Iron-Responsive Repression of Urease Expression in Helicobacter hepaticus Is Mediated by the Transcriptional Regulator Fur

Bacterial strains, plasmids, and growth conditions. H. hepaticus strain ATCC 51449 (22, 41) and its isogenic fur and perR mutants (this study) were routinely cultured at 37°C under microaerobic conditions (10% CO₂, 5% O₂, and 85% N₂) on Dent agar (Oxoid) (44). Liquid growth was performed in brucella broth (Difco) supplemented with 0.2% (wt/vol) β-cyclodextrins (Fluka) (BBC). Cultures were inoculated with a starting optical density at 600 nm (OD₆₀₀) of 0.05 and were shaken at 40 rpm. For growth under iron-restricted conditions, 20 μM desferoxamine mesylate (Desferal; Sigma) was added to brucella broth before addition of the cyclodextrins. Iron-replete media were prepared by the addition of 100 μM FeCl₃ (Sigma) to the iron-restricted medium. In case of transcriptional analyses (primer extension analysis), iron restriction was achieved by the addition of 2,2′-dipyridyl (Sigma) or 2,2′-dipyridyl and FeCl₃ to final concentrations of 100 μM to an overnight culture of H. hepaticus, followed by incubation for 30 min at 37°C (14). The reason for the use of different chelators was that H. hepaticus RNA isolated after overnight growth in the presence of Desferal was of lower quality than RNA isolated from cells incubated with dipyridyl (data not shown) and could not be used for primer extension analysis, whereas incubation with 2,2′-dipyridyl was too short to result in phenotypic changes (protein profile and urease activity). Both chelators have been used previously for iron restriction in H. pylori and gave similar results for iron-responsive regulation of iron uptake and iron storage genes (6, 14, 16). Incubation of H. hepaticus with 2,2′-dipyridyl resulted in the absence of growth, probably due to the toxicity of the chelator (data not shown). Escherichia coli strains ER1793 and DH5α were grown aerobically in Luria-Bertani medium (36) at 37°C. When indicated, growth media were supplemented with chloramphenicol and ampicillin to final concentrations of 20 μg/ml and 100 μg/ml, respectively.

Urease enzyme assay.Urease enzyme activity was determined in freshly sonicated lysates by measuring ammonia production from the hydrolysis of urea, as described previously (44). The concentration of ammonia in the samples was inferred from a standard NH₄Cl concentration curve. Enzyme activity was expressed as micromoles of urea substrate hydrolyzed per minute per milligram of protein.

Primer extension analysis.RNA was isolated from H. hepaticus with TRIzol reagent (Invitrogen), according to the manufacturer's instructions. Primer extension analyses were performed with the reverse primer HhureA-DIG (Table 1) essentially as described previously (17). The digoxigenin-labeled primer was annealed to 10 μg of total RNA from H. hepaticus strain ATCC 51449, and cDNA was synthesized after the addition of 5 units of avian myeloblastosis virus reverse transcriptase (Promega) and incubation for 1 h at 42°C. The cDNA product was separated on a 7% acrylamide-8 M urea sequencing gel and then blotted onto a nylon membrane (Roche), followed by chemiluminescent digoxigenin (DIG) detection (16). A sequencing reaction made with the same primer was used to determine the transcriptional start site.

Construction of H. hepaticus fur and perR mutants.The fur gene of H. hepaticus strain ATCC 51449 was amplified with primers Hhfur-mutF1 and Hhfur-mutR1 (Table 1) and cloned in pGEM-T Easy (Promega), resulting in plasmid pCB1. The fur gene was subsequently interrupted by insertion of the chloramphenicol resistance gene from pAV35 (47) in the unique BsmI site, resulting in plasmid pCB2. The perR gene of H. hepaticus strain ATCC 51449 was amplified with primers HhperR-mutF1 and HhperR-mutR1 (Table 1) and cloned in pGEM-T Easy (Promega), resulting in plasmid pCB3. The perR gene was subsequently interrupted by insertion of the chloramphenicol resistance gene from pAV35 (47) in the unique HindIII site, resulting in plasmid pCB4. Both pCB2 and pCB4 were first introduced into E. coli ER1793, and plasmids isolated from E. coli ER1793 were subsequently used for natural transformation of H. hepaticus 51449 (5). Chloramphenicol-resistant colonies isolated were designated 51449fur and 51449perR, respectively. Two colonies derived from independent transformations were tested, and both colonies gave identical results in all experiments. Correct allelic replacement of the wild-type fur and perR genes with the interrupted version was confirmed by PCR using combinations of the primers Hhfur-outF1/Hhfur-outR1 and HhperR-outF1/HhperR-outR1, respectively (Table 1).

Protein analysis. H. hepaticus ATCC 51449 was grown for 24 h in iron-restricted and iron-replete BBC, centrifuged at 4,000 × g for 10 min at room temperature, and resuspended in phosphate-buffered saline, pH 7.4, to a final OD₆₀₀ of 10. H. hepaticus cells were subsequently lysed by sonication for 15 s on ice with an MSE Soniprep 150 at amplitude 6. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% (wt/vol) polyacrylamide gel and stained with Coomassie brilliant blue. Immunoblotting was performed after electrotransfer of proteins from a 10% SDS-polyacrylamide gel to a nitrocellulose membrane (Roche). The blot was subsequently probed with antibodies raised in rabbits to Helicobacter felis UreA or UreB (Intervet International BV, Boxmeer, The Netherlands) (5). Bound antibodies were visualized with swine anti-rabbit antibodies labeled with alkaline phosphatase (Promega), with BCIP (5-bromo-4-chloro-3-indolylphosphate) and Nitro Blue Tetrazoleum (Promega) used as a substrate (5).

Recombinant expression of H. hepaticus Fur protein.The fur gene was amplified from H. hepaticus ATCC 51449 with primers Hhfur-overF1 and Hhfur-overR1 (Table 1). The resulting fragment was digested with BamHI and PstI and ligated into pASK-IBA7 (IBA, Gottingen, Germany) to create pCB10. The wild-type sequence of the fur gene was confirmed by DNA sequencing. H. hepaticus Fur was expressed with an N-terminal Streptag, which has been used with the H. pylori Fur, NikR, and CrdR proteins and which does not influence DNA-binding activity (17, 18, 48, 49); therefore, the Streptag was not removed prior to use. The recombinant protein was purified as described in the manufacturer's instructions (48). The recombinant protein was over 90% pure, as determined by staining with Coomassie brilliant blue following electrophoresis on 12% SDS-polyacrylamide gels. Purified protein was dialyzed and stored in 50 mM Tris (pH 8.0), 100 mM NaCl, 1 mM MgCl₂, 2 mM dithiothreitol (DTT), and 50% glycerol and was used without further purification for electrophoretic mobility shift and DNase I footprinting assays.

Electrophoretic mobility shift assays.DIG-labeled ureA promoter DNA was amplified with primers HhureAF1 and HhureA-DIG (Table 1). As a negative control, the promoter of the ksgA gene (HH1174) (41) was amplified with primers HhksgA-F1 and HhksgA-DIG (Table 1). The binding buffer used for the gel shift assays contained 24% glycerol, 40 mM Tris (pH 8.0), 150 mM KCl, 2 mM DTT, 600 μg/ml bovine serum albumin, and 1 μg denatured salmon sperm DNA. Gel shift assays were performed with 0.5 nM DIG-labeled ureA or ksgA promoter DNA, which was mixed with increasing concentrations of recombinant Fur protein and was subsequently incubated for 30 min in binding buffer at 37°C. Reactions were separated for 35 min at 200 V on an 8% polyacrylamide gel. The assay was performed in the presence of either 200 μM MnCl₂ or 200 μM EDTA in binding buffer and PAGE buffer (2.5 M Tris and 0.19 M glycine). Gels were blotted onto a positively charged nylon membrane (Roche), followed by chemiluminescent DIG detection (44).

DNase I footprinting assays.DNase footprint assays were performed with 500 nM recombinant Fur protein and 50 nM DIG-labeled ureA promoter DNA or different concentrations of recombinant Fur in combination with 20 nM DIG-labeled ureA promoter DNA. Binding buffer consisted of 50 mM Tris (pH 8.0), 250 mM NaCl, 50 mM KCl, 5 mM DTT, 0.5% NP-40, 50% glycerol, and 100 μg/ml denatured salmon DNA in the presence or absence of 200 μM MnCl₂. DNase I digestion was carried out by the addition of 0.25 units of DNase I (Promega) for 1 minute at room temperature, and the reaction was stopped by the addition of 2.5 M sodium acetate, 20 mM EDTA, and 10 μg denatured salmon DNA. Subsequently the DNA was ethanol precipitated and resuspended in 6 μl of loading buffer (0.05% bromophenol blue [Sigma] and 0.05% xylene cyanol [Merck] in deionized formamide [Sigma]), denatured, and separated on a 7% polyacrylamide gel containing 8 M urea. Gels were blotted onto a positively charged nylon membrane (Roche), followed by chemiluminescent DIG detection (44).

Induction of H. hepaticus urease expression and activity by iron restriction.The effect of iron on H. hepaticus growth was determined by growing bacteria in BBC (standard conditions) or BBC supplemented with either desferoxamine mesylate (iron-restricted conditions) or desferoxamine mesylate and 100 μM FeCl₃ (iron-replete conditions). Iron restriction resulted in reduced growth of H. hepaticus, and cells entered late log phase between 16 and 24 h of growth (Fig. 1A). In contrast, growth under standard conditions and iron-replete conditions saw cells in late log phase after approximately 24 h. At 24 h, cells were still viable, and examination of cells by Gram staining indicated that after 24 h in iron-restricted and iron-replete conditions the cultures still contained >95% bacillary forms of H. hepaticus. Changes in gene expression after 24 h are therefore likely to be a result of changes in iron availability rather than being growth phase dependent (16, 31); thus, we decided to use the 24-h time point for phenotypic analyses.

• Open in new tab
• Download powerpoint

FIG. 1.

Induction of H. hepaticus urease transcription, expression, and activity under iron-restricted conditions. (A) H. hepaticus ATCC5 1449 was grown in standard BBC (gray squares), iron-restricted medium (white diamonds), and iron-replete medium (black triangles), and the OD₆₀₀ was monitored over a time period of 36 h. The arrow indicates the 24-h time point, representing late log phase, which was chosen for comparison of iron-restricted and iron-replete conditions. (B) Urease activity in H. hepaticus is iron repressed, as measured after 24 h of growth in iron-restricted medium (white bar) or iron-replete medium (black bar). Each bar represents data from seven independent experiments; error bars denote standard deviations. An asterisk indicate a significant increase in urease activity after growth under iron-restricted conditions compared to growth under iron-replete conditions (P < 0.05 [Mann-Whitney U test]). (C) The increase in H. hepaticus urease activity under iron-restricted conditions is associated with increased expression of the UreA and UreB enzyme subunits, as shown by SDS-PAGE (left panel) and immunoblotting (right panel) with antibodies to H. felis urease. The left panel displays the relevant section of the protein profile of the lysates used for immunoblots, stained with Coomassie brilliant blue. Relevant marker sizes are indicated on the left. (D) Identification of the H. hepaticus ureA transcription start site by primer extension analysis using RNA purified from H. hepaticus ATCC 51449 grown under iron-restricted (−Fe) and iron-replete (+Fe) conditions. The sequence of the corresponding promoter region is displayed on the right, with the +1 residue and the −10 promoter sequence indicated. Note that the primer extension product displays iron-responsive repression of transcription.

Growth of H. hepaticus under iron-restricted conditions resulted in a twofold increase in urease activity compared to cells grown under iron-replete conditions (Fig. 1B). This iron-responsive urease activity resulted from increased expression of the urease subunit proteins UreA and UreB, as demonstrated by SDS-PAGE and immunoblot assay using both UreA and UreB antibodies (Fig. 1C). These findings contrast with nickel induction of urease expression in H. hepaticus, as this is mediated solely by increased enzyme activity, without additional UreA or UreB protein expression (5). Induction of urease activity and expression by iron restriction was mediated at the transcriptional level and mediated from the ureA promoter, as shown by primer extension analysis (Fig. 1D). The transcription start site of the ureA gene was identified to be the G residue 47 bp upstream of the ureA ATG start codon (Fig. 1D). There was only a single transcription start site, which displayed iron-responsive repression (Fig. 1D) similar to the regulation of expression and activity of H. hepaticus urease (Fig. 1B and C). Upstream of the transcription start site there is a TACAAT hexamer sequence identical to the −10 sequence of the H. pylori ureA promoter (12), which suggests that the H. hepaticus urease cluster is transcribed via the σ⁸⁰ sigma factor. Further comparison of the H. hepaticus ureA promoter sequence with that of H. pylori indicated the absence of the recently identified NikR binding sequence (13, 18, 19), consistent with the absence of nickel-responsive transcription of the urease genes in H. hepaticus (5).

Fur, but not PerR, mediates iron-responsive regulation of H. hepaticus urease.Since iron-responsive regulation of urease expression is mediated at the transcriptional level, this suggested the involvement of an iron-responsive regulatory protein. The H. hepaticus genome encodes two orthologs of such regulators: the peroxide stress regulator PerR (HH0942) and the iron-responsive regulator Fur (HH0893) (41). Both genes are predicted to be transcribed as a monocistronic mRNA (41); thus, mutation of these genes is unlikely to result in polar effects on surrounding genes. To determine whether Fur or PerR was involved in the iron-responsive regulation of urease expression, isogenic H. hepaticus fur and perR mutants were created by insertion of a chloramphenicol resistance gene.

Mutation of either the perR or fur gene did not significantly affect the growth of H. hepaticus, since growth of the mutants was similar to that of the wild-type strain under both iron-restricted and iron-replete conditions (data not shown). Mutation of the perR gene did not affect iron-responsive regulation of urease activity, and the resulting urease activities were similar to that of the wild-type strain (Fig. 2), while mutation of fur abolished regulation of urease activity at the tested conditions (Fig. 2). It should be noted that urease activity in the fur mutant does not show complete derepression compared to the wild-type strain under iron-restricted conditions; thus, it is possible that other regulatory mechanisms play a role in urease regulation in H. hepaticus.

• Open in new tab
• Download powerpoint

FIG. 2.

Fur, but not PerR, mediates iron-responsive regulation of urease expression in H. hepaticus. Urease activity of H. hepaticus strain ATCC 51449 and the fur and perR mutants was assessed after 24 h of growth in iron-restricted medium (white bars) or iron-replete medium (black bars). Each bar represents data from three independent experiments for each strain; error bars denote standard deviations. An asterisk indicate a significant increase in urease activity after growth under iron-resticted conditions compared to growth under iron-replete conditions (P ≤ 0.05 [Mann-Whitney U test]); NS, not significant.

Fur binds to the H. hepaticus ureA promoter in a metal-dependent matter.To determine whether the Fur-mediated regulation of urease transcription in H. hepaticus is direct or indirect, the H. hepaticus Fur protein was expressed as recombinant protein in E. coli. Recombinant H. hepaticus Fur protein was mixed with DIG-labeled ureA promoter and tested by electrophoretic mobility shift assay in the presence and absence of metal. Fur displayed concentration- and metal-dependent binding to the ureA promoter region of H. hepaticus, as the ureA promoter fragment shifted in the presence of Fur and MnCl₂ (Fig. 3, top panel), whereas this shift was not detected in the presence of Fur and EDTA (Fig. 3, middle panel). As a negative control, Fur protein was also incubated with the promoter region of the ksgA gene of H. hepaticus in the presence of metal; this did not result in a shift of the DNA fragment (lower panel), indicating that binding of Fur to the ureA promoter is sequence specific.

• Open in new tab
• Download powerpoint

FIG. 3.

H. hepaticus Fur displays metal-dependent binding to the ureA promoter. Shown is an electrophoretic mobility shift assay using recombinant H. hepaticus Fur protein and DIG-labeled ureA promoter DNA (P_ureA). In the absence of the iron substitution manganese (+EDTA, middle panel), Fur is unable to complex with the ureA promoter region and a shift is not observed. Only in the presence of manganese (+MnCl₂, top panel) is Fur able to bind to the ureA promoter region and cause a mobility shift (indicated as P_ureA + Fur). No shift is observed when the Fur protein is incubated with the promoter region of the ksgA gene (P_ksgA) in the presence of manganese (+MnCl₂, bottom panel). The concentration of Fur is indicated above the lanes; the concentration of DNA was 0.5 nM ureA or ksgA promoter DNA.

Fur binds to an operator sequence in the H. hepaticus ureA promoter.The Fur-binding site in the ureA promoter was subsequently identified by DNase I footprinting. Addition of recombinant Fur resulted in the protection of a single region in the ureA promoter only in the presence of metal. Absence of either Fur or metal did not result in protection of the ureA promoter DNA (Fig. 4A), a finding consistent with the data obtained by electrophoretic mobility shift assays (Fig. 3). Fur-mediated protection of the region in the ureA promoter was concentration dependent (Fig. 4B). The protected sequence was located between positions −35 and −75 upstream of the ureA transcription start site (Fig. 4 and 5A), and the operator sequence contained a single sequence resembling a Fur box, consisting of three NAT(A/T)AT hexamers in an F-F-x-R formation (26).

• Open in new tab
• Download powerpoint

FIG. 4.

Identification of the Fur operator sequence in the H. hepaticus ATCC 51449 ureA promoter. (A) DNase I footprinting assay performed in the presence and absence of the iron substitute MnCl₂, using 50 nM DIG-labeled H. hepaticus ATCC 51449 ureA promoter DNA and 500 nM Fur. On the left side are indicated the positions relative to the ureA transcription start site. On the right-hand side is indicated the position of the protected region (located from −35 to −75 relative to the transcription start site). Note that a protected region is observed only in the presence of both Fur and MnCl₂. (B) Binding of ureA promoter DNA is dependent on the Fur concentration. Increasing concentrations of Fur were mixed with 20 nM DIG-labeled H. pylori ATCC 51449 ureA promoter in the presence of 200 μM MnCl₂. The concentration of Fur protein used is indicated above the lanes. Positions relative to the ureA transcription start site are indicated on the left; the protected region is indicated on the right.

• Open in new tab
• Download powerpoint

FIG. 5.

(A) Schematic representation of the H. hepaticus ureA promoter region with the location and sequence of the Fur-binding site indicated. The proposed ribosome-binding site (RBS), −10 promoter sequence, and transcription start site (+1) are indicated. (B) Annotation of the proposed Fur box present in the Fur-bound sequence in the H. hepaticus ureA promoter region, according to the F-F-x-R hexamer consensus sequence. Residues indicated in bold type match the 5′-NAT(A/T)AT consensus hexamer sequence.