoRV site of pUI1087 (40) to create plasmid pBRO86. This plasmid was then used for oligonucleotide-directed mutagenesis with primer PrrAD63KUP (5'-GCCTATGCAGTGGTGAAGCTGCGGCTCGAGGAC-3') and the complementary primer, PrrAD63KDOWN, to create a mutant prrA gene coding for the D63K amino-acid-substituted protein. Using the same plasmid template, a truncated prrA' gene that lacked sequences coding for the last 24 amino acids of PrrA was generated by PCR performed with primers PrrA5'UP and PrrATrunc (5'-AAGCTTCAGACATTGCGGTCGCACATTTC-3'). Subsequent oligonucleotide-directed mutagenesis of the prrA' gene with primers PrrAD63KUP and PrrAD63KDOWN further altered the sequences so that they encoded a D63K amino acid substitution and truncated protein. All of the prrA mutant alleles were excised from the vector plasmids using the flanking HindIII and BamHI sites and ligated into HindIII-BamHI-restricted pRK415, creating plasmids pD63K, pPrrA', and pD63K-PrrA'.

Construction of R. sphaeroides mutant strains. For construction of the prrA null mutant BRO107 and fnrL prrA double null mutant BRO108, in which the PrrA coding sequences were deleted, the cre-lox site-specific recombinase vectors (25) were used. To generate the target strain containing the loxP sites for the cre recombinase, suicide vector pBRO81 was constructed, in which 491 bp upstream and 515 bp downstream of the PrrA coding sequences were inserted into pCM184 so that they were correctly oriented 5' to 3' with respect to each other, with the loxP-Kn^r-loxP DNA sequences located between them. This plasmid was mobilized into wild-type strain 2.4.1 and fnrL null mutant strain JZ1691 (Table 1). Plasmid pCM154, which carried the cre site-specific recombinase gene, was then mobilized into suitable recombinant candidates that were Kn^r but Tc^s. For several of the Tc^r exconjugants, which were now Kn^s, the pCM154 plasmid was segregated away by culturing the cells in the absence of Tc. Recombination events in both BRO107 and BRO108 were confirmed by PCR amplification of the genomic DNA sequences using primers corresponding to sequences flanking the prrA deletion and subsequently sequencing the PCR products. Additionally, the absence of PrrA was confirmed by immunoblot analysis.

For construction of the wild-type and prrA-D63K "knock-in" strains JZ4148 and JZ4141, the wild-type or mutant genes were isolated from pBRO86 and its mutagenized derivative plasmid using PvuII and then ligated to the suicide vector fragment of pSUP202 generated using ScaI. The resulting recombinant pSUP-WT and pSUP-D63K plasmids were mobilized into prrA null mutant strain BRO107, and plasmid integrants were selected for using Tc^r. PCR and immunoblot analyses were used to confirm the presence and functionality of the wild-type and mutant prrA alleles.

PrrA purification and modification. The PrrA protein was purified from ER2566 with plasmid pJC407 (4), an E. coli plasmid that expresses a PrrA intein/chitin-binding domain fusion protein, using the IMPACT T7 one-step protein purification system (New England Biolabs, Waltham, MA). Binding of the fusion protein to chitin beads and release of the PrrA protein by the intein cleavage reaction were carried out under conditions described previously (4). The purified protein was then concentrated, dialyzed against storage buffer (40 mM Tris-HCl [pH 7.9], 50 mM KCl, 5 mM MgCl₂, 1 mM dithiothreitol), and stored as described previously (4).

PrrA was treated with acetyl phosphate as previously described (4); treatment with the phosphate analogue BeF₃^– (37) was performed as described by Laguri et al. (21). BeF₃^– was generated in situ using reaction mixtures (total volume, 20 µl) containing 30 µM PrrA, 2 mM BeCl₂, 6 mM NaF, and 20 mM MgCl₂ that were incubated for 30 min at 30°C.

Phosphoprotein enrichment. Affinity chromatography of phosphoproteins with BD phosphoprotein or TALON PMAC phosphoprotein enrichment kits (Clontech, Mountain View, CA) was performed according to the manufacturer's instructions, except that an additional final wash with 20 mM Tris-HCl-0.5 M NaCl (pH 7.5) was included prior to elution to improve specificity. The samples used were crude lysates of cells in buffer A (a component of the enrichment kits) and protease inhibitor cocktail (Sigma Chemical Co.) at the concentrations recommended by the manufacturers. The crude cell lysates were prepared by passaging the cells through an SLM-Aminco French pressure cell (Spectronic Instruments Inc., Rochester, NY) at 700 lb/in², and insoluble material was pelleted by centrifugation at 13,000 x g for 15 min at 4°C.

Immunoblot analysis. Nitrocellulose membranes (Micron Separations Inc., Westboro, MA) were prepared by electrophoretic transfer of proteins resolved by polyacrylamide gel electrophoresis using 12% gels from Invitrogen (Carlsbad, CA). The membranes were then probed using standard procedures (15) and a 1:5,000 dilution of primary PrrA antisera or a 1:10,000 dilution of primary HemA rabbit antisera. In all cases, the secondary antibody was alkaline phosphatase-conjugated goat anti-rabbit antisera (Sigma Chemical Co.). Detection of immunocomplexes was carried out using the ImmunO alkaline phosphatase substrate 5-bromo-4-chloro-3-indolylphosphate/nitroblue tetrazolium (MP Biomedicals, LLC, Aurora, OH). Relative band intensities were determined using the Kodak Image Analysis software.

EMSAs. End-labeled DNA was generated by PCR using primer pairs in which one of the primers was 5' end labeled with biotin. Suitable concentrations of end-labeled DNA were determined by dot blot analysis as described previously (30). The binding reaction mixtures (total volume, 20 µl) contained approximately 150 nM biotin-labeled target DNA, 1.33 µM PrrA or BeF₃^–-PrrA, 2.5% glycerol, 50 ng/µl poly(dA-dT), 40 mM Tris (pH 7.9), 50 mM MgCl₂, 2 mM EDTA, and various amounts of unlabeled competitor DNA, and they were incubated for 20 min at room temperature, which was followed by addition of 5 µl loading buffer (Pierce chemiluminescent electrophoretic mobility shift assay [EMSA] kit). The mobility shift assays were carried out according to the manufacturer's instructions, except that poly(dA-dT) (Sigma Chemical Co.) was substituted for poly(dC-dG), which is recommended to reduce nonspecific binding of high-G+C-content genomes. Sample electrophoresis, transfer, and detection of the biotin-labeled DNA were performed as previously described (30).

β-Galactosidase activity assays. Assays to determine β-galactosidase activities in crude cell lysates were performed as previously described (36). The lysates were prepared as described above for phosphoprotein affinity chromatography, except that the cells were lysed in phosphate buffer (0.1 M NaPO₄, pH 7.7).

Protein concentrations. Protein concentrations were determined with Pierce bicinchoninic acid protein assay reagents or the Bio-Rad protein assay dye reagent concentrate (Hercules, CA), and bovine serum albumin was used as a standard in all cases.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In vivo analysis of the phosphorylation state of PrrA versus oxygen tensions. The binding affinity of PrrA for either or both of its target sites within the hemA upstream sequences (Fig. 1) may be affected by the phosphorylation state of the protein. To examine this possibility, we needed to first establish the conditions under which phosphorylated PrrA is present in the cell. We used a combination of phosphoprotein affinity chromatography and immunoblot analyses to examine samples prepared from wild-type strain 2.4.1 cells grown under different conditions, and we found that as the growth conditions became increasingly anaerobic, the amount of immunodetectable PrrA eluted from the affinity columns increased (Fig. 2). Thus, in agreement with measurements of the kinase activity of the cognate sensor partner protein PrrB, which is inhibited under aerobic growth conditions (29), our results indicate that the level of phosphorylated PrrA in the cell increased as oxygen tensions decreased.

View larger version (47K): [in this window] [in a new window]

FIG. 2. Immunoblots of protein samples probed with anti-PrrA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from a phosphoprotein affinity column (Eluate) prepared from cultures of R. sphaeroides wild-type strain 2.4.1 grown under dark and highly aerobic (30%), semiaerobic (2%), or anaerobic (0%) conditions. A 4.1-mg protein sample of each lysate was applied to the affinity columns. Immunoblots were prepared from the crude lysates (18.0 µg protein) or equal sample volumes of the peak fractions eluted from each column.

In vivo analysis of the phosphorylation state of PrrA in the presence of oxygen in a cytochrome cbb₃ oxidase^– mutant. Cytochrome cbb₃ oxidase^– mutant strain JZ722 was isolated following transposon mutagenesis using selection for mutants having increased transcription from hemA upstream sequences under aerobic conditions (39). The hypothesis that the presence and activity of the oxidase are linked to PrrBA activity (39) was based on studies of this oxidase mutant. The most recent model of the relationship between cytochrome cbb₃ oxidase and PrrBA proposes that an aerobic inhibitory signal directed toward the kinase activity of PrrB is generated from the oxidase (28). This inhibition of PrrB kinase activity would then be absent in cells lacking the cbb₃ oxidase. Consequently, there should be more phosphorylated PrrA under aerobic conditions in mutant strain JZ722 than in wild-type strain 2.4.1.

We used our phosphoaffinity/immunodetection technique to compare the levels of PrrA eluted from a column following application of lysates of wild-type strain 2.4.1 and mutant JZ722 grown under highly aerobic conditions. We found that while PrrA was detected in the eluate from the JZ722 lysate, no PrrA was detected in the eluate from the 2.4.1 lysate (Fig. 3). Also, the immunodetectable levels of the HemA protein correlated with the amount of phosphorylated PrrA, as the levels were higher in lysate prepared from mutant strain JZ722 than in lysate prepared from wild-type strain 2.4.1 (Fig. 3).

View larger version (33K): [in this window] [in a new window]

FIG. 3. Immunoblots of protein samples probed with anti-PrrA or anti-HemA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from phosphoaffinity columns (Eluate) prepared from cultures of R. sphaeroides wild-type strain 2.4.1 or cbb3 oxidase– mutant strain JZ722 grown under dark and highly aerobic (30% oxygen) conditions. The antisera used to probe the immunoblots are indicated at the top. A 2.1-mg protein sample of each lysate was applied to affinity columns. The immunoblot probed with anti-HemA antisera was prepared using samples of crude lysate (21 µg protein). The immunoblots probed with anti-PrrA antisera were prepared using samples of the crude lysates (2.0 µg protein) or equal sample volumes of the peak fractions eluted from each column. For further details see Materials and Methods.

Thus, when either oxygen tensions were decreased or functional cbb₃ oxidase was removed, the amount of phosphorylated PrrA apparently increased in the cell. Since PrrA is directly involved in hemA transcription (30), these results also indicate that the higher level of hemA transcription in transposon mutant strain JZ722, which was the basis for our mutant isolation process (39), was attributable to the higher intracellular concentration of phosphorylated PrrA.

In vitro analysis of the phosphorylation state of PrrA versus binding affinities for sites I and II within the hemA upstream sequences. We used EMSAs to examine the affinities of PrrA and phosphorylated PrrA for the two binding sites within the hemA upstream sequences. Previously, phosphorylation of PrrA was achieved by incubation of the purified protein with acetyl phosphate (30). However, it has been reported that the efficiency of modification of PrrA with BeF₃^–, which emulates phosphorylation of the D63 residue (22), is estimated to be 90%, compared to the approximately 20% efficiency of acetyl phosphate treatment (22). Therefore, we expected BeF₃^– modification of PrrA would increase the sensitivity of the EMSAs. The relative affinities of PrrA and BeF₃^–-PrrA for each binding site were evaluated by performing the mobility shift assays in the presence of various amounts of unlabeled competitor DNA having sequences containing both binding site I and binding site II (Fig. 4).

View larger version (20K): [in this window] [in a new window]

FIG. 4. Competition EMSA results for PrrA, target biotin-labeled DNA containing binding site I or binding site II sequences, and competitor unlabeled DNA containing both sites I and II. The DNA used in the assays was generated by PCR performed with the primers shown in Fig. 1, and either 1.33 µM PrrA or BeF3–-PrrA was used. The amounts of competitor DNA used are indicated at the bottom. For further details see Materials and Methods.

For binding site I, the unlabeled DNA competed more effectively with complexes involving BeF₃^–-PrrA than with unmodified PrrA, while the opposite was true for binding site II. Since the same concentrations of protein and competitor DNA were used in the assays involving equivalent amounts of labeled binding site I or II DNA, the results indicate that BeF₃^–-PrrA behaves differently than unmodified PrrA toward the two binding sites. They also indicate that unphosphorylated PrrA binds DNA and that it binds with higher affinity for binding site I than for binding site II, while BeF₃^–-PrrA binds with higher affinity for site II than for site I. To confirm that the method used to modify PrrA did not contribute to the outcome, we also performed this analysis using acetyl phosphate-treated PrrA. While more protein was required, since modification was less efficient, we again found that the modified protein had greater affinity for binding site II than for binding site I (results not shown).

While in vitro transcription assays involving both hemA (30) and cycA (the P2 promoter [4]) have demonstrated that unphosphorylated PrrA can bind DNA and activate transcription, other studies were unable to detect DNA binding of unmodified PrrA to the cycA sequences in the absence of RNA polymerase (22). Our comparisons of the two hemA binding sites indicated that unphosphorylated PrrA has greater affinity for certain DNA sequences than for others. Comparison of the relative dissociation constants (K_d) of BeF₃^–-PrrA and untreated PrrA for binding site I estimated from our EMSAs and the relative K_d reported for the cycA PrrA binding site (22) reinforced this conclusion. The K_d of BeF₃^–-PrrA for site I is approximately 3.30 µM, and since unmodified PrrA binds to hemA binding site I with greater affinity than BeF₃^–-PrrA, the K_d of PrrA for site I is less than 3.30 µM. In contrast, the K_d of BeF₃^–-PrrA for its target within the cycA upstream sequences has been reported to be approximately 5 µM, while the K_d of unmodified PrrA for the same sequence is approximately 1 mM (22). Thus, while the affinity of unphosphorylated PrrA for one of its target sequences, hemA binding site I, is greater than the affinity of phosphorylated PrrA, the affinity of unphosphorylated PrrA for another target sequence, the cycA binding site, is 200-fold less than the affinity of phosphorylated PrrA.

In vivo analysis of the role of phosphorylated and unphosphorylated PrrA in hemA transcription regulation. Previously, we used hemA::lacZ transcription reporter plasmids to examine the consequences of altering one or both of the PrrA binding sites within the hemA upstream sequences that are otherwise intact (i.e., the plasmids contain both P1 and P2 promoter sequences). These studies demonstrated that the two binding sites are equally important for transcription under aerobic conditions, while binding site II is more important under anaerobic conditions (30). Interpreted in the context of our in vitro results, which indicate that the amount of phosphorylated PrrA increases with decreasing oxygen tensions, the difference in importance of the two binding sites correlates with the phosphorylation state of PrrA. Our in vitro transcription assays also identified the upstream P1 promoter as the target for PrrA- and phosphorylated PrrA-mediated activation (30). Therefore, the relative importance of PrrA binding sites I and II versus oxygen availability should persist even in the absence of the downstream P2 promoter. We examined this hypothesis using a set of reporter plasmids which did not contain the P2 promoter sequences but otherwise were equivalent (Fig. 5). In the presence of oxygen, the β-galactosidase activities were reduced to similar extents when either binding site I or binding site II was altered; the activities were 0.3 and 0.2 times those obtained with the intact sequences, respectively. In contrast, in the absence of oxygen, the effects of altering the two binding sites differed; the activities when binding site I was altered versus the activities when binding site II was altered were reduced 0.8- and 0.3-fold compared to the intact sequences. Therefore, regardless of the presence of the P2 promoter, the relative importance of the two binding sites depends on the phosphorylation state of PrrA.

View larger version (22K): [in this window] [in a new window]

FIG. 5. β-Galactosidase activities in extracts of R. sphaeroides wild-type strain 2.4.1 with hemA(P1)::lacZ reporter plasmids having intact or altered PrrA binding sites that had been grown under highly aerobic (30% oxygen) or anaerobic (0% oxygen) conditions. The reporter plasmids used were pBRO75 (none), pBRO100 (I), pBRO109 (II), and pBRO61 (I & II). Alterations in the hemA sequences present on the plasmids are shown in Fig. 1, and additional information about the plasmids is shown in Table 1. The error bars indicate the standard deviations from the means. The values are the average values for duplicate assays of a minimum of three independent growth experiments. One unit of enzyme activity was defined as 1 µmol of o-nitrophenyl-β-D-galactopyranoside hydrolyzed per min. For further details concerning the growth conditions used, see Materials and Methods.

Since our phosphoafffinity/immunodetection results indicated that more phosphorylated PrrA was present even under highly aerobic conditions in cytochrome cbb₃ oxidase^– mutant strain JZ722 (Table 1), we expected that the pattern of β-galactosidase activities with the reporters having alterations within the PrrA binding sites would follow suit; i.e., in the presence of oxygen, alterations to binding site II should have been more important in mutant strain JZ722 than in wild-type strain 2.4.1. We measured the levels of β-galactosidase activity in extracts of JZ722 with reporter plasmids having alterations in site I or site II (30), and the results were as predicted; in the JZ722 background, the β-galactosidase activities were 1.3-fold higher (89 ± 1 versus 68 ± 7 U mg^–1 protein) when binding site II was altered but 2.3-fold higher when binding site I was altered (369 ± 10 versus 158 ± 5 U mg^–1 protein). Therefore, regardless of whether it is due to a reduction in oxygen tensions or to mutation, the more phosphorylated PrrA that is present in a cell, the greater the relative importance of binding site II.

To directly examine the role of unphosphorylated PrrA in vivo, we needed to construct a mutant strain having a prrA gene coding for a PrrA mutant protein that could not be phosphorylated. In constructing this mutant, we considered the following findings. While Comolli et al. (4) demonstrated that residue D63 of PrrA is the target for PrrB-mediated phosphorylation, a D63A substitution led to a loss of DNA binding activity, as well as a loss of phosphorylation. However, Hemschemeier et al. (16) showed that a D-to-K substitution at the same residue in the Rhodobacter capsulatus PrrA homolog RegA results in a loss of phosphorylation, but the mutant protein is still capable of binding to DNA in vitro. We predicted that this would also be the case for the D63K mutant PrrA protein, as RegA and PrrA share 91% similarity and 83% identity and the sequences are absolutely conserved within the helix-turn-helix motif of their DNA binding domains.

A prrA null mutant having no antibiotic resistance markers was also required. Therefore, we used the cre-lox system (25) to construct mutant strain BRO107, in which the prrA coding sequences were completely excised and a single loxP site was left behind (Table 1) (see Materials and Methods). We then generated otherwise isogenic strains coding for wild-type PrrA or D63K-PrrA by introducing the suicide vector pSUP202 carrying either the wild-type prrA or prrA-D63K mutant gene. In this way, "knock-in" strains JZ4141 (wild-type prrA) and JZ4148 (prrA-D63K) were obtained. Using our phosphoaffinity/immunodetection technique with samples of cultures grown under anaerobic-dark conditions, we confirmed that the prrA genes were present and functional (i.e., the PrrA or D63K-PrrA protein was produced and PrrA was phosphorylated, but D63K-PrrA was not phosphorylated) (Fig. 6). Note that this result also confirmed the specificity of the phosphoaffinity enrichment protocol for the phosphorylated form of PrrA.

View larger version (52K): [in this window] [in a new window]

FIG. 6. Immunoblots of protein samples probed with anti-PrrA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from TALON PMAC magnetic beads (Eluate). The crude lysates were prepared from cultures of R. sphaeroides mutant strain JZ4141 or JZ4148 (Table 1) grown under anaerobic-dark conditions with DMSO, and the eluate was obtained by processing samples of crude lysates containing 5.6 mg of total protein. The immunoblots were prepared using samples of the crude lysates (24.0 µg protein) or equal sample volumes of the eluates. For further details, see Materials and Methods.

Our hemA(P1)::lacZ transcription reporter plasmids having alterations in one or the other of the two PrrA binding sites were then mobilized into JZ4141 and JZ4148, as well as into the parent prrA null mutant strain BRO107, and β-galactosidase activities in extracts of the exconjugants grown aerobically or anaerobically were measured (Table 2). First, even in the prrA null mutant strain BRO107, the β-galactosidase activities were higher in extracts of cells grown in the absence of oxygen than in extracts of cells grown in the presence of oxygen. The indirect activity of FnrL with respect to the P1 promoter (11) could account for this anaerobic induction. Second, since activities were higher in cells having D63K-PrrA than in cells lacking PrrA, the unphosphorylated protein was capable of binding DNA and activating transcription. Third, and consistent with all our other transcription reporter plasmid data, under phosphorylating conditions (no oxygen), alterations to site II were more important for the wild-type PrrA protein than for D63K-PrrA, which could not be phosphorylated.

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

TABLE 2. β-Galactosidase activities in extracts of R. sphaeroides with hemA(P1)::lacZ reporter plasmids

Can unphosphorylated PrrA dimerize? The physical studies of Laguri et al. (22) demonstrated that the PrrA protein used by these workers was predominantly monomeric, while their BeF₃^–-PrrA was predominantly dimeric, so the presence of dimeric protein was also correlated with the ability to detect BeF₃^–-PrrA but not with PrrA binding to the target site within the cycA upstream sequences. However, our mobility shift assays indicated that unmodified PrrA had a higher affinity for binding site I of hemA than BeF₃^–-PrrA had. These results suggest either that unphosphorylated PrrA can dimerize and bind DNA or, far less likely, that monomeric PrrA binds to site I with affinities comparable to those of dimeric phosphorylated PrrA. Since the PrrA protein used in our in vitro studies differed from the protein used by Laguri et al. (22) in that the protein of Laguri et al. was a PrrA protein in which 21 amino acids, including a polyhistidine tag, were added to the N terminus, while the purified protein used here had a Pro-Gly extension at its C terminus (4), and since the importance of the cytoplasmic environment for PrrA or phosphorylated PrrA dimer formation is not known, we used a genetic approach with R. sphaeroides in order to evaluate PrrA dimerization in the cell.

For this analysis, we constructed plasmid pD63K-PrrA coding for full-length PrrA having a D63K substitution, plasmid pPrrA' containing a truncated prrA gene lacking sequences coding for the DNA binding domain of PrrA (missing amino acid residues 160 to 184, including the entire helix-turn-helix motif), and plasmid pD63K-PrrA', which carried a truncated prrA gene and coded for a D63K-substituted polypeptide that also lacked the DNA binding domain. These plasmids were mobilized into prrA null mutant BRO107 and wild-type strain 2.4.1.

The permissive conditions for growth of PrrA^– mutants include aerobic and anaerobic-dark conditions (with DMSO), whereas phototrophic conditions are nonpermissive (9). prrA null mutant strain BRO107 containing any of the plasmids could not grow under phototrophic conditions (results not shown), which indicates that PrrA in the phosphorylated form is essential for growth under these conditions. Therefore, if wild-type polypeptides dimerized with the products of the mutant alleles, the effective concentration of dimeric phosphorylated full-length PrrA would be reduced, and we expected that this might manifest itself as a diminished capacity for phototrophic growth.

We found that under dark-aerobic and anaerobic (with DMSO) conditions the growth of wild-type strain 2.4.1 cells with the plasmids carrying the mutant prrA genes was indistinguishable from the growth of cells containing the plasmid vector alone. However, the growth of cells containing plasmids with any of the mutant prrA genes was reduced under photosynthetic conditions compared to the growth of cells with the empty vector (Fig. 7), which suggests that the defective proteins are able to interact with wild-type PrrA. These results are consistent with those previously reported for a multicopy analysis of a longer prrA' allele coding for a protein truncated at amino acid residue 176 (9).

View larger version (49K): [in this window] [in a new window]

FIG. 7. Growth of R. sphaeroides wild-type strain 2.4.1 with the vector or plasmids coding for the PrrA mutant proteins indicated following incubation under the conditions indicated at the top.

It could be argued that the dominant-negative behavior of the prrA-D63K gene is due to interference of D63K-PrrA with phosphorylation of the wild-type protein. We examined this possibility using our phosphoaffinity/immunodetection protocol (Fig. 8). Notwithstanding the presence of the D63K protein, the wild-type PrrA polypeptide could be phosphorylated, and digital image analysis indicated that the ratio of the amounts of phospho-PrrA detected in the two eluates was 1.0:1.0. Since our in vivo and in vitro results indicated that unphosphorylated PrrA can bind DNA, we propose that the most parsimonious interpretation of the dominant-negative behavior of prrA-D63K is that D63K-PrrA can dimerize with wild-type PrrA.

View larger version (53K): [in this window] [in a new window]

FIG. 8. Immunoblots of protein samples probed with anti-PrrA antisera. The samples examined were either crude lysates (Lysate) or protein eluted from TALON PMAC magnetic beads (Eluate). The crude lysates were prepared from cultures of R. sphaeroides wild-type strain 2.4.1 with either the pRK415 vector or plasmid pD63K-PrrA from cultures grown under anaerobic-dark conditions with DMSO, and the eluate was obtained by processing samples of crude lysates containing 3.8 mg of total protein. The immunoblots were prepared using samples of the crude lysates (23.0 µg protein) and equal sample volumes of the eluates. For further details see Materials and Methods.

Dual regulation of hemA transcription by PrrA and FnrL. Previously, we have shown that FnrL is required for anaerobic induction of the downstream P2 promoter and that FnrL has an indirect role in transcription from the upstream P2 promoter as well (11). Results presented here and elsewhere (30) indicate that PrrA activity is directed toward the upstream P1 promoter. To further examine the role of FnrL and PrrA in transcription emanating from the P1 promoter, we determined the β-galactosidase activities in wild-type and FnrL^– cells containing our set of hemA(P1)::lacZ transcription reporter plasmids (Fig. 9). The results indicated that regardless of whether the PrrA binding sites were intact or not, P1 transcription was lower in the absence of a functional fnrL gene. Therefore, both FnrL and PrrA are required for normal regulated P1 transcription.

View larger version (21K): [in this window] [in a new window]

FIG. 9. β-Galactosidase activities in extracts of R. sphaeroides wild-type strain 2.4.1 or FnrL– mutant strain JZ1678 with hemA(P1)::lacZ reporter plasmids having intact or altered PrrA binding sites that had been grown under semiaerobic (2% oxygen) conditions. The reporter plasmids used were the plasmids described in the legend to Fig. 5. Additional information about the strains and plasmids is shown in Table 1. The error bars indicate the standard deviations from the means for duplicate assays of a minimum of three independent growth experiments. For an explanation of the units see the legend to Fig. 5. For further details concerning the growth conditions used, see Materials and Methods.

Oxygen control of hemA transcription: testing for other factors. To determine whether PrrA and FnrL fully account for the regulated response of hemA transcription to oxygen availability, we constructed double-mutant strain BRO108, which cannot produce either PrrA or FnrL (Table 1). We then evaluated β-galactosidase activity using reporter plasmids having both the P1 and P2 promoters or having P1 alone. While we found that the activity levels were somewhat higher in extracts of cells grown under highly aerobic and semiaerobic conditions when both promoters were present (55 ± 2 and 65 ± 5 U mg^–1 protein) than in extracts of cells when only P1 was present (1 and 3 U mg^–1 protein), there was no appreciable response to decreasing oxygen tensions. Thus, it appears that removing PrrA and FnrL from the cytoplasm essentially abolishes all oxygen control of hemA transcription.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
These studies were directed primarily toward understanding how PrrA acts differently toward its two binding sites within the hemA upstream sequences. We have shown previously by using in vitro transcription that unphosphorylated PrrA activates transcription, but the activation is much higher in the presence of phosphorylated PrrA (30). Therefore, we hypothesized that alteration of phosphorylation could cause the difference in PrrA behavior toward the two binding sites. To test this possibility, we first showed that the levels of phosphorylated PrrA protein in the cell increase when oxygen tensions decrease. We then examined how the phosphorylation state affects PrrA affinity for binding sites I and II. Our in vitro measurements indicated that phosphorylated PrrA has greater binding affinity for site II than for site I and that the opposite is true for unphosphorylated PrrA. The in vivo results suggest that the in vitro measurements reflect what takes place in the cell, since β-galactosidase activity assays indicated that the levels of transcription under aerobic conditions are affected equally by alterations to binding site I and alterations to binding site II, while anaerobic induction relies on intact binding site II. The finding that the D63K-PrrA mutant protein, which cannot be phosphorylated, affects transcriptional events further confirmed that unphosphorylated PrrA can bind DNA.

Based on our findings for binding sites I and II of hemA, we propose that the PrrA binding sites of R. sphaeroides can be subdivided into two types, those for which unphosphorylated PrrA has greater affinity, such as hemA binding site I, and those for which phosphorylated PrrA has greater affinity, such as hemA binding site II. The existence of two types of sites provides a plausible explanation for why analyses to detect DNA binding by unphosphorylated PrrA in vitro have not always been successful since our findings indicate that the results should depend on the DNA sequences that are used. For hemA, the presence of both kinds of targets explains how PrrA in the unphosphorylated form can assist in transcription under noninducing conditions and how phosphorylation of PrrA can increase transcription.

The broader significance of these findings is that the same two modes of regulation of hemA mediated by unphosphorylated and phosphorylated PrrA may apply to additional genes in R. sphaeroides. Recently, Eraso et al. provided an elegant description of the transcriptome and proteome of cells lacking PrrA (10). We hope that the present study will provide the impetus for further transcriptome and proteome investigations directed toward resolving the gene sets subject to each of the two modi operandi of PrrA.

The ability to detect phosphorylated PrrA using the phosphoaffinity/immunodetection technique also made it possible to evaluate the current model as to how the PrrBA two-component system senses and responds to changes in oxygen availability. A signal inhibiting PrrB kinase activity toward PrrA is thought to emanate from the rate of electron flow through cytochrome cbb₃ oxidase; high flow rates generate high levels of the inhibitory signal, like those that would occur under aerobic conditions (28). The model then predicts that if the oxidase is absent from the cell because of mutation, phosphorylated PrrA should be present regardless of the presence of oxygen. We found that, while phosphorylated PrrA could not be detected in wild-type cells grown under highly aerobic conditions, it could be detected in cells lacking the oxidase, which is consistent with the model. However, it has not escaped our notice that as the level of phosphoprotein increases, either because of decreasing oxygen tensions or through the absence of the oxidase, apparently the total amount of PrrA also increases. To what degree the increase in PrrB substrate availability contributes to the increase in phosphorylated PrrA present in the cell is not known yet. Irrespective of how this comes about, the increase in the phosphorylated PrrA level precisely explains why a mutant strain lacking functional cytochrome cbb₃ oxidase would be identified in a selection for transposon mutants having higher levels of hemA transcription in the presence of oxygen (39). Furthermore, consistent with our other measurements, the presence of higher levels of phosphorylated PrrA in the cytochrome cbb₃ oxidase^– mutant was found to be more significant for binding site II than for binding site I of hemA.

These studies also addressed the question of whether unphosphorylated PrrA protein can dimerize in the cell. Our demonstration that unphosphorylated protein can bind to DNA with a K_d close to that of the phosphorylated protein for site I of hemA supports the hypothesis that the protein binds to DNA in dimeric form. Further, multicopy analysis of a prrA gene coding for the D63K-PrrA protein revealed that it behaves in a dominant-negative fashion in wild-type cells. Laguri et al. (22) demonstrated that the monomer-dimer equilibrium is shifted toward the dimeric form for the phosphoprotein, while the amount of dimeric unphosphorylated PrrA was below the level that could be detected using their methods. We have not examined the relative concentrations of the dimer and the monomer. Importantly, however, since we have shown that the level of hemA P1 transcription in vivo is higher in the presence of prrA-D63K than in the absence of prrA, our findings demonstrate that the amount of dimeric unphosphorylated PrrA present in the cell is physiologically relevant.

We now consider the role of FnrL in P1 transcription, whose importance was made starkly apparent by altering the PrrA binding sites such that the PrrA contribution was diminished or eliminated. In fact, the results probably underestimate the full scope of the contribution of FnrL in anaerobic induction of P1, as we were limited to culturing the cells under semiaerobic conditions, which are permissive for the growth of the FnrL^– mutant bacteria (38). We believe that the contribution of FnrL to P1 transcription is indirect, and we have therefore postulated that there is a factor that directly modulates P1 transcription. Our results indicate that PrrA and FnrL are the only transcription factors involved in oxygen responsiveness of hemA and so suggest that PrrA may be this factor. A central question is how this might come about. It may be that FnrL regulates prrA transcription, that FnrL affects the degree of PrrA phosphorylation, or that FnrL influences the PrrA or phosphorylated PrrA monomer-dimer equilibrium. It is also conceivable that FnrL interacts with PrrA and thereby alters its ability to bind to DNA. Certain findings suggest that FnrL probably does not regulate the total amount of PrrA in the cell; the prrA upstream sequences do not contain an FNR consensuslike sequence, and the transcriptome profiles of wild-type versus FnrL^– mutant bacteria grown under semiaerobic conditions indicate that the levels of prrA transcripts are the same (unpublished results). Further work is required to confirm or eliminate other possibilities.

While these studies were limited to investigations of hemA regulation by both PrrA and FnrL, it is known that the overlap between the PrrA and FnrL regulons encompasses many genes in addition to hemA. Therefore, insights into how the combined actions of these DNA binding proteins achieve the correct transcriptional response for hemA should contribute to our understanding of the regulation of these other genes as well.

 
We thank Y. Fedotova and B. Gliga for their technical assistance and T. Donohue and colleagues for generously providing the PrrA intein/chitin-binding domain fusion protein expression system. We also express our gratitude for the outstanding technical support for the phosphoenrichment kits from Clontech.

This work was supported by MCB award 0320550 from the National Science Foundation.

 
* Corresponding author. Mailing address: Department of Biological Sciences, 217 Life Sciences Building, Bowling Green State University, Bowling Green, OH 43403. Phone: (419) 372-2872. Fax: (419) 372-2024. E-mail: jzeilst{at}bgnet.bgsu.edu

Published ahead of print on 8 August 2008.

Present address: Department of Biological Sciences, Lake Superior State University, 615 W. Easterday Avenue, Sault Ste. Marie, MI 49783.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A. Smith, and K. Struhl. 1989. Current protocols in molecular biology. John Wiley and Sons, New York, NY.
2
2. Barber, R. D., and T. J. Donohue. 1998. Pathways for transcriptional activation of a glutathione-dependent formaldehyde dehydrogenase gene. J. Mol. Biol. 280:775-784.[CrossRef][Medline]
3
3. Cauthen, S. E., J. R. Pattison, and J. Lascelles. 1967. Vitamin B₁₂ in photosynthetic bacteria and methionine synthesis in Rhodopseudomonas spheroides. Biochem. J. 102:774-781.[Medline]
4
4. Comolli, J. C., A. J. Carl, C. Hall, and T. Donohue. 2002. Transcriptional activation of the Rhodobacter sphaeroides cytochrome c₂ gene P2 promoter by the response regulator PrrA. J. Bacteriol. 184:390-399.[Abstract/Free Full Text]
5
5. Davis, J., T. J. Donohue, and S. Kaplan. 1988. Construction, characterization, and complementation of a puf mutant of Rhodobacter sphaeroides. J. Bacteriol. 170:320-329.[Abstract/Free Full Text]
6
6. Ditta, G., S. Stanfield, D. Corbin, and D. R. Helinski. 1980. Broad host range DNA cloning system for gram-negative bacteria; construction of a gene bank of Rhizobium meliloti. Proc. Natl. Acad. Sci. USA 77:7347-7351.[Abstract/Free Full Text]
7
7. Donohue, T. J., A. G. McEwan, and S. Kaplan. 1986. Cloning, DNA sequence, and expression of the Rhodobacter sphaeroides cytochrome c₂ gene. J. Bacteriol. 168:962-972.[Abstract/Free Full Text]
8
8. Eraso, J. M., and S. Kaplan. 1995. Oxygen-insensitive synthesis of the photosynthetic membranes of Rhodobacter sphaeroides: a mutant histidine kinase. J. Bacteriol. 177:2695-2706.[Abstract/Free Full Text]
9
9. Eraso, J. M., and S. Kaplan. 1994. prrA, a putative response regulator involved in oxygen regulation of photosynthesis gene expression in Rhodobacter sphaeroides. J. Bacteriol. 176:32-43.[Abstract/Free Full Text]
10
10. Eraso, J. M., J. H. Roh, X. Zeng, S. J. Callister, M. S. Lipton, and S. Kaplan. 2008. Role of the global transcriptional regulator PrrA in Rhodobacter sphaeroides 2.4.1: combined transcriptome and proteome analysis. J. Bacteriol. 190:4831-4848.[Abstract/Free Full Text]
11
11. Fales, L., L. Nogaj, and J. Zeilstra-Ryalls. 2002. Analysis of the upstream sequences of Rhodobacter sphaeroides 2.4.1 hemA gene: in vivo evidence for the presence of two promoters that are both regulated by fnrL. Photosynth. Res. 74:143-151.[CrossRef][Medline]
12
12. Figurski, D. H., and D. R. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA 76:1648-1652.[Abstract/Free Full Text]
13
13. Gibson, J. L., J. M. Dubbs, and F. R. Tabita. 2002. Differential expression of the CO₂ fixation operons of Rhodobacter sphaeroides by the Prr/Reg two-component system during chemoautotrophic growth. J. Bacteriol. 184:6654-6664.[Abstract/Free Full Text]
14
14. Gough, S. P., B. O. Petersen, and J. O. Duus. 2000. Anaerobic chlorophyll isocyclic ring formation in Rhodobacter capsulatus requires a cobalamin cofactor. Proc. Natl. Acad. Sci. USA 97:6908-6913.[Abstract/Free Full Text]
15
15. Harlow, E., and D. Lane. 1999. Using antibodies: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
16
16. Hemschemeier, S. K., U. Ebel, A. Jager, A. Balzer, M. Kirndorfer, and G. Klug. 2000. In vivo and in vitro analysis of RegA response regulator mutants of Rhodobacter capsulatus. J. Mol. Microbiol. Biotechnol. 2:291-300.[Medline]
17
17. Jessee, J. 1986. New subcloning efficiency competent cells: >1 x 10⁶ transformants/µg. Focus 8:9.
18
18. Joshi, H. M., and F. R. Tabita. 1996. A global two component signal transduction system that integrates the control of photosynthesis, carbon dioxide assimilation, and nitrogen fixation. Proc. Natl. Acad. Sci. USA 93:14515-14520.[Abstract/Free Full Text]
19
19. Keen, N. T., S. Tamaki, D. Kobayashi, and D. Trollinger. 1988. Improved broad-host-range plasmids for DNA cloning in gram-negative bacteria. Gene 70:191-197.[CrossRef][Medline]
20
20. Kiley, P. J., and H. Beinert. 2003. The role of Fe-S proteins in sensing and regulation in bacteria. Curr. Opin. Microbiol. 6:181-185.[CrossRef][Medline]
21
21. Laguri, C., M. K. Phillips-Jones, and M. P. Williamson. 2003. Solution structure and DNA binding of the effector domain from the global regulator PrrA (RegA) from Rhodobacter sphaeroides: insights into DNA binding specificity. Nucleic Acids Res. 31:6778-6787.[Abstract/Free Full Text]
22
22. Laguri, C., R. A. Stenzel, T. J. Donohue, M. K. Phillips-Jones, and M. P. Williamson. 2006. Activation of the global gene regulator PrrA (RegA) from Rhodobacter sphaeroides. Biochemistry 45:7872-7881.[CrossRef][Medline]
23
23. Lascelles, J. 1964. Tetrapyrrole biosynthesis and its regulation. W. A. Benjamin, Inc., New York, NY.
24
24. Lee, J. K., and S. Kaplan. 1995. Transcriptional regulation of puc operon expression in Rhodobacter sphaeroides: analysis of the cis-acting downstream regulatory sequence. J. Biol. Chem. 270:20453-20458.[Abstract/Free Full Text]
25
25. Marx, C. J., and M. E. Lidstrom. 2002. Broad-host-range cre-lox system for antibiotic marker recycling in Gram-negative bacteria. BioTechniques 33:1062-1067.[Medline]
26
26. Neidle, E. L., and S. Kaplan. 1993. Expression of the Rhodobacter sphaeroides hemA and hemT genes, encoding two 5-aminolevulinic acid synthase isozymes. J. Bacteriol. 175:2292-2303.[Abstract/Free Full Text]
27
27. Oh, J.-I., J. M. Eraso, and S. Kaplan. 2000. Interacting regulatory circuits involved in orderly control of photosynthesis gene expression in Rhodobacter sphaeroides 2.4.1. J. Bacteriol. 182:3081-3087.[Abstract/Free Full Text]
28
28. Oh, J.-I., and S. Kaplan. 2000. Redox signaling: globalization of gene expression. EMBO J. 19:4237-4247.[CrossRef][Medline]
29
29. Oh, J.-I., I.-J. Ko, and S. Kaplan. 2001. The default state of the membrane-localized histidine kinase PrrB of Rhodobacter sphaeroides 2.4.1 is in the kinase-positive mode. J. Bacteriol. 183:6807-6814.[Abstract/Free Full Text]
30
30. Ranson-Olson, B., D. F. Jones, T. J. Donohue, and J. H. Zeilstra-Ryalls. 2006. In vitro and in vivo analysis of the role of PrrA in Rhodobacter sphaeroides 2.4.1 hemA gene expression. J. Bacteriol. 188:3208-3218.[Abstract/Free Full Text]
31
31. Romagnoli, S., H. L. Packer, and J. P. Armitage. 2002. Tactic responses to oxygen in the phototrophic bacterium Rhodobacter sphaeroides WS8N. J. Bacteriol. 184:5590-5598.[Abstract/Free Full Text]
32
32. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
33
33. Simon, R., U. Priefer, and A. Puhler. 1983. A broad host range mobilization system for in vivo genetic engineering: transposon mutagenesis in Gram-negative bacteria. Bio/Technology 1:37-45.
34
34. Sistrom, W. R. 1960. A requirement for sodium in the growth of Rhodopseudomonas spheroides. J. Gen. Microbiol. 22:778-785.[Abstract/Free Full Text]
35
35. Spiro, S., and J. R. Guest. 1990. FNR and its role in oxygen-regulated gene expression in Escherichia coli. FEMS Microbiol. Rev. 75:399-428.[CrossRef]
36
36. Tai, T.-N., M. D. Moore, and S. Kaplan. 1988. Cloning and characterization of the 5-aminolevulinate synthase gene(s) from Rhodobacter sphaeroides. Gene 70:139-151.[CrossRef][Medline]
37
37. Yan, D., H. S. Cho, C. A. Hastings, M. M. Igo, S.-Y. Lee, J. G. Pelton, V. Stewart, D. E. Wemmer, and S. Kustu. 1999. Beryllofluoride mimics phosphorylation of NtrC and other bacterial response regulators. Proc. Natl. Acad. Sci. USA 96:14789-14794.[Abstract/Free Full Text]
38
38. Zeilstra-Ryalls, J. H., and S. Kaplan. 1995. Aerobic and anaerobic regulation in Rhodobacter sphaeroides 2.4.1: the role of the fnrL gene. J. Bacteriol. 177:6422-6431.[Abstract/Free Full Text]
39
39. Zeilstra-Ryalls, J. H., and S. Kaplan. 1996. Control of hemA expression in Rhodobacter sphaeroides 2.4.1: regulation through alterations in the cellular redox state. J. Bacteriol. 178:985-993.[Abstract/Free Full Text]
40
40. Zeilstra-Ryalls, J. H., and S. Kaplan. 1995. Regulation of 5-aminolevulinic acid synthesis in Rhodobacter sphaeroides 2.4.1: the genetic basis of mutant H-5 auxotrophy. J. Bacteriol. 177:2760-2768.[Abstract/Free Full Text]
41
41. Zeilstra-Ryalls, J. H., and K. L. Schornberg. 2006. Analysis of hemF gene function and expression in Rhodobacter sphaeroides 2.4.1. J. Bacteriol. 188:801-804.[Abstract/Free Full Text]

---

Journal of Bacteriology, October 2008, p. 6769-6778, Vol. 190, No. 20
0021-9193/08/$08.00+0     doi:10.1128/JB.00828-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Ranson-Olson, B. Articles by Zeilstra-Ryalls, J. H. Search for Related Content PubMed PubMed Citation Articles by Ranson-Olson, B. Articles by Zeilstra-Ryalls, J. H.