Campylobacter jejuni gene expression in response to iron limitation and the role of Fur

Kathryn Holmes1, Francis Mulholland1, Bruce M. Pearson1, Carmen Pin1, Johanna McNicholl-Kennedy2, Julian M. Ketley2 and Jerry M. Wells1,{dagger}

1 Department of Food Safety Science, BBSRC Institute of Food Research, Norwich Laboratory, Colney Lane, Norwich Research Park, Colney, Norwich NR4 7UA, UK
2 Department of Genetics, University of Leicester, University Road, Leicester LE1 7RH, UK

Correspondence
Jerry Wells
jwells{at}science.uva.nl


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Campylobacter jejuni is a zoonotic pathogen and the most common cause of bacterial foodborne diarrhoeal illness worldwide. To establish intestinal colonization prior to either a commensal or pathogenic interaction with the host, C. jejuni will encounter iron-limited niches where there is likely to be intense competition from the host and normal microbiota for iron. To gain a better understanding of iron homeostasis and the role of ferric uptake regulator (Fur) in iron acquisition in C. jejuni, a proteomic and transcriptome analysis of wild-type and fur mutant strains in iron-rich and iron-limited growth conditions was carried out. All of the proposed iron-transport systems for haemin, ferric iron and enterochelin, as well as the putative iron-transport genes p19, Cj1658, Cj0177, Cj0178 and cfrA, were expressed at higher levels in the wild-type strain under iron limitation and in the fur mutant in iron-rich conditions, suggesting that they were regulated by Fur. Genes encoding a previously uncharacterized ABC transport system (Cj1660–Cj1663) also appeared to be Fur regulated, supporting a role for these genes in iron uptake. Several promoters containing consensus Fur boxes that were identified in a previous bioinformatics search appeared not to be regulated by iron or Fur, indicating that the Fur box consensus needs experimental refinement. Binding of purified Fur to the promoters upstream of the p19, CfrA and CeuB operons was verified using an electrophoretic mobility shift assay (EMSA). These results also implicated Fur as having a role in the regulation of several genes, including fumarate hydratase, that showed decreased expression in response to iron limitation. The known PerR promoters were also derepressed in the C. jejuni Fur mutant, suggesting that they might be co-regulated in response to iron and peroxide stress. These results provide new insights into the effects of iron on metabolism and oxidative stress response as well as the regulatory role of Fur.


Abbreviations: EMSA, electrophoretic mobility shift assay

The raw microarray data are available as a supplement to the online version of this paper (at http://mic.sgmjournals.org), which also has links to Supplementary Tables A–C (available with the online version of the paper only).

{dagger}Present address: Swammerdam Institute for Life Sciences, Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Iron is an essential micronutrient for all living organisms and its acquisition from the environment is vital to bacteria (Wooldridge & Williams, 1993). Iron plays an essential role in cellular metabolism, participating, for example, in respiration, enzyme catalysis and protein structure stabilization (Wooldridge & Williams, 1993). At physiological pH in the presence of oxygen, iron is found in its ferric state [Fe(III)] as an insoluble hydroxide. Within animal tissues the majority of Fe(III) is tightly bound to high-affinity iron-binding and transport proteins, e.g. haemoglobin, transferrin, lactoferrin and ferritin (Ratledge & Dover, 2000). As only a minor fraction (10–18 M) is available, iron restriction provides a non-specific defence system against microbial infection (Ratledge & Dover, 2000). Iron acquisition is an important determinant for a pathogen to colonize a host, and the detection of an iron-limited niche serves as an environmental cue to express virulence determinants (Litwin & Calderwood, 1993).

To ensure that iron is available to the cell in iron-restricted environments, bacteria have evolved high-affinity iron-scavenging and uptake systems (Ratledge & Dover, 2000; Wooldridge & Williams, 1993). There are two general strategies for iron uptake: the production and uptake of siderophores, and the direct utilization of host iron compounds such as transferrin, lactoferrin or haem-containing molecules (Wooldridge & Williams, 1993). Iron–siderophore complexes are transported back into the cell via membrane-associated transport systems. Gram-negative Fe(III) acquisition systems usually consist of an outer-membrane receptor, with transport of the iron compound across the outer membrane energized by a TonB/ExbB/ExbD complex, a periplasmic binding protein and an inner-membrane ABC transporter. A single inner-membrane protein energized by ATP hydrolysis usually mediates ferrous iron transport.

Campylobacter jejuni is the most common cause of bacterial foodborne diarrhoeal illness worldwide. Poultry and poultry products are reported as major sources of infection in developed countries (Oberhelman & Taylor, 2000), but many other sources and vehicles have been noted (Sopwith et al., 2003). In avian species Campylobacter spp. behave as commensals and can reach exceptionally high numbers (up to 108–1010 c.f.u. g–1) in the caecum of chickens (Jacobs-Reitsma et al., 1995), with large numbers shed into the environment (Wesley et al., 2000). Campylobacter is also the most common antecedent to the peripheral neuropathies Guillain–Barré syndrome (GBS) and Miller–Fisher syndrome (MFS) (Nachamkin et al., 2000). Campylobacters are fastidious microaerobes with a minimum growth temperature of 30 °C (Vandamme, 2000). In contrast to their apparent sensitivity under laboratory conditions, infectivity does not appear to be compromised by exposure to stressful environmental conditions outside of the host.

Although we have only a cursory outline of the pathogenic processes involved in campylobacter-mediated disease (van Vliet & Ketley, 2001), colonization of the intestine is fundamental to both commensalism and pathogenesis. Colonization of the nutritionally complex intestinal tract requires effective mechanisms for acquiring iron (Ratledge & Dover, 2000; Wooldridge & Williams, 1993). Although campylobacters produce few, if any, siderophores, they are able to use exogenous ferrichrome and enterochelin from other bacteria and haem compounds, which might be released in the site of inflammation (Baig et al., 1986; Field et al., 1986; Pickett et al., 1992). Several iron-acquisition systems have been identified in campylobacters, including a Fe(II)-transport system (FeoB) (Raphael & Joens, 2003; van Vliet et al., 2002), a haemin/haemoglobin-uptake system (ChuA) (Rock et al., 2001), an enterochelin-transport system (CeuBCDE) (Richardson & Park, 1995) and a ferric enterobactin uptake receptor (CfrA) (Guerry et al., 1997; Palyada et al., 2004). Additionally, a predicted ferrichrome-uptake system (CfhuABD) has been described (Galindo et al., 2001) and there are two putative uptake systems: Cj0178–0181 (van Vliet et al., 2002) and p19 (van Vliet et al., 2002), for which specificity remains unknown.

Iron, in combination with oxygen, can generate reactive oxygen species such as peroxides and hydroxyl radicals via the Fenton and Haber–Weiss reactions that damage proteins, lipids and DNA (Ratledge & Dover, 2000). Thus iron uptake and assimilation need to be tightly controlled in order to avoid iron-associated oxidative-stress-mediated damage to the cell. In C. jejuni, the iron-binding proteins Dps and ferritin play an important role in protection against the generation of reactive oxygen species by sequestering intracellular free iron (Ishikawa et al., 2003; Wai et al. 1996).

In many bacteria, expression of iron transport and storage systems is controlled by the negative regulator Fur (ferric uptake regulator), with ferrous iron acting as a co-repressor (Hantke, 2001). In C. jejuni, Fur was found to regulate the expression of several proposed iron-uptake systems, including Ceu, CfrA, ChuA and p19 (van Vliet et al., 1998), but unlike the fur gene of many other bacteria, C. jejuni fur is not autoregulated (van Vliet et al., 2000). In some bacteria iron levels are linked to oxidative stress responses via a Fur-like iron-responsive regulator, PerR (Bsat et al., 1998; Horsburgh et al., 2001). In C. jejuni, the expression of two iron-regulated proteins (AhpC and KatA; Baillon et al., 1999) was found to be only partially affected by the mutation of fur; and a second Fur homologue, the peroxide stress regulator PerR, was identified and shown to co-regulate ahpC and katA (van Vliet et al., 1999).

During infection pathogens are likely to be in a state of near-continual iron deficiency in the face of fierce competition from the host and normal microbiota. In order to gain a better understanding of the role of iron in the regulation of iron acquisition and homeostasis in C. jejuni we have carried out a proteomic and transcriptome analysis of the response to iron stress in a wild-type and a fur mutant. These results have provided new insights into the effects of iron on metabolism, oxidative stress defence and the regulatory role of Fur.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Growth of C. jejuni strains and iron stress conditions.
The bacterial strains and plasmids used in this study are listed in Table 1. C. jejuni strain NCTC 11168 was routinely grown at 42 °C under microaerobic conditions (10 % CO2, 85 % N2, 5 % O2) on Skirrow agar plates or in Mueller–Hinton broth (MHB; Oxoid) on an orbital shaker (380 r.p.m.) inside a MACS-MG-1000 controlled-atmosphere workstation (DW Scientific). When required, kanamycin at a final concentration of 25 µg ml–1 (Sigma-Aldrich) was added to the medium. To study the effects of iron stress, 50 ml MHB was inoculated with C. jejuni grown on Skirrow agar and incubated overnight (16–18 h). A 3 % inoculum was then used to inoculate a fresh 50 ml broth containing either 20 µM of the iron chelator deferoxamine mesylate (iron-limited conditions) (van Vliet et al., 1998) or 40 µM Fe2(SO4)3 (iron-rich conditions) and grown to mid–late exponential phase (approx. 7·5 h; determined by viable counts and OD600).


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains and plasmids

 
Growth of the wild-type strain NCTC 11168 was slower in the presence of deferoxamine mesylate than in iron-rich conditions and almost 200-fold fewer campylobacters were recovered after 8 h when the wild-type reached stationary phase. As previously reported (van Vliet et al., 1998), growth of the fur mutant strain AV17 was markedly slower than that of the wild-type strain in normal iron-rich conditions (data not shown). However, the growth rate of AV17 was not dissimilar to that of the wild-type strain grown under iron-limited conditions and viability was maintained over a period of about 10 h (data not shown).

RNA isolation and purification.
Cultures were grown in triplicate under both iron-rich and iron-limited conditions as described above and bacteria were harvested by centrifugation at 3000 g for 20 min. Pellets were resuspended in 1 ml Tri Reagent (Sigma-Aldrich), then 0·2 ml chloroform (Sigma-Aldrich) was added, vortexed, and equilibrated at room temperature for 10 min. After centrifugation at 12 000 g for 15 min the aqueous phase was removed and applied to Qiagen's RNeasy Mini columns for RNA purification according to the protocol supplied by the manufacturer. DNA removal was ensured by treatment with DNA-free (Ambion) and the quality and quantity of RNA was checked using the Agilent 2100 Bioanalyser (Agilent Technologies).

Construction of the C. jejuni DNA microarray.
Internal DNA fragments corresponding to unique segments of the individual open reading frames (ORFs) in the annotated genome sequence of strain NCTC 11168 (Parkhill et al., 2000) were amplified by PCR using gene-specific primers (Sigma Genosys ORFmer set), then purified and spotted on GAPSII slides (Corning) using an in-house Stanford-designed microarrayer as described by Pearson et al. (2003).

Transcriptome analysis.
Labelled cDNA was prepared from 15 µg RNA using Stratascript RT (Stratagene) with the direct incorporation of Cy3 and Cy5 dyes (Amersham). Labelled cDNA was purified using a Qiaquick purification kit (Qiagen) and dried before being resuspended in 19·5 µl water, 2·25 µl human Cot1 DNA (Invitrogen), 4·5 µl 20x SSC (20x SSC consisted of 0·7 µl 1 M HEPES pH 7·0, 0·7 µl 10 % SDS and 3 µl Denhardt's solution). Samples were boiled for 3 min, cooled at room temperature for 5 min and centrifuged at maximum speed in a microfuge for 2 min to remove any solid particles from the hybridization mixture. This mixture was put onto the microarray slide and sealed with a coverslip in a GeneMachine hybridization chamber (Anachem) and incubated for 18 h at 63 °C. Following hybridization, microarray slides were washed briefly in pre-warmed (60 °C) 2x SSC, 0·1 % SDS to remove the coverslip and then washed for 2x5 min in each of the following buffers: (a) 2x SSC, 0·1 % SDS (pre-warmed), (b) 1x SSC, and finally (c) 0·2x SSC. Microarray slides were dried by centrifugation at 300 g for 15 min before scanning. DNA microarrays were scanned using an Axon GenePix 4000A microarray laser scanner (Axon Instruments) and the data from detected features initially processed using the GenePix 3.0 software.

Microarray data normalization and analysis.
Re and Ge denote the Cy5 and Cy3 fluorescence signals respectively for the genes equally expressed (=e) in both samples. The relationship between the natural logarithms of the two signals is described as lnRe={alpha}e+{beta}e lnGe, where {alpha}e (intercept) and {beta}e (slope) are estimated by regression using the mean fluorescence intensities for each gene and the complete set of equally expressed genes.

A control experiment involving comparison of two mRNA samples from independent replicates was used to estimate the boundaries between genes that were equally and not equally expressed. These boundaries determined a region for the lnRe={alpha}e+{beta}e lnGt±k, being k=0·767, which is equivalent to approximately twofold greater intensity in either the Cy3 or the Cy5 fluorescence values. With these boundaries the error in classification on the control dataset was 0 %.

Each test hybridization dataset was normalized by applying a modified algorithm (C. Pin and others, unpublished). Briefly, normalization was based on the parameters {alpha}e and {beta}e, calculated using only the equally expressed genes. To calculate these parameters, in the first step, the parameters {alpha}e and {beta}e are assigned with the initial values 0 and 1, respectively. In the second step, the genes that potentially have increased or decreased expression levels are identified as those with mean intensities that lie outside the boundaries, i.e. lnRt<{alpha}e+{beta}e lnGtk or lnRt>{alpha}e+{beta}e lnGt+k, where Rt and Gt are the mean values of the intensities for the test gene in the red and green channels. The genes that potentially have increased or decreased expression levels are tested using the regression models lnRt={alpha}t+{beta}e lnGt and lnRt={alpha}e+{beta}t ln Gt. F-tests were carried out on the hypothesis {alpha}e={alpha}t and {beta}e={beta}t; if one of these was rejected, the tested gene was classified as a gene that had increased or decreased expression levels, otherwise it was classified as equally expressed. In the third step, the values for {alpha}e and {beta}e were recalculated by regression according to the genes classified as equally expressed in the second step. The second and third steps were iterated until the parameters {alpha}e and {beta}e converged.

All the hybridization datasets were normalized by correcting the intensities to obtain {alpha}e=0 and {beta}e=1 for the regression line of the equally expressed genes. The normalized data from each array were unified in one single dataset and reanalysed to identify the differentially expressed genes. The analysis of the unified datasets was carried out by first identifying the genes that potentially had increased or decreased expression levels as those with mean intensities that lay outside the boundaries, i.e. lnRt<lnGt–0·767 or lnRt>lnGt+0·767. F-tests were carried out on the hypothesis {alpha}t=0 and {beta}t=1; if one of these was rejected, the tested gene was classified as having increased or decreased levels of transcripts, otherwise the gene was classified as having equal amounts of transcript in both samples.

The raw microarray data are available as a supplement to the online version of this paper (at http://mic.sgmjournals.org).

Proteome analysis by 2D gel electrophoresis.
C. jejuni cells were harvested in the late exponential phase by centrifugation (3000 g for 10 min) and washed with Tris-buffered saline pH 7·5 prior to lysis by 4x1 min glass bead beating (106 µm or finer) in a lysis buffer containing 50 mM Tris, pH 7·5, 0·3 % SDS, 0·2 M DTT, 3·3 mM MgCl2, 16·7 µg RNase ml–1 and 1·67 U DNase ml–1. Following beating the extract was kept on ice for 20 min before centrifuging at 18 500 g for 20 min; the supernatant was retained for further analysis. Protein concentrations for 2D gel samples were determined using the 2D Quant Kit (Amersham) according to the manufacturer's instructions.

For the first dimension 100–125 µg protein of the cell-free extract (in 50 µl maximum volume) was mixed with an IPG Strip rehydration buffer, containing 7 M urea, 2 M thiourea, 2 % CHAPS, bromophenol blue, 18·2 mM DTT, 2 % pH 3–10NL non-linear IPG Buffer (400 µl final volume) prior to loading on 18 cm 3–10NL Immobiline DryStrips (Amersham Biosciences). Following overnight rehydration IEF was performed for 80 kVh at 20 °C over 24 h using the pHaser system (Genomic Solutions). Prior to the second dimension the focused strips were conditioned in filtered (0·45 µm) equilibration buffer prepared as 5 % SDS and 0·01 % bromophenol blue in 0·122 M Tris/acetate (Tris Acetate Equilibration Buffer, Genomic Solutions), modified with 5 mg ml–1 final volume SDS, 360 mg ml–1 urea and 300 mg ml–1 99 % glycerol). To reduce and alkylate cysteines the strips were treated first with 8 mg ml–1 DTT in the equilibration buffer (9 ml; 30 min with gentle shaking) before transferring into 25 mg ml–1 iodoacetamide in the equilibration buffer (9 ml; 30 min with gentle shaking). Ten per cent Duracryl gels (28x23 cm; 1 mm thick) were prepared for use in the Investigator 2nd Dimension Running System (Genomic Solutions); cathode buffer (200 mM Tris base, 200 mM Tricine, 14 mM SDS,) and anode buffer (25 mM Tris/acetate buffer, pH 8·3). Electrophoresis was carried out using either a maximum voltage of 500 V or a maximum power of 20 W per gel.

Proteins were stained by Sypro-Ruby (Bio-Rad) according to the manufacturer's instructions and the gels were imaged at 100 µm resolution using the ProXPRESS Proteomics Imaging System with ProFinder imaging software (Perkin Elmer Life Sciences). Top illumination was used with a 480/30 excitation filter and a 630/30 emission filter. Gel images were compared using ProteomWeaver analysis software (Definiens). Protein spots with altered levels of expression under iron-rich versus iron-limited conditions were excised from the gel using the ProPick excision robot (Genomic Solutions), and in-gel trypsin digested using a ProGest Protein Digester (Genomic Solutions). Gel plugs were conditioned with two 20 min incubations in 200 mM ammonium bicarbonate (ABC) in 50 % acetonitrile (50 µl) followed by 10 min incubations with acetonitrile (50 µl). The gel plugs were then conditioned for 15 min with 25 mM ABC (50 µl) followed by 10 min in acetonitrile (50 µl). A final 5 min incubation of acetonitrile (50 µl) preceded the trypsin digestion at 37 °C (3 h) using 50 ng (5 µl per well) sequencing-grade porcine trypsin (Promega) dissolved in 25 mM ABC. Digestion was stopped and peptides extracted using formic acid (5 %; 5 µl per tube).

The tryptic digests were analysed using a Reflex III MALDI-TOF instrument (Bruker) with a Scout 384 ion source using a nitrogen laser (wavelength 337 nm) to desorb/ionize the matrix/analyte material from the sample substrate. The matrix used was a mixture of {alpha}-cyano-4-hydroxycinnamic acid and nitrocellulose, applied to the target as a thin film. Sample (0·3 µl) was then spotted on to the target and dried before washing with 10 % formic acid. All spectra were acquired in a positive-ion reflector. The acceleration voltage was set to 25 kV, the reflection voltage to 28·7 kV, the ion source acceleration voltage to 20·9 kV, and the reflector-detector voltage to 1·65 kV.

Proteins were identified from the peptide mass peak list by the Protein Mass Fingerprint technique using the Mascot search tool (Matrix Science; http://www.matrixscience.com/). An initial mass tolerance of 75 p.p.m. was used in the searches, with up to one missed trypsin cleavage allowed. Carbamidomethyl modification of cysteine (as a fixed modification) and oxidation of methionine (as a variable modification) were permitted. Mowse scores (Pappin, 2003) greater than 46 were considered to be statistically significant when specifically searching C. jejuni protein sequences in the National Center for Biotechnology Information (NCBI) database (http://www.ncbi.nlm.nih.gov/).

Electrophoretic mobility shift assay (EMSA) with Fur.
The EMSA was used to demonstrate the binding of purified C. jejuni Fur to Fur-regulated promoters seen by microarray expression profiling. C. jejuni Fur was expressed as a fusion protein and purified by affinity chromatography using the Strep-tag II system (IBA, Germany). The fur gene was amplified (fur-F and fur-R; see Supplementary Table A with the online version of this paper) and cloned into the BsaI site of pASK-IBA7 to form pJMcK1. Expression of the Strep-tag : : fur gene fusion in Escherichia coli XL-1 Blue (pJMcK1) was induced using anhydrotetracycline (final concentration of 0·2 µg ml–1) at 37 °C in Luria Broth with ampicillin (100 µg ml–1) as directed by the manufacturer. The cells were lysed by sonication and the cell-free extract was subjected to affinity purification using the Strep-Tactin affinity column (IBA) according to the manufacturer's instructions. In order to remove the tag, purified Strep-tag–Fur protein was treated with biotinylated factor Xa (Boehringer Mannheim) and tag-free Fur was separated from factor Xa with a Strep-Tactin affinity column. Factor Xa digestion was verified by Western blotting using anti-Strep-tag antibody (IBA). Expressed, purified and factor-Xa-digested Fur was analysed by PAGE and transferred to a PVDF membrane; the sequence of the N-terminal 16 amino acids was confirmed by Edman degradation (ABI 476 protein sequencer; Applied Biosystems).

The DNA fragments used in the EMSA were generated by PCR with primers (Supplementary Table A) designed to amplify the putative Fur binding sites (van Vliet et al., 2002). DNA amplification was carried out using Taq polymerase (Abgene) with Pfu proofreading enzyme (Stratagene) on an Eppendorf Thermal Cycler using standard protocols. Amplified fragments were purified (Qiagen) and labelled with DIG-ddUTP (Roche) according to the manufacturer's instructions. The final amount of labelled DNA was 0·155 pmol, which was diluted as required.

The EMSA binding reactions were carried out in a 10 µl volume containing binding buffer (20 mM Bistris pH 7·6, 40 mM KCl, 1 mM MgCl2, 2 mM DTT, 1 µg salmon sperm DNA µl–1, 100 µg BSA ml–1, 100 µM MnSO4), 1·5–5 fmol labelled DNA and varying concentrations of Fur. If unlabelled competitor DNA was being used, this was added to the binding reaction immediately before the addition of Fur and labelled DNA. Reactions were mixed on ice and loading buffer (0·5x TBE, 10 % glycerol) added before the samples were separated on 8 % native PAGE gels in 0·5x TBE. Gels were pre-run at 12 V cm–1 for 10 min and then run at 12 V cm–1 for 60 min. Following electrophoresis, samples were transferred to nylon membranes (Amersham) and the DNA fixed by UV cross-linking. Labelled DNA was detected with anti-DIG antibody (1 : 10 000 dilution) using a standard DIG detection protocol (Roche). Membranes treated with chemiluminescent substrate (CSPD; Roche) were exposed to X-ray film for 30 min at room temperature.


   RESULTS AND DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Effects of iron limitation on the transcriptome of C. jejuni
To investigate the response to iron stress in C. jejuni, the transcription profiles of wild-type strain NCTC 11168 were compared in iron-rich and iron-limited conditions during the exponential growth phase. Under the growth conditions used in this study approximately 83 % of the gene features gave a fluorescence signal above background following transcriptome analysis. In three independent microarray experiments the relative transcript levels were significantly increased for 117 genes and decreased for 30 genes in response to iron limitation (Table 2). Almost half of the genes induced under iron limitation are predicted to encode cell-surface proteins and iron-binding and transport systems, including chuA, cfrA and p19, which were shown previously to be iron regulated (van Vliet et al., 1998, 2002).


View this table:
[in this window]
[in a new window]
 
Table 2. Comparison of relative transcript levels in iron-rich and iron-limited growth conditions for wild-type strain NCTC 11168

 
Iron limitation also led to increased transcript levels of three regulatory genes, a two-component system regulator (Cj1227c), a putative transcriptional regulator (Cj0466), both of unknown function, and the peroxide stress regulator, perR (Cj0322) (Table 2). Iron-responsive autoregulation of the fur gene was not observed as previously reported (van Vliet et al., 1999). Of those genes associated with toxin production or host cell invasion, only ciaB, a gene that encodes a protein required for invasion (Konkel et al., 1999), had increased expression in iron-limited conditions. Additionally, there were 36 genes with increased transcript levels that encode proteins with no known function or with homology to hypothetical proteins found in other bacteria, implicating them as having a potential role in iron transport or metabolism.

Iron homeostasis is important to bacteria because iron, in combination with oxygen, can generate reactive oxygen species (ROS) such as peroxides and hydroxyl radicals via the Fenton and Haber–Weiss reactions (van Vliet et al., 2002). These ROS can damage lipids, proteins and DNA. Oxidative stress enzymes are used by bacteria to remove ROS, and consequently the expression of these enzymes is often coordinated with the expression of iron-transport systems in anticipation of the influx of iron into the cell. In C. jejuni, for example, production of catalase (KatA) and alkyl hydroperoxide reductase (AhpC) is regulated by the PerR regulator, a homologue of Fur that also utilizes iron and possibly managanese as a cofactor for DNA binding. Studies in Bacillus subtilis suggest that PerR might sense peroxide stress by oxidation of the metal cofactor or oxidation of the protein itself (Herbig & Helmann, 2001). One model proposes that iron-bound PerR is sensitive to oxidation whereas manganese-bound PerR is relatively resistant; thus the balance between iron and manganese availability determines the response of PerR to peroxide stress (Jakubovics & Jenkinson, 2001).

It was not surprising therefore to find that several members of the peroxide stress regulon (i.e. KatA, AhpC and PerR itself) were expressed at higher levels in iron-limited conditions (Table 2). We also found that Cj1386, which encodes an ankyrin-repeat-containing protein of unknown function immediately downstream of katA (Cj1385), and two genes (Cj1383c and Cj1384c) that are transcribed divergently from the katA promoter, had increased transcript levels under iron limitation. Interestingly, in Pseudomonas aeruginosa a gene encoding an ankyrin-repeat-containing protein (AnkB) lies downstream of the catalase gene and has been demonstrated to increase the catalytic activity of catalase (Howell et al., 2000). It is possible that the co-regulated Cj1386 gene has a similar function in C. jejuni. Thioredoxins, thioredoxin reductases and thiol peroxidases also play important roles in defence against oxidative stress. Under iron limitation, two thioredoxins encoded by genes Cj1664 and Cj1665, the thioredoxin TrxB (Cj0146c), the thioredoxin reductase TrxA (Cj0147c) and a probable thiol peroxidase known as Tpx (Cj0779) also had increased expression (Table 2).

Transcript levels of several non-essential iron-containing proteins were decreased in iron-limited conditions, presumably as a mechanism to make more iron available for vital cellular processes. For example, a non-haem protein (Cj0012c; also observed in 2D gels; see Table 4), ferredoxin (Cj0333c), cytochrome c peroxidase (Cj1358) and members of the succinate dehydrogenase operon (Cj0437–0438) were all decreased in expression (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 4. Proteomic analysis of iron-regulated protein expression, excluding the known iron-transport proteins shown in Table 3

The fold changes in protein expression that were considered significant according to the ProteomWeaver replicate quality test are underlined. Protein identifications shown in bold were identified in the sections of 2D gels shown in Fig. 2(C) and (D).

 
Proteomic analysis of the response to iron limitation
2D gel electrophoresis is unable to reveal every protein in the cell due to problems with solubility, the low abundance of some proteins and difficulties in detecting proteins at the boundaries of the isoelectric and molecular mass ranges. This is especially true of integral membrane proteins, which are typically insoluble or poorly soluble in the solutions used for isoelectric focusing in the first dimension. One important application of proteomics is the detection of post-translational modifications of proteins, something that would not be apparent from mRNA expression data or gene sequences. Protein phosphorylation, glycosylation and proteolytic processing can be extremely important factors influencing the activity, stability and localization of proteins in both prokaryotic and eukaryotic cells.

Here, proteomics was used to investigate the response of C. jejuni strain NCTC 11168 to iron limitation by comparing 2D gels of cell protein extracts from exponentially growing bacteria in iron-rich and iron-limited conditions. Proteins were visualized by Sypro-Ruby staining and quantified by the use of a multiwave fluoroimager (ProExpress) and the Proteome Weaver 2D gel analysis software package. Digital images from 2D gels of iron-rich and iron-limited samples were differently coloured and computer aligned to highlight the differentially expressed proteins (Fig. 1). Analysis of the gels using ProteomWeaver 2D gel analysis software followed by manual filtering and editing identified 595 active spots on the iron-limited gel and 699 active spots on the iron-rich gel, with 496 protein spots being common to both conditions. Using the replicate quality test of the gels within the ProteomWeaver software a regulation factor of greater than 2·123 was considered to be significant for these gels. Thus, 40 of the matched spots had significantly increased intensity in the iron-limited gels and 27 spots had significantly increased intensity in the iron-rich gels. A number of these spots, however, were then manually filtered out of the analysis as spurious, caused by edge effects or unable to provide peptide mass fingerprint (PMF) data sufficient to allow identification of the protein.



View larger version (76K):
[in this window]
[in a new window]
 
Fig. 1. Overlay of 2D gels of C. jejuni 11168 grown under iron-rich and iron-limited conditions. Blue indicates increased expression of the protein spot under iron-limited conditions. Orange indicates increased expression of the protein spot under iron-rich conditions. Black indicates similar expression of the protein spot in both iron-rich and iron-limited conditions. Boxes A–D have been enlarged in Fig. 2(A–D) to show individual gels in more detail together with relevant protein spot identification. Spots circled with a red line were significantly increased in iron-rich conditions: 1, Putative periplasmic protein Cj0092; 2, major outer-membrane protein (fragment) Cj1259.

 
The proteomics data supported the microarray studies by showing that the major protein spots induced under iron limitation (increased 2·3- to 11·8-fold) were also components of known iron-uptake systems (Table 3, Fig. 2A–D). These proteins included ChuA (Cj1614) (Fig. 2B), CfrA (Cj0755) (Fig. 2A), p19 (Cj1659) (Fig. 2C) and components of two iron-regulated ABC transport systems, Cj0175c (Fig. 2D) and Cj1663 (Fig. 2C). In the case of CfrA, two protein spots of different charge were detected in iron-limited conditions, possibly linked to the binding of an iron compound (Fig. 2A, Table 3). Some outer-membrane iron-binding proteins were identified by proteomics, but not the more hydrophobic integral membrane components of the iron-transport systems. There was no obvious correlation between protein abundance and mRNA abundance for these genes (Table 3), but this could be due to the existence of a variety of possible post-transcriptional and translational control mechanisms in bacteria.


View this table:
[in this window]
[in a new window]
 
Table 3. Increased expression of iron-transport proteins in iron-limited growth conditions

 


View larger version (45K):
[in this window]
[in a new window]
 
Fig. 2. Enlarged images of the boxed sections A–D of the 2D gel shown in Fig 1.

 
Apart from the iron-transport proteins, 42 other spots appeared to be differentially expressed under iron-limited conditions (Fig. 1), many of which are indicated in the enlarged areas shown in Fig. 2(C) and Fig. 2(D) and listed in Table 4. Multiple protein spots were detected for catalase (KatA, Cj1385), alkyl hydroperoxide reductase (AhpC, Cj0334), ferritin (Cft, Cj0612c), bacterioferritin (Bfr/Dps, Cj1534c) and a probable thiol peroxidase protein (Tpx, Cj0779), bringing the total number of affected proteins to 35. In total, 13 of the different proteins listed in Table 4 were shown to be regulated at the level of transcription in response to iron stress (Table 3) but only six of these, comprising 12 protein spots (i.e. AhpC, Tpx, KatA, Cft, hypothetical protein Cj1613c and a 10 kDa heat-shock chaperonin, Cj1220), were statistically significant according to the criteria mentioned above (Table 4).

C. jejuni genes with altered expression in a fur mutant in iron-rich conditions
The Fur regulator has previously been shown to repress the transcription of at least seven genes in C. jejuni, five of which were identified as ceuE, chuD, p19, chuA and cfrA (van Vliet et al., 1998). In order to identify which of the iron-regulated genes listed in Table 2 were regulated by Fur we compared the transcriptome of a fur mutant with the wild-type strain grown in iron-rich conditions. Transcript levels of all the proposed iron-transport systems of C. jejuni that were expressed at higher levels in iron-limited conditions were elevated in the fur deletion mutant, suggesting that they are Fur regulated and indeed play a role in the uptake of iron (Tables 2 and 5). For example, the gene encoding ChuA, a putative outer-membrane receptor for haemin, showed a dramatic increase (76-fold) in transcript levels in the fur mutant strain grown in iron-rich conditions (Supplementary Table B). The genes downstream of chuA in the chuABCD operon encode a probable ABC transporter which was also induced in iron-limited conditions and derepressed in the fur mutant; however, mutations in chuB, chuC and chuD do not seem to affect haemin utlilization (Rock et al., 2001), and the cognate ABC transporter for this transport receptor remains unknown. Gene Cj1613c, which encodes a hypothetical protein predicted to be transcribed in the opposite direction to the chuA operon, also appeared to be Fur regulated, thus implying a possible role for this gene in iron metabolism and possibly haemin transport. The previously uncharacterized ABC transporter encoded by genes Cj16601663 was also derepressed in a fur mutant, providing evidence for its role in iron transport. This operon lies downstream of Cj1658, a putative inner-membrane protein, and p19 (Cj1659), a periplasmic protein previously shown to be iron regulated (van Vliet et al., 1998). Thus it is possible that Cj1660–1663 might function as an ABC transport system for Cj1658/p19 or alternatively for the ChuA receptor. Another candidate transport protein is the putative ABC transporter gene Cj1587c, which was also induced under iron limitation and derepressed in the fur mutant. This study also indicated a possible role for several other uncharacterized genes in iron transport or metabolism. For example, Cj0379c, Cj0567 and Cj0819, which encode hypothetical proteins, a putative integral membrane protein (Cj1560) and a putative periplasmic protein (Cj0909), were all induced under iron limitation and derepressed in a fur mutant.


View this table:
[in this window]
[in a new window]
 
Table 5. The Fur regulon

 
We showed that the expression of a possible haemin-like receptor system (Cj0177 and Cj0178) and the tonB1/exbB1/exbD1 gene complex downstream (Cj0179–0181) were derepressed in a fur mutant. It is therefore probable that these genes are co-transcribed from the promoter upstream of Cj0177 that was previously shown to contain a Fur-box-like sequence (van Vliet et al., 2002). Interestingly, the genes divergently transcribed from this promoter encoding a putative ferric iron transport system (Cfbp system, Cj0173c-0176c) were also derepressed in a fur mutant, suggesting that the Fur box is involved in the regulation of both promoters. Apart from tonB (Cj0179) mentioned above, there are two other tonB gene homologues in C. jejuni, Cj0753c (tonB3) and Cj1630 (tonB2), that were both shown to be iron regulated. There are two further sets of exbB/D genes, Cj1628–9 (exbB2/D2) and Cj0109–10 (exbB3/D3), with only the former being iron and Fur responsive. Thus, it is likely that these iron- and Fur-responsive genes are acting in concert to express another TonB/ExbB/ExbD energy-transduction complex for iron transport.

The expression of a pseudogene, Cj0444, was also found to be Fur regulated, although the promoter does not contain a consensus Fur box (van Vliet et al., 2002). In some other strains of C. jejuni this gene is present as an uninterrupted open reading frame with homology to other outer-membrane iron-uptake proteins (unpublished; GenBank accession AY656248).

Seven genes that were expected to be derepressed in the fur mutant because of the presence of a consensus Fur box in the upstream promoter (van Vliet et al., 2002) appeared not to be iron or Fur regulated in these experiments. This suggests that the consensus Fur box sequence postulated from the E. coli consensus (van Vliet et al., 2002) will need experimental refinement before it will be possible to predict all members of the operon using a bioinformatics approach.

A significant number of other hypothetical proteins and proteins predicted to be periplasmic or membrane associated had increased expression in iron-limited conditions, but were not affected by the fur mutation in iron-rich conditions. It is possible therefore that these also play a role in iron metabolism and are regulated independently of Fur, perhaps even by PerR. However, we also recognize the possibility that transcription of these genes is altered due to the physiological and metabolic changes that take place in iron-limited conditions (Supplementary Table B compared with Table 2).

In iron-rich conditions, 30 genes were transcribed at lower levels in the fur mutant than the wild-type strain (Supplementary Table B). Of these 30 genes, only fumarate hydratase (Cj1364) and Cj0859c (hypothetical) were previously shown to have reduced expression in the wild-type strain under iron limitation and thus might be directly or indirectly regulated by Fur. Fumarase activity has also been shown to be iron responsive in Pseudomonas aeruginosa, Pasteurella multocida and E. coli (Hassett et al., 1997; Paustian et al., 2001; Tseng, 1997).

In C. jejuni, PerR regulation has been demonstrated for only the katA (Cj1385) and ahpC (Cj0334) genes, but the situation is complicated by the fact that Fur also seems be involved in regulation of these genes (van Vliet et al., 1999, 2002). Thus, several other genes containing a PerR/Fur box consensus sequence, such as thioredoxin reductase (trxB, Cj0146c), bacterioferritin (bfr/dps, Cj1534c), cytochrome c peroxidase (ccp, Cj0358), glutamyl-tRNA reductase (hemA, Cj0542), superoxide dismutase (sodB, Cj0169) and the peroxide stress regulator perR (Cj0322) itself, might also be potentially derepressed in a fur mutant. However, only ahpC, trxB, katA and the downstream gene Cj1386 encoding an ankyrin-repeat-containing protein were increased in expression in the fur mutant grown in iron-rich conditions. Interestingly, this study identified two putative thioredoxins (Cj1664 and Cj1665) other than TrxB (Cj0146c), as being regulated by Fur. The transcript levels of these two putative thioredoxins were also strongly increased in the wild-type strain under iron limitation (see above), supporting a role for Fur in their regulation and a potential role for these enzymes in defence against oxidative stress induced by elevated levels of intracellular iron. The similarity between the predicted Fur and PerR boxes (van Vliet et al., 2002), and the apparent co-regulation of Fur and PerR shown by our results, makes it difficult to establish a separate set of genes controlled by each regulator. Thus binding studies with purified proteins will be needed to investigate this phenomenon for individual promoters. The potential for cooperative binding or competition between Fur and PerR as suggested by van Vliet et al. (2002), coupled with their own regulated expression, would possibly allow for control of target gene expression in response to multiple environmental stimuli.

In E. coli, a group of genes encoding iron-storage and iron-containing proteins, including the succinate dehydrogenase operon, is repressed in iron-limited conditions (Masse & Gottesman, 2002). This repression was shown to be due to transcription repression by a small non-coding regulatory RNA called RyhB that is itself repressed by Fur (Masse & Gottesman, 2002). Thus in E. coli, low iron availability results in loss of Fur binding and expression of RyhB, which then represses transcription of a set of genes encoding iron-binding proteins. Although transcription of the C. jejuni succinate dehydrogenase operon was decreased under iron limitation it was not repressed in the fur mutant grown in iron-rich conditions, suggesting that a RyhB-like mechanism is not involved in the regulation of this operon.

C. jejuni genes with altered expression in a fur mutant grown in iron-limited conditions
By comparing the transcriptome of our fur mutant with the wild-type under iron-limited conditions we were able to identify genes that were differentially expressed in the fur mutant that were not a consequence of iron limitation and derepression by Fur.

Only 18 genes were expressed at higher levels in AV17 as compared to the wild-type strain in iron-limited conditions (Supplementary Table C). Six of these were previously shown to be affected by iron limitation in the wild-type strain and presumably reflect strain differences in the transcript levels of genes within the iron stimulon. A further four of the 18 affected genes (i.e. Cj0424, Cj0425, Cj0571c, Cj0949c) were not part of the iron stimulon identified above, and were also transcribed at higher levels in AV17 compared to wild-type strain in iron-rich conditions. Thus these changes might be due to polar effects of the fur mutation (van Vliet et al., 1998) or an adaptation to the permanent loss of Fur regulation.

It was also evident from comparing the transcriptome of the fur mutant strain AV17 with the wild-type parent strain grown in iron-rich or iron-limited conditions that the fur mutation substantially decreases transcription of two housekeeping genes downstream of fur, lysS (Cj0401) and glyA (Cj0402), 2·7- and 4·3-fold respectively (Supplementary Tables B and C). This might be the cause of the slower growth rate of strain AV17 compared to the parent strain and also the reason for the iron-independent differences observed between the transcriptome of strain AV17 and the parent strain. For example, many flagellar genes are differentially expressed between the two strains regardless of iron availability. The lack of flagellar expression might be directly attributable to the reduced growth capabilities of the mutant as expression of flagella is an energy-expensive process.

EMSA of Fur-regulated promoters
Iron-uptake systems are primary members of the Fur regulon, and members of these systems in C. jejuni serve to corroborate the pattern of genome-wide gene expression under iron stress revealed by microarray. Fur-dependent iron-responsive expression of iron-uptake-associated systems would indicate that the Fur protein binds to the promoter element controlling transcription of each system. In order to demonstrate Fur-promoter binding directly, purified C. jejuni Fur protein was expressed and used in an EMSA with established members of the Fur regulon. The complete open reading frame of the fur gene was cloned into the expression vector pASK-IBA-7. A Strep-tag–Fur fusion protein was expressed and purified by affinity chromatography. The Strep-tag was removed by factor Xa cleavage and the repurified Fur protein used for the EMSA. In this study we verified by EMSA that purified Fur binds to the p19, cfrA and ceuB promoters (Fig. 3). As illustrated by the p19-associated promoter, the mobility change due to Fur could be inhibited competitively by the unlabelled p19 promoter fragment derived from just upstream of Cj1658. In contrast, a similarly sized DNA fragment located between the waaF and gmhA genes did not act as a competitor for Fur (data not shown). The Ceu system (Cj1352–1355) previously identified as a specific transport system for the siderophore enterochelin (Richardson & Park, 1995) was only partly induced under iron-limited conditions, with only ceuE (Cj1355) having significantly elevated transcript levels (Table 2). An EMSA with purified ceuB promoter region was used to confirm Fur binding, but the concentration of Fur needed to cause a band shift was around 300 times higher than with the most strongly induced promoter upstream of p19 (Fig. 3). Although this might be an explanation for the weak induction of the ceuBCD genes, cfrA, which is also expressed from a highly induced promoter (Table 2), requires approximately 50-fold more Fur to achieve a shift equivalent to that seen with p19 with 0·05 nM Fur (Fig. 3 and data not shown).



View larger version (98K):
[in this window]
[in a new window]
 
Fig. 3. Gel images of the EMSA assays carried out with the different promoter probes indicated. All binding reactions contained 1·5 fmol of the respective promoter fragment. p19: lanes 1–5 contained 0, 0·025, 0·05, 0·1 and 0·25 nM of purified Fur in the binding reaction respectively. cfrA: lanes 1–5 contained 0, 10, 12·5, 15 and 17·5 nM of purified Fur in the binding reactions respectively. ceuB: lanes 1–8 contained 0, 10, 12·5, 15, 17·5, 20, 22·5 and 26 nM of purified Fur in the binding reactions respectively. In the p19 competition assay, 1·5 fmol of labelled p19 promoter fragment was incubated with 1·5 nM purified Fur in the presence of increasing concentrations of an unlabelled p19 competitor fragment as follows: lane 1 contained only the promoter fragment; lane 2, no competitor; lane 3, 0·55 pmol competitor (366-fold excess); lane 4, 1·1 pmol competitor (733-fold excess); lane 5, 1·5 pmol competitor (1099-fold excess); lane 6, 2·2 pmol competitor (1466-fold excess).

 
Concluding remarks
This study (i) describes the C. jejuni iron stimulon at the level of both transcription and translation, (ii) defines members of the Fur regulon and (iii) verifies that Fur binds to a set of promoters identified in this study to be Fur regulated. After submission of our manuscript a paper was published also using microarrays to identify C. jejuni genes that have their transcript abundance altered by iron concentration (Palyada et al., 2004). In steady-state experiments comparing the transcriptome profiles of C. jejuni in iron-limited MEM{alpha} and iron-rich MEM{alpha} containing 40 µM ferrous sulphate, 208 genes were differentially expressed. Hierarchical clustering of the temporal response of these genes to addition of iron identified six major clusters, A–F, with gene transcripts from cluster A being altered more substantially and more rapidly than those from clusters C, D and E. The genes in clusters A and B play a role in iron acquisition and detoxification and correlate very well with the list of iron-transport and oxidative-stress genes identified in this study (Table 2). In Palyada et al. (2004) Table 3 indicates 45 genes from clusters A–E that show substantial reduction in expression after addition of iron to Campylobacter grown in iron-limited medium. In contrast, in our study 10 of these genes (mainly encoding putative transmembrane proteins and cofactor biosynthesis enzymes) were not found to have their transcript levels affected by iron availability. However, this disparity could reflect different experimental design and/or growth medium. It was also reassuring to find that the majority of the genes suggested to be repressed by Fur in the presence of iron (cluster D in Fig. 4 in Palyada et al., 2004) were also identified in our study as being Fur regulated.

Similar studies were also recently conducted in the closely related bacterium Helicobacter pylori to identify genes that were differentially regulated upon a growth-inhibiting shift to iron-starvation conditions (Merrell et al., 2003). Many known iron-regulated genes were identified as well as a number of genes whose regulation by iron concentration was not previously predicted, including the virulence genes cagA, vacA and napA. A bioinformatic analysis of the predicted promoter regions of the differentially regulated genes led to identification of several putative Fur boxes, suggesting a direct role for Fur in iron-dependent regulation of these genes (Merrell et al., 2003).

Our future research is focused on extending the EMSA to all of the Fur-regulated promoters identified in this study and to experimentally define the Fur box consensus. Comparison of the fur mutant with the parent strain under different physiological conditions was a useful approach to identify the subset of iron-regulated genes. We aim to define the PerR regulon using similar approaches and to investigate the mechanism of co-regulation by PerR and Fur.


   ACKNOWLEDGEMENTS
 
This research was supported by a grant from the Biotechnology and Biological Sciences Research Council (BBSRC) UK to J. W. and J. K. entitled ‘Iron responsive gene regulation in Campylobacter jejuni’. J. McN.-K. is supported by a doctoral studentship from BBSRC.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS AND DISCUSSION
REFERENCES
 
Baig, B. H., Wachsmuth, I. K. & Morris, G. K. (1986). Utilization of exogenous siderophores by Campylobacter species. J Clin Microbiol 23, 431–433.[Medline]

Baillon, M. L., van Vliet, A. H., Ketley, J. M., Constantinidou, C. & Penn, C. W. (1999). An iron-regulated alkyl hydroperoxide reductase (AhpC) confers aerotolerance and oxidative stress resistance to the microaerophilic pathogen Campylobacter jejuni. J Bacteriol 181, 4798–4804.[Abstract/Free Full Text]

Bsat, N., Herbig, A., Casillas-Martinez, L., Setlow, P. & Helmann, J. D. (1998). Bacillus subtilis contains multiple Fur homologues: identification of the iron uptake (Fur) and peroxide regulon (PerR) repressors. Mol Microbiol 29, 189–198.[CrossRef][Medline]

Field, L. H., Headley, V. L., Payne, S. M. & Berry, L. J. (1986). Influence of iron on growth, morphology, outer membrane protein composition, and synthesis of siderophores in Campylobacter jejuni. Infect Immun 54, 126–132.[Medline]

Galindo, M. A., Day, W. A., Raphael, B. H. & Joens, L. A. (2001). Cloning and characterization of a Campylobacter jejuni iron-uptake operon. Curr Microbiol 42, 139–143.[Medline]

Guerry, P., Perez-Casal, J., Yao, R., McVeigh, A. & Trust, T. J. (1997). A genetic locus involved in iron utilization unique to some Campylobacter strains. J Bacteriol 179, 3997–4002.[Abstract]

Hantke, K. (2001). Iron and metal regulation in bacteria. Curr Opin Microbiol 4, 172–177.[CrossRef][Medline]

Hassett, D. J., Howell, M. L., Sokol, P. A., Vasil, M. L. & Dean, G. E. (1997). Fumarase C activity is elevated in response to iron deprivation and in mucoid, alginate-producing Pseudomonas aeruginosa: cloning and characterization of fumC and purification of native FumC. J Bacteriol 179, 1442–1451.[Abstract]

Herbig, A. F. & Helmann, J. D. (2001). Roles of metal ions and hydrogen peroxide in modulating the interaction of the Bacillus subtilis PerR peroxide regulon repressor with operator DNA. Mol Microbiol 41, 849–859.[CrossRef][Medline]

Horsburgh, M. J., Clements, M. O., Crossley, H., Ingham, E. & Foster, S. J. (2001). PerR controls oxidative stress resistance and iron storage proteins and is required for virulence in Staphylococcus aureus. Infect Immun 69, 3744–3754.[Abstract/Free Full Text]

Howell, M. L., Alsabbagh, E., Ma, J. F. & 10 other authors (2000). AnkB, a periplasmic ankyrin-like protein in Pseudomonas aeruginosa, is required for optimal catalase B (KatB) activity and resistance to hydrogen peroxide. J Bacteriol 182, 4545–4556.[Abstract/Free Full Text]

Ishikawa, T., Mizunoe, Y., Kawabata, S., Takade, A., Harada, M., Wai, S. N. & Yoshida, S. (2003). The iron-binding protein Dps confers hydrogen peroxide stress resistance to Campylobacter jejuni. J Bacteriol 185, 1010–1017.[Abstract/Free Full Text]

Jacobs-Reitsma, W. F., Maas, H. M. E. & Jansen, W. H. (1995). Penner serotyping of campylobacter isolates from poultry, with absorbed pooled antisera. J Appl Bacteriol 79, 286–291.[Medline]

Jakubovics, N. S. & Jenkinson, H. F. (2001). Out of the iron age: new insights into the critical role of manganese homeostasis in bacteria. Microbiology 147, 1709–1718.[Medline]

Konkel, M. E., Kim, B. J., Rivera-Amill, V. & Garvis, S. G. (1999). Bacterial secreted proteins are required for the internalization of Campylobacter jejuni into cultured mammalian cells. Mol Microbiol 32, 691–701.[CrossRef][Medline]

Litwin, C. M. & Calderwood, S. B. (1993). Role of iron in regulation of virulence genes. Clin Microbiol Rev 6, 137–149.[Abstract]

Masse, E. & Gottesman, S. (2002). A small RNA regulates the expression of genes involved in iron metabolism in Escherichia coli. Proc Natl Acad Sci U S A 99, 4620–4625.[Abstract/Free Full Text]

Merrell, D. S., Thompson, L. J., Kim, C. C., Mitchell, H., Tompkins, L. S., Lee, A. & Falkow, S. (2003). Growth phase-dependent response of Helicobacter pylori to iron starvation. Infect Immun 71, 6510–6525.[Abstract/Free Full Text]

Nachamkin, I., Allos, B. M. & Ho, T. W. (2000). Campylobacter jejuni infection and the association with Guillain-Barre syndrome. In Campylobacter, pp. 155–176. Edited by I. Nachamkin & M. J. Blaser. Washington, DC: American Society for Microbiology.

Oberhelman, R. A. & Taylor, D. N. (2000). Campylobacter infections in developing countries. In Campylobacter, pp. 139–153. Edited by I. Nachamkin & M. J. Blaser. Washington, DC: American Society for Microbiology.

Palyada, K., Threadgill, D. & Stintzi, A. (2004). Iron acquisition and regulation in Campylobacter jejuni. J Bacteriol 186, 4714–4729.[Abstract/Free Full Text]

Pappin, D. J. (2003). Peptide mass fingerprinting using MALDI-TOF mass spectrometry. Methods Mol Biol 211, 211–219.[Medline]

Parkhill, J., Wren, B. W., Mungall, K. & 18 other authors (2000). The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665–668.[CrossRef][Medline]

Paustian, M. L., May, B. J. & Kapur, V. (2001). Pasteurella multocida gene expression in response to iron limitation. Infect Immun 69, 4109–4115.[Abstract/Free Full Text]

Pearson, B. M., Pin, C., Wright, J., I'Anson, K., Humphrey, T. & Wells, J. M. (2003). Comparative genome analysis of Campylobacter jejuni using whole genome DNA microarrays. FEBS Lett 554, 224–230.[CrossRef][Medline]

Pickett, C. L., Auffenberg, T., Pesci, E. C., Sheen, V. L. & Jusuf, S. S. (1992). Iron acquisition and hemolysin production by Campylobacter jejuni. Infect Immun 60, 3872–3877.[Abstract]

Raphael, B. H. & Joens, L. A. (2003). FeoB is not required for ferrous iron uptake in Campylobacter jejuni. Can J Microbiol 49, 727–731.[CrossRef][Medline]

Ratledge, C. & Dover, L. G. (2000). Iron metabolism in pathogenic bacteria. Annu Rev Microbiol 54, 881–941.[CrossRef][Medline]

Richardson, P. T. & Park, S. F. (1995). Enterochelin acquisition in Campylobacter coli: characterization of components of a binding-protein-dependent transport system. Microbiology 141, 3181–3191.[Medline]

Rock, J. D., van Vliet, A. H. M. & Ketley, J. M. (2001). Haemin uptake in Campylobacter jejuni. Int J Med Microbiol 291 (S31), 125.

Sopwith, W., Ashton, M., Frost, J. A., Tocque, K., O'Brien, S., Regan, M. & Syed, Q. (2003). Enhanced surveillance of campylobacter infection in the North West of England 1997–1999. J Infect 46, 35–45.[CrossRef][Medline]

Tseng, C. P. (1997). Regulation of fumarase (fumB) gene expression in Escherichia coli in response to oxygen, iron and heme availability: role of the arcA, fur, and hemA gene products. FEMS Microbiol Lett 157, 67–72.[CrossRef][Medline]

Vandamme, P. (2000). Taxonomy of the family Campylobacteraceae. In Campylobacter. Edited by I. Nachamkin & M. J. Blaser. Washington, DC: American Society for Microbiology.

van Vliet, A. H. M. & Ketley, J. M. (2001). Pathogenesis of enteric Campylobacter infection. J Appl Microbiol 90, 45S–56S.[CrossRef]

van Vliet, A. H., Wooldridge, K. G. & Ketley, J. M. (1998). Iron-responsive gene regulation in a Campylobacter jejuni Fur mutant. J Bacteriol 180, 5291–5298.[Abstract/Free Full Text]

van Vliet, A. H., Baillon, M. L., Penn, C. W. & Ketley, J. M. (1999). Campylobacter jejuni contains two fur homologs: characterization of iron-responsive regulation of peroxide stress defense genes by the PerR repressor. J Bacteriol 181, 6371–6376.[Abstract/Free Full Text]

van Vliet, A. H. M., Rock, J. D., Madeleine, L. N. & Ketley, J. M. (2000). The iron-responsive regulator Fur of Campylobacter jejuni is expressed from two separate promoters. FEMS Microbiol Lett 188, 115–118.[CrossRef][Medline]

van Vliet, A. H. M., Ketley, J. M., Park, S. F. & Penn, C. W. (2002). The role of iron in Campylobacter gene regulation, metabolism and oxidative stress defense. FEMS Microbiol Rev 26, 173–186.[CrossRef][Medline]

Wai, S. N., Nakayama, K., Umene, K., Moriya, T. & Amako, K. (1996). Construction of a ferritin-deficient mutant of Campylobacter jejuni: contribution of ferritin to iron storage and protection against oxidative stress. Mol Microbiol 20, 1127–1134.[CrossRef][Medline]

Wesley, I. V., Wells, S. J., Harmon, K. M., Green, A., Schroeder-Tucker, L., Glover, M. & Siddique, I. (2000). Fecal shedding of Campylobacter and Arcobacter spp. in dairy cattle. Appl Environ Microbiol 66, 1994–2000.[Abstract/Free Full Text]

Wooldridge, K. G. & Williams, P. H. (1993). Iron uptake mechanisms of pathogenic bacteria. FEMS Microbiol Rev 12, 325–348.[CrossRef][Medline]

Received 17 June 2004; revised 14 September 2004; accepted 23 September 2004.



This Article
Abstract
Full Text (PDF)
Supplementary data
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Holmes, K.
Articles by Wells, J. M.
Articles citing this Article
PubMed
PubMed Citation
Articles by Holmes, K.
Articles by Wells, J. M.
Agricola
Articles by Holmes, K.
Articles by Wells, J. M.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2005 Society for General Microbiology.