Diverse roles for HspR in Campylobacter jejuni ... - Semantic Scholar

5 downloads 0 Views 312KB Size Report
expression is likely to be responsible for the reduced motility and ability to autoagglutinate that was observed for hspR mutant cells. Electron micrographs ...
Microbiology (2005), 151, 905–915

DOI 10.1099/mic.0.27513-0

Diverse roles for HspR in Campylobacter jejuni revealed by the proteome, transcriptome and phenotypic characterization of an hspR mutant Marianne Thorup Andersen,1 Lone Brøndsted,1 Bruce M. Pearson,2 Francis Mulholland,2 Mary Parker,2 Carmen Pin,2 Jerry M. Wells2 and Hanne Ingmer1 1

Department of Veterinary Pathobiology, The Royal Veterinary and Agricultural University, Stigbøjlen 4, DK-1958 Frederiksberg C, Denmark

Correspondence Hanne Ingmer

2

[email protected]

Institute of Food Research, Colney, Norwich NR4 7UA, UK

Received 25 July 2004 Revised 12 November 2004 Accepted 18 November 2004

Campylobacter jejuni is a leading cause of bacterial gastroenteritis in the developed world. The role of a homologue of the negative transcriptional regulatory protein HspR, which in other organisms participates in the control of the heat-shock response, was investigated. Following inactivation of hspR in C. jejuni, members of the HspR regulon were identified by DNA microarray transcript profiling. In agreement with the predicted role of HspR as a negative regulator of genes involved in the heat-shock response, it was observed that the transcript amounts of 13 genes were increased in the hspR mutant, including the chaperone genes dnaK, grpE and clpB, and a gene encoding the heat-shock regulator HrcA. Proteomic analysis also revealed increased synthesis of the heat-shock proteins DnaK, GrpE, GroEL and GroES in the absence of HspR. The altered expression of chaperones was accompanied by heat sensitivity, as the hspR mutant was unable to form colonies at 44 6C. Surprisingly, transcriptome analysis also revealed a group of 17 genes with lower transcript levels in the hspR mutant. Of these, eight were predicted to be involved in the formation of the flagella apparatus, and the decreased expression is likely to be responsible for the reduced motility and ability to autoagglutinate that was observed for hspR mutant cells. Electron micrographs showed that mutant cells were spiral-shaped and carried intact flagella, but were elongated compared to wild-type cells. The inactivation of hspR also reduced the ability of Campylobacter to adhere to and invade human epithelial INT-407 cells in vitro, possibly as a consequence of the reduced motility or lower expression of the flagellar export apparatus in hspR mutant cells. It was concluded that, in C. jejuni, HspR influences the expression of several genes that are likely to have an impact on the ability of the bacterium to successfully survive in food products and subsequently infect the consumer.

INTRODUCTION Campylobacter jejuni is one of the most frequently diagnosed food-borne human pathogens to cause gastroenteritis, and contaminated poultry is thought to be one of the main food vehicles. In rare cases, C. jejuni triggers Guillain-Barre´ syndrome, an autoimmune disorder of the peripheral nervous system (Nachamkin et al., 1998). The pathogenesis of C. jejuni is not fully understood, but may involve damage to the intestinal epithelium by invasion of host cells and toxin production (Wassenaar, 1997; Wooldridge & Ketley, 1997). Characterized virulence factors include those involved in adhesion and invasion, as well as motility, which enables the spiral-shaped bacterium to move into the mucus layer,

presenting the bacterium to the intestinal epithelial cells (Szymanski et al., 1995; Guerry et al., 2000). In addition, the ability to cope with the unfavourable conditions encountered during transfer from the intestinal tract of poultry to the human body is crucial to the success of C. jejuni as a human food-borne pathogen. Despite the apparent sensitivity of Campylobacter to stresses in the laboratory, their infectivity does not seem to be compromised by exposure to environmental stresses during transmission from animals to humans (Ketley, 1997). Accordingly, a better understanding of the mechanisms by which C. jejuni overcomes the stresses encountered in the food chain is important for the identification, assessment and control of food hazards.

Abbreviations: 2D, two dimensional; HAIR, HspR-associated inverted repeat.

In general, exposure of bacteria to stressful conditions leads to induction of a set of heat-shock proteins that often fall

0002-7513 G 2005 SGM

Printed in Great Britain

905

M. Thorup Andersen and others

into one of two categories, the chaperones and the ATPdependent proteases. Inspection of the C. jejuni NCTC 11168 genome sequence has revealed the presence of several heat-shock protein homologues also found in other bacteria, including the major chaperones GroEL, GroES, GrpE, DnaK and DnaJ, and several ATP-dependent proteases (Parkhill et al., 2000). In many other Gram-negative bacteria, the response to heat stress is mediated by a specialized set of heat-shock sigma factors, such as s32, but in C. jejuni, there are only two alternative sigma factors, RpoN (s54) and FliA (s28), which both control expression of flagella genes (Nuijten et al., 1990; Guerry et al., 1991; Kinsella et al., 1997; Hendrixson et al., 2001; Hendrixson & DiRita, 2003; Jagannathan et al., 2001; Carrillo et al., 2004). Instead, C. jejuni possesses homologues of the HrcA and HspR regulators that negatively control the heat-stress response in other bacteria (Narberhaus, 1999). HrcA has been identified in more than 40 eubacteria, including proteobacteria and cyanobacteria (Zuber & Schumann, 1994; Narberhaus, 1999). Studies of Bacillus subtilis have shown that the transcriptional repressor HrcA requires the GroE chaperonin system for binding to the well-conserved CIRCE DNA element (Mogk et al., 1998; Reischl et al., 2002). Following exposure to stress, GroE levels are decreased due to its association with misfolded proteins, resulting in decreased binding of HrcA to DNA and activation of heat-shock gene expression. The HspR/HAIR (HspR-associated inverted repeat) repressor/operator system is less widely utilized in the bacterial kingdom, but has been described in actinomycetes and in Helicobacter pylori, an organism closely related to Campylobacter (Tomb et al., 1997). Like HrcA, HspR is a transcriptional repressor that binds to at least three HAIR sequences in the promoter region of regulated genes (Bucca et al., 1997; Grandvalet et al., 1997, 1999). Similarly, environmental stress releases HspR from its DNA-binding element, resulting in expression of the target genes. In Streptomyces coelicolor, DnaK seems to function as a co-repressor in a complex with HspR, and the sequestering of DnaK by denatured or partially unfolded proteins under stress conditions may result in release of HspR and induction of gene expression (Bucca et al., 2003). In Streptomyces, the HspR regulon includes the chaperones dnaK, clpB and dnaJ, as well as the lon protease gene, while in H. pylori the regulon comprises dnaK, groES, groEL and cbpA (Grandvalet et al., 1997; Spohn & Scarlato, 1999; Servant & Mazodier, 2001; Bucca et al., 2003). Interestingly, disruption of hspR in Streptomyces seems to have a limited impact on cell physiology (Grandvalet et al., 1997), whereas an hspR mutant in H. pylori is non-motile and has reduced levels of urease enzyme, suggesting that HspR or other components of the HspR regulon might influence cellular processes, in addition to the heat-shock response (Spohn & Scarlato, 1999). The aim of the present study was to characterize the phenotypic effects of an hspR mutation and identify members of the regulon by comparison of the proteome 906

and transcriptome of an hspR mutant and a wild-type strain of C. jejuni.

METHODS Bacterial strains and growth conditions. Campylobacter jejuni

strain NCTC 11168 was obtained from NCTC. Unless otherwise stated, C. jejuni was grown at 37 uC under microaerobic conditions (10 % CO2, 85 % N2, 5 % O2) inside a MACS-MG-1000 controlledatmosphere workstation (DW Scientific) on brain heart infusion (BHI) agar (Oxoid) or in BHI broth (Oxoid). Alternatively, the microaerobic environment was generated by CampyGen (Oxoid) and Blood Agar Base II (Oxoid) supplemented with 5 % calf blood (Base II) was used. Escherichia coli DH5a (w80lacZDM15 D(lacZYAargF)U169 recA1 endA1 hsdR17 supE44 thi-1 gyrA96 relA1) was grown in Luria–Bertani broth or on Luria–Bertani agar (Difco). When appropriate, media were supplemented with 100 mg ampicillin ml21 or 20 mg chloramphenicol ml21. Transformation. C. jejuni NCTC 11168 was transformed by electroporation essentially as described by Wassenaar et al. (1993a). To produce competent cells, C. jejuni NCTC 11168 was harvested from overnight-incubated agar plates with 2 ml ice-cold wash buffer (272 mM sucrose, 15 %, v/v, glycerol) and subjected to four wash steps with repeated centrifugations (4 uC, 10 000 r.p.m., 10 min). After the final centrifugation, cells were resuspended in 1/10 volume wash buffer, resulting in a concentration of approximately 109 c.f.u. ml21. A cell volume of 50 ml was electroporated (1?80 kV, 200 V, 25 F) with 1–5 ml plasmid DNA. Immediately after electroporation, 1 ml of recovery broth (10 % glycerol, 2/3 Mueller–Hinton broth, 1/3 Brucella broth) was added and cells were plated on non-selective plates. After incubation overnight at 37 uC, cells were harvested with recovery broth and plated on selective plates. E. coli DH5a was transformed by standard methods (Cohen et al., 1972). DNA manipulations. Extraction of chromosomal DNA from C.

jejuni was performed using Fast Prep DNA (Bio 101), with the modification that cells were incubated in phosphate buffer, pH 7?0, containing 10 mg lysozyme ml21 and 20 % sucrose for 80 minutes at 37 uC prior to DNA purification. Recombinant plasmid DNA from E. coli was isolated by using QIAGEN columns, as recommended by the supplier (Qiagen). Biolabs supplied restriction endonuclease enzymes and Klenow DNA polymerase. All enzymes were used as recommended by the supplier. PCR reactions were performed using Bioline taq DNA polymerase and buffer (DNA Technology). The following oligonucleotides were used in this study: HspR-A-F, 59-CAATCTAGAGCTTCACATTCCT-39; HspR-B-R, 59GACGTCGCCTTAAGGTTTGTGGATGTA-39; HspR-C-F, 59-CACAAACCTTAAGGCGACGTCATGAACTACGCTTAGAATTAAGCA-39; HspR-D-R, 59-GCAAAGGAATAACGGCTATATC-39; CAT-up, 59GTCCTGAACTCTTCATGTCG-39; CAT-down, 59-CGTGGACAAGCTTTGAAGG-39. Construction of C. jejuni hspR mutant. C. jejuni NCTC 11168

chromosomal DNA was used as template for amplification. For construction of the hspR mutant, a 593 bp DNA fragment containing the 59 end of the hspR gene and upstream sequences (primers HspRA-F and HspR-B-R) and a 593 bp fragment carrying the 39 end of the hspR gene and downstream sequences (primers HspR-C-F and HspR-D-R) were amplified. In a second round of PCR, the hspR fragments were joined by the splicing by overlap extension PCR method (Horton et al., 1993), creating a PCR fragment containing an in-frame deletion of 187 bp in hspR and introducing an AatII site between the upstream and downstream fragments. The DhspR PCR fragment was cloned in the TOPO TA cloning vector pCR2.1 (Invitrogen), resulting in plasmid pHspR-AD. Finally, pHspR-AD was digested with AatII, followed by treatment with Klenow, and Microbiology 151

HspR in Campylobacter jejuni ligated to a DNA fragment containing the chloramphenicol acetyl transferase (cat) gene obtained from pRY109 (Yao et al., 1993) to generate pHspR-ADcat, in which the cat gene was transcribed in the same direction as hspR. C. jejuni NCTC 11168 was transformed with pHspR-ADcat (DhspR : : cat), and several chloramphenicol-resistant colonies were isolated. Chromosomal DNA was isolated from six different chloramphenicol-resistant colonies and used as template in PCR reactions to verify that the mutations were transferred by a double crossover event to the chromosome of C. jejuni NCTC 11168. Primers that annealed to sequences upstream and downstream of the region cloned in pHspR-ADcat (primers HspR-E-F and HspR-F-R) were combined with primers CAT-up and CATdown annealing internally in the cat gene. In each case, PCR fragments were obtained that verified the double crossover event, and one of these mutants, C. jejuni DhspR : : cat (MTA55), was used in subsequent experiments. Construction of the C. jejuni DNA microarray. Internal DNA

fragments corresponding to unique segments of the individual ORFs in the annotated genome sequence of C. jejuni strain NCTC 11168 were amplified using gene-specific primers, as described in Pearson et al. (2003). DNA probes were spotted on GAPSII slides (Corning) using an in-house Stanford-designed arrayer and employing the recommended software and protocols (for further details, see http:// cmgm.stanford.edu/pbrown/mguide/). Each glass slide contained two arrays each, with duplicate probes (or features) for each gene. RNA isolation and purification. Cultures (50 ml in BHI) were

grown in triplicate to an OD600 of 0?3, and bacteria were harvested by centrifugation at 3000 g for 20 min, resuspended in 1 ml TriReagent (Sigma-Aldrich) 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 RNeasy Mini columns for RNA purification, according to the manufacturer’s protocol. DNA removal was ensured by treatment with DNA-free (Ambion), and the quality and quantity of RNA were checked using the Agilent 2100 Bioanalyser (Agilent Technologies, www.agilent.com/chem/ labonachip). Microarray transcriptome analysis. Two independent RNA pre-

parations (biological replicates) of each sample were labelled and hybridized to glass-slide microarrays. Labelled cDNA was prepared from 15 mg RNA using Stratascript RT (Stratagene) for 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 ml water, 2?25 ml human Cot1 DNA (Invitrogen), 4?5 ml 206 SSC, 0?72 ml 1 M HEPES, pH 7?0, 0?68 ml 10 % SDS and 3 ml Denhardt. Samples were heated for 3 min in a boiling water bath, 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, sealed with a coverslip in a GeneMachine hybridization chamber (Anachem) and incubated for 18 h at 63 uC. Following hybridization, microarray slides were washed briefly in pre-warmed (60 uC) 16 SSC/0?03 % SDS to remove the coverslip and then washed twice for 5 min in each of the following buffers: a) 16 SSC/0?03 % SDS (pre-warmed to 60 uC), b) 0?26 SSC and finally c) 0?056 SSC. Microarray slides were dried by centrifugation at 300 g for 15 min before scanning. Microarray data analysis. Microarrays were scanned using an

Axon 4000A microarray scanner and images were acquired using GenePixPro 3.0 software (Axon). All microarray data were filtered to remove poor-quality data using four sequential cut-off values (Mark Reuter, IFR, personal communication). All data values were from features above 50 mm in diameter, with sum of medians above 50, regression coefficient squared values above 0?2 and with a sum of the signal : noise ratio greater than 3. http://mic.sgmjournals.org

A control experiment (comparison of two mRNA samples from replicated independent cultures of the wild-type strain) was used to estimate the boundaries between genes that were equally and differentially expressed in the two samples (Holmes et al., 2005). For the analyses described here, this boundary would detect changes equivalent to about 3?3-fold greater intensity in one of the fluorescence channels. Applying this boundary to the control dataset gave an error rate for misclassification of approximately 0?43 % of the gene features that give a fluorescence signal above background. Approximately 84 % of the 1654 annotated gene probes gave a fluorescence signal above background under these experimental conditions. All test hybridization datasets (hspR mutant versus wild-type) were normalized (Holmes et al., 2005). The normalized data from each independent array were then unified in one single dataset and reanalysed. The genes that potentially had increased or decreased expression levels were first identified as those with mean intensities outside the boundaries specified above, and were further tested statistically by an F test. According to this test, the genes were classified as either differently or equally expressed in both populations. Genes identified in these analyses were annotated using on-line databases (http://www.sanger.ac.uk/Projects/C_jejuni/0). Proteome analysis by two-dimensional (2D) gel electrophoresis. Campylobacter cells were harvested in late exponential phase

by centrifugation (3000 g for 10 min) and washed with Tris-buffered saline, pH 7?5, prior to lysis by four times 1 min glass bead beating (106 mm 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 mg RNase ml21 and 1?67 U Dnase ml21. 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 were determined using the 2D Quant Kit (Amersham) according to the manufacturer’s instructions. Proteomic analysis of the cell-free extracts, including 2D electrophoresis, imaging, spot picking, digestion and MALDI-TOF analysis, was carried out as described in Holmes et al. (2005), using 100–125 mg protein per IPG strip. Physiological studies. C. jejuni strains were grown overnight on

Base II agar plates at 37 uC under microaerobic conditions (Campygen, Oxoid), harvested in BHI medium and adjusted to an OD600 of 0?1. Serial dilutions were made, and 10 ml of each dilution was spotted onto three Base II agar plates, which were incubated microaerobically (Campygen, Oxoid) for 3 days at 37, 42 or 44 uC. Alternatively, 10 ml of each dilution was spotted onto a Base II agar plate containing 10, 15 or 20 mg puromycin ml21 and incubated microaerobically (Campygen, Oxoid) for 5 days at 37 uC. Growth at 17–18 % O2 was obtained by incubating in a candle light atmosphere (Wang et al., 1983). Growth in liquid culture of C. jejuni strains was examined under microaerophilic conditions (5 % O2 influx) by inoculating 50 ml prewarmed BHI (37 or 42 uC) to a final OD600 of 0?1. Two flasks were incubated at each growth temperature (i.e. 37 and 42 uC), one of which was shaken continuously and the other kept static, except for mixing at hourly intervals when aliquots were taken for OD600 measurement. Light and electron microscopy. C. jejuni was grown overnight in sterile filtered Mueller–Hinton broth (MH, Oxoid) under microaerophilic conditions (5 % O2 influx) at 37 or 42 uC. Bacterial cells were examined either under the light microscope (1006 magnification) or by electron microscopy. For the latter, bacteria were negatively stained with saturated aqueous uranyl acetate and photographed in a JEOL EX/B transmission electron microscope at 80 kV. Motility/chemotaxis assay. C. jejuni strains were grown overnight, harvested in BHI medium and adjusted to an OD600 ml21 of 0?1. One microlitre was spotted onto the centre of heart infusion

907

M. Thorup Andersen and others broth (Difco) motility plates containing a low concentration (0?25 %) of agar, and the motility of C. jejuni was examined after 48 h of microaerobic incubation at 37 uC. Autoagglutination assay. Autoagglutination assays with the wild-

type and hspR mutant strains were performed essentially as described by Misawa & Blaser (2000). Briefly, C. jejuni cells were grown overnight on Base II agar plates, harvested in MilliQ water and washed once by centrifugation before being resuspended in 10 mM PBS, pH 7?2, to a final concentration of 1 OD600 ml21. Cell suspensions were then incubated under aerobic conditions at 25 uC, and 1 ml from the top of the suspension was collected to measure OD600. C. jejuni cells normally agglutinate and precipitate, causing a decrease in the OD. The mean percentage decrease in OD600 and standard deviation were calculated using data from three independent assays. Adherence and invasion assay. INT 407 cells were cultured in

Eagle’s minimal essential medium (EMEM) supplemented with 10 % fetal calf serum (FCS) at 37 uC in a humidified 5 % CO2 incubator. For the adherence and invasion assay, approximately 16108 bacterial cells were added to a monolayer of 56105 INT407 epithelial cells (m.o.i. 200) and incubated for 2 h. In order to determine the number of adherent bacteria, the epithelial cell monolayers were washed three times with 10 mM PBS, pH 7?2, and then lysed with a solution of 0?1 % (v/v) Triton X-100. The suspensions were serially diluted and the number of viable adherent bacteria was determined by enumeration of bacterial colonies after plating on Base II agar. To measure the number of internalized bacteria, the epithelial cell monolayers were incubated with bacteria, as described above, and then treated with EMEM containing 250 mg gentamicin ml21 for 2 h at 37 uC in 5 % CO2 in order to kill extracellular bacteria. The epithelial cells were then washed three times with PBS, lysed with 0?1 % (v/v) Triton X-100 and plated to enumerate the internalized viable bacteria, as described above. The values obtained represented the mean count and standard deviation calculated using samples from four replicate wells.

RESULTS Inactivation of HspR affects cell morphology in C. jejuni

growth rate of the DhspR mutant was identical to wild-type cells at 37 uC, as determined both by OD600 measurement and by colony forming units (data not shown); however, when examined by microscopy, we found that the cells of the HspR mutant strain were highly elongated, and this was more pronounced at 42 uC than at 37 uC (Fig. 1). Genome-wide transcriptional profile of C. jejuni mutant DhspR Since HspR is a negative regulator of the heat-shock response in other bacteria, we predicted that a comparison of the transcriptome of the hspR mutant with the wildtype strain would reveal gene members of the HspR regulon that had been depressed in the mutant. The transcriptome of the hspR mutant and wild-type strain were compared in exponential-phase growth by DNA microarray analysis, as described in Methods. These experiments were repeated twice using independently isolated samples of RNA (i.e. two biological replicates), and hybridization was performed in duplicate (i.e. two technical replicates). The results presented in Table 1 show that the transcript levels of 13 and 17 genes were increased and decreased, respectively, in the hspR mutant strain compared to the wild-type strain. Transcript levels of the putative heatshock operon genes hrcA, grpE and dnaK, as well as two downstream genes encoding hypothetical proteins (Cj0760 and Cj0761), were substantially elevated (4?8- to 28?2-fold) in the hspR mutant, suggesting that this operon is negatively regulated by HspR (Table 1). While the functions of the hypothetical proteins are unknown, Cj0760 shows homology to hydrolases of the metallo-beta-lactamase superfamily. Interestingly, the gene encoding the ATP-dependent chaperone ClpB, which interacts with DnaK to reactivate proteins that have become aggregated after heat shock

In the genome sequence of Campylobacter jejuni NCTC 11168, gene Cj1230 is annotated as hspR. Sequence analysis (BLASTP) confirmed that Cj1230 encodes a 124 amino acid product with extensive homology (63 % identity over 120 amino acids) to the transcriptional regulator HspR from H. pylori. In C. jejuni, hspR is located 19 bp downstream of cbpA (Cj1229), which encodes a predicted protein with homology to the DnaJ chaperone (Ueguchi et al., 1994). While no obvious promoter-like sequences are present immediately upstream of hspR, the gene is preceded by a Shine–Dalgarno sequence (AAGGAA) located 8 nucleotides upstream of the start codon. In H. pylori, cbpA and hspR form an operon, and the promoter elements identified upstream of H. pylori cbpA are also conserved in C. jejuni (data not shown). With the aim of determining the role of HspR and members of the HspR regulon in C. jejuni, we substituted the central 186 bp corresponding to nucleotides 83 to 268 of hspR with a chloramphenicol resistance gene marker that is transcribed in the same direction as hspR (see Methods). The 908

Fig. 1. Electron micrographs of wild-type and hspR mutant cells. Electron microscopy of (a) C. jejuni NTCT 11168 (wildtype), (b) C. jejuni MTA55 (DhspR), both grown at 37 6C, and (c) C. jejuni MTA55 (DhspR) grown at 42 6C. All cells were collected in the late-exponential growth phase. Bars, (a) and (b), 500 nm; (c), 1000 nm. Microbiology 151

HspR in Campylobacter jejuni

Table 1. Results of a DNA microarray-based comparison of relative transcript levels in the hspR mutant and the wild-type strain The gene number is according to the Sanger Centre annotation for strain NCTC 11168. The fold change in the transcript level of hspR mutant cells relative to wild-type cells is shown; a negative fold change value indicates that expression is decreased in the mutant relative to the wild-type. Gene no.

Gene name

Fold change in hspR mutant

Proposed gene function

Cj0043 Cj0044c Cj0453 Cj0687c Cj0697 Cj0887c Cj0986c Cj1242 Cj1338c Cj1339c Cj1349c Cj1364c Cj1450 Cj1462 Cj1502c Cj1537c Cj1729c Cj0191c Cj0257 Cj0378c Cj0509c Cj0515 Cj0757 Cj0758 Cj0759 Cj0760 Cj0761 Cj0892c Cj1202 Cj1230

flgE 2 thiC flgH flgG2 flaD 2 2 flaB flaA 2 fumC 2 flgI putP acs flgE2 def dgkA 2 clpB 2 hrcA grpE dnaK 2 2 2 metF hspR

24?7 23?5 23?9 24?8 24?9 26?6 23?8 23?4 24?2 25?1 23?5 24?7 24?0 26?7 23?6 24?8 23?5 4?0 3?3 3?3 21?8 3?4 4?8 11 11?8 28?2 15?4 3?9 3?4 4?0

Flagellar hook protein flgE Hypothetical protein Cj0044c Thiamin biosynthesis protein ThiC Putative flagellar L-ring protein precursor Putative flagellar basal-body rod protein Putative flagellin Putative integral membrane protein Hypothetical protein Cj1242 Flagellin Flagellin Possible fibronectin/fibrinogen-binding protein Fumarate hydratase Putative ATP/GTP-binding protein Flagellar P-ring protein Sodium/proline symporter Acetyl-coenzyme A synthetase Flagellar hook subunit protein Polypeptide deformylase Diacylglycerol kinase Putative integral membrane protein ATP-dependent CLP protease ATP-binding subunit Putative periplasmic protein Putative heat-shock regulator Heat-shock protein grpE Heat-shock protein dnaK Hypothetical protein Cj0760 Hypothetical protein Cj0761 Putative periplasmic protein 5,10-Methylenetetrahydrofolate reductase Putative heat-shock transcriptional regulator

(Goloubinoff et al., 1999), was also transcribed at substantially higher levels (21?8-fold) in the hspR mutant. In the C. jejuni strain NCTC 11168, clpB appears to be transcribed as a single cistron from its own promoter. In addition, higher amounts of hspR transcripts were present in the hspR mutant (4?0-fold) compared to the wild-type strain. In the hspR mutant strain, hspR was inactivated by replacement of an internal fragment with the cat gene; thus, increased transcription from either the upstream cbpA promoter or the cat gene promoter could increase the amount of labelled cDNA hybridizing to the hspR probe on the microarray. Polypeptide deformylase, the essential enzyme that cleaves the formyl group of the N terminus of nascent polypeptide chains, was also expressed at 4?0-fold higher levels in the hspR mutant. Of the genes that were reduced in expression in the absence of HspR, eight were predicted to be involved in assembly http://mic.sgmjournals.org

of the flagellar apparatus, including genes encoding flagellin (FlaA and FlaB) and a putative flagellin (FlaD), as well as the ring protein (FlgH), hook protein (FlgE, FlgE2) and basal body protein (FlgG2) (Table 1). In addition, genes encoding acetyl-coenzyme A synthetase and a sodium/proline transporter were also transcribed at lower levels in the hspR mutant compared to the wild-type. Proteome analysis of the hspR mutant The effect of HspR on protein expression was investigated by comparing the proteomes of hspR mutant and wildtype cells. Proteins were visualized by Sypro-Ruby staining and quantified by the use of a multiwave fluoroimager (ProExpress) and the Proteomweaver 2D gel analysis software package. Digital images from 2D gels of protein samples of the hspR mutant and wild-type strain indicating the differentially expressed proteins are shown in Fig. 2. 909

M. Thorup Andersen and others

(a)

(b)

Fig. 2. Two-dimensional protein gel analysis of C. jejuni strain 11168 (a) and the hspR mutant (b) derived from this strain. Proteins were extracted from bacteria cultivated in BHI at 37 6C collected when OD600 reached 0?4.

Analysis of the gels using Proteomweaver 2D gel analysis software followed by manual filtering and editing identified six of the matched spots as being significantly higher in the hspR mutant (Table 2). Apart from the cat gene, which used a selectable marker for inactivation of hspR, all other proteins increased in amount in the hspR mutant were linked to the heat-shock response (i.e. DnaK, GrpE, GroEL, GroES and ClpB). The measured increase in these

heat-shock proteins ranged from 3?7- to 35-fold (Table 2). Curiously, none of the proteins encoded by the flagellaassociated genes transcribed at lower levels in the hspR mutant were identified by proteomic analysis. Possible reasons for this include poor solubility of the flagella proteins due to high stability of the flagella apparatus, or the hydrophobic nature of the membrane-associated flagella components.

Table 2. Proteomic analysis of hspR mutant cells Mol. wt, molecular weight; WT, wild-type. Mowse scores greater than 46 are considered significant when searching against Campylobacter jejuni sequences. Spot intensity values show the geometric mean of two independent gels run together under the same conditions. Protein identification

ATP-dependent clpB protease ATPbinding subunit GrpE protein (HSP-70 cofactor) Heat-shock protein DnaK 10 kDa chaperonin (cpn10) GroES 60 kDa chaperonin (cpn60) GroEL Chloramphenicol acetyltransferase (Cat)

910

Cj no.

Mol. wt

pI

Mowse score for ID

Sequence coverage for ID

WT 37 6C, mean spot intensity

hspR mutant 37 6C, mean spot intensity

hspR mutant 37 6C, induction compared to WT (-fold)

Cj0509C

95485

5?47

82

12 %

0?807

24?773

30?7085

Cj0758 Cj0759 Cj1220 Cj1221

20412 67491 9452 57991 24504

4?83 4?98 5?38 5?02 5?52

144 205 70 320 98

51 % 28 % 58 % 54 % 23 %

0?604 3?087 4?356 31?434

9?893 107?732 16?026 141?34 2?071

16?3914 34?8996 3?6794 4?4964

Microbiology 151

HspR in Campylobacter jejuni

Temperature and oxygen influence growth in the absence of HspR

sensitivity is not a result of reduced tolerance to misfolded proteins.

Previous studies have shown that increased expression of chaperones in some cases improves the ability of cells to adapt to heat stress (Stewart et al., 2001). As several chaperones were expressed more highly in the hspR mutant compared to the wild-type strain, we investigated the heat tolerance of the two strains. After 3 days growth at the optimum growth temperature of 42 uC and at the lower temperature of 37 uC, both strains produced similar numbers of colony forming units (c.f.u.) on agar plates (Fig. 3a). However, after growth at 44 uC, significantly fewer colonies were recovered from the hspR mutant compared to the wild-type, indicating that in C. jejuni HspR is important for growth at high temperature (Fig. 3a).

During the course of these experiments, we also noted that the temperature sensitivity of the hspR mutant was related to oxygen concentration. The growth rate of wild-type and hspR mutant cells was identical at 42 and 37 uC when cultivated in shaken liquid cultures with 5 % O2. However, we found that the hspR mutation conferred oxygen sensitivity, as mutant cells formed colonies at a significantly lower frequency than the wild-type at 37 uC in the presence of 17–18 % O2 (Fig. 3b). These results show that HspR is required for oxygen tolerance in C. jejuni.

To determine if HspR is important for growth under other conditions that lead to the accumulation of misfolded proteins, we examined the ability of the hspR mutant and the wild-type strain to form colonies in the presence of the aminoacyl-tRNA analogue pyromycin, which causes premature termination of protein synthesis, leading to puromycyl-containing polypeptides (VanBogelen et al., 1987; Goldberg, 1972). When exposed to various concentrations of puromycin, we found no differences between the mutant and the wild-type in the ability to form colonies (data not shown), suggesting that the heat

DNA microarray analysis revealed that the transcript levels of eight flagella-associated genes were reduced in the hspR mutant. This finding prompted us to investigate the presence of flagella in the mutant cells by electron microscopy. DNA micrographs revealed that both strains were spiralshaped and possessed flagella of a similar length (Fig. 1, and data not shown). However, on agar motility plates there was an obvious reduction in the motility of the hspR mutant compared to wild-type cells both at 37 uC (Fig. 4) and at 42 uC (data not shown). This result was reproducibly obtained in independent experiments and also confirmed

HspR is required for normal motility/chemotaxis

(a)

(b) 37 ºC WT

42 ºC DhspR

WT

DhspR

WT

44 ºC DhspR

WT

DhspR

Fig. 3. Temperature- and oxygen-dependent growth of wild-type and hspR mutant cells. (a) Growth was evaluated by plating dilutions of C. jejuni wild-type and hspR mutant cells on Base II agar plates and incubating at 37, 42 and 44 6C under microaerophilic conditions (CampyGen). (b) Growth in a candle-jar atmosphere (17–18 % O2) was evaluated after 3 days at 42 6C. http://mic.sgmjournals.org

911

M. Thorup Andersen and others

DhspR

WT

Fig. 4. Inactivation of hspR affects motility. Motility/chemotaxis was evaluated for wild-type and hspR mutant cells grown on soft agar for 2 days at 37 6C. Plate diameter, 8?5 cm.

by examination of the motility of the strains by light microscopy (data not shown). In C. jejuni, expression of flagella has been correlated with the ability to autoagglutinate (Misawa & Blaser, 2000; Golden & Acheson, 2002). When autoagglutination was examined in static cultures, wild-type cells formed a distinct pellet at the bottom of the tube that was not apparent with the hspR mutant. The OD600 of the wild-type culture was reduced to 9?5 %±1?2 of the initial value, while that of the mutant was only reduced to 14?9 %±0?9 %. Thus, we concluded that the hspR mutation results in a minor reduction in the ability of C. jejuni to autoagglutinate. The adherence and invasion ability of the hspR mutant is impaired In C. jejuni, the flagella and motility are critical both for invasion of gastrointestinal epithelial cells and for colonization of the mucus lining of the gastrointestinal tract (Yao et al., 1994). As the hspR mutation reduced expression of a number of flagella genes, we determined the ability of mutant and wild-type cells to adhere to and invade INT-407 epithelial cells. Using an m.o.i. of 200 to maximize the number of internalized bacteria (Hu & Kopecko, 1999), we calculated that the number of hspR mutant bacteria adhering to the epithelial cells was only 6 % of that obtained with the wild-type strain (data not shown). Furthermore, the invasion capacity of the mutant was reduced to only 4 % of that measured for the wild-type strain. Thus, both adherence and invasion properties were reduced in the hspR mutant.

DISCUSSION Heat-shock proteins are among the most highly conserved proteins in nature, and diverse regulatory mechanisms have evolved for controlling their synthesis. We have examined the role of a C. jejuni homologue of the putative negative 912

heat-shock regulatory protein HspR in controlling gene expression in C. jejuni. Microarray comparison of the transcriptome of hspR mutant cells and the wild-type strain revealed that the transcript levels of 13 and 17 genes were increased and decreased, respectively, in the hspR mutant. In the absence of HspR, transcript levels of all genes in the hrcA–grpE–dnaK operon were substantially increased, suggesting that this heat-shock operon is normally regulated negatively by HspR. The hspR gene was also more highly expressed in the mutant, suggesting that it may be auto-regulated, as previously observed in H. pylori (Spohn & Scarlato, 1999). The promoter for hspR probably lies upstream of cbpA (Cj1229) as there is only a 19 bp intergenic region between cbpA and hspR. Although the cbpA gene was not classified as being differentially expressed in the hspR mutant, the relative transcript amounts were 2?2fold higher in the hspR mutant than in the wild-type. This value is under the statistical cut-off used for differentially expressed genes (see Methods), but might possibly be biologically significant. The proteome and transcriptome results also indicated that, in contrast to H. pylori, the Campylobacter HspR controls expression of clpB; the latter encodes a putative molecular chaperone that in E. coli cooperates with DnaK to remove aggregated proteins (Goloubinoff et al., 1999). Furthermore, the proteome analysis revealed that the GroES and GroEL proteins were also produced in higher amounts in the hspR mutant, although they were not classified as being differentially expressed by DNA microarray analysis of the transcriptomes. However, there was a 1?9-fold increase in the amount of transcript of both the groEL and groES genes in the hspR mutant compared to the wild-type. Thus, as in the case of cbpA, it is possible that these changes in transcript levels were biologically relevant, even if below our statistical cut-off. This notion was supported by inspection of the C. jejuni NCTC 11168 genome for sequences that resemble the HspR binding site, Microbiology 151

HspR in Campylobacter jejuni

HAIR, reported for Streptomyces albus G (CTTGAGT-N7ACTCAAG, Grandvalet et al., 1999) and for H. pylori (Delany et al., 2002), as we found putative HspR binding sites in the promoter region of groEL–groES (CTTGGCTTTATCAAACTCAAC) as well as in front of hrcA–grpE– dnaK (ATCAAGTGGAATTCATTCAAG) and cbpA–hspR (CTTGAATGAATAAGACTTAAT). In contrast, we were not able to identify HAIR-like sequences in association with other of the regulated genes. The biological importance of HspR at elevated temperature was revealed by the finding that C. jejuni is unable to form colonies at 44 uC in the absence of HspR. In contrast, hspR mutant cells of Mycobacterium tuberculosis were more temperature resistant than wild-type cells (Stewart et al., 2001), while disruption of hspR in Streptomyces albus G gave a marginal growth defect at 30 uC but not at 37 uC (Grandvalet et al., 1997). Interestingly, the major heatshock proteins are overexpressed in all three cases, suggesting that they are not responsible for the observed differences. Additional evidence for the role of HspR in the C. jejuni heat-shock response comes from a recent DNA microarray analysis of cells shifted from 37 to 42 uC, which shows that the heat-shock gene clusters with altered transcript levels overlap with those that we showed to be de-repressed in the absence of HspR (Stintzi, 2003). In C. jejuni, HspR controls expression of the other negative heat-shock regulator, HrcA, and putative binding sites for HrcA (Schulz & Schumann, 1996) precede dnaK and groEL (data not shown), indicating a complex interrelationship between the HrcA and HspR regulons. Likewise, it has been shown that both dnaK and groEL are co-regulated by HspR and HrcA in H. pylori (Spohn et al., 2004). Coregulation of heat-shock genes has also been observed in Staphylococcus aureus, and it was hypothesized that the cooperative activity of both HrcA and CtsR is required to maintain low heat-shock gene expression in the absence of stress (Chastanet et al., 2003). Similarly, certain heatshock genes in C. jejuni may require both HspR and HrcA to ensure a low level of expression in the absence of stress. To address the respective roles of the two regulatory proteins in controlling gene expression in C. jejuni, we also set out to delete the hrcA gene, but numerous attempts proved unsuccessful (data not shown), suggesting that hrcA or a gene product encoded by the hrcA regulon may be essential for growth in the laboratory. We conclude that the C. jejuni HspR homologue encodes a negative regulator of the expression of several major chaperones in C. jejuni. The motility of Campylobacter is conferred by polar flagella, and approximately 40 genes are predicted to be involved in flagella biosynthesis or function (Yao et al., 1994; Parkhill et al., 2000). Interestingly, our DNA microarray analysis revealed that transcription of eight flagella-associated genes was reduced in hspR mutant cells when compared to wild-type cells. By microscopy and motility assays, we showed that independently isolated hspR mutants had reduced motility. Motility is correlated with the ability to http://mic.sgmjournals.org

self-associate or auto-agglutinate, and mutations affecting motility often influence auto-agglutination (Misawa & Blaser, 2000; Golden & Acheson, 2002). Likewise, we observed that the reduction in motility conferred by the hspR mutation was accompanied by a moderate decrease in the ability to auto-agglutinate. In general, flagellum biosynthesis is dependent on expression of the major structural component, flagellin, in addition to gene products constituting the flagella basal body, and switch- and hook-associated proteins (Aldridge & Hughes, 2002). In C. jejuni, flaA encodes the major flagellin, although the minor flagellin component FlaB is also required for normal motility (Guerry et al., 1991). Interestingly, the expression of both flaA (Cj1339c) and flaB (Cj1338c), as well as a third putative flagellin gene, flaD (Cj0887c), was reduced in the absence of HspR. In addition, genes involved in formation of the flagella hook (flgE, Cj0043; flgE2, Cj1729), flagella Lring (flgH, Cj0687), P-ring (flgI, Cj1462) and basal body (flgG2, Cj0697) were all expressed at substantially lower levels in the mutant (3?5- to 6?7-fold). Several of these gene products are required for normal motility, as studies have shown that motility is reduced by mutations in flaA, flgD, flgH, motA (Hendrixson et al., 2001), flaD, flaA, flgI (Golden & Acheson, 2002), flgE (Kinsella et al., 1997), flaD, flgE (Colegio et al., 2001) flaA, flaB (Guerry et al., 1991) and flgE2 (Konkel et al., 2004). Thus, the expression of a number of gene products involved in flagella formation is reduced in the absence of HspR, and this reduction is likely to be the reason for the impaired motility. In H. pylori, it was recently shown that the absence of hspR impaired motility, suggesting that HspR in this organism also is a central regulator of flagella biosynthesis (Spohn & Scarlato, 1999). In C. jejuni, several factors are involved in expression of flagella genes (Jagannathan et al., 2001). Of the genes determined by our DNA microarray as being controlled by HspR, expression of flaA is dependent on s28 (FliA), while flaB and flgDE are expressed from a s54-associated RNA polymerase, (Guerry et al., 1991; Kinsella et al., 1997; Lu¨neberg et al., 1998; Hendrixson et al., 2001). Additionally, sequence analysis of the C. jejuni genome revealed putative s54 promoter sequences upstream of flgH, flgG2, flgI and flgE2 (Hendrixson & DiRita, 2003; Wo¨sten et al., 2004). The finding that both s28- and s54-controlled genes are affected by the hspR mutation suggests that the effect of HspR on flagella gene expression is not mediated via a s54- or s28specific factor, but rather involves a regulator affecting both of these regulons. Since HspR has not previously been shown to directly stimulate gene expression, we believe that expression of a putative negative regulator is controlled by HspR. Such a regulator might also be involved in the heat-induced expression of several flagella genes that has observed in C. jejuni cells shifted from 37 to 42 uC (Stintzi, 2003). Interestingly, the positive effect of HspR on gene expression may be unique to C. jejuni, and perhaps H. pylori, as DNA microarray analysis of M. tuberculosis and S. coelicolor failed to reveal genes with expression stimulated by HspR (Stewart et al., 2002; Bucca et al., 2003). 913

M. Thorup Andersen and others

In C. jejuni, motility is associated with pathogenesis, and colonization of the gastro-intestinal tract depends on FlaA expression (Wassenaar et al., 1993b; Morooka et al., 1985; Nachamkin et al., 1993). Accordingly, we found that both adherence and invasion of hspR mutant cells in INT407 epithelial cells were substantially reduced when compared to wild-type cells. Since our electron micrographs revealed intact flagella, the defective interaction with host cells is likely due to reduced motility rather than a lack of FlaA expression. In addition to the altered expression of flagella gene products that are known to function as adhesins (Yao et al., 1993) and to secrete virulence proteins (Konkel et al., 2004; Song et al., 2004), we also found that expression of a putative fibronectin/fibrinogen binding protein (Cj1349) was reduced in the absence of HspR. Previously, fibronectin binding by C. jejuni has been shown to be mediated by CadF, which also is required for maximal adherence to and invasion of INT407 cells (Monteville et al., 2003). However, the role of Cj1349 in these processes will be the focus of future investigations. In conclusion, C. jejuni HspR controls the expression of genes involved in diverse processes, including motility, stress tolerance, morphology and virulence, suggesting that either directly or indirectly it is a key regulator of adaptive responses both in the environment and inside host organisms.

Colegio, O. R., Griffin, T. J., Grindley, N. D. & Galan, J. E. (2001). In

vitro transposition system for efficient generation of random mutants of Campylobacter jejuni. J Bacteriol 183, 2384–2388. Delany, I., Spohn, G., Rappuoli, R. & Scarlato, V. (2002). In vitro

selection of high affinity HspR-binding sites within the genome of Helicobacter pylori. Gene 283, 63–69. Goldberg, A. L. (1972). Degradation of abnormal proteins in

Escherichia coli (protein breakdown-protein structure-mistranslationamino acid analogs-puromycin). Proc Natl Acad Sci U S A 69, 422–426. Golden, N. J. & Acheson, D. W. (2002). Identification of motility and autoagglutination Campylobacter jejuni mutants by random transposon mutagenesis. Infect Immun 70, 1761–1771. Goloubinoff, P., Mogk, A., Zvi, A. P., Tomoyasu, T. & Bukau, B. (1999). Sequential mechanism of solubilization and refolding of

stable protein aggregates by a bichaperone network. Proc Natl Acad Sci U S A 96, 13732–13737. Grandvalet, C., Servant, P. & Mazodier, P. (1997). Disruption of

hspR, the repressor gene of the dnaK operon in Streptomyces albus G. Mol Microbiol 23, 77–84. Grandvalet, C., Crecy-Lagard, V. & Mazodier, P. (1999). The ClpB

ATPase of Streptomyces albus G belongs to the HspR heat shock regulon. Mol Microbiol 31, 521–532. Guerry, P., Alm, R. A., Power, M. E., Logan, S. M. & Trust, T. J. (1991). Role of two flagellin genes in Campylobacter motility.

J Bacteriol 173, 4757–4764. Guerry, P., Alm, R., Szymanski, C. M. & Trust, T. J. (2000). Structure,

ACKNOWLEDGEMENTS We would like to thank P. Guerry for supplying plasmid pRY109. K. Jørgensen is thanked for helpful discussions. H. Ø. Jarmer is acknowledged for assisting the computational analysis. The Danish Directorate for Food, Fisheries and Agro Business and The Royal Veterinary and Agricultural University, Denmark, supported M. T. Andersen and L. Brøndsted. J. W., B. P., F. M. and M. P. gratefully acknowledge financial support from the Biotechnology and Biological Sciences Research Council (BBSRC) UK for a Core Strategic Grant for research on Campylobacter.

REFERENCES Aldridge, P. & Hughes, K. T. (2002). Regulation of flagellar assembly.

Curr Opin Microbiol 5, 160–165. Bucca, G., Hindle, Z. & Smith, C. P. (1997). Regulation of the dnaK

operon of Streptomyces coelicolor A3(2) is governed by HspR, an autoregulatory repressor protein. J Bacteriol 179, 5999–6004. Bucca, G., Brassington, A. M., Hotchkiss, G., Mersinias, V. & Smith, C. P. (2003). Negative feedback regulation of dnaK, clpB and lon

expression by the DnaK chaperone machine in Streptomyces coelicolor, identified by transcriptome and in vivo DnaK-depletion analysis. Mol Microbiol 50, 153–166.

function, and antigenicity of Campylobacter flagella. In Campylobacter, pp. 405–422. Edited by I. Nachamkin & M. J. Blaser. Washington, DC: American Society for Microbiology. Hendrixson, D. R. & DiRita, V. J. (2003). Transcription of s54dependent but not s28-dependent flagellar genes in Campylobacter

jejuni is associated with formation of the flagellar secretory apparatus. Mol Microbiol 50, 687–702. Hendrixson, D. R., Akerley, B. J. & DiRita, V. J. (2001). Transposon

mutagenesis of Campylobacter jejuni identifies a bipartite energy taxis system required for motility. Mol Microbiol 40, 214–224. Holmes, K., Mulholland, F., Pearson, B. M., Pin, C., McNichollKennedy, J., Ketley, J. & Wells, J. (2005). Campylobacter jejuni gene

expression in response to iron limitation and the role of Fur. Microbiology 151, 243–257. Horton, R. M., Ho, S. N., Pullen, J. K., Hunt, H. D., Cai, Z. & Pease, L. R. (1993). Gene splicing by overlap extension. Methods Enzymol

217, 270–279. Hu, L. & Kopecko, D. J. (1999). Campylobacter jejuni 81-176

associates with microtubules and dynein during invasion of human intestinal cells. Infect Immun 67, 4171–4182. Jagannathan, A., Constantinidou, C. & Penn, C. W. (2001). Roles

of rpoN, fliA, and flgR in expression of flagella in Campylobacter jejuni. J Bacteriol 183, 2937–2942.

Carrillo, C. D., Taboada, E., Nash, J. H. & 15 other authors (2004).

Ketley, J. M. (1997). Pathogenesis of enteric infection by Campylobacter. Microbiology 143, 5–21.

Genome-wide expression analyses of Campylobacter jejuni NCTC 11168 reveals coordinate regulation of motility and virulence by flhA. J Biol Chem 279, 20327–20338.

Kinsella, N., Guerry, P., Cooney, J. & Trust, T. J. (1997). The flgE gene of Campylobacter coli is under the control of the alternative sigma factor s54. J Bacteriol 179, 4647–4653.

Chastanet, A., Fert, J. & Msadek, T. (2003). Comparative genomics

Konkel, M. E., Klena, J. D., Rivera-Amill, V., Monteville, M. R., Biswas, D., Raphael, B. & Mickelson, J. (2004). Secretion of

reveal novel heat shock regulatory mechanisms in Staphylococcus aureus and other Gram-positive bacteria. Mol Microbiol 47, 1061–1073. Cohen, S. N., Chang, A. C. & Hsu, L. (1972). Nonchromosomal

virulence proteins from Campylobacter jejuni is dependent on a functional flagellar export apparatus. J Bacteriol 186, 3296–3303.

antibiotic resistance in bacteria: genetic transformation of Escherichia coli by R-factor DNA. Proc Natl Acad Sci U S A 69, 2110–2114.

Lu¨neberg, E., Glenn-Calvo, E., Hartmann, M., Bar, W. & Frosch, M. (1998). The central, surface-exposed region of the flagellar hook

914

Microbiology 151

HspR in Campylobacter jejuni protein FlgE of Campylobacter jejuni shows hypervariability among strains. J Bacteriol 180, 3711–3714. Misawa, N. & Blaser, M. J. (2000). Detection and characterization of

autoagglutination activity by Campylobacter jejuni. Infect Immun 68, 6168–6175. Mogk, A., Vo¨lker, A., Engelmann, S., Hecker, M., Schumann, W. & Vo¨lker, U. (1998). Nonnative proteins induce expression of the

Stewart, G. R., Snewin, V. A., Walzl, G. & 7 other authors (2001).

Overexpression of heat-shock proteins reduces survival of Mycobacterium tuberculosis in the chronic phase of infection. Nat Med 7, 732–737. Stewart, G. R., Wernisch, L., Stabler, R., Mangan, J. A., Hinds, J., Laing, K. G., Young, D. B. & Butcher, P. D. (2002). Dissection of the

Bacillus subtilis CIRCE regulon. J Bacteriol 180, 2895–2900.

heat-shock response in Mycobacterium tuberculosis using mutants and microarrays. Microbiology 148, 3129–3138.

Monteville, M. R., Yoon, J. E. & Konkel, M. E. (2003). Maximal

Stintzi, A. (2003). Gene expression profile of Campylobacter jejuni

adherence and invasion of INT 407 cells by Campylobacter jejuni requires the CadF outer-membrane protein and microfilament reorganization. Microbiology 149, 153–165.

Szymanski, C. M., King, M., Haardt, M. & Armstrong, G. D. (1995).

Morooka, T., Umeda, A. & Amako, K. (1985). Motility as an

intestinal colonization factor for Campylobacter jejuni. J Gen Microbiol 131, 1973–1980. Nachamkin, I., Yang, X. H. & Stern, N. J. (1993). Role of

Campylobacter jejuni flagella as colonization factors for three-dayold chicks: analysis with flagellar mutants. Appl Environ Microbiol 59, 1269–1273. Nachamkin, I., Allos, B. M. & Ho, T. (1998). Campylobacter species

and Guillain-Barre´ syndrome. Clin Microbiol Rev 11, 555–567. Narberhaus, F. (1999). Negative regulation of bacterial heat shock

genes. Mol Microbiol 31, 1–8. Nuijten, P. J., van Asten, F. J., Gaastra, W. & van der Zeijst, B. A. (1990). Structural and functional analysis of two Campylobacter

jejuni flagellin genes. J Biol Chem 265, 17798–17804. Parkhill, J. B. W., Wren, K., Mungall, J. M. & 18 other authors (2000).

in response to growth temperature variation. J Bacteriol 185, 2009–2016. Campylobacter jejuni motility and invasion of Caco-2 cells. Infect Immun 63, 4295–4300. Tomb, J.-F., White, O., Kerlavage, A. R. & 39 other authors (1997).

The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388, 539–547. Ueguchi, C., Kakeda, M., Yamada, H. & Mizuno, T. (1994). An

analogue of the DnaJ molecular chaperone in Escherichia coli. Proc Natl Acad Sci U S A 91, 1054–1058. VanBogelen, R. A., Kelley, P. M. & Neidhardt, F. C. (1987).

Differential induction of heat shock, SOS, and oxidation stress regulons and accumulation of nucleotides in Escherichia coli. J Bacteriol 169, 26–32. Wang, W. L., Luechtefeld, N. W., Blaser, M. J. & Reller, L. B. (1983).

Effect of incubation atmosphere and temperature on isolation of Campylobacter jejuni from human stools. Can J Microbiol 29, 468–470.

The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403, 665–668.

Wassenaar, T. M. (1997). Toxin production by Campylobacter spp.

Pearson, B. M., Pin, C., Wright, J., I’Anson, K., Humphrey, T. & Wells, J. M. (2003). Comparative genome analysis of Campylobacter jejuni

Wassenaar, T. M., Fry, B. N. & van der Zeijst, B. A. (1993a). Genetic

using whole genome DNA microarrays. FEBS Lett 554, 224–230. Reischl, S., Wiegert, T. & Schumann, W. (2002). Isolation and

analysis of mutant alleles of the Bacillus subtilis HrcA repressor with reduced dependency on GroE function. J Biol Chem 277, 32659–32667. Schulz, A. & Schumann, W. (1996). hrcA, the first gene of the

Bacillus subtilis dnaK operon encodes a negative regulator of class I heat shock genes. J Bacteriol 178, 1088–1093. Servant, P. & Mazodier, P. (2001). Negative regulation of the heat

Clin Microbiol Rev 10, 466–476. manipulation of Campylobacter: evaluation of natural transformation and electro-transformation. Gene 132, 131–135. Wassenaar, T. M., van der Zeijst, B. A., Ayling, R. & Newell, D. G. (1993b). Colonization of chicks by motility mutants of Campylo-

bacter jejuni demonstrates the importance of flagellin A expression. J Gen Microbiol 139, 1171–1175. Wooldridge, K. G. & Ketley, J. M. (1997). Campylobacter-host cell

interactions. Trends Microbiol 5, 96–102. Wo¨sten, M. M., Wagenaar, J. A. & van Putten, J. P. (2004). The

shock response in Streptomyces. Arch Microbiol 176, 237–242.

FlgS/FlgR two-component signal transduction system regulates the fla regulon in Campylobacter jejuni. J Biol Chem 279, 16214–16222.

Song, Y. C., Jin, S., Louie, H. & 8 other authors (2004). FlaC, a

Yao, R., Alm, R. A., Trust, T. J. & Guerry, P. (1993). Construction of

protein of Campylobacter jejuni TGH9011 (ATCC43431) secreted through the flagellar apparatus, binds epithelial cells and influences cell invasion. Mol Microbiol 53, 541–553.

new Campylobacter cloning vectors and a new mutational cat cassette. Gene 130, 127–130.

Spohn, G. & Scarlato, V. (1999). The autoregulatory HspR repressor

protein governs chaperone gene transcription in Helicobacter pylori. Mol Microbiol 34, 663–674. Spohn, G., Danielli, A., Roncarati, D., Delany, I., Rappuoli, R. & Scarlato, V. (2004). Dual control of Helicobacter pylori heat shock

gene transcription by HspR and HrcA. J Bacteriol 186, 2956–2965.

http://mic.sgmjournals.org

Yao, R., Burr, D. H., Doig, P., Trust, T. J., Niu, H. & Guerry, P. (1994).

Isolation of motile and non-motile insertional mutants of Campylobacter jejuni: the role of motility in adherence and invasion of eukaryotic cells. Mol Microbiol 14, 883–893. Zuber, U. & Schumann, W. (1994). CIRCE, a novel heat shock

element involved in regulation of heat shock operon dnaK of Bacillus subtilis. J Bacteriol 176, 1359–1363.

915