Transcriptional Control of the rhuIR-bhuRSTUV Heme Acquisition Locus in Bordetella avium
http://www.100md.com
感染与免疫杂志 2005年第3期
Witebsky Center for Microbial Pathogenesis and Immunology
Department of Microbiology and Immunology
Department of Oral Biology, University at Buffalo, State University of New York, Buffalo, New York
ABSTRACT
Iron (Fe) is an essential nutrient for most bacterial pathogens. In these organisms, a variety of regulatory systems that respond to specific Fe complexes found within their vertebrate hosts have evolved. In Bordetella avium, the heme utilization locus encoded by rhuIR-bhuRSTUV mediates efficient acquisition of Fe from heme and hemoproteins. Control of bhuRSTUV expression is promulgated at two levels. When Fe is abundant, expression is repressed in a Fur-dependent manner which is partially relieved when Fe is limiting. In the presence of heme or hemoproteins, expression of the bhuRSTUV operon is induced via a three-component signal transduction cascade composed of RhuI, RhuR, and BhuR. Herein, we report the identification of two promoters (PrhuI and PbhuR) that control expression of the rhuIR-bhuRSTUV cluster. Primer extension analysis identified the transcriptional start site of PrhuI within a putative Fur box. Transcriptional initiation of PbhuR mapped within the rhuR-bhuR intergenic region. Maximal transcription from PbhuR required Fe-limiting conditions, the presence of heme (or hemoglobin), and rhuI; however, analysis of transcripts produced from the rhuIR-bhuRSTUV locus revealed a pattern of low-level bhuR transcription in the absence of heme which originated from both PbhuR and PrhuI. Transcription from PrhuI was repressed by Fe in the presence of fur and somewhat enhanced by the addition of hemin to Fe-limited media. The nature of this hemin-associated PrhuI stimulation was rhuI independent and therefore not induced by heme via the BhuR-RhuR-RhuI signal cascade. Fe also repressed transcription from PbhuR in a fur-dependent manner; however, activation from this promoter, in the presence or absence of heme, did not occur without rhuI.
INTRODUCTION
Iron (Fe), an essential element for most pathogenic bacteria, must be acquired from the infected host. Within that host, however, Fe is usually sequestered within high-affinity ferromolecules such as transferrin, lactoferrin, and ferritin. Sequestration by these molecules maintains the free-Fe concentration in the tissue and fluids of vertebrates below 10–18 M (30), a level at which the growth of most bacteria is inhibited. To survive in these extremely Fe-limited host environments, bacteria express high-affinity Fe acquisition systems (reviewed in reference 19). In most cases, these acquisition systems are modulated in response to the availability of Fe in the local environment. Global Fe-dependent regulation is achieved primarily by Fur, a repressor protein whose regulation is dependent upon intracellular Fe concentration. Fur, which is expressed by a wide range of both gram-positive and gram-negative bacteria, has the capacity to reversibly bind Fe. In its repressor mode, Fe-bound Fur has affinity for the Fur box, a consensus motif that is located within or near Fur-regulated promoters (7). Binding of Fe-Fur to the Fur box is believed to interfere with binding of RNA polymerase (RNAP) to the promoter. Alternatively, binding of Fur might inhibit promoter escape of RNAP once the polymerase is bound to the promoter (10). When intracellular Fe levels decline to critical levels, the intracellular pool of Fur is no longer saturated with Fe. Since apo-Fur has little affinity for Fur boxes, Fur-dependent promoters are subsequently released from repression (10).
Successful pathogens have evolved a variety of mechanisms to acquire one or more of the biological Fe sources which are expressed by vertebrates (e.g., transferrin, ferritin, heme, hemoglobin, etc.) (3). Control of these systems solely by Fur would likely be a wasteful expenditure of resources, as it is unlikely that invading or colonizing bacteria will simultaneously encounter all of these sources of Fe in a single microhabitat. Thus, bacteria evolved additional regulatory elements that coordinately interact with Fur to rapidly and specifically respond to the presence of specific Fe-containing molecules. Various Fur-augmented regulatory systems have been described previously: siderophore-responsive transcription is dependent on AraC-type regulators in both Pseudomonas aeruginosa (14) and Bordetella spp. (2), LysR-type regulators have been implicated in the heme-responsive gene expression of Vibrio vulnificus (20) and Bradyrhizobium japonicum (12), and classical two-component sensor-kinase pairs have been described for Corynebacterium diphtheriae (25) and Pseudomonas aeruginosa (6). Sigma () factors, particularly those denoted as extracytoplasmic function (ECF) factors, also participate in Fur-dependent gene regulation. In Pseudomonas putida (18), Escherichia coli (29), Bordetella pertussis (27), and Bordetella avium (17), specific Fe acquisition systems are maximally expressed only during Fe-limited growth in the presence of the cognate stimulatory ligand which, by binding to an extracellular receptor, stimulates the activity of the ECF factor for an associated ECF-dependent promoter.
ECF factor-based control of gene expression has been most extensively described in E. coli. In that bacterium, uptake of extracellular ferric dicitrate is governed by fecIRABCDE which is transcribed only when the bacterium is simultaneously stressed for Fe and ferric dicitrate is locally available. Responsiveness of E. coli to ferric dicitrate is mediated by the interaction of a three-component ECF regulatory cascade that includes FecA, the outer membrane receptor for ferric dicitrate (13); FecR, an inner membrane regulatory protein (23); and FecI, an ECF factor (1). Mechanistically, when FecA binds ferric dicitrate, a signal is transduced across the periplasm to FecR which transmits the signal to FecI (9). Acquisition of the signal by FecI elicits a change in its activation state which stimulates association of FecI to core RNAP. Recruitment of RNAP holoenzyme to PfecA initiates direct transcription of the fecABCDE operon (1). Examination of the transcripts produced from the fecIRABCDE locus revealed two major mRNAs: a 1.5-kb mRNA corresponding to fecIR and a 2.5-kb mRNA corresponding to fecA. It has been proposed that the 2.5-kb fecA-bearing transcript is likely a degradation product of a 6-kb polycistronic transcript that contains sequences derived from the fecABCDE genes (8). In the absence of ferric dicitrate, basal expression of fecABCDE in Fe-stressed cells is regulated by Fur (29).
Nevertheless, an interesting conundrum was evident in the regulation of FecA. FecA is a requisite component of the FecIR-regulatory cascade and thus is required as a receptor of ferric dicitrate for initiating the signal cascade. In fact, low levels of FecA were detected in the outer membrane of Fe-stressed E. coli in the absence of ferric dicitrate. How, then, is FecA expressed in the absence of the stimulatory ligand The answer to this question is yet to be revealed for E. coli. A mechanism to explain this conundrum, however, was recently described for B. pertussis (28). In B. pertussis, two promoters, PhurI and PbhuR, control expression of the heme utilization gene cluster hurIR-bhuRSTUV. Transcription from PhurI, located upstream from hurI, is Fe regulated and heme independent. Transcription from PbhuR, a promoter located within the hurR-bhuR intergenic region, is also Fe regulated and is induced by heme via HurI, the cognate ECF factor (27). Under Fe-stressed conditions in the absence of heme, expression of bhuR arises via transcriptional initiation at PhurI in B. pertussis. Essentially, transcription does not terminate at the 3' end of hurIR but proceeds infrequently into bhuR (28). This rare "readthrough" transcription is believed to be sufficient to express amounts of BhuR in the outer membrane for initial triggering of heme-dependent regulation of PbhuR when heme becomes available.
B. avium, a gram-negative pathogen that infects domestic and wild fowl, elicits bordetellosis (or coryza), an upper respiratory illness defined by a loss of ciliated tracheal cells and a persistent, aggravated cough. The clinical presentation of coryza is highly similar to that of human pertussis (whooping cough) caused by taxonomically related B. pertussis. For that reason, B. avium has been used as a model for human pertussis. B. avium encodes the heme utilization locus rhuIR-bhuRSTUV (22), a heme uptake system with homology to hurIR-bhuRSTUV of B. pertussis. Expression levels of rhuIR-bhuRSTUV and hurIR-bhuRSTUV are regulated in similar manners in response to heme; i.e., in B. avium, expression of bhuR from PbhuR is Fe regulated and heme responsive in an rhuIR-dependent manner (17). In this paper, the transcriptional architecture of the heme utilization locus in B. avium is described to compare and contrast the regulation of the heme uptake systems of B. avium and B. pertussis. The results confirm that unlike the regulation reported for B. pertussis (28), heme-independent expression of BhuR in B. avium is controlled by transcription which initiates at both the PrhuI and PbhuR promoters.
MATERIALS AND METHODS
Media, strains, and growth conditions. Bacterial strains and plasmids used in this study are listed in Table 1. Strains of B. avium were maintained on brain heart infusion (BHI) agar, in BHI broth (Difco Laboratories, Detroit, Mich.), or in Chelex-treated complete defined medium (CDM) (5). Strains of E. coli were cultured on Luria-Bertani agar. For Fe-replete growth conditions, BHI or CDM was supplemented with 36 μM FeSO4. Conditions of Fe limitation were achieved in CDM by reducing Fe by treatment with technical-grade Chelex-20 resin (Bio-Rad, Hercules, Calif.). Fe-limited conditions were achieved in BHI by supplementing the agar or broth with 100 μM ethylene-di-o-hydroxyphenylacetic acid (EDDHA). Unless otherwise noted, antibiotics used were as follows: ampicillin (200 μg/ml), rifampin (10 μg/ml), streptomycin (200 μg/ml), spectinomycin (10 μg/ml), tetracycline (10 μg/ml), kanamycin (50 μg/ml), and gentamicin (10 μg/ml). Antibiotics were obtained from Sigma Biochemicals (St. Louis, Mo.) and Amresco (Solon, Ohio). Biochemical reagents were purchased from Life Technologies, Inc. (Frederick, Md.) and Sigma Biochemicals. Restriction enzymes and DNA-modifying enzymes were obtained from MBI Fermentas, Inc. (Hanover, Md.). Deionized water with an electrical resistance of >18 M was used to prepare all solutions.
Isolation of B. avium RNA. Total RNA was prepared from cultured cells by using a method developed by Chauhan and O'Brian (4). Cells grown to an optical density at 600 nm (OD600) of 0.4 to 0.6 were harvested by centrifugation at 4°C. Cell pellets from 12.5 ml of culture were resuspended in 600 μl of lysis buffer (10 mM NaCl, 10 mM Tris [pH 8], 5% sodium dodecyl sulfate, 200 μg of proteinase K per ml) and incubated at 37°C for 5 min. Three hundred microliters of 5 M NaCl was added to the lysed cells. Mixtures were incubated on ice for 10 min, after which the debris was removed by centrifugation at 4°C. RNA in the supernatant was precipitated by centrifugation after the addition of 3 volumes of ethanol and incubation at –80°C for 1 h. RNA was resuspended in 1 mM EDTA, extracted with acid phenol-chloroform, and reprecipitated with ethanol. Purified RNA was treated with DNase I in the presence of human placental RNase inhibitor to eliminate contaminating DNA.
Primer extension analysis. Primers APbhuR1 (5'-CCGCTGCGTGAGAACAGACGAAAAG-3') and APrhuI1 (5'-GCTTTTTGCGCAGCCAGCCATTCAG-3') were end labeled in a reaction mixture containing 10 pmol of primer, 1 mM spermidine, 4 μl of [-32P]ATP (3,000 Ci/mmol, 10 μCi/ml), 10 U of T4 polynucleotide kinase, and 1x labeling buffer (MBI Fermentas). Reaction mixtures were incubated at 37°C for 30 min, and reactions were terminated by heating to 95°C for 2 min. Labeled primers were purified by using a Sephadex G-25 column according to the manufacturer's instructions (Roche, Indianapolis, Ind.). Samples containing 20 μg of total RNA were ethanol precipitated, and pellets were resuspended in mixtures containing 0.2 pmol of labeled primer and 1x Superscript II reverse transcriptase reaction buffer (Invitrogen, Carlsbad, Calif.). RNA was denatured by heating to 95°C for 1 min, and labeled primers were annealed to the RNA by cooling samples to 70°C for 5 min and then to 30°C over a 10-min time span. Reaction mixtures were supplemented with 100 mM dithiothreitol, 10 mM deoxynucleoside triphosphates (dNTPs), Superscript II reverse transcriptase, and 1x Superscript II reverse transcriptase reaction buffer (Invitrogen). Reaction mixtures were incubated at 45°C for 30 min, terminated by the addition of EDTA to 100 μM, extracted with phenol-chloroform, precipitated with ethanol, and resuspended in Tris-EDTA buffer. Extended primers were resolved on 8 M urea-6% sequencing-grade polyacrylamide. For sizing purposes, a ladder was prepared by sequencing an appropriate template using the same primer.
Sequencing ladder for primer extension analysis. A total of 3 μg of pAD3 was denatured with 0.2 N NaOH, neutralized with 2 M NH4Ac, and ethanol precipitated. Pelleted DNA was resuspended with 0.5 pmol of primer (APbhuR1 or APrhuI1) in 1x Sequenase buffer (USB Corp., Cleveland, Ohio). Primer and denatured template were annealed by heating to 65°C for 2 min and cooled slowly to room temperature. Labeling and termination reactions were performed according to the manufacturer's instructions (USB Corp.).
Construction of 4169rifrhuI. The rhuI open reading frame (ORF) was deleted from the chromosome of wild-type (wt) B. avium 4169rif using allelic exchange (16). Splicing of overlapping ends (SOEing) by PCR was used to construct the allelic exchange construct pAEK22.1. The 507 bp immediately upstream from the rhuI GTG start codon were PCR amplified by using primers rhuI-a2 (5'-GCTCTAGATAGATGTGATGCGGTTCTT-3' [the XbaI site is underlined]) and rhuI-b2 (5'-ACGCCGGCTGCGCTCATTTTTTGCAAGAAATATAA-3' [the underlined region is complementary to the 5' end of rhuR]) (PCR conditions were as follows: the PCR mixture contained 1x ThermoPol buffer, 10% dimethyl sulfoxide [DMSO], 250 μM dNTPs, 400 nM each oligonucleotide primer, and 1 U of Vent polymerase [Invitrogen], and PCR was carried out at 94°C for 1 min, 45°C for 1 min, and 72°C for 1 min for 30 cycles). Primers rhuI-c (5'-ATGAGCGCAGCCGGCGT-3') and rhuI-d (5'-AGGGAGCTCTCTATTTAGTAACAGGAATCATTTA-3' [the SacI site is underlined]) were used to amplify the 944 bp located immediately downstream from the rhuI ORF beginning at the ATG start codon of rhuR (PCR conditions were as follows: the PCR mixture contained 1x PCRx buffer [Promega, Madison, Wis.], 10% DMSO, 150 μM dNTPs, 1 μM each oligonucleotide primer, and 1 to 10 U of Taq polymerase, and PCR was carried out at 94°C for 1 min, 45°C for 1 min, and 72°C for 1 min for 25 cycles). PCR products, purified using a GFX PCR DNA and gel band purification kit (Amersham Pharmacia Biotech, Inc., Piscataway, N.J.), were used as templates for a modified PCR SOEing reaction in which 1 μl of each PCR product was added to an annealing reaction mixture containing 1x ThermoPol buffer (Invitrogen), 10% DMSO, and 250 μM dNTPs. This mixture was heated to 95°C for 5 min and cooled to 25°C over a period of 3 min to allow the ends of each PCR product to anneal, thus generating partially double-stranded templates. The double-stranded templates were repaired by using 1 U of Klenow fragment (MBI Fermentas, Inc.). After 10 min at 25°C, Klenow fragment was heat inactivated by incubation at 95°C for 5 min. Primers (rhuI-a2 and rhuI-d, at a concentration of 1 μM each) and 1 U of Vent polymerase (Invitrogen) were added to the mixture to initiate PCR amplification (94°C for 1 min, 45°C for 1 min, 72°C for 3 min, 25 cycles). The PCR SOEing product was ligated into pBluescript KS(+) at the XbaI and SacI sites to engineer pAEK22. After the sequence of pAEK22 was confirmed by automated DNA sequencing, the insert was released from the plasmid by digestion with XbaI and SacI, gel purified, and ligated into pCVD442tet, which had been digested with the same enzymes. The resulting plasmid, pAEK22.1, was moved into B. avium 4169rif by conjugation (22). Transconjugants were selected for growth on BHI agar containing tetracycline. Cells that had undergone a second recombination event to remove the plasmid sequences were selected for growth on BHI agar containing 20% sucrose. Deletion of rhuI in 4169rifrhuI was confirmed by colony PCR and Southern hybridization.
Construction of 4169rif, Pho20, and Pho20fur. An omega () fragment containing bidirectional transcriptional and translational terminators was inserted directly downstream from the rhuR ORF in the chromosome of B. avium strains 4169rif, Pho20, and Pho20fur using allelic exchange. In brief, the plasmid pAEK37.3, used for allelic exchange, was constructed using the following cloning scheme. A 993-bp fragment containing the rhuR-bhuR intergenic region and a portion of the bhuR ORF was amplified from pERM1 by PCR using primers rhuR (5'-GGAATTCGATTCCTGTTACTAAATAGATTCGTAAAAAC-3' [the EcoRI site is underlined]) and bhuRD (5'-TCCGAGCTCCGCCTGAATCAACCATGTCGTGTTGT-3' [the SacI site is underlined]) (PCR conditions were as follows: the PCR mixture contained 1x Fermentas buffer without Mg2+, 1.5 mM MgCl2, 10% DMSO, 250 μM dNTPs, 400 nM each oligonucleotide primer, and 1U of Taq polymerase [MBI Fermentas], and PCR was carried out at 95°C for 1 min, 55°C for 1 min, and 72°C for 2 min for 30 cycles). This fragment was cloned into the EcoRI- and SacI-digested plasmid pAEK26.1, which contains the rhuR ORF immediately upstream from an EcoRI site, to generate plasmid pAEK37.2. The fragment from pHP45 was cloned into the EcoRI site of pAEK37.2 to engineer pAEK37.2. This plasmid was digested with XbaI and SacI to isolate a fragment containing rhuR--bhuR which was inserted into the same sites of plasmid pCVD442tet to produce pAEK37.3. The construction of pAEK37.3 was confirmed by automated DNA sequencing. The plasmid was transformed into SM10pir and subsequently conjugated into B. avium. Transconjugants were selected for growth on BHI agar containing tetracycline. Cells that had undergone the second recombination event were selected for growth on BHI agar containing 20% sucrose. The mutant strains 4169rif, Pho20, and Pho20fur were isolated, and their mutant genotypes were confirmed by Southern hybridization.
-Galactosidase assay. Expression of the lacZYA reporter gene in pDJM41 was determined by measuring -galactosidase activity (17). Briefly, cultures of Fe-limited cells grown overnight were used to inoculate secondary cultures. The secondary cultures were incubated at 37°C for 16 to 20 h. Bacteria in 1.0 ml of culture were pelleted by centrifugation for 5 min at 5,000 x g. Cells were resuspended in Z buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4, 38 mM -mercaptoethanol), and the OD600 of the cell suspension was adjusted to 0.28 to 0.70. After diluting 400 μl of the suspension with 400 μl of Z buffer, cells were permeabilized by the addition of 45 μl of 0.1% sodium dodecyl sulfate and 90 μl of chloroform. Cell suspensions were vortexed for 10 s, followed by incubation at 30°C for 15 min. Enzymatic reactions were initiated by the addition of 160 μl of a solution containing 4 mg of o-nitrophenyl--D-galactopyranoside per ml. Following incubation at 30°C, samples were observed for the development of a yellow color, at which point 400 μl of 1 M Na2CO3 was added to terminate the reaction. After a brief centrifugation to pellet debris and chloroform, the OD420 and OD550 of each reaction were recorded, and the relative -galactosidase activity was calculated by use of the following formula (21): {1,000[OD420 – 1.75(OD550)]}/(t)(0.4)(OD600), where t is the time of the reaction in minutes. Relative enzyme activities are reported as the means of triplicate assays.
Alkaline phosphatase assay. Expression of the bhuR::phoA fusion of B. avium mutants Pho20, Pho20, Pho20fur, and Pho20fur was determined by using a modified p-nitrophenyl phosphate assay (11). Cultures of Fe-limited cells grown overnight were used to inoculate secondary cultures in BHI broth under either Fe-replete (36 μM FeSO4) or Fe-limited (100 μM EDDHA) conditions. At stationary phase, 1 ml of the culture was centrifuged, and the cell pellet was resuspended in a solution containing 10 mM Tris (pH 8) and 100 mM NaCl. Cell densities of the resuspensions were adjusted to an OD600 of 0.4 to 0.8. An aliquot containing 500 μl of the cell suspension was added to an equal volume of 1 M Tris (pH 8), and the reaction was initiated by the addition of 100 μl of 4% p-nitrophenyl phosphate. Following incubation at room temperature, samples were observed for development of a yellow color, at which point 100 μl of 1 M KH2PO4 was added to terminate the reaction. After a brief centrifugation to pellet debris, the OD420 and OD550 of each reaction were recorded, and the relative alkaline phosphatase activity was calculated by use of the following formula (21): {1,000[OD420 – 1.75(OD550)]}/(t)(0.5)(OD600), where t is the time of the reaction in minutes. Relative enzyme activities are reported as the means of triplicate assays.
RNase protection assay (RPA). Construction of the DNA templates for the bhuR and bhuRS probes (pERM29 and pERM30, respectively) was described previously (22). PCR was used to construct the templates for the rhuI and rhuIR probes. Primers rhuIRPA1 (5'-CCGCTCGAGTCCATCGCTTCTATGCT-3' [the XhoI site is underlined]) and rhuIRPA2 (5'-CCCAAGCTTACGGTAGTGATTGAGCA-3' [the HindIII site is underlined]) amplified a 215-bp fragment within the rhuI ORF. A 254-bp fragment which contains the last 171 bp of rhuI and the first 86 bp of the rhuR 5' end was amplified by primers IRRPA1 (5'-CCGCTCGAGTGTCTGCTCTGCCCC-3' [the XhoI site is underlined]) and IRRPA2 (5'-CCCAAGCTTCTGCCCCGAAGCAAGA-3' [the HindIII site is underlined]). For both primer sets, the PCR conditions were as follows: the PCR mixture contained 1x PCRx buffer (Promega), 10% DMSO, 100 μM dNTPs, 1 μM each oligonucleotide primer, and 1 to 10 U of Taq polymerase, and PCR was carried out at 95°C for 45 s, 45°C for 45 s, and 72°C for 60 s for 30 cycles. The resulting amplicons were ligated into pBluescript KS(–) at the XhoI and HindIII sites. The sequences of pAEK32, the rhuI probe template, and pAEK33, the rhuIR probe template, were confirmed by automated DNA sequencing. To construct the PbhuR template, pAEK31, a 195-bp fragment, was released from pDJM31 by digestion with EcoRI and BamHI. This fragment, containing 31 bp of the 3' end of rhuR, the entire 102-bp intergenic region, and 47 bp of the 5' end of bhuR, was ligated into the EcoRI and BamHI sites of pBluescript KS(–).
Antisense RNA probes for RPAs were generated by using a MaxiScript in vitro transcription kit (Ambion Inc., Austin, Tex.). Plasmids pERM29 and pERM30 were linearized with XhoI and transcribed using T3 polymerase. Plasmids pAEK31, pAEK32, and pAEK33 were linearized with EcoRI and transcribed using T7 polymerase. The probes, generated by incorporating [-32P]CTP, were labeled according to the manufacturer's protocol, gel purified using an 8 M urea-5% polyacrylamide gel, and eluted from the gel at 37°C overnight using probe elution buffer (HybeSpeed RPA kit; Ambion). The activity of each probe was determined by using a Wallac (Turku, Finland) 1409 liquid scintillation counter. RPAs were performed by using a HybSpeed RPA kit (Ambion) according to manufacturer's protocol. A total of 15 μg of RNA was used in assays with the PbhuR, rhuI, and rhuIR probes and 5 μg of RNA was used in assays with the bhuR and bhuRS probes. Control reactions, in which pAEK31 was linearized with BamHI and transcribed with T3 polymerase generating labeled sense-strand RNA, were used to ensure that protected fragments were not the products of contaminating DNA in the RNA samples. The final reactions were separated by electrophoresis through an 8 M urea-5% polyacrylamide gel. Gels were dried and exposed to blue sensitive autoradiographic film (Marsh Bioproducts, Rochester, N.Y.) at –80°C using an intensifying screen.
Nucleotide sequencing and analysis. Nucleotide sequencing was performed at the Biopolymer Facility at Roswell Park Cancer Institute (Buffalo, N.Y.). Sequence analysis was performed by using the Wisconsin Package version 9.0 (Genetics Computer Group, Madison, Wis.) and ClustalW (http://www.ebi.ac.uk/clustalw).
RESULTS
Primer extension analysis of PrhuI and PbhuR. As an initial step to define the transcriptional architecture of the rhuIR-bhuRSTUV (rhu-bhu) locus, the transcriptional start sites of PrhuI and PbhuR were identified by use of primer extension. For the extension reactions, total RNA was isolated from cultures of B. avium 4169rif which had been cultured in CDM under conditions of Fe limitation. For comparison, total RNA was also obtained from 4169rif cultured under Fe-replete conditions. To evaluate the capacity of the cells to respond to heme as a putative source of Fe and as an inducer of the rhu-bhu locus, total RNA was also isolated from cultures of Fe-starved 4169rif to which heme had been added. In those cultures, the broth was supplemented with EDDHA to remove any contaminating Fe that might have been present in the hemin stock solution.
Primer extension experiments revealed that transcription from PrhuI was strongly repressed in Fe-replete cultures (Fig. 1A). In Fe-limiting growth, transcription from PrhuI initiated from a G residue located 28 bases upstream from the rhuI start codon. Analysis of the region of DNA which contained this G residue showed that this transcriptional start site was located within a putative Fur box which is optimally spaced from –35 and –10 elements that are predicted to be 70 dependent (17). These results indicated that transcription from PrhuI is regulated by Fe, perhaps in a Fur-dependent manner. The observations were also consistent with those reported for B. pertussis (28). Unexpectedly, when hemin was the sole source of Fe, transcriptional initiation in 4169rif was significantly increased from PrhuI and suggests that heme is an inducer of this promoter.
As was observed for PrhuI, transcription from PbhuR in cells obtained from Fe-replete cultures was not detected (Fig. 1B). Hemin supplementation elicited initiation of transcription at an A residue located 37 bases upstream from the bhuR start codon, a pattern of transcriptional initiation which was consistent with previous experiments that demonstrated hemin-dependent expression and induction of bhuR (17). Surprisingly, when cells were cultured under Fe-limiting growth conditions in the absence of hemin, the level of transcription from PbhuR in those cells was similar to the level of transcription from PrhuI in cells cultured under equivalent conditions of Fe limitation. This pattern of transcription was in contrast to the pattern of transcriptional initiation reported for B. pertussis and Bordetella bronchiseptica. In those species, heme-independent bhuR expression originated solely from PhurI (28).
To discount the possibility that PbhuR-dependent extension products in the primer extension experiments were artifacts caused by a premature termination of the extension reactions generated by inhibitory secondary structures in the RNA located proximal to the putative start site, primer extension analysis was also performed using an artificial RNA substrate. RNA for this control experiment was obtained from in vitro transcription of pAEK31, a plasmid containing a DNA fragment encompassing the 3' end of rhuR, the rhuR-bhuR intergenic region, and the 5' end of bhuR cloned into pBluescript. When primer extension reactions were performed on this artificial RNA substrate, the extension reaction proceeded through the putative start site and into the rhuR sequences (data not shown). In brief, these results indicated that the start site revealed by the primer extension product for the PbhuR site was not an artifact caused by premature termination of reverse transcriptase but is likely to be the actual initiation site for PbhuR transcriptional initiation in vivo.
Comparison of the B. avium rhuR-bhuR and B. pertussis hurR-bhuR intergenic regions. Since B. avium and B. pertussis exhibited differences in PbhuR transcriptional patterns, the nucleotide sequences of the regions surrounding these promoters in the two species were compared in an attempt to identify structural motifs which might mediate those differences. While the stop codon of the rhuI ORF and the start codon of bhuR are separated in B. avium by 105 bp of DNA, the two corresponding sites in B. pertussis are separated by 207 bp. Homology between these two intergenic regions is most apparent when the PbhuR start site of B. avium and the PbhuR start site of B. pertussis (28) are aligned. In this orientation, the putative –35 and –10 elements are identical. Seventy-two nucleotides (nt) in the rhuR-bhuR intergenic region of B. pertussis, which are located downstream from the bhuR transcriptional start site, are absent in the corresponding intergenic region of B. avium. Computer analysis of the 72-bp region of B. pertussis did not reveal known regulatory motifs. While experiments have yet to reveal a new regulatory structure within this segment of DNA, it is reasonable to hypothesize that the 72-nt region contributes to the apparent differences in heme-independent bhuR transcription and polypeptide expression observed between B. avium and B. pertussis.
rhuI contributes to heme-independent expression of bhuR. When overexpressed from a plasmid in B. avium, recombinant rhuI induces both Fe-independent and heme-independent expression of bhuR (17). To further assess the role of rhuI on PbhuR activity, a mutant of 4169rif which had a precise deletion of the rhuI ORF was engineered. Engineering of the rhuI mutant was such that the adjacent sequences encoding rhuR and the sequences encoding the rhuR-bhuR intergenic region were undisturbed. PbhuR activity was monitored by measuring the activity of an extrachromosomal PbhuR::lacZYA reporter in pDJM41 (Fig. 2A). Constitutive expression of recombinant rhuI in those strains was produced by introduction of pERM26, as needed. All strains were cultured in BHI broth under Fe-replete (36 μM FeSO4), Fe-limited (100 μM EDDHA), or Fe-limited plus hemin (100 μM EDDHA plus 5 μM hemin) growth conditions.
Analysis of 4169rif(pDJM41, pRK415) cultured under Fe-replete growth conditions revealed only low-level expression of -galactosidase activity which increased only twofold when cells were cultured in Fe-limited broth (Fig. 2B). Supplementation of the Fe-limited broth with hemin, however, enhanced -galactosidase activity fivefold relative to Fe-replete conditions (Fig. 2B) (22), a pattern of expression consistent with prior reports which indicated that heme was a positive inducer of PbhuR (17, 26). In comparison to PbhuR activity in cells cultured under Fe-replete growth conditions, -galactosidase activity of 4169rifrhuI(pDJM41, pRK415) was also enhanced twofold when exposed to Fe limitation (Fig. 2B). Yet -galactosidase activity of 4169rifrhuI(pDJM41, pRK415) was not elevated when the Fe-limited culture broth was supplemented with hemin (Fig. 2B). Introduction of pERM26 into 4169rifrhuI(pDJM41) highly enhanced -galactosidase activity under all growth conditions (Fig. 2B).
These results indicated that rhuI was essential for heme-dependent activation of PbhuR. Fe-dependent expression of bhuR, however, was independent of rhuI (17). Yet it was also clear from these experiments that overexpression of a recombinant rhuI constitutively activated PbhuR in the absence of heme, which suggested that RhuI functions as a PbhuR-dependent sigma factor in the absence of heme.
Assessing the role of PrhuI on Fe-dependent, heme-independent bhuR expression. Fe-dependent, heme-independent expression of bhuR in B. pertussis and B. bronchiseptica is hypothesized to occur by readthrough transcription initiating from PhurI (28). Thus, expression of bhuR in cells cultured under Fe-limited, heme-free growth conditions would be expected to be absent in a mutant in which transcription is terminated at rhuR. To test this hypothesis, various mutants were constructed in which an cassette, a DNA cassette containing multiple translational and transcription terminators, was inserted into the chromosomes of 4169rif and Pho20 at a position 2 bp downstream of the TAA translational terminator of rhuR (5). Pho20 is a reporter mutant containing an in-frame phoA insertion into bhuR (Fig. 3A).
Using Pho20 and Pho20, Fe-dependent expression of bhuR was assessed by measuring alkaline phosphatase activity of cultures grown under Fe-replete or Fe-limited conditions. As previously reported, alkaline phosphatase activity of Pho20 was low when cells were cultured under Fe-replete conditions and increased slightly when cells were cultured under Fe-limiting conditions (Fig. 3B). These data were consistent with prior reports (17) which demonstrated that bhuR expression responded to Fe-dependent repression. When the PhoA activity of Pho20 was evaluated, it was found that the level of alkaline phosphatase activity of Pho20 was similar to the alkaline phosphatase activity of Pho20 when the mutants were cultured under Fe-limiting growth conditions. Remarkably, the alkaline phosphatase activity of Pho20 was essentially unchanged, even when Pho20 was cultured under Fe-replete conditions (Fig. 3B). This unexpected result suggested that the cassette likely imposed an artificial Fe-independent transcriptional activity that promoted expression of the bhuR::phoA fusion. It should be noted that similar problematical results were observed when a mutant containing an cassette was incorporated into the hurR-bhuR intergenic region of a plasmid-borne copy of the B. pertussis hurIR-bhuR-lacZ locus (28). In that case, introduction of the cassette stimulated an increase in reporter activity when the cells were cultured under Fe-replete conditions. Thus, these data obtained in B. avium and in B. pertussis using cassette mutants neither supported nor refuted the hypothesis that bhuR expression under conditions of Fe-limited growth was promulgated by readthrough transcription from PrhuI. Rather, it is obvious that the cassette is not a useful tool for experiments to artificially terminate transcription in either B. avium or B. pertussis and that results from such experiments need to be interpreted with care.
The role of Fur in expression of bhuR. Alkaline phosphatase activity of Pho20fur, a mutant of Pho20 in which the genomic copy of fur is genetically disrupted (17), was significantly higher than its Pho20 parent when cultured under Fe-replete conditions (Fig. 3B). These data supported a model of Fur regulation of Fe-dependent repression of bhuR (17). Under Fe-replete conditions, the alkaline phosphatase activity of Pho20fur was also significantly higher than the alkaline phosphatase activity of Pho20 (Fig. 3B). This pattern of expression of the reporter suggests that the artifactual problems inherent in cassette mutants can be overcome by evaluating the effects of the cassette on readthrough transcription in fur backgrounds in B. avium. Indeed, the results show that alkaline phosphatase activity of Pho20fur was reduced compared to the activity of Pho20fur (Fig. 3B) and indicated that Fe-dependent and Fur-dependent activity from PrhuI is at least partly responsible for bhuR expression.
Evaluation of PbhuR transcriptional activity. Primer extension analysis indicated that transcription initiated from both PrhuI and PbhuR when cells were cultured under conditions of Fe limitation in the presence or absence of hemin (Fig. 1). From these data, it was inferred that PbhuR is responsible for heme-independent expression of bhuR. These observations are in contrast to results observed for B. pertussis (28) in which PbhuR activity was solely heme dependent. In the Pho20fur and Pho20fur mutant strains, PrhuI activity was at least partially responsible for Fe-dependent bhuR expression (Fig. 3B). To further analyze heme-independent expression of bhuR in B. avium, the effects of the fur, rhuI, and cassette mutations on PbhuR activity were examined using primer extension as a direct measurement of transcriptional activity from that promoter.
Primer extension reactions using total RNA harvested from BHI cultures of 4169rif, 4169rifrhuI, 4169rif, and 4169riffur revealed a previously unknown activity of RhuI. As previously reported, PbhuR was inactive when 4169rif was cultured under conditions in which Fe was abundant (Fig. 1B and 4B). Under Fe-limiting conditions, however, significant transcription from PbhuR was detected regardless of the presence or absence of hemin (Fig. 1B and 4B). Transcription from PbhuR was absent in the rhuI mutant under any condition tested (Fig. 4B), a result which was consistent with a model in which rhuI is essential for Fe-dependent PbhuR activity whether or not the cell is grown in the presence or absence of hemin. In contrast, the PbhuR transcription profile of the cassette mutant was indistinguishable from that of wt 4169rif (Fig. 4B), thus indicating that the transcripts detected by primer extension did not originate from PrhuI or other upstream promoters. Interestingly, a modest transcriptional activity was observed in RNA obtained from Fe-replete cultures of the fur mutation (Fig. 4B). This observation suggested that Fe-dependent repression of PbhuR activity is mediated by Fur. Taken together, these results supported the hypothesis that PbhuR is active under Fe-limiting conditions irrespective of heme and that PbhuR requires a functional rhuI for transcriptional activity. These results also strongly implied that PbhuR activity is repressed by Fe in a Fur-dependent manner.
Evaluation of PbhuR activity by RNase protection analysis. As a complementary approach, the transcriptional architecture of the genomic rhu-bhu locus was also evaluated using RNase protection. Results from primer extension analysis of 4169rif were consistent with a model in which transcription from PbhuR initiates under Fe-limiting growth conditions in the absence of heme (Fig. 1B and 4B). While results from promoter-reporter analyses of 4169rifrhuI (Fig. 2B), Pho20, and Pho20fur (Fig. 3B) were inconclusive with respect to PbhuR activation, primer extension results from these mutants strongly suggested that Fe-dependent, heme-independent transcription originates from PbhuR, an observation which is not consistent with previous reports of transcriptional analysis of hurIR-bhuR of B. pertussis (28). Furthermore, promoter-reporter analyses of Pho20fur and Pho20fur (Fig. 3B) clearly showed that bhuR expression originated, at least in part, from PrhuI. To explicate the details of PbhuR activity in B. avium, the effects of rhuI and cassette mutations on transcription of rhuIR-bhuRSTUV were examined using RNase protection analysis. Antisense probes directed against three different regions of the rhu-bhu locus were designed to evaluate expression of the following different regions: (i) a region encompassing the 3' end of rhuI and the 5' end of rhuR (rhuIR), (ii) the entire rhuR-bhuR intergenic region (PbhuR), and (iii) a region encompassing the 5' end of bhuR (bhuR) (Fig. 5A).
None of the three antisense probes were protected from RNase A digestion by the addition of total RNA isolated which was obtained from Fe-replete cultures of 4169rif, indicating that transcripts from the rhu-bhu locus are highly repressed under these growth conditions (Fig. 5B). All three antisense probes were, however, protected by total RNA isolated from 4169rif cultured under conditions of Fe limitation (Fig. 5B). Sizes of the protected fragments in each case were determined by comparison to a commercial RNA standard ladder (data not shown). A fragment of approximately 237 nt, equivalent in size to the full-length antisense probe of the rhuIR region, was protected by total RNA from 4169rif, demonstrating that rhuI and rhuR are cotranscribed on a polycistronic mRNA. Likewise, total RNA from 4169rif cultured under Fe-limiting conditions protected a fragment of approximately 182 nt in length, consistent with the full-length antisense probe of PbhuR which targeted the 3' end of rhuR, the rhuR-bhuR intergenic region, and the 5' end of bhuR. Protection of the entire length of this second antisense probe indicated that transcription proceeded from rhuR and continued through the intergenic region and into bhuR. The intensity of the 182-nt PbhuR-protected fragment was considerably lower than the intensity of the rhuIR-protected fragment, suggesting that only some of the transcripts originating upstream from rhuI continued through the intergenic region and into bhuR. A 230-nt protected fragment representing the full-length antisense probe of a region within the bhuR ORF exhibited a signal which was higher in intensity than that of the fragment from the probe homologous to the rhuR-bhuR intergenic region. These results suggested that transcripts originating upstream from rhuI were not the sole source of bhuR-encoding transcripts produced by the cell. In fact, a shorter, 80-nt protected fragment was evident in experiments using the PbhuR antisense probe (Fig. 5B). The length of this shorter protected fragment corresponded to the PbhuR start site as determined by primer extension (Fig. 1B). Taken together, the combined intensities of the shorter 80-nt and full-length 182-nt protected fragments from the PbhuR antisense probe were essentially equivalent to the signal produced by protection of the bhuR antisense probe. When hemin was added to the culture broth, the intensities of the shorter 80-nt PbhuR-protected fragment and the bhuR-protected fragment increased significantly (Fig. 5B). This observation was consistent with the patterns of heme induction of bhuR from PbhuR reported by Kirby et al. (17). A slight increase in the intensity of the rhuIR-protected fragment was also detected upon addition of hemin to the culture medium (Fig. 5B). These data supported the results shown in Fig. 1A that demonstrated that PrhuI activity was induced by heme.
Total RNA isolated from the rhuI mutant provided the same degree of protection of rhuIR-specific probes as did total RNA from 4169rif when cultured under similar growth conditions (Fig. 5B). These data confirmed that rhuIR expression was not autoregulated. These results also indicate that unlike PbhuR, PrhuI is induced by heme but not via the BhuR-RhuR-RhuI signal cascade. These data suggest the presence of an additional heme-dependent regulatory mechanism. The smaller size of the rhuIR-protected fragment was consistent with the size of the fragment predicted to be protected by total RNA obtained from the mutant in which the rhuI ORF had been removed (i.e., 4169rifrhuI). Protection of the antisense bhuR probe was greatly reduced in RNA isolated from Fe-limited cultures of the rhuI mutant cultured in the presence or absence of hemin. The pattern of protection of the antisense PbhuR probe was also influenced by the rhuI mutation. The shorter 80-nt fragment of PbhuR was not protected by RNA isolated from the rhuI mutant grown under Fe-limited conditions in the presence or absence of hemin. The larger 182-nt fragment of PbhuR, however, was protected to a degree equal to the protection of the bhuR antisense probe (Fig. 5B). The results again confirmed that some expression of bhuR is dependent upon PrhuI activity. The results also showed that transcription from PbhuR under Fe-limiting conditions not only is rhuI-dependent and hemin-induced but also occurs in the absence of hemin, which is consistent with the primer extension results (Fig. 1B and 4B).
Total RNA isolated from the 4169rif protected the rhuIR and bhuR antisense probes in a manner similar to that of RNA from 4169rif cells cultured under Fe-limiting conditions in either the presence or absence of hemin (Fig. 5B). Analysis of the PbhuR region showed that protection of the shorter 80-nt fragment was unaffected by the cassette, whereas protection of the 182-nt full-length PbhuR antisense probe was eliminated (Fig. 5B). These results demonstrated that transcriptional initiation at PbhuR is not dependent upon heme or PrhuI and indicated that Fe-dependent, heme-independent expression of bhuR originates from two promoters (PrhuI and PbhuR). Furthermore, it is clear that PbhuR activity requires rhuI.
Fe-dependent regulation of PrhuI and PbhuR activities in a B. avium fur mutant strain. rhuI is Fur regulated in B. avium (17). Also, Fe-dependent repression of bhuR was found to be influenced by fur (17). Consistent with those observations, the rhuR-bhuR intergenic region was reported to contain a nucleotide sequence with weak homology to the E. coli Fur box consensus. This region also exhibited weak but significant titration activity in the E. coli Fur titration assay (FURTA) strain. Thus, experiments were designed to establish whether or not Fur had a direct role in Fe-dependent regulation of PbhuR (17). Promoter-reporter assays using the Pho20fur and Pho20fur strains indicated that PrhuI contributed to Fur-dependent bhuR expression (Fig. 3). In contrast, primer extension analysis showed that both PrhuI and PbhuR were inactive under Fe-replete conditions and induced only when Fe was limiting (Fig. 1). Further analysis showed that PbhuR in a fur mutant was slightly activated under Fe-replete conditions (Fig. 4B).
To further evaluate expression of the rhu-bhu locus in a fur mutant, RNase protection assays were conducted. Total RNA isolated from Fe-replete and Fe-limited cultures of 4169rif and 4169riffur were utilized for these RNase protection analyses. In addition to the antisense probes used in the previous RNase protection assays (Fig. 5A), two additional probes were employed to broaden the transcriptional analysis of the rhu-bhu locus: an rhuI antisense probe which was directed against the 5' end of rhuI and a bhuRS antisense probe which was directed against a region encompassing the 3' end of bhuR and the 5' end of bhuS (Fig. 6A). All five antisense probes were protected from RNase A digestion by total RNA isolated from 4169rif cultures grown under Fe-limiting conditions but were not protected by total RNA isolated from 4169rif cultured under Fe-replete conditions (Fig. 6B). These data, along with the data reported in Fig. 5B, supported a model in which the rhu-bhu locus including rhuI, rhuR, bhuR, and bhuS was expressed in an Fe-dependent manner. The presence of the 182- and 80-nt protected fragments of the antisense PbhuR probe in reactions containing total RNA isolated from Fe-limited cultures is likely the result of readthrough transcription from PrhuI and of transcription from PbhuR, respectively. This pattern of expression indicated that transcriptional activities of both PrhuI and PbhuR were Fe-regulated events. Surprisingly, the intensity of the 182-nt protected fragment of the PbhuR antisense probe was equal to the intensities of the protected fragments of the rhuI and rhuIR antisense probes (Fig. 6B). This observation was interpreted as strongly supportive of a model in which all transcripts from PrhuI extended past the rhuR-bhuR intergenic region. This result is contrary to data reported in Fig. 5 which indicated that not all transcripts from PrhuI proceeded through the intergenic region into bhuR. Although the discrepancy between the two experiments has not been resolved, it is possible that the difference is indicative of an unknown mechanism controlling differential transcriptional termination in the rhu-bhu locus for which the regulatory conditions are yet to be determined.
Total RNA isolated from Fe-replete cultures of 4169riffur protected the antisense probes directed against regions of the rhu-bhu locus. Also, Fe-dependent repression from rhuI to bhuS was relieved in the fur mutant. All fragments of antisense probes protected by RNA from Fe-replete cultures of the fur mutant are equally protected by RNA from 4169rif cells grown under Fe-limited conditions, with one exception. The 80-nt protected fragment, which likely is representative of transcription from PbhuR, was only slightly evident in the fur mutant when cells were cultured under Fe-replete conditions. Longer exposure of the autoradiograph revealed that this fragment was protected by RNA from Fe-replete cultures of 4169riffur but not by RNA from Fe-replete cultures of 4169rif (data not shown). Additionally, -galactosidase activity from the PbhuR::lacZYA reporter in pDJM41 was measured in 4169rif and 4169riffur strains, both cultured under Fe-replete and Fe-limited growth conditions (Fig. 7B). The results show that Fe-dependent repression of PbhuR is relieved in the absence of fur. This evidence, taken together with previous results reported by Kirby et al. (17), strongly implicated fur in regulating Fe-dependent PbhuR activity.
DISCUSSION
In B. avium, heme-dependent expression of bhuRSTUV requires the outer membrane heme receptor BhuR, the ECF factor RhuI, and the ECF factor activator RhuR (16, 17, 22). Paradoxically, this regulatory mechanism requires bhuR to be expressed in the absence of heme inducer. In B. pertussis and B. bronchiseptica, heme-independent expression of bhuR was found to originate from PhurI, an Fe-regulated promoter located immediately upstream from the hurIR-bhuRSTUV gene cluster (28). This gene cluster is orthologous to the B. avium heme utilization locus rhuIR-bhuRSTUV (22). Here, we present evidence that in B. avium, heme-independent transcription of bhuR originates from PbhuR, a proximal RhuI-dependent promoter, as well as from PrhuI, a distal RhuI-independent promoter. This mechanism ensures the synthesis of adequate amounts of BhuR in the absence of heme induction for subsequent heme sensing. Data also point toward the likelihood that both promoters are repressed by Fur under Fe-sufficient conditions and suggest the necessity for swift repression of the heme acquisition system once the demand for Fe has been met.
Transcription from PrhuI initiates 28 bases upstream from the rhuI ORF and is evident only when Fe is limiting (Fig. 1A). Transcription from PrhuI is unaffected by a rhuI mutation, indicating that this promoter is not autoregulated and, therefore, not likely to be activated by heme via the BhuR-RhuR-RhuI signal cascade (Fig. 5B). The observed effect that the addition of hemin to the medium slightly enhanced PrhuI activity was unexpected and suggests the presence of an additional heme-dependent signal cascade. An alternative explanation, however, in which PrhuI activity is indirectly affected by heme cannot be ruled out. Further examination of PrhuI is needed to assess the role of heme in rhuI-independent expression from this promoter. The PrhuI transcriptional start site is optimally spaced from 70-like promoter elements and is located within a putative Fur box in a region that exhibited positive FURTA activity (17). Alkaline phosphatase activities of the Pho20, Pho20fur, and Pho20fur strains support Fur-dependent regulation of PrhuI and indicated that at least a portion of the transcripts from PrhuI read through the rhuR-bhuR intergenic region (Fig. 3B). RNase protection analysis provides evidence that the readthrough transcripts include the coding regions of bhuR and bhuS, at the very least, and likely include the other genes of the bhuR locus (Fig. 5B and 6B). These results, consistent with those reported for B. pertussis and B. bronchiseptica (28), describe a suitable mechanism for the heme-independent expression of bhuR observed in all three species.
Transcriptional analyses of the rhuR-bhuR intergenic region indicated that PrhuI is not the sole active promoter driving bhuR expression. Transcription from PbhuR starts 37 bases upstream from bhuR (Fig. 1B). RNase protection analysis using an antisense probe directed against the rhuR-bhuR intergenic region yielded, in addition to a 182-nt protected fragment that represents readthrough transcription, a shorter 80-nt protected fragment whose length is coincident with the PbhuR transcriptional start site (Fig. 5B). Surprisingly, transcription from PbhuR occursunder Fe-limiting conditions in the absence of heme, although to a much lesser degree than when heme is present (Fig. 2B, 4B, and 5B). Previous studies have shown that heme and heme-containing proteins are inducers of bhuR expression (22). Mutational studies confirmed that RhuI is required for PbhuR activity and showed that RhuI can also function in the absence of heme inducer (16, 17) (Fig. 3B, 4B, and 5B). Previous experiments have shown that, when overexpressed, RhuI constitutively activates bhuR expression (17). It is possible that RhuI has a basal activity for PbhuR induction that is enhanced in the presence of heme through signaling via BhuR and RhuR.
In B. avium, both PrhuI and PbhuR participate in heme-independent bhuR expression. A similar rhuI-dependent, heme-independent PbhuR activity, observed in primer extension analysis of B. bronchiseptica, was reported in a previous study, but no explanations for this activity were proffered (28). Instead, heme-independent expression of bhuR in B. bronchiseptica and B. pertussis was reported to originate exclusively from PhurI (28). It is unexpected that transcription from PbhuR in B. avium should be controlled differently from that in B. pertussis and B. bronchiseptica. In all three organisms, expression of bhuRSTUV is induced by heme via a system that requires BhuR, the ECF factor (HurI), and the coregulator (HurR) (RhuI and RhuR for B. avium), the expression of which are Fe regulated in a Fur-dependent manner. The differences in the regulation of bhuR in these organisms could be attributable to a region of DNA found in the hurIR-bhuR intergenic region of B. pertussis and B. bronchiseptica but absent in the rhuR-bhuR intergenic region of B. avium. This span of additional DNA could contain structural cis-acting elements that proffer additional regulation.
Employing two promoters to control heme-independent bhuR expression would appear redundant for B. avium. It is feasible, however, that the regulatory system involving the two promoters evolved to enable the bacterium to respond to different sources of heme. Several observations support this model: (i) acquisition of both heme and heme-containing proteins (e.g., hemoglobin, myoglobin, catalase, etc.) requires bhuR (22); (ii) growth of B. avium with hemin as its sole Fe source does not require rhuR, but a rhuR mutant has reduced capacity to utilize hemoglobin (16); and (iii) BhuR is expressed in two forms in the outer membrane, a 91-kDa protein encoded by the entire bhuR ORF and a truncated 82-kDa protein in which the N-terminal 104 amino acids of the full-size protein have been removed by a proteolytic cleavage (22). The 82-kDa form is preferentially expressed when B. avium encounters heme. The interplay between PrhuI and PbhuR may be necessary to facilitate a response by B. avium to different biological sources of heme. Experiments to evaluate this model are ongoing.
Transcriptional analysis using RNA from Fe-replete cultures of strain 4169riffur revealed that Fe-dependent repression of the rhuIR-bhuRSTUV locus was relieved in a fur mutant (Fig. 6B). Protected fragments representing transcripts from rhuI and rhuR from 4169riffur cultured under Fe-replete conditions were equally abundant, thus confirming that rhuI and rhuR are transcribed on a single message that is repressed by fur in an Fe-dependent manner. It is clear from the RNase protection experiments that synthesis of this polycistronic mRNA sometimes extends into bhuR and bhuS. Although not determined in these experiments, it is likely that synthesis of the polycistronic mRNA extends into bhuTUV, thus encoding all proteins of the heme uptake system. However, RNase protection experiments also demonstrated that transcripts originating from PrhuI sometimes terminate after rhuI (Fig. 5B). The conditions and the cognate signal(s) that regulate whether polymerization of transcripts continues or terminates after synthesis of rhuI and rhuR have not been established. There is no evidence of a bhuR-associated feedback regulation. However, it is possible that one of the genes located downstream of bhuR may participate in regulating synthesis of the polycistronic mRNA. The role of downstream genes (bhuSTUV) in the control of readthrough transcription, however, is yet to be investigated.
It is clear that Fur has a potent role in governing expression of rhuI, rhuR, bhuR, and bhuS. A region with weak (8 of 19 nucleotides) homology to the consensus Fur box was located between the putative –35 and –10 elements of PbhuR (17). The Fur box appears to be functionally active, since a segment of DNA containing this region exhibited weak FURTA activity (17), implying that it has the capacity to functionally interact with Fur. Transcriptional analyses of RNA isolated from Fe-replete cultures of a fur mutant of B. avium supported the hypothesis that Fur represses the RhuI-dependent PbhuR (Fig. 4B). Furthermore, -galactosidase expressed from PbhuR on pDJM41, which does not harbor DNA containing PrhuI, is still Fe responsive, even in the rhuI mutant (Fig. 3B). Although not unequivocal, these data provided strong evidence that PbhuR activity is highly dependent upon Fe, likely via an interaction with Fur. It was predicted that experiments utilizing the cassette would be useful in isolating PbhuR from PrhuI for evaluating the effect of Fur on PbhuR activity. Unfortunately, any effects that Fur exerted on PbhuR were masked in these experiments by an apparent artifactual Fe-independent transcriptional activity originating from within the cassette. For that reason, the cassette is not a particularly desirable tool for isolating transcriptional units in the bordetellae. As a better alternative strategy for investigating putative interactions of Fur with PbhuR, future experiments will evaluate PbhuR activity in an engineered mutant in which PrhuI is absent.
The need for Fur-dependent regulation at PrhuI and PbhuR sites may indicate that rapid downregulation of expression of bhuRSTUV is desirable when heme and the other components of the signal transduction cascade are still present in the cell but when the Fe requirements of the cell have already been met. Fur may be a strong participant in a regulatory scheme which determines whether transcription from PrhuI terminates after rhuR or reads through to bhuRSTUV. Given the multiple forms of BhuR, the multiple inducers that stimulate bhuR expression, the multiple promoters that drive this expression, and the complexity of bhuR regulation, it is apparent that a complete description of the regulation and functioning of the rhuIR-bhuRSTUV heme utilization locus is yet to be fully revealed. Elucidation of the details of the RhuI-RhuR-BhuR signaling cascade and the characterization of the genes downstream from bhuR may provide valuable insights into Fur regulation, Fe acquisition, and virulence of the bordetellae.
ACKNOWLEDGMENTS
Funds to support this investigation were made available to T.D.C. from the School of Medicine and Biomedical Sciences, University of Buffalo, State University of New York. A.E.K. was supported by an NIH training grant (AIO7614) awarded to the Witebsky Center for Microbial Pathogenesis and Immunology and by a Presidential Fellowship administered by the Office of the Provost, the University of Buffalo. N.D.K. was supported by an NIH training grant (T32 DE007034) awarded to the Department of Oral Biology in the School of Dental Medicine at the University of Buffalo.
REFERENCES
1. Angerer, A., S. Enz, M. Ochs, and V. Braun. 1995. Transcriptional regulation of ferric citrate transport in Escherichia coli K-12. Fecl belongs to a new subfamily of sigma 70-type factors that respond to extracytoplasmic stimuli. Mol. Microbiol. 18:163-174.
2. Beaumont, F. C., H. Y. Kang, T. J. Brickman, and S. K. Armstrong. 1998. Identification and characterization of alcR, a gene encoding an AraC-like regulator of alcaligin siderophore biosynthesis and transport in Bordetella pertussis and Bordetella bronchiseptica. J. Bacteriol. 180:862-870.
3. Braun, V., and H. Killmann. 1999. Bacterial solutions to the iron-supply problem. Trends Biochem. Sci. 24:104-109.
4. Chauhan, S., and M. R. O'Brian. 1997. Transcriptional regulation of delta-aminolevulinic acid dehydratase synthesis by oxygen in Bradyrhizobium japonicum and evidence for developmental control of the hemB gene. J. Bacteriol. 179:3706-3710.
5. Connell, T. D., A. Dickenson, A. J. Martone, K. T. Militello, M. J. Filiatraut, M. L. Hayman, and J. Pitula. 1998. Iron starvation of Bordetella avium stimulates expression of five outer membrane proteins and regulates a gene involved in acquiring iron from serum. Infect. Immun. 66:3597-3605.
6. Dean, C. R., S. Neshat, and K. Poole. 1996. PfeR, an enterobactin-responsive activator of ferric enterobactin receptor gene expression in Pseudomonas aeruginosa. J. Bacteriol. 178:5361-5369.
7. de Lorenzo, V., S. Wee, M. Herrero, and J. B. Neilands. 1987. Operator sequences of the aerobactin operon of plasmid ColV-K30 binding the ferric uptake regulation (fur) repressor. J. Bacteriol. 169:2624-2630.
8. Enz, S., V. Braun, and J. H. Crosa. 1995. Transcription of the region encoding the ferric dicitrate-transport system in Escherichia coli: similarity between promoters for fecA and for extracytoplasmic function sigma factors. Gene 163:13-18.
9. Enz, S., S. Mahren, U. H. Stroeher, and V. Braun. 2000. Surface signaling in ferric citrate transport gene induction: interaction of the FecA, FecR, and FecI regulatory proteins. J. Bacteriol. 182:637-646.
10. Escolar, L., J. Perez-Martin, and V. de Lorenzo. 1999. Opening the iron box: transcriptional metalloregulation by the Fur protein. J. Bacteriol. 181:6223-6229.
11. Gutierrez, C., J. Barondess, C. Manoil, and J. Beckwith. 1987. The Use of TnphoA to detect genes for cell envelope proteins subject to a common regulatory stimulus. Analysis of osmotically related genes in Escherichia coli. J. Mol. Biol. 19:289-297.
12. Hamza, I., Z. Qi, N. D. King, and M. R. O'Brian. 2000. Fur-independent regulation of iron metabolism by Irr in Bradyrhizobium japonicum. Microbiology 146:669-676.
13. Harle, C., I. Kim, A. Angerer, and V. Braun. 1995. Signal transfer through three compartments: transcription initiation of the Escherichia coli ferric citrate transport system from the cell surface. EMBO J. 14:1430-1438.
14. Heinrichs, D. E., and K. Poole. 1996. PchR, a regulator of ferripyochelin receptor gene (fptA) expression in Pseudomonas aeruginosa, functions both as an activator and as a repressor. J. Bacteriol. 178:2586-2592.
15. 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.
16. Kirby, A. E., N. D. King, and T. D. Connell. 2004. RhuR, an extracytoplasmic function sigma factor activator, is essential for heme-dependent expression of the outer membrane heme and hemoprotein receptor of Bordetella avium. Infect. Immun. 72:896-907.
17. Kirby, A. E., D. J. Metzger, E. R. Murphy, and T. D. Connell. 2001. Heme utilization in Bordetella avium is regulated by RhuI, a heme-responsive extracytoplasmic function sigma factor. Infect. Immun. 69:6951-6961.
18. Koster, M., W. van Klompenburg, W. Bitter, J. Leong, and P. Weisbeek. 1994. Role for the outer membrane ferric siderophore receptor PupB in signal transduction across the bacterial cell envelope. EMBO J. 13:2805-2813.
19. Litwin, C. M., and S. B. Calderwood. 1993. Role of iron in regulation of virulence genes. Clin. Microbiol. Rev. 6:137-149.
20. Litwin, C. M., and J. Quackenbush. 2001. Characterization of a Vibrio vulnificus LysR homologue, HupR, which regulates expression of the haem uptake outer membrane protein, HupA. Microb. Pathog. 31:295-307.
21. Miller, J. H. 1992. A short course in bacterial genetics. Cold Spring Harbor Laboratory Press, Plainview, N.Y.
22. Murphy, E. R., R. E. Sacco, A. Dickenson, D. J. Metzger, Y. Hu, P. E. Orndorff, and T. D. Connell. 2002. BhuR, a virulence-associated outer membrane protein of Bordetella avium, is required for the acquisition of iron from heme and hemoproteins. Infect. Immun. 70:5390-5403.
23. Ochs, M., S. Veitinger, I. Kim, D. Welz, A. Angerer, and V. Braun. 1995. Regulation of citrate-dependent iron transport of Escherichia coli: fecR is required for transcription activation by FecI. Mol. Microbiol. 15:119-132.
24. Prentki, P., and H. M. Krisch. 1984. In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29:303-313.
25. Schmitt, M. P. 1999. Identification of a two-component signal transduction system from Corynebacterium diphtheriae that activates gene expression in response to the presence of heme and hemoglobin. J. Bacteriol. 181:5330-5340.
26. Vanderpool, C. K., and S. K. Armstrong. 2001. The Bordetella bhu locus is required for heme iron utilization. J. Bacteriol. 183:4278-4287.
27. Vanderpool, C. K., and S. K. Armstrong. 2003. Heme-responsive transcriptional activation of Bordetella bhu genes. J. Bacteriol. 185:909-917.
28. Vanderpool, C. K., and S. K. Armstrong. 2004. Integration of environmental signals controls expression of Bordetella heme utilization genes. J. Bacteriol. 186:938-948.
29. Van Hove, B., H. Staudenmaier, and V. Braun. 1990. Novel two-component transmembrane transcription control: regulation of iron dicitrate transport in Escherichia coli K-12. J. Bacteriol. 172:6749-6758.
30. Weinberg, E. D. 1993. The development of awareness of iron-withholding defense. Perspect. Biol. Med. 36:215-221.(Natalie D. King, Amy E. K)
Department of Microbiology and Immunology
Department of Oral Biology, University at Buffalo, State University of New York, Buffalo, New York
ABSTRACT
Iron (Fe) is an essential nutrient for most bacterial pathogens. In these organisms, a variety of regulatory systems that respond to specific Fe complexes found within their vertebrate hosts have evolved. In Bordetella avium, the heme utilization locus encoded by rhuIR-bhuRSTUV mediates efficient acquisition of Fe from heme and hemoproteins. Control of bhuRSTUV expression is promulgated at two levels. When Fe is abundant, expression is repressed in a Fur-dependent manner which is partially relieved when Fe is limiting. In the presence of heme or hemoproteins, expression of the bhuRSTUV operon is induced via a three-component signal transduction cascade composed of RhuI, RhuR, and BhuR. Herein, we report the identification of two promoters (PrhuI and PbhuR) that control expression of the rhuIR-bhuRSTUV cluster. Primer extension analysis identified the transcriptional start site of PrhuI within a putative Fur box. Transcriptional initiation of PbhuR mapped within the rhuR-bhuR intergenic region. Maximal transcription from PbhuR required Fe-limiting conditions, the presence of heme (or hemoglobin), and rhuI; however, analysis of transcripts produced from the rhuIR-bhuRSTUV locus revealed a pattern of low-level bhuR transcription in the absence of heme which originated from both PbhuR and PrhuI. Transcription from PrhuI was repressed by Fe in the presence of fur and somewhat enhanced by the addition of hemin to Fe-limited media. The nature of this hemin-associated PrhuI stimulation was rhuI independent and therefore not induced by heme via the BhuR-RhuR-RhuI signal cascade. Fe also repressed transcription from PbhuR in a fur-dependent manner; however, activation from this promoter, in the presence or absence of heme, did not occur without rhuI.
INTRODUCTION
Iron (Fe), an essential element for most pathogenic bacteria, must be acquired from the infected host. Within that host, however, Fe is usually sequestered within high-affinity ferromolecules such as transferrin, lactoferrin, and ferritin. Sequestration by these molecules maintains the free-Fe concentration in the tissue and fluids of vertebrates below 10–18 M (30), a level at which the growth of most bacteria is inhibited. To survive in these extremely Fe-limited host environments, bacteria express high-affinity Fe acquisition systems (reviewed in reference 19). In most cases, these acquisition systems are modulated in response to the availability of Fe in the local environment. Global Fe-dependent regulation is achieved primarily by Fur, a repressor protein whose regulation is dependent upon intracellular Fe concentration. Fur, which is expressed by a wide range of both gram-positive and gram-negative bacteria, has the capacity to reversibly bind Fe. In its repressor mode, Fe-bound Fur has affinity for the Fur box, a consensus motif that is located within or near Fur-regulated promoters (7). Binding of Fe-Fur to the Fur box is believed to interfere with binding of RNA polymerase (RNAP) to the promoter. Alternatively, binding of Fur might inhibit promoter escape of RNAP once the polymerase is bound to the promoter (10). When intracellular Fe levels decline to critical levels, the intracellular pool of Fur is no longer saturated with Fe. Since apo-Fur has little affinity for Fur boxes, Fur-dependent promoters are subsequently released from repression (10).
Successful pathogens have evolved a variety of mechanisms to acquire one or more of the biological Fe sources which are expressed by vertebrates (e.g., transferrin, ferritin, heme, hemoglobin, etc.) (3). Control of these systems solely by Fur would likely be a wasteful expenditure of resources, as it is unlikely that invading or colonizing bacteria will simultaneously encounter all of these sources of Fe in a single microhabitat. Thus, bacteria evolved additional regulatory elements that coordinately interact with Fur to rapidly and specifically respond to the presence of specific Fe-containing molecules. Various Fur-augmented regulatory systems have been described previously: siderophore-responsive transcription is dependent on AraC-type regulators in both Pseudomonas aeruginosa (14) and Bordetella spp. (2), LysR-type regulators have been implicated in the heme-responsive gene expression of Vibrio vulnificus (20) and Bradyrhizobium japonicum (12), and classical two-component sensor-kinase pairs have been described for Corynebacterium diphtheriae (25) and Pseudomonas aeruginosa (6). Sigma () factors, particularly those denoted as extracytoplasmic function (ECF) factors, also participate in Fur-dependent gene regulation. In Pseudomonas putida (18), Escherichia coli (29), Bordetella pertussis (27), and Bordetella avium (17), specific Fe acquisition systems are maximally expressed only during Fe-limited growth in the presence of the cognate stimulatory ligand which, by binding to an extracellular receptor, stimulates the activity of the ECF factor for an associated ECF-dependent promoter.
ECF factor-based control of gene expression has been most extensively described in E. coli. In that bacterium, uptake of extracellular ferric dicitrate is governed by fecIRABCDE which is transcribed only when the bacterium is simultaneously stressed for Fe and ferric dicitrate is locally available. Responsiveness of E. coli to ferric dicitrate is mediated by the interaction of a three-component ECF regulatory cascade that includes FecA, the outer membrane receptor for ferric dicitrate (13); FecR, an inner membrane regulatory protein (23); and FecI, an ECF factor (1). Mechanistically, when FecA binds ferric dicitrate, a signal is transduced across the periplasm to FecR which transmits the signal to FecI (9). Acquisition of the signal by FecI elicits a change in its activation state which stimulates association of FecI to core RNAP. Recruitment of RNAP holoenzyme to PfecA initiates direct transcription of the fecABCDE operon (1). Examination of the transcripts produced from the fecIRABCDE locus revealed two major mRNAs: a 1.5-kb mRNA corresponding to fecIR and a 2.5-kb mRNA corresponding to fecA. It has been proposed that the 2.5-kb fecA-bearing transcript is likely a degradation product of a 6-kb polycistronic transcript that contains sequences derived from the fecABCDE genes (8). In the absence of ferric dicitrate, basal expression of fecABCDE in Fe-stressed cells is regulated by Fur (29).
Nevertheless, an interesting conundrum was evident in the regulation of FecA. FecA is a requisite component of the FecIR-regulatory cascade and thus is required as a receptor of ferric dicitrate for initiating the signal cascade. In fact, low levels of FecA were detected in the outer membrane of Fe-stressed E. coli in the absence of ferric dicitrate. How, then, is FecA expressed in the absence of the stimulatory ligand The answer to this question is yet to be revealed for E. coli. A mechanism to explain this conundrum, however, was recently described for B. pertussis (28). In B. pertussis, two promoters, PhurI and PbhuR, control expression of the heme utilization gene cluster hurIR-bhuRSTUV. Transcription from PhurI, located upstream from hurI, is Fe regulated and heme independent. Transcription from PbhuR, a promoter located within the hurR-bhuR intergenic region, is also Fe regulated and is induced by heme via HurI, the cognate ECF factor (27). Under Fe-stressed conditions in the absence of heme, expression of bhuR arises via transcriptional initiation at PhurI in B. pertussis. Essentially, transcription does not terminate at the 3' end of hurIR but proceeds infrequently into bhuR (28). This rare "readthrough" transcription is believed to be sufficient to express amounts of BhuR in the outer membrane for initial triggering of heme-dependent regulation of PbhuR when heme becomes available.
B. avium, a gram-negative pathogen that infects domestic and wild fowl, elicits bordetellosis (or coryza), an upper respiratory illness defined by a loss of ciliated tracheal cells and a persistent, aggravated cough. The clinical presentation of coryza is highly similar to that of human pertussis (whooping cough) caused by taxonomically related B. pertussis. For that reason, B. avium has been used as a model for human pertussis. B. avium encodes the heme utilization locus rhuIR-bhuRSTUV (22), a heme uptake system with homology to hurIR-bhuRSTUV of B. pertussis. Expression levels of rhuIR-bhuRSTUV and hurIR-bhuRSTUV are regulated in similar manners in response to heme; i.e., in B. avium, expression of bhuR from PbhuR is Fe regulated and heme responsive in an rhuIR-dependent manner (17). In this paper, the transcriptional architecture of the heme utilization locus in B. avium is described to compare and contrast the regulation of the heme uptake systems of B. avium and B. pertussis. The results confirm that unlike the regulation reported for B. pertussis (28), heme-independent expression of BhuR in B. avium is controlled by transcription which initiates at both the PrhuI and PbhuR promoters.
MATERIALS AND METHODS
Media, strains, and growth conditions. Bacterial strains and plasmids used in this study are listed in Table 1. Strains of B. avium were maintained on brain heart infusion (BHI) agar, in BHI broth (Difco Laboratories, Detroit, Mich.), or in Chelex-treated complete defined medium (CDM) (5). Strains of E. coli were cultured on Luria-Bertani agar. For Fe-replete growth conditions, BHI or CDM was supplemented with 36 μM FeSO4. Conditions of Fe limitation were achieved in CDM by reducing Fe by treatment with technical-grade Chelex-20 resin (Bio-Rad, Hercules, Calif.). Fe-limited conditions were achieved in BHI by supplementing the agar or broth with 100 μM ethylene-di-o-hydroxyphenylacetic acid (EDDHA). Unless otherwise noted, antibiotics used were as follows: ampicillin (200 μg/ml), rifampin (10 μg/ml), streptomycin (200 μg/ml), spectinomycin (10 μg/ml), tetracycline (10 μg/ml), kanamycin (50 μg/ml), and gentamicin (10 μg/ml). Antibiotics were obtained from Sigma Biochemicals (St. Louis, Mo.) and Amresco (Solon, Ohio). Biochemical reagents were purchased from Life Technologies, Inc. (Frederick, Md.) and Sigma Biochemicals. Restriction enzymes and DNA-modifying enzymes were obtained from MBI Fermentas, Inc. (Hanover, Md.). Deionized water with an electrical resistance of >18 M was used to prepare all solutions.
Isolation of B. avium RNA. Total RNA was prepared from cultured cells by using a method developed by Chauhan and O'Brian (4). Cells grown to an optical density at 600 nm (OD600) of 0.4 to 0.6 were harvested by centrifugation at 4°C. Cell pellets from 12.5 ml of culture were resuspended in 600 μl of lysis buffer (10 mM NaCl, 10 mM Tris [pH 8], 5% sodium dodecyl sulfate, 200 μg of proteinase K per ml) and incubated at 37°C for 5 min. Three hundred microliters of 5 M NaCl was added to the lysed cells. Mixtures were incubated on ice for 10 min, after which the debris was removed by centrifugation at 4°C. RNA in the supernatant was precipitated by centrifugation after the addition of 3 volumes of ethanol and incubation at –80°C for 1 h. RNA was resuspended in 1 mM EDTA, extracted with acid phenol-chloroform, and reprecipitated with ethanol. Purified RNA was treated with DNase I in the presence of human placental RNase inhibitor to eliminate contaminating DNA.
Primer extension analysis. Primers APbhuR1 (5'-CCGCTGCGTGAGAACAGACGAAAAG-3') and APrhuI1 (5'-GCTTTTTGCGCAGCCAGCCATTCAG-3') were end labeled in a reaction mixture containing 10 pmol of primer, 1 mM spermidine, 4 μl of [-32P]ATP (3,000 Ci/mmol, 10 μCi/ml), 10 U of T4 polynucleotide kinase, and 1x labeling buffer (MBI Fermentas). Reaction mixtures were incubated at 37°C for 30 min, and reactions were terminated by heating to 95°C for 2 min. Labeled primers were purified by using a Sephadex G-25 column according to the manufacturer's instructions (Roche, Indianapolis, Ind.). Samples containing 20 μg of total RNA were ethanol precipitated, and pellets were resuspended in mixtures containing 0.2 pmol of labeled primer and 1x Superscript II reverse transcriptase reaction buffer (Invitrogen, Carlsbad, Calif.). RNA was denatured by heating to 95°C for 1 min, and labeled primers were annealed to the RNA by cooling samples to 70°C for 5 min and then to 30°C over a 10-min time span. Reaction mixtures were supplemented with 100 mM dithiothreitol, 10 mM deoxynucleoside triphosphates (dNTPs), Superscript II reverse transcriptase, and 1x Superscript II reverse transcriptase reaction buffer (Invitrogen). Reaction mixtures were incubated at 45°C for 30 min, terminated by the addition of EDTA to 100 μM, extracted with phenol-chloroform, precipitated with ethanol, and resuspended in Tris-EDTA buffer. Extended primers were resolved on 8 M urea-6% sequencing-grade polyacrylamide. For sizing purposes, a ladder was prepared by sequencing an appropriate template using the same primer.
Sequencing ladder for primer extension analysis. A total of 3 μg of pAD3 was denatured with 0.2 N NaOH, neutralized with 2 M NH4Ac, and ethanol precipitated. Pelleted DNA was resuspended with 0.5 pmol of primer (APbhuR1 or APrhuI1) in 1x Sequenase buffer (USB Corp., Cleveland, Ohio). Primer and denatured template were annealed by heating to 65°C for 2 min and cooled slowly to room temperature. Labeling and termination reactions were performed according to the manufacturer's instructions (USB Corp.).
Construction of 4169rifrhuI. The rhuI open reading frame (ORF) was deleted from the chromosome of wild-type (wt) B. avium 4169rif using allelic exchange (16). Splicing of overlapping ends (SOEing) by PCR was used to construct the allelic exchange construct pAEK22.1. The 507 bp immediately upstream from the rhuI GTG start codon were PCR amplified by using primers rhuI-a2 (5'-GCTCTAGATAGATGTGATGCGGTTCTT-3' [the XbaI site is underlined]) and rhuI-b2 (5'-ACGCCGGCTGCGCTCATTTTTTGCAAGAAATATAA-3' [the underlined region is complementary to the 5' end of rhuR]) (PCR conditions were as follows: the PCR mixture contained 1x ThermoPol buffer, 10% dimethyl sulfoxide [DMSO], 250 μM dNTPs, 400 nM each oligonucleotide primer, and 1 U of Vent polymerase [Invitrogen], and PCR was carried out at 94°C for 1 min, 45°C for 1 min, and 72°C for 1 min for 30 cycles). Primers rhuI-c (5'-ATGAGCGCAGCCGGCGT-3') and rhuI-d (5'-AGGGAGCTCTCTATTTAGTAACAGGAATCATTTA-3' [the SacI site is underlined]) were used to amplify the 944 bp located immediately downstream from the rhuI ORF beginning at the ATG start codon of rhuR (PCR conditions were as follows: the PCR mixture contained 1x PCRx buffer [Promega, Madison, Wis.], 10% DMSO, 150 μM dNTPs, 1 μM each oligonucleotide primer, and 1 to 10 U of Taq polymerase, and PCR was carried out at 94°C for 1 min, 45°C for 1 min, and 72°C for 1 min for 25 cycles). PCR products, purified using a GFX PCR DNA and gel band purification kit (Amersham Pharmacia Biotech, Inc., Piscataway, N.J.), were used as templates for a modified PCR SOEing reaction in which 1 μl of each PCR product was added to an annealing reaction mixture containing 1x ThermoPol buffer (Invitrogen), 10% DMSO, and 250 μM dNTPs. This mixture was heated to 95°C for 5 min and cooled to 25°C over a period of 3 min to allow the ends of each PCR product to anneal, thus generating partially double-stranded templates. The double-stranded templates were repaired by using 1 U of Klenow fragment (MBI Fermentas, Inc.). After 10 min at 25°C, Klenow fragment was heat inactivated by incubation at 95°C for 5 min. Primers (rhuI-a2 and rhuI-d, at a concentration of 1 μM each) and 1 U of Vent polymerase (Invitrogen) were added to the mixture to initiate PCR amplification (94°C for 1 min, 45°C for 1 min, 72°C for 3 min, 25 cycles). The PCR SOEing product was ligated into pBluescript KS(+) at the XbaI and SacI sites to engineer pAEK22. After the sequence of pAEK22 was confirmed by automated DNA sequencing, the insert was released from the plasmid by digestion with XbaI and SacI, gel purified, and ligated into pCVD442tet, which had been digested with the same enzymes. The resulting plasmid, pAEK22.1, was moved into B. avium 4169rif by conjugation (22). Transconjugants were selected for growth on BHI agar containing tetracycline. Cells that had undergone a second recombination event to remove the plasmid sequences were selected for growth on BHI agar containing 20% sucrose. Deletion of rhuI in 4169rifrhuI was confirmed by colony PCR and Southern hybridization.
Construction of 4169rif, Pho20, and Pho20fur. An omega () fragment containing bidirectional transcriptional and translational terminators was inserted directly downstream from the rhuR ORF in the chromosome of B. avium strains 4169rif, Pho20, and Pho20fur using allelic exchange. In brief, the plasmid pAEK37.3, used for allelic exchange, was constructed using the following cloning scheme. A 993-bp fragment containing the rhuR-bhuR intergenic region and a portion of the bhuR ORF was amplified from pERM1 by PCR using primers rhuR (5'-GGAATTCGATTCCTGTTACTAAATAGATTCGTAAAAAC-3' [the EcoRI site is underlined]) and bhuRD (5'-TCCGAGCTCCGCCTGAATCAACCATGTCGTGTTGT-3' [the SacI site is underlined]) (PCR conditions were as follows: the PCR mixture contained 1x Fermentas buffer without Mg2+, 1.5 mM MgCl2, 10% DMSO, 250 μM dNTPs, 400 nM each oligonucleotide primer, and 1U of Taq polymerase [MBI Fermentas], and PCR was carried out at 95°C for 1 min, 55°C for 1 min, and 72°C for 2 min for 30 cycles). This fragment was cloned into the EcoRI- and SacI-digested plasmid pAEK26.1, which contains the rhuR ORF immediately upstream from an EcoRI site, to generate plasmid pAEK37.2. The fragment from pHP45 was cloned into the EcoRI site of pAEK37.2 to engineer pAEK37.2. This plasmid was digested with XbaI and SacI to isolate a fragment containing rhuR--bhuR which was inserted into the same sites of plasmid pCVD442tet to produce pAEK37.3. The construction of pAEK37.3 was confirmed by automated DNA sequencing. The plasmid was transformed into SM10pir and subsequently conjugated into B. avium. Transconjugants were selected for growth on BHI agar containing tetracycline. Cells that had undergone the second recombination event were selected for growth on BHI agar containing 20% sucrose. The mutant strains 4169rif, Pho20, and Pho20fur were isolated, and their mutant genotypes were confirmed by Southern hybridization.
-Galactosidase assay. Expression of the lacZYA reporter gene in pDJM41 was determined by measuring -galactosidase activity (17). Briefly, cultures of Fe-limited cells grown overnight were used to inoculate secondary cultures. The secondary cultures were incubated at 37°C for 16 to 20 h. Bacteria in 1.0 ml of culture were pelleted by centrifugation for 5 min at 5,000 x g. Cells were resuspended in Z buffer (60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4, 38 mM -mercaptoethanol), and the OD600 of the cell suspension was adjusted to 0.28 to 0.70. After diluting 400 μl of the suspension with 400 μl of Z buffer, cells were permeabilized by the addition of 45 μl of 0.1% sodium dodecyl sulfate and 90 μl of chloroform. Cell suspensions were vortexed for 10 s, followed by incubation at 30°C for 15 min. Enzymatic reactions were initiated by the addition of 160 μl of a solution containing 4 mg of o-nitrophenyl--D-galactopyranoside per ml. Following incubation at 30°C, samples were observed for the development of a yellow color, at which point 400 μl of 1 M Na2CO3 was added to terminate the reaction. After a brief centrifugation to pellet debris and chloroform, the OD420 and OD550 of each reaction were recorded, and the relative -galactosidase activity was calculated by use of the following formula (21): {1,000[OD420 – 1.75(OD550)]}/(t)(0.4)(OD600), where t is the time of the reaction in minutes. Relative enzyme activities are reported as the means of triplicate assays.
Alkaline phosphatase assay. Expression of the bhuR::phoA fusion of B. avium mutants Pho20, Pho20, Pho20fur, and Pho20fur was determined by using a modified p-nitrophenyl phosphate assay (11). Cultures of Fe-limited cells grown overnight were used to inoculate secondary cultures in BHI broth under either Fe-replete (36 μM FeSO4) or Fe-limited (100 μM EDDHA) conditions. At stationary phase, 1 ml of the culture was centrifuged, and the cell pellet was resuspended in a solution containing 10 mM Tris (pH 8) and 100 mM NaCl. Cell densities of the resuspensions were adjusted to an OD600 of 0.4 to 0.8. An aliquot containing 500 μl of the cell suspension was added to an equal volume of 1 M Tris (pH 8), and the reaction was initiated by the addition of 100 μl of 4% p-nitrophenyl phosphate. Following incubation at room temperature, samples were observed for development of a yellow color, at which point 100 μl of 1 M KH2PO4 was added to terminate the reaction. After a brief centrifugation to pellet debris, the OD420 and OD550 of each reaction were recorded, and the relative alkaline phosphatase activity was calculated by use of the following formula (21): {1,000[OD420 – 1.75(OD550)]}/(t)(0.5)(OD600), where t is the time of the reaction in minutes. Relative enzyme activities are reported as the means of triplicate assays.
RNase protection assay (RPA). Construction of the DNA templates for the bhuR and bhuRS probes (pERM29 and pERM30, respectively) was described previously (22). PCR was used to construct the templates for the rhuI and rhuIR probes. Primers rhuIRPA1 (5'-CCGCTCGAGTCCATCGCTTCTATGCT-3' [the XhoI site is underlined]) and rhuIRPA2 (5'-CCCAAGCTTACGGTAGTGATTGAGCA-3' [the HindIII site is underlined]) amplified a 215-bp fragment within the rhuI ORF. A 254-bp fragment which contains the last 171 bp of rhuI and the first 86 bp of the rhuR 5' end was amplified by primers IRRPA1 (5'-CCGCTCGAGTGTCTGCTCTGCCCC-3' [the XhoI site is underlined]) and IRRPA2 (5'-CCCAAGCTTCTGCCCCGAAGCAAGA-3' [the HindIII site is underlined]). For both primer sets, the PCR conditions were as follows: the PCR mixture contained 1x PCRx buffer (Promega), 10% DMSO, 100 μM dNTPs, 1 μM each oligonucleotide primer, and 1 to 10 U of Taq polymerase, and PCR was carried out at 95°C for 45 s, 45°C for 45 s, and 72°C for 60 s for 30 cycles. The resulting amplicons were ligated into pBluescript KS(–) at the XhoI and HindIII sites. The sequences of pAEK32, the rhuI probe template, and pAEK33, the rhuIR probe template, were confirmed by automated DNA sequencing. To construct the PbhuR template, pAEK31, a 195-bp fragment, was released from pDJM31 by digestion with EcoRI and BamHI. This fragment, containing 31 bp of the 3' end of rhuR, the entire 102-bp intergenic region, and 47 bp of the 5' end of bhuR, was ligated into the EcoRI and BamHI sites of pBluescript KS(–).
Antisense RNA probes for RPAs were generated by using a MaxiScript in vitro transcription kit (Ambion Inc., Austin, Tex.). Plasmids pERM29 and pERM30 were linearized with XhoI and transcribed using T3 polymerase. Plasmids pAEK31, pAEK32, and pAEK33 were linearized with EcoRI and transcribed using T7 polymerase. The probes, generated by incorporating [-32P]CTP, were labeled according to the manufacturer's protocol, gel purified using an 8 M urea-5% polyacrylamide gel, and eluted from the gel at 37°C overnight using probe elution buffer (HybeSpeed RPA kit; Ambion). The activity of each probe was determined by using a Wallac (Turku, Finland) 1409 liquid scintillation counter. RPAs were performed by using a HybSpeed RPA kit (Ambion) according to manufacturer's protocol. A total of 15 μg of RNA was used in assays with the PbhuR, rhuI, and rhuIR probes and 5 μg of RNA was used in assays with the bhuR and bhuRS probes. Control reactions, in which pAEK31 was linearized with BamHI and transcribed with T3 polymerase generating labeled sense-strand RNA, were used to ensure that protected fragments were not the products of contaminating DNA in the RNA samples. The final reactions were separated by electrophoresis through an 8 M urea-5% polyacrylamide gel. Gels were dried and exposed to blue sensitive autoradiographic film (Marsh Bioproducts, Rochester, N.Y.) at –80°C using an intensifying screen.
Nucleotide sequencing and analysis. Nucleotide sequencing was performed at the Biopolymer Facility at Roswell Park Cancer Institute (Buffalo, N.Y.). Sequence analysis was performed by using the Wisconsin Package version 9.0 (Genetics Computer Group, Madison, Wis.) and ClustalW (http://www.ebi.ac.uk/clustalw).
RESULTS
Primer extension analysis of PrhuI and PbhuR. As an initial step to define the transcriptional architecture of the rhuIR-bhuRSTUV (rhu-bhu) locus, the transcriptional start sites of PrhuI and PbhuR were identified by use of primer extension. For the extension reactions, total RNA was isolated from cultures of B. avium 4169rif which had been cultured in CDM under conditions of Fe limitation. For comparison, total RNA was also obtained from 4169rif cultured under Fe-replete conditions. To evaluate the capacity of the cells to respond to heme as a putative source of Fe and as an inducer of the rhu-bhu locus, total RNA was also isolated from cultures of Fe-starved 4169rif to which heme had been added. In those cultures, the broth was supplemented with EDDHA to remove any contaminating Fe that might have been present in the hemin stock solution.
Primer extension experiments revealed that transcription from PrhuI was strongly repressed in Fe-replete cultures (Fig. 1A). In Fe-limiting growth, transcription from PrhuI initiated from a G residue located 28 bases upstream from the rhuI start codon. Analysis of the region of DNA which contained this G residue showed that this transcriptional start site was located within a putative Fur box which is optimally spaced from –35 and –10 elements that are predicted to be 70 dependent (17). These results indicated that transcription from PrhuI is regulated by Fe, perhaps in a Fur-dependent manner. The observations were also consistent with those reported for B. pertussis (28). Unexpectedly, when hemin was the sole source of Fe, transcriptional initiation in 4169rif was significantly increased from PrhuI and suggests that heme is an inducer of this promoter.
As was observed for PrhuI, transcription from PbhuR in cells obtained from Fe-replete cultures was not detected (Fig. 1B). Hemin supplementation elicited initiation of transcription at an A residue located 37 bases upstream from the bhuR start codon, a pattern of transcriptional initiation which was consistent with previous experiments that demonstrated hemin-dependent expression and induction of bhuR (17). Surprisingly, when cells were cultured under Fe-limiting growth conditions in the absence of hemin, the level of transcription from PbhuR in those cells was similar to the level of transcription from PrhuI in cells cultured under equivalent conditions of Fe limitation. This pattern of transcription was in contrast to the pattern of transcriptional initiation reported for B. pertussis and Bordetella bronchiseptica. In those species, heme-independent bhuR expression originated solely from PhurI (28).
To discount the possibility that PbhuR-dependent extension products in the primer extension experiments were artifacts caused by a premature termination of the extension reactions generated by inhibitory secondary structures in the RNA located proximal to the putative start site, primer extension analysis was also performed using an artificial RNA substrate. RNA for this control experiment was obtained from in vitro transcription of pAEK31, a plasmid containing a DNA fragment encompassing the 3' end of rhuR, the rhuR-bhuR intergenic region, and the 5' end of bhuR cloned into pBluescript. When primer extension reactions were performed on this artificial RNA substrate, the extension reaction proceeded through the putative start site and into the rhuR sequences (data not shown). In brief, these results indicated that the start site revealed by the primer extension product for the PbhuR site was not an artifact caused by premature termination of reverse transcriptase but is likely to be the actual initiation site for PbhuR transcriptional initiation in vivo.
Comparison of the B. avium rhuR-bhuR and B. pertussis hurR-bhuR intergenic regions. Since B. avium and B. pertussis exhibited differences in PbhuR transcriptional patterns, the nucleotide sequences of the regions surrounding these promoters in the two species were compared in an attempt to identify structural motifs which might mediate those differences. While the stop codon of the rhuI ORF and the start codon of bhuR are separated in B. avium by 105 bp of DNA, the two corresponding sites in B. pertussis are separated by 207 bp. Homology between these two intergenic regions is most apparent when the PbhuR start site of B. avium and the PbhuR start site of B. pertussis (28) are aligned. In this orientation, the putative –35 and –10 elements are identical. Seventy-two nucleotides (nt) in the rhuR-bhuR intergenic region of B. pertussis, which are located downstream from the bhuR transcriptional start site, are absent in the corresponding intergenic region of B. avium. Computer analysis of the 72-bp region of B. pertussis did not reveal known regulatory motifs. While experiments have yet to reveal a new regulatory structure within this segment of DNA, it is reasonable to hypothesize that the 72-nt region contributes to the apparent differences in heme-independent bhuR transcription and polypeptide expression observed between B. avium and B. pertussis.
rhuI contributes to heme-independent expression of bhuR. When overexpressed from a plasmid in B. avium, recombinant rhuI induces both Fe-independent and heme-independent expression of bhuR (17). To further assess the role of rhuI on PbhuR activity, a mutant of 4169rif which had a precise deletion of the rhuI ORF was engineered. Engineering of the rhuI mutant was such that the adjacent sequences encoding rhuR and the sequences encoding the rhuR-bhuR intergenic region were undisturbed. PbhuR activity was monitored by measuring the activity of an extrachromosomal PbhuR::lacZYA reporter in pDJM41 (Fig. 2A). Constitutive expression of recombinant rhuI in those strains was produced by introduction of pERM26, as needed. All strains were cultured in BHI broth under Fe-replete (36 μM FeSO4), Fe-limited (100 μM EDDHA), or Fe-limited plus hemin (100 μM EDDHA plus 5 μM hemin) growth conditions.
Analysis of 4169rif(pDJM41, pRK415) cultured under Fe-replete growth conditions revealed only low-level expression of -galactosidase activity which increased only twofold when cells were cultured in Fe-limited broth (Fig. 2B). Supplementation of the Fe-limited broth with hemin, however, enhanced -galactosidase activity fivefold relative to Fe-replete conditions (Fig. 2B) (22), a pattern of expression consistent with prior reports which indicated that heme was a positive inducer of PbhuR (17, 26). In comparison to PbhuR activity in cells cultured under Fe-replete growth conditions, -galactosidase activity of 4169rifrhuI(pDJM41, pRK415) was also enhanced twofold when exposed to Fe limitation (Fig. 2B). Yet -galactosidase activity of 4169rifrhuI(pDJM41, pRK415) was not elevated when the Fe-limited culture broth was supplemented with hemin (Fig. 2B). Introduction of pERM26 into 4169rifrhuI(pDJM41) highly enhanced -galactosidase activity under all growth conditions (Fig. 2B).
These results indicated that rhuI was essential for heme-dependent activation of PbhuR. Fe-dependent expression of bhuR, however, was independent of rhuI (17). Yet it was also clear from these experiments that overexpression of a recombinant rhuI constitutively activated PbhuR in the absence of heme, which suggested that RhuI functions as a PbhuR-dependent sigma factor in the absence of heme.
Assessing the role of PrhuI on Fe-dependent, heme-independent bhuR expression. Fe-dependent, heme-independent expression of bhuR in B. pertussis and B. bronchiseptica is hypothesized to occur by readthrough transcription initiating from PhurI (28). Thus, expression of bhuR in cells cultured under Fe-limited, heme-free growth conditions would be expected to be absent in a mutant in which transcription is terminated at rhuR. To test this hypothesis, various mutants were constructed in which an cassette, a DNA cassette containing multiple translational and transcription terminators, was inserted into the chromosomes of 4169rif and Pho20 at a position 2 bp downstream of the TAA translational terminator of rhuR (5). Pho20 is a reporter mutant containing an in-frame phoA insertion into bhuR (Fig. 3A).
Using Pho20 and Pho20, Fe-dependent expression of bhuR was assessed by measuring alkaline phosphatase activity of cultures grown under Fe-replete or Fe-limited conditions. As previously reported, alkaline phosphatase activity of Pho20 was low when cells were cultured under Fe-replete conditions and increased slightly when cells were cultured under Fe-limiting conditions (Fig. 3B). These data were consistent with prior reports (17) which demonstrated that bhuR expression responded to Fe-dependent repression. When the PhoA activity of Pho20 was evaluated, it was found that the level of alkaline phosphatase activity of Pho20 was similar to the alkaline phosphatase activity of Pho20 when the mutants were cultured under Fe-limiting growth conditions. Remarkably, the alkaline phosphatase activity of Pho20 was essentially unchanged, even when Pho20 was cultured under Fe-replete conditions (Fig. 3B). This unexpected result suggested that the cassette likely imposed an artificial Fe-independent transcriptional activity that promoted expression of the bhuR::phoA fusion. It should be noted that similar problematical results were observed when a mutant containing an cassette was incorporated into the hurR-bhuR intergenic region of a plasmid-borne copy of the B. pertussis hurIR-bhuR-lacZ locus (28). In that case, introduction of the cassette stimulated an increase in reporter activity when the cells were cultured under Fe-replete conditions. Thus, these data obtained in B. avium and in B. pertussis using cassette mutants neither supported nor refuted the hypothesis that bhuR expression under conditions of Fe-limited growth was promulgated by readthrough transcription from PrhuI. Rather, it is obvious that the cassette is not a useful tool for experiments to artificially terminate transcription in either B. avium or B. pertussis and that results from such experiments need to be interpreted with care.
The role of Fur in expression of bhuR. Alkaline phosphatase activity of Pho20fur, a mutant of Pho20 in which the genomic copy of fur is genetically disrupted (17), was significantly higher than its Pho20 parent when cultured under Fe-replete conditions (Fig. 3B). These data supported a model of Fur regulation of Fe-dependent repression of bhuR (17). Under Fe-replete conditions, the alkaline phosphatase activity of Pho20fur was also significantly higher than the alkaline phosphatase activity of Pho20 (Fig. 3B). This pattern of expression of the reporter suggests that the artifactual problems inherent in cassette mutants can be overcome by evaluating the effects of the cassette on readthrough transcription in fur backgrounds in B. avium. Indeed, the results show that alkaline phosphatase activity of Pho20fur was reduced compared to the activity of Pho20fur (Fig. 3B) and indicated that Fe-dependent and Fur-dependent activity from PrhuI is at least partly responsible for bhuR expression.
Evaluation of PbhuR transcriptional activity. Primer extension analysis indicated that transcription initiated from both PrhuI and PbhuR when cells were cultured under conditions of Fe limitation in the presence or absence of hemin (Fig. 1). From these data, it was inferred that PbhuR is responsible for heme-independent expression of bhuR. These observations are in contrast to results observed for B. pertussis (28) in which PbhuR activity was solely heme dependent. In the Pho20fur and Pho20fur mutant strains, PrhuI activity was at least partially responsible for Fe-dependent bhuR expression (Fig. 3B). To further analyze heme-independent expression of bhuR in B. avium, the effects of the fur, rhuI, and cassette mutations on PbhuR activity were examined using primer extension as a direct measurement of transcriptional activity from that promoter.
Primer extension reactions using total RNA harvested from BHI cultures of 4169rif, 4169rifrhuI, 4169rif, and 4169riffur revealed a previously unknown activity of RhuI. As previously reported, PbhuR was inactive when 4169rif was cultured under conditions in which Fe was abundant (Fig. 1B and 4B). Under Fe-limiting conditions, however, significant transcription from PbhuR was detected regardless of the presence or absence of hemin (Fig. 1B and 4B). Transcription from PbhuR was absent in the rhuI mutant under any condition tested (Fig. 4B), a result which was consistent with a model in which rhuI is essential for Fe-dependent PbhuR activity whether or not the cell is grown in the presence or absence of hemin. In contrast, the PbhuR transcription profile of the cassette mutant was indistinguishable from that of wt 4169rif (Fig. 4B), thus indicating that the transcripts detected by primer extension did not originate from PrhuI or other upstream promoters. Interestingly, a modest transcriptional activity was observed in RNA obtained from Fe-replete cultures of the fur mutation (Fig. 4B). This observation suggested that Fe-dependent repression of PbhuR activity is mediated by Fur. Taken together, these results supported the hypothesis that PbhuR is active under Fe-limiting conditions irrespective of heme and that PbhuR requires a functional rhuI for transcriptional activity. These results also strongly implied that PbhuR activity is repressed by Fe in a Fur-dependent manner.
Evaluation of PbhuR activity by RNase protection analysis. As a complementary approach, the transcriptional architecture of the genomic rhu-bhu locus was also evaluated using RNase protection. Results from primer extension analysis of 4169rif were consistent with a model in which transcription from PbhuR initiates under Fe-limiting growth conditions in the absence of heme (Fig. 1B and 4B). While results from promoter-reporter analyses of 4169rifrhuI (Fig. 2B), Pho20, and Pho20fur (Fig. 3B) were inconclusive with respect to PbhuR activation, primer extension results from these mutants strongly suggested that Fe-dependent, heme-independent transcription originates from PbhuR, an observation which is not consistent with previous reports of transcriptional analysis of hurIR-bhuR of B. pertussis (28). Furthermore, promoter-reporter analyses of Pho20fur and Pho20fur (Fig. 3B) clearly showed that bhuR expression originated, at least in part, from PrhuI. To explicate the details of PbhuR activity in B. avium, the effects of rhuI and cassette mutations on transcription of rhuIR-bhuRSTUV were examined using RNase protection analysis. Antisense probes directed against three different regions of the rhu-bhu locus were designed to evaluate expression of the following different regions: (i) a region encompassing the 3' end of rhuI and the 5' end of rhuR (rhuIR), (ii) the entire rhuR-bhuR intergenic region (PbhuR), and (iii) a region encompassing the 5' end of bhuR (bhuR) (Fig. 5A).
None of the three antisense probes were protected from RNase A digestion by the addition of total RNA isolated which was obtained from Fe-replete cultures of 4169rif, indicating that transcripts from the rhu-bhu locus are highly repressed under these growth conditions (Fig. 5B). All three antisense probes were, however, protected by total RNA isolated from 4169rif cultured under conditions of Fe limitation (Fig. 5B). Sizes of the protected fragments in each case were determined by comparison to a commercial RNA standard ladder (data not shown). A fragment of approximately 237 nt, equivalent in size to the full-length antisense probe of the rhuIR region, was protected by total RNA from 4169rif, demonstrating that rhuI and rhuR are cotranscribed on a polycistronic mRNA. Likewise, total RNA from 4169rif cultured under Fe-limiting conditions protected a fragment of approximately 182 nt in length, consistent with the full-length antisense probe of PbhuR which targeted the 3' end of rhuR, the rhuR-bhuR intergenic region, and the 5' end of bhuR. Protection of the entire length of this second antisense probe indicated that transcription proceeded from rhuR and continued through the intergenic region and into bhuR. The intensity of the 182-nt PbhuR-protected fragment was considerably lower than the intensity of the rhuIR-protected fragment, suggesting that only some of the transcripts originating upstream from rhuI continued through the intergenic region and into bhuR. A 230-nt protected fragment representing the full-length antisense probe of a region within the bhuR ORF exhibited a signal which was higher in intensity than that of the fragment from the probe homologous to the rhuR-bhuR intergenic region. These results suggested that transcripts originating upstream from rhuI were not the sole source of bhuR-encoding transcripts produced by the cell. In fact, a shorter, 80-nt protected fragment was evident in experiments using the PbhuR antisense probe (Fig. 5B). The length of this shorter protected fragment corresponded to the PbhuR start site as determined by primer extension (Fig. 1B). Taken together, the combined intensities of the shorter 80-nt and full-length 182-nt protected fragments from the PbhuR antisense probe were essentially equivalent to the signal produced by protection of the bhuR antisense probe. When hemin was added to the culture broth, the intensities of the shorter 80-nt PbhuR-protected fragment and the bhuR-protected fragment increased significantly (Fig. 5B). This observation was consistent with the patterns of heme induction of bhuR from PbhuR reported by Kirby et al. (17). A slight increase in the intensity of the rhuIR-protected fragment was also detected upon addition of hemin to the culture medium (Fig. 5B). These data supported the results shown in Fig. 1A that demonstrated that PrhuI activity was induced by heme.
Total RNA isolated from the rhuI mutant provided the same degree of protection of rhuIR-specific probes as did total RNA from 4169rif when cultured under similar growth conditions (Fig. 5B). These data confirmed that rhuIR expression was not autoregulated. These results also indicate that unlike PbhuR, PrhuI is induced by heme but not via the BhuR-RhuR-RhuI signal cascade. These data suggest the presence of an additional heme-dependent regulatory mechanism. The smaller size of the rhuIR-protected fragment was consistent with the size of the fragment predicted to be protected by total RNA obtained from the mutant in which the rhuI ORF had been removed (i.e., 4169rifrhuI). Protection of the antisense bhuR probe was greatly reduced in RNA isolated from Fe-limited cultures of the rhuI mutant cultured in the presence or absence of hemin. The pattern of protection of the antisense PbhuR probe was also influenced by the rhuI mutation. The shorter 80-nt fragment of PbhuR was not protected by RNA isolated from the rhuI mutant grown under Fe-limited conditions in the presence or absence of hemin. The larger 182-nt fragment of PbhuR, however, was protected to a degree equal to the protection of the bhuR antisense probe (Fig. 5B). The results again confirmed that some expression of bhuR is dependent upon PrhuI activity. The results also showed that transcription from PbhuR under Fe-limiting conditions not only is rhuI-dependent and hemin-induced but also occurs in the absence of hemin, which is consistent with the primer extension results (Fig. 1B and 4B).
Total RNA isolated from the 4169rif protected the rhuIR and bhuR antisense probes in a manner similar to that of RNA from 4169rif cells cultured under Fe-limiting conditions in either the presence or absence of hemin (Fig. 5B). Analysis of the PbhuR region showed that protection of the shorter 80-nt fragment was unaffected by the cassette, whereas protection of the 182-nt full-length PbhuR antisense probe was eliminated (Fig. 5B). These results demonstrated that transcriptional initiation at PbhuR is not dependent upon heme or PrhuI and indicated that Fe-dependent, heme-independent expression of bhuR originates from two promoters (PrhuI and PbhuR). Furthermore, it is clear that PbhuR activity requires rhuI.
Fe-dependent regulation of PrhuI and PbhuR activities in a B. avium fur mutant strain. rhuI is Fur regulated in B. avium (17). Also, Fe-dependent repression of bhuR was found to be influenced by fur (17). Consistent with those observations, the rhuR-bhuR intergenic region was reported to contain a nucleotide sequence with weak homology to the E. coli Fur box consensus. This region also exhibited weak but significant titration activity in the E. coli Fur titration assay (FURTA) strain. Thus, experiments were designed to establish whether or not Fur had a direct role in Fe-dependent regulation of PbhuR (17). Promoter-reporter assays using the Pho20fur and Pho20fur strains indicated that PrhuI contributed to Fur-dependent bhuR expression (Fig. 3). In contrast, primer extension analysis showed that both PrhuI and PbhuR were inactive under Fe-replete conditions and induced only when Fe was limiting (Fig. 1). Further analysis showed that PbhuR in a fur mutant was slightly activated under Fe-replete conditions (Fig. 4B).
To further evaluate expression of the rhu-bhu locus in a fur mutant, RNase protection assays were conducted. Total RNA isolated from Fe-replete and Fe-limited cultures of 4169rif and 4169riffur were utilized for these RNase protection analyses. In addition to the antisense probes used in the previous RNase protection assays (Fig. 5A), two additional probes were employed to broaden the transcriptional analysis of the rhu-bhu locus: an rhuI antisense probe which was directed against the 5' end of rhuI and a bhuRS antisense probe which was directed against a region encompassing the 3' end of bhuR and the 5' end of bhuS (Fig. 6A). All five antisense probes were protected from RNase A digestion by total RNA isolated from 4169rif cultures grown under Fe-limiting conditions but were not protected by total RNA isolated from 4169rif cultured under Fe-replete conditions (Fig. 6B). These data, along with the data reported in Fig. 5B, supported a model in which the rhu-bhu locus including rhuI, rhuR, bhuR, and bhuS was expressed in an Fe-dependent manner. The presence of the 182- and 80-nt protected fragments of the antisense PbhuR probe in reactions containing total RNA isolated from Fe-limited cultures is likely the result of readthrough transcription from PrhuI and of transcription from PbhuR, respectively. This pattern of expression indicated that transcriptional activities of both PrhuI and PbhuR were Fe-regulated events. Surprisingly, the intensity of the 182-nt protected fragment of the PbhuR antisense probe was equal to the intensities of the protected fragments of the rhuI and rhuIR antisense probes (Fig. 6B). This observation was interpreted as strongly supportive of a model in which all transcripts from PrhuI extended past the rhuR-bhuR intergenic region. This result is contrary to data reported in Fig. 5 which indicated that not all transcripts from PrhuI proceeded through the intergenic region into bhuR. Although the discrepancy between the two experiments has not been resolved, it is possible that the difference is indicative of an unknown mechanism controlling differential transcriptional termination in the rhu-bhu locus for which the regulatory conditions are yet to be determined.
Total RNA isolated from Fe-replete cultures of 4169riffur protected the antisense probes directed against regions of the rhu-bhu locus. Also, Fe-dependent repression from rhuI to bhuS was relieved in the fur mutant. All fragments of antisense probes protected by RNA from Fe-replete cultures of the fur mutant are equally protected by RNA from 4169rif cells grown under Fe-limited conditions, with one exception. The 80-nt protected fragment, which likely is representative of transcription from PbhuR, was only slightly evident in the fur mutant when cells were cultured under Fe-replete conditions. Longer exposure of the autoradiograph revealed that this fragment was protected by RNA from Fe-replete cultures of 4169riffur but not by RNA from Fe-replete cultures of 4169rif (data not shown). Additionally, -galactosidase activity from the PbhuR::lacZYA reporter in pDJM41 was measured in 4169rif and 4169riffur strains, both cultured under Fe-replete and Fe-limited growth conditions (Fig. 7B). The results show that Fe-dependent repression of PbhuR is relieved in the absence of fur. This evidence, taken together with previous results reported by Kirby et al. (17), strongly implicated fur in regulating Fe-dependent PbhuR activity.
DISCUSSION
In B. avium, heme-dependent expression of bhuRSTUV requires the outer membrane heme receptor BhuR, the ECF factor RhuI, and the ECF factor activator RhuR (16, 17, 22). Paradoxically, this regulatory mechanism requires bhuR to be expressed in the absence of heme inducer. In B. pertussis and B. bronchiseptica, heme-independent expression of bhuR was found to originate from PhurI, an Fe-regulated promoter located immediately upstream from the hurIR-bhuRSTUV gene cluster (28). This gene cluster is orthologous to the B. avium heme utilization locus rhuIR-bhuRSTUV (22). Here, we present evidence that in B. avium, heme-independent transcription of bhuR originates from PbhuR, a proximal RhuI-dependent promoter, as well as from PrhuI, a distal RhuI-independent promoter. This mechanism ensures the synthesis of adequate amounts of BhuR in the absence of heme induction for subsequent heme sensing. Data also point toward the likelihood that both promoters are repressed by Fur under Fe-sufficient conditions and suggest the necessity for swift repression of the heme acquisition system once the demand for Fe has been met.
Transcription from PrhuI initiates 28 bases upstream from the rhuI ORF and is evident only when Fe is limiting (Fig. 1A). Transcription from PrhuI is unaffected by a rhuI mutation, indicating that this promoter is not autoregulated and, therefore, not likely to be activated by heme via the BhuR-RhuR-RhuI signal cascade (Fig. 5B). The observed effect that the addition of hemin to the medium slightly enhanced PrhuI activity was unexpected and suggests the presence of an additional heme-dependent signal cascade. An alternative explanation, however, in which PrhuI activity is indirectly affected by heme cannot be ruled out. Further examination of PrhuI is needed to assess the role of heme in rhuI-independent expression from this promoter. The PrhuI transcriptional start site is optimally spaced from 70-like promoter elements and is located within a putative Fur box in a region that exhibited positive FURTA activity (17). Alkaline phosphatase activities of the Pho20, Pho20fur, and Pho20fur strains support Fur-dependent regulation of PrhuI and indicated that at least a portion of the transcripts from PrhuI read through the rhuR-bhuR intergenic region (Fig. 3B). RNase protection analysis provides evidence that the readthrough transcripts include the coding regions of bhuR and bhuS, at the very least, and likely include the other genes of the bhuR locus (Fig. 5B and 6B). These results, consistent with those reported for B. pertussis and B. bronchiseptica (28), describe a suitable mechanism for the heme-independent expression of bhuR observed in all three species.
Transcriptional analyses of the rhuR-bhuR intergenic region indicated that PrhuI is not the sole active promoter driving bhuR expression. Transcription from PbhuR starts 37 bases upstream from bhuR (Fig. 1B). RNase protection analysis using an antisense probe directed against the rhuR-bhuR intergenic region yielded, in addition to a 182-nt protected fragment that represents readthrough transcription, a shorter 80-nt protected fragment whose length is coincident with the PbhuR transcriptional start site (Fig. 5B). Surprisingly, transcription from PbhuR occursunder Fe-limiting conditions in the absence of heme, although to a much lesser degree than when heme is present (Fig. 2B, 4B, and 5B). Previous studies have shown that heme and heme-containing proteins are inducers of bhuR expression (22). Mutational studies confirmed that RhuI is required for PbhuR activity and showed that RhuI can also function in the absence of heme inducer (16, 17) (Fig. 3B, 4B, and 5B). Previous experiments have shown that, when overexpressed, RhuI constitutively activates bhuR expression (17). It is possible that RhuI has a basal activity for PbhuR induction that is enhanced in the presence of heme through signaling via BhuR and RhuR.
In B. avium, both PrhuI and PbhuR participate in heme-independent bhuR expression. A similar rhuI-dependent, heme-independent PbhuR activity, observed in primer extension analysis of B. bronchiseptica, was reported in a previous study, but no explanations for this activity were proffered (28). Instead, heme-independent expression of bhuR in B. bronchiseptica and B. pertussis was reported to originate exclusively from PhurI (28). It is unexpected that transcription from PbhuR in B. avium should be controlled differently from that in B. pertussis and B. bronchiseptica. In all three organisms, expression of bhuRSTUV is induced by heme via a system that requires BhuR, the ECF factor (HurI), and the coregulator (HurR) (RhuI and RhuR for B. avium), the expression of which are Fe regulated in a Fur-dependent manner. The differences in the regulation of bhuR in these organisms could be attributable to a region of DNA found in the hurIR-bhuR intergenic region of B. pertussis and B. bronchiseptica but absent in the rhuR-bhuR intergenic region of B. avium. This span of additional DNA could contain structural cis-acting elements that proffer additional regulation.
Employing two promoters to control heme-independent bhuR expression would appear redundant for B. avium. It is feasible, however, that the regulatory system involving the two promoters evolved to enable the bacterium to respond to different sources of heme. Several observations support this model: (i) acquisition of both heme and heme-containing proteins (e.g., hemoglobin, myoglobin, catalase, etc.) requires bhuR (22); (ii) growth of B. avium with hemin as its sole Fe source does not require rhuR, but a rhuR mutant has reduced capacity to utilize hemoglobin (16); and (iii) BhuR is expressed in two forms in the outer membrane, a 91-kDa protein encoded by the entire bhuR ORF and a truncated 82-kDa protein in which the N-terminal 104 amino acids of the full-size protein have been removed by a proteolytic cleavage (22). The 82-kDa form is preferentially expressed when B. avium encounters heme. The interplay between PrhuI and PbhuR may be necessary to facilitate a response by B. avium to different biological sources of heme. Experiments to evaluate this model are ongoing.
Transcriptional analysis using RNA from Fe-replete cultures of strain 4169riffur revealed that Fe-dependent repression of the rhuIR-bhuRSTUV locus was relieved in a fur mutant (Fig. 6B). Protected fragments representing transcripts from rhuI and rhuR from 4169riffur cultured under Fe-replete conditions were equally abundant, thus confirming that rhuI and rhuR are transcribed on a single message that is repressed by fur in an Fe-dependent manner. It is clear from the RNase protection experiments that synthesis of this polycistronic mRNA sometimes extends into bhuR and bhuS. Although not determined in these experiments, it is likely that synthesis of the polycistronic mRNA extends into bhuTUV, thus encoding all proteins of the heme uptake system. However, RNase protection experiments also demonstrated that transcripts originating from PrhuI sometimes terminate after rhuI (Fig. 5B). The conditions and the cognate signal(s) that regulate whether polymerization of transcripts continues or terminates after synthesis of rhuI and rhuR have not been established. There is no evidence of a bhuR-associated feedback regulation. However, it is possible that one of the genes located downstream of bhuR may participate in regulating synthesis of the polycistronic mRNA. The role of downstream genes (bhuSTUV) in the control of readthrough transcription, however, is yet to be investigated.
It is clear that Fur has a potent role in governing expression of rhuI, rhuR, bhuR, and bhuS. A region with weak (8 of 19 nucleotides) homology to the consensus Fur box was located between the putative –35 and –10 elements of PbhuR (17). The Fur box appears to be functionally active, since a segment of DNA containing this region exhibited weak FURTA activity (17), implying that it has the capacity to functionally interact with Fur. Transcriptional analyses of RNA isolated from Fe-replete cultures of a fur mutant of B. avium supported the hypothesis that Fur represses the RhuI-dependent PbhuR (Fig. 4B). Furthermore, -galactosidase expressed from PbhuR on pDJM41, which does not harbor DNA containing PrhuI, is still Fe responsive, even in the rhuI mutant (Fig. 3B). Although not unequivocal, these data provided strong evidence that PbhuR activity is highly dependent upon Fe, likely via an interaction with Fur. It was predicted that experiments utilizing the cassette would be useful in isolating PbhuR from PrhuI for evaluating the effect of Fur on PbhuR activity. Unfortunately, any effects that Fur exerted on PbhuR were masked in these experiments by an apparent artifactual Fe-independent transcriptional activity originating from within the cassette. For that reason, the cassette is not a particularly desirable tool for isolating transcriptional units in the bordetellae. As a better alternative strategy for investigating putative interactions of Fur with PbhuR, future experiments will evaluate PbhuR activity in an engineered mutant in which PrhuI is absent.
The need for Fur-dependent regulation at PrhuI and PbhuR sites may indicate that rapid downregulation of expression of bhuRSTUV is desirable when heme and the other components of the signal transduction cascade are still present in the cell but when the Fe requirements of the cell have already been met. Fur may be a strong participant in a regulatory scheme which determines whether transcription from PrhuI terminates after rhuR or reads through to bhuRSTUV. Given the multiple forms of BhuR, the multiple inducers that stimulate bhuR expression, the multiple promoters that drive this expression, and the complexity of bhuR regulation, it is apparent that a complete description of the regulation and functioning of the rhuIR-bhuRSTUV heme utilization locus is yet to be fully revealed. Elucidation of the details of the RhuI-RhuR-BhuR signaling cascade and the characterization of the genes downstream from bhuR may provide valuable insights into Fur regulation, Fe acquisition, and virulence of the bordetellae.
ACKNOWLEDGMENTS
Funds to support this investigation were made available to T.D.C. from the School of Medicine and Biomedical Sciences, University of Buffalo, State University of New York. A.E.K. was supported by an NIH training grant (AIO7614) awarded to the Witebsky Center for Microbial Pathogenesis and Immunology and by a Presidential Fellowship administered by the Office of the Provost, the University of Buffalo. N.D.K. was supported by an NIH training grant (T32 DE007034) awarded to the Department of Oral Biology in the School of Dental Medicine at the University of Buffalo.
REFERENCES
1. Angerer, A., S. Enz, M. Ochs, and V. Braun. 1995. Transcriptional regulation of ferric citrate transport in Escherichia coli K-12. Fecl belongs to a new subfamily of sigma 70-type factors that respond to extracytoplasmic stimuli. Mol. Microbiol. 18:163-174.
2. Beaumont, F. C., H. Y. Kang, T. J. Brickman, and S. K. Armstrong. 1998. Identification and characterization of alcR, a gene encoding an AraC-like regulator of alcaligin siderophore biosynthesis and transport in Bordetella pertussis and Bordetella bronchiseptica. J. Bacteriol. 180:862-870.
3. Braun, V., and H. Killmann. 1999. Bacterial solutions to the iron-supply problem. Trends Biochem. Sci. 24:104-109.
4. Chauhan, S., and M. R. O'Brian. 1997. Transcriptional regulation of delta-aminolevulinic acid dehydratase synthesis by oxygen in Bradyrhizobium japonicum and evidence for developmental control of the hemB gene. J. Bacteriol. 179:3706-3710.
5. Connell, T. D., A. Dickenson, A. J. Martone, K. T. Militello, M. J. Filiatraut, M. L. Hayman, and J. Pitula. 1998. Iron starvation of Bordetella avium stimulates expression of five outer membrane proteins and regulates a gene involved in acquiring iron from serum. Infect. Immun. 66:3597-3605.
6. Dean, C. R., S. Neshat, and K. Poole. 1996. PfeR, an enterobactin-responsive activator of ferric enterobactin receptor gene expression in Pseudomonas aeruginosa. J. Bacteriol. 178:5361-5369.
7. de Lorenzo, V., S. Wee, M. Herrero, and J. B. Neilands. 1987. Operator sequences of the aerobactin operon of plasmid ColV-K30 binding the ferric uptake regulation (fur) repressor. J. Bacteriol. 169:2624-2630.
8. Enz, S., V. Braun, and J. H. Crosa. 1995. Transcription of the region encoding the ferric dicitrate-transport system in Escherichia coli: similarity between promoters for fecA and for extracytoplasmic function sigma factors. Gene 163:13-18.
9. Enz, S., S. Mahren, U. H. Stroeher, and V. Braun. 2000. Surface signaling in ferric citrate transport gene induction: interaction of the FecA, FecR, and FecI regulatory proteins. J. Bacteriol. 182:637-646.
10. Escolar, L., J. Perez-Martin, and V. de Lorenzo. 1999. Opening the iron box: transcriptional metalloregulation by the Fur protein. J. Bacteriol. 181:6223-6229.
11. Gutierrez, C., J. Barondess, C. Manoil, and J. Beckwith. 1987. The Use of TnphoA to detect genes for cell envelope proteins subject to a common regulatory stimulus. Analysis of osmotically related genes in Escherichia coli. J. Mol. Biol. 19:289-297.
12. Hamza, I., Z. Qi, N. D. King, and M. R. O'Brian. 2000. Fur-independent regulation of iron metabolism by Irr in Bradyrhizobium japonicum. Microbiology 146:669-676.
13. Harle, C., I. Kim, A. Angerer, and V. Braun. 1995. Signal transfer through three compartments: transcription initiation of the Escherichia coli ferric citrate transport system from the cell surface. EMBO J. 14:1430-1438.
14. Heinrichs, D. E., and K. Poole. 1996. PchR, a regulator of ferripyochelin receptor gene (fptA) expression in Pseudomonas aeruginosa, functions both as an activator and as a repressor. J. Bacteriol. 178:2586-2592.
15. 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.
16. Kirby, A. E., N. D. King, and T. D. Connell. 2004. RhuR, an extracytoplasmic function sigma factor activator, is essential for heme-dependent expression of the outer membrane heme and hemoprotein receptor of Bordetella avium. Infect. Immun. 72:896-907.
17. Kirby, A. E., D. J. Metzger, E. R. Murphy, and T. D. Connell. 2001. Heme utilization in Bordetella avium is regulated by RhuI, a heme-responsive extracytoplasmic function sigma factor. Infect. Immun. 69:6951-6961.
18. Koster, M., W. van Klompenburg, W. Bitter, J. Leong, and P. Weisbeek. 1994. Role for the outer membrane ferric siderophore receptor PupB in signal transduction across the bacterial cell envelope. EMBO J. 13:2805-2813.
19. Litwin, C. M., and S. B. Calderwood. 1993. Role of iron in regulation of virulence genes. Clin. Microbiol. Rev. 6:137-149.
20. Litwin, C. M., and J. Quackenbush. 2001. Characterization of a Vibrio vulnificus LysR homologue, HupR, which regulates expression of the haem uptake outer membrane protein, HupA. Microb. Pathog. 31:295-307.
21. Miller, J. H. 1992. A short course in bacterial genetics. Cold Spring Harbor Laboratory Press, Plainview, N.Y.
22. Murphy, E. R., R. E. Sacco, A. Dickenson, D. J. Metzger, Y. Hu, P. E. Orndorff, and T. D. Connell. 2002. BhuR, a virulence-associated outer membrane protein of Bordetella avium, is required for the acquisition of iron from heme and hemoproteins. Infect. Immun. 70:5390-5403.
23. Ochs, M., S. Veitinger, I. Kim, D. Welz, A. Angerer, and V. Braun. 1995. Regulation of citrate-dependent iron transport of Escherichia coli: fecR is required for transcription activation by FecI. Mol. Microbiol. 15:119-132.
24. Prentki, P., and H. M. Krisch. 1984. In vitro insertional mutagenesis with a selectable DNA fragment. Gene 29:303-313.
25. Schmitt, M. P. 1999. Identification of a two-component signal transduction system from Corynebacterium diphtheriae that activates gene expression in response to the presence of heme and hemoglobin. J. Bacteriol. 181:5330-5340.
26. Vanderpool, C. K., and S. K. Armstrong. 2001. The Bordetella bhu locus is required for heme iron utilization. J. Bacteriol. 183:4278-4287.
27. Vanderpool, C. K., and S. K. Armstrong. 2003. Heme-responsive transcriptional activation of Bordetella bhu genes. J. Bacteriol. 185:909-917.
28. Vanderpool, C. K., and S. K. Armstrong. 2004. Integration of environmental signals controls expression of Bordetella heme utilization genes. J. Bacteriol. 186:938-948.
29. Van Hove, B., H. Staudenmaier, and V. Braun. 1990. Novel two-component transmembrane transcription control: regulation of iron dicitrate transport in Escherichia coli K-12. J. Bacteriol. 172:6749-6758.
30. Weinberg, E. D. 1993. The development of awareness of iron-withholding defense. Perspect. Biol. Med. 36:215-221.(Natalie D. King, Amy E. K)