Selenium-Dependent Biogenesis of Formate Dehydrogenase in ...

Selenium-Dependent Biogenesis of Formate Dehydrogenase in Campylobacter jejuni Is Controlled by the fdhTU Accessory Genes Frances L. Shaw, Francis Mulholland, Gwénaëlle Le Gall, Ida Porcelli, Dave J. Hart, Bruce M. Pearson, and Arnoud H. M. van Vliet Institute of Food Research, Norwich Research Park, Norwich, United Kingdom

The food-borne bacterial pathogen Campylobacter jejuni efficiently utilizes organic acids such as lactate and formate for energy production. Formate is rapidly metabolized via the activity of the multisubunit formate dehydrogenase (FDH) enzyme, of which the FdhA subunit is predicted to contain a selenocysteine (SeC) amino acid. In this study we investigated the function of the cj1500 and cj1501 genes of C. jejuni, demonstrate that they are involved in selenium-controlled production of FDH, and propose the names fdhT and fdhU, respectively. Insertional inactivation of fdhT or fdhU in C. jejuni resulted in the absence of FdhA and FdhB protein expression, reduced fdhABC RNA levels, the absence of FDH enzyme activity, and the lack of formate utilization, as assessed by 1H nuclear magnetic resonance. The fdhABC genes are transcribed from a single promoter located two genes upstream of fdhA, and the decrease in fdhABC RNA levels in the fdhU mutant is mediated at the posttranscriptional level. FDH activity and the ability to utilize formate were restored by genetic complementation with fdhU and by supplementation of the growth media with selenium dioxide. Disruption of SeC synthesis by inactivation of the selA and selB genes also resulted in the absence of FDH activity, which could not be restored by selenium supplementation. Comparative genomic analysis suggests a link between the presence of selA and fdhTU orthologs and the predicted presence of SeC in FdhA. The fdhTU genes encode accessory proteins required for FDH expression and activity in C. jejuni, possibly by contributing to acquisition or utilization of selenium.

T

he food-borne bacterial pathogen Campylobacter jejuni is the most prevalent bacterial cause of human gastroenteritis in the developed world (7). The organism colonizes the avian cecum to high densities and is highly adapted to life in this environmental niche. C. jejuni is a microaerophilic organism that cannot ferment sugars for energy production but instead relies on organic acids and amino acids (22). One of the preferred substrates for C. jejuni is formate (15), which is produced by anaerobic fermentation by the local microflora in the avian and mammalian intestine (22). Formate is metabolized by the formate dehydrogenase (FDH) enzyme via oxidation (21), with electrons being donated to the menaquinone pool (22, 24). The C. jejuni NCTC 11168 genome sequence (10, 25) contains four genes annotated as FDH subunits (fdhABCD; cj1511c to cj1508c). FdhA (Cj1511c) is a large 110-kDa protein, which is predicted to contain a selenocysteine amino acid (SeC), whereas FdhB (24 kDa) and FdhC (35 kDa) are an iron-sulfur protein and a cytochrome b-containing protein, respectively (22). FdhD acts as sulfurtransferase in Escherichia coli, where it is required for FDH activity (37), but its function in C. jejuni is unknown. Upstream of fdhABCD, there are two more genes involved in FDH biosynthesis and activity: cj1514c was shown to be specifically required for the activity of FDH and was designated fdhM (14), whereas cj1513c encodes a TAT-exported protein of unknown function that is translationally coupled to the downstream fdhA gene and is specific to the epsilonproteobacteria (14). The C. jejuni FDH has been reported to be a tungstoenzyme (31, 35), and the Cj1513c protein has been suggested to be required for formation of a tungsten-pterin cofactor (14). FDH plays an important role in C. jejuni, since the combined absence of FDH and hydrogenase activity significantly reduced the ability of C. jejuni to colonize the chicken cecum (42), whereas the inactivation of either fdhD or the formic acid chemoreceptor cj0952c gene resulted in reduced immunopathology in gnotobiotic mice containing a humanized in-

3814

jb.asm.org

Journal of Bacteriology

testinal microflora (2). The natural flora of the chicken cecum is dominated by anaerobic fermentative organisms that produce formate and hydrogen, and a mutant unable to oxidize either of these substrates in this ecological niche is most likely disadvantaged. The SeC amino acid, as predicted to be present in C. jejuni FdhA, is the 21st amino acid (5). It has the same structure as cysteine, except that the sulfur atom is replaced by selenium and has both a lower pKa and a higher redox potential than cysteine. In selenoproteins, the SeC amino acid is coded for by an in-frame UGA codon (which is normally a stop codon) (4, 5, 13). This process is controlled via a specialized element present in the mRNA, called SECIS (SElenoCysteine Insertion Sequence), which is located immediately downstream of the UGA codon and, together with SelB, ensures the incorporation of the SeC into the polypeptide chain (11). Synthesis of tRNA-SeC (selC) is mediated by SelA, and selAB orthologs are present in the C. jejuni genome (cj1378-cj1379) (6, 13, 41). In general, when cells are grown in the absence of selenium, translation of selenoproteins terminates at the SeC UGA codon, resulting in a truncated, nonfunctional enzyme (1). Although the biosynthesis of selenoproteins has been extensively studied, there are still questions relating to which transport systems, in what form selenium is taken up into bacterial cells, and what factors other than the currently known sel genes are involved in the biogenesis of selenoproteins. One class of proteins that has

Received 6 December 2011 Accepted 8 May 2012 Published ahead of print 18 May 2012 Address correspondence to Arnoud H. M. van Vliet, [email protected]. Copyright © 2012, American Society for Microbiology. All Rights Reserved. doi:10.1128/JB.06586-11

p. 3814 –3823

August 2012 Volume 194 Number 15

Selenium and C. jejuni Formate Dehydrogenase

TABLE 1 C. jejuni strains used in this investigation

TABLE 2 Oligonucleotide primers used for 5= RACE

C. jejuni straina

Details and/or referencea,b

Primer (gene)

Sequence (5=–3=)

NCTC 11168 81116 11168 ⌬fdhT 11168 ⌬fdhU 11168 ⌬fdhTU 11168 ⌬fdhT::fdhUC

Wild-type strain (25) Wild-type strain (NCTC 11828) (26) cj1500::Kanr cj1501::Kanr cj1500-cj1501::Kanr cj1500::Kanr complemented with Cj0046:: (PfdxA-Cj1501 Catr) cj1501::Kanr complemented with Cj0046:: (PfdxA-Cj1501 Catr) cj1500-cj1501::Kanr complemented with Cj0046::(PfdxA-Cj1501 Catr) NCTC 11168 with Cj0046::(PfdxA-Cj1501 Catr), overexpressing fdhU cj1378::Kanr cj1379::Kanr cj1378-cj1379::Kanr cj1511c::Kanr C8J_1404::Kanr C8J_1404::Kanr complemented with cj0046:: (PfdxA-cj1501 Catr)

RNA adaptor

AUAUGCGCGAAUUCCUGUAGAACG AACACUAGAAGAAA GCGCGAATTCCTGTAGA AACCTATAAACATACCTATG TTGGCTTCTATTTGTCCTTG ATTCACCCTCACCCACTTTG GGTTGATCTAAACATTTCTG ATCGATATTGGCAGTATGTG TTTGAAGTTCCTACAGTATG CCACCTGAACTCACAGGATG TCCCAATTAGCACTTCTTTG AAGAAAAAGTCGTTGGAATG

11168 ⌬fdhU::fdhUC 11168 ⌬fdhTU::fdhUC 11168 wild type::fdhUC 11168 ⌬selA 11168 ⌬selB 11168 ⌬selAB 11168 ⌬fdhA 81116 ⌬fdhTU* 81116 ⌬fdhTU*::fdhUC

a Complementation and overexpression of fdhU was performed using the cj0046 pseudogene (36), with the gene expressed from the fdxA promoter (PfdxA) (39). b Kanr is the kanamycin resistance cassette (40); Catr is the chloramphenicol resistance cassette (40).

recently been implicated in microbial selenium metabolism are those containing a SirA-like domain (23, 47), which are present in many microbial and archaeal genomes. This family shares an Nterminal domain, modeled by pfam01206, with the sulfurtransferase TusA (also called SirA-like) (19), but experimental validation of a role of SirA-like domain containing proteins in selenium metabolism is lacking to date. In the present study we have investigated the function of the C. jejuni cj1501 gene, which encodes a SirA-like protein that we have named FdhU, and report on the role of the cj1500-cj1501 (fdhTU) and selAB genes in selenium-dependent FDH biogenesis in C. jejuni. MATERIALS AND METHODS Bacterial strains, media, and growth conditions. The C. jejuni strains NCTC 11168 and 81116 (NCTC 11828) and their isogenic mutants (Table 1) were routinely grown in brucella medium at 37°C under microaerobic conditions (85% N2, 5% O2, 10% CO2). Escherichia coli TOP10 (Novagen) was grown aerobically in Luria-Bertani medium (30) at 37°C. Where appropriate, media were supplemented with ampicillin (final concentration, 100 ␮g ml⫺1), kanamycin (final concentration, 20 ␮g ml⫺1), or chloramphenicol (final concentration, 20 ␮g ml⫺1). Filter-sterilized selenium dioxide solution in water was used to supplement brucella medium to a final concentration of 5 ␮M. Filter-sterilized formate solution in water was used to supplement the medium to a final concentration of 1 mM. Genetic manipulation of C. jejuni. Inactivation and complementation of C. jejuni genes was performed essentially as described previously (36). Specific C. jejuni genes were inactivated by the replacement and insertion of antibiotic resistance cassettes (kanamycin or chloramphenicol). The mutants were confirmed by PCR and sequencing. To complement or supplement the cj1501 gene in C. jejuni strains, the gene was inserted together with a chloramphenicol resistance gene into the pseudogene cj0046 under the control of a native C. jejuni promoter, fdxA (36, 39). This construct was then introduced into the C. jejuni wild-type strain and its isogenic cj1501 mutant by electroporation, followed by selection for chloramphenicol resistance. Correct integration of the constructs into


Adaptor-specific primer Cj1500R (fdhT) Cj1503cR (putA) Cj1505cR (cj1505c) Cj1506cR (ccaA) Cj1508cR (fdhD) Cj1511cR-SeC (fdhA)a Cj1511cR (fdhA)b Cj1513cR (cj1513c) Cj1514cR (fdhM) a b

Primer located downstream (3=) of the SeC UGA codon in the fdhA gene. Primer located upstream (5=) of the SeC UGA codon in the fdhA gene.

the genomic copy of the cj0046 pseudogene of C. jejuni NCTC 11168 was validated by PCR (36). Preparation of C. jejuni genomic DNA and RNA. Genomic DNA was isolated from C. jejuni cells grown microaerobically for 24 h on brucella agar by using a DNeasy kit (Qiagen) according to the manufacturer’s instructions. RNA was extracted from C. jejuni cultures grown to an optical density at 600 nm (OD600) of ⬃0.4 according to previously described protocols (18). Briefly, 0.1 volume of 5% phenol in ethanol was mixed with the broth culture, and the RNA was isolated from pelleted cells with Tri-Reagent (Sigma) and chloroform. RNA was further purified using an RNeasy kit (Qiagen) and Turbo DNA-free treatment (Ambion) according to the manufacturer’s instructions. The purity of the RNA was determined using an RNA 6000 Nano kit (Agilent) according to the manufacturer’s instructions. The concentration of the RNA was determined by using a NanoDrop NS-1000 spectrophotometer (Thermo Scientific). Transcription start site determination by 5= RACE. Transcription start sites in the cj1500-cj1514c region of C. jejuni NCTC 11168 were determined using 5= RACE (rapid amplification of cDNA ends), essentially as described previously (9). Briefly, 12 ␮g of RNA, isolated from a mid-log-phase culture of C. jejuni NCTC 11168 using the RNeasy kit (Qiagen, United Kingdom), was treated with tobacco acid pyrophosphatase (TAP), and RNA oligonucleotide adaptor (Table 2) was ligated to the 5= ends of the treated RNA (9). TAP cleaves the 5=-triphosphate of primary transcripts to a monophosphate, thus making them available for ligation of the RNA adaptor. This results in an enrichment of 5=-RACE products for primary transcripts in TAP-treated RNA compared to an untreated control. First-strand cDNA synthesis was performed using random hexamers, followed by PCR amplification with gene-specific primers and a 5=-adaptor-specific DNA primer (Table 2). The resulting PCR products were cloned into the pGEM-T Easy cloning vector (Promega, United Kingdom), and the nucleotide sequence of the inserts was determined according to standard protocols. Transcriptomic analysis by whole-genome microarray. An 8⫻15K microarray slide (Agilent) specific for the genome of C. jejuni NCTC 11168 was used to visualize relative mRNA levels in the wild-type strain compared to the fdhU mutant, the complemented fdhU (fdhUC) mutant, and the wild-type strain overexpressing fdhU (Table 1). An Affinity Script kit (GE Healthcare) was used to produce Cy3- or Cy5-labeled cDNA from the aforementioned RNA samples. These were then hybridized to the microarray slide according to the manufacturer’s instructions (Agilent Hi-RPM gene expression hybridization kit). DNA microarrays were scanned using an Axon GenePix 4000A microarray laser scanner (Axon Instruments), and the data from detected features were initially processed using GenePix 3.0 software. Statistical analysis of the data was performed with Marray as described previously (17). The microarray design and data have been deposited in the GEO database under accession numbers

jb.asm.org 3815

Shaw et al.

GPL13841 (microarray design) and GSE35016 (data from the present study). Proteomics. C. jejuni cells were harvested from 50 ml of brucella broth by centrifugation (4,000 ⫻ g, 10 min, room temperature). The cell pellets were resuspended in 500 ␮l of lysis buffer (50 mM Tris (pH 7.5), 0.3% sodium dodecyl sulfate (SDS), 0.2 M dithiothreitol, 3.3 mM MgCl2, 16.7 ␮g of RNase ml⫺1, and 1.67 U of DNase ml⫺1) and then sonicated (Soniprep 150 MSE; Sanyo) for 10 s six times with incubation on ice between sonications and 20 min of incubation on ice following sonication. The samples were centrifuged (20,000 ⫻ g, 20 min, 4°C). The total cell protein was quantified by using a 2D Quant kit (GE Healthcare) according to the manufacturer’s instructions. A total of 100 ␮g of protein was separated by isoelectric focusing on a pH 3-11 NL 24-cm IPG strip (GE Healthcare) for 44.7 kV·h at 20°C over 8.75 h using the IPGphor (GE Healthcare) (17). The focused IPG strips were then conditioned, and the size separation on the second dimension was performed using SDS-PAGE as described previously (17). The proteins were fixed, stained by Sypro-Ruby (Invitrogen), and then destained according to the manufacturer’s instructions. The gel images were captured with a Pharos FX⫹ molecular imager with Quantity One imaging software (v4.6.1; Bio-Rad). A 532-nm excitation laser was used with a 605-nm band-pass emission filter, and the gels were scanned at a 100-␮m resolution to produce a 16 bit image. The laser strength was adjusted for each image to give the maximum signal without saturation on the strongest spot. Gel images were compared using ProteomWeaver analysis software (v3.0.1; Definiens). Protein identification using LC-MS/MS. Proteins of interest were removed from the gel by a using ProPick excision robot (Genomic Solutions) and in-gel trypsin digested using a ProGest protein digester (Genomic Solutions) essentially as described previously (17). Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis was performed using a LTQ-Orbitrap mass spectrometer (Thermo Electron) and a Nanoflow-HPLC system (nanoACQUITY; Waters). Peptides were trapped on line to a Symmetry C18 trap (5 ␮m, 180 ␮m by 20 mm) which was then switched in-line to a UPLC BEH C18 column (1.7 ␮m, 75 ␮m by 250 mm) held at 45°C. Peptides were eluted by a gradient of 0 to 80% acetonitrile in 0.1% formic acid over 50 min at a flow rate of 250 nl min⫺1. The mass spectrometer was operated in positive-ion mode with a nanospray source at a capillary temperature of 200°C. The Orbitrap was run with a resolution of 60,000 over the mass range m/z 300 to 2,000, with an MS target of 106 and a 1-s maximum scan time. The MS/MS was triggered by a minimal signal of 2,000 with an automatic gain control target of 30,000 ions and maximum scan time of 150 ms. For MS/MS events, the selection of 2⫹ and 3⫹ charge states was used. Dynamic exclusion was set to 1 count and a 30-s exclusion time with an exclusion mass window of ⫾20 ppm. Proteins were identified by searching the Thermo RAW files converted to Mascot generic format by DTA supercharger (http://www .msquant.sourceforge.net) against C. jejuni protein sequences in a monthly updated copy of the SPtrEMBL database, using an in-house version (v2.2) of the Mascot search tool (Matrix Science, Inc.). FDH activity assay. FDH activity was measured using a benzylviologen-coupled colorimetric assay (14). Bacterial cells were grown overnight in brucella broth, and the OD600 was measured. Portions (5 ml) of culture were spun down (10 min, 10,000 ⫻ g), washed twice with 1 ml of 25 mM phosphate buffer (pH 7.4), and centrifuged for 1 min at 10,000 ⫻ g. The cells remained on ice until the start of the enzyme assay. The cells were resuspended in 1 ml of 25 mM phosphate buffer (pH 7.4), and 200 ␮l was aliquoted into anaerobic cuvettes, followed by the addition of 200 ␮l of 10 mM benzylviologen and 1,600 ␮l of 25 mM phosphate buffer (pH 7.4). The cuvettes were individually sparged with nitrogen for 8 min. The cuvette was then heated to 37°C, and a reading was taken at 578 nm. Subsequently, substrate, which consisted of 20 ␮l of 1 M sodium formate sparged with nitrogen for ⬃20 min prior to use, was added. The reaction was monitored at 578 nm at one read per s for ⬃200 s or until the reading was ⬎2.0. The FDH activity was calculated by determination of the nmol

3816

jb.asm.org

of viologen reduced per min per mg of protein. The protein concentration was determined using a Bradford assay (Bio-Rad). 1 H NMR. 1H nuclear magnetic resonance (1H NMR) was used to identify the presence, absence, and concentration of several metabolites in C. jejuni growth medium. The spent growth medium was filter sterilized (0.2-␮m pore size) to remove the cells. Supernatant samples were thawed at room temperature and prepared for 1H NMR spectroscopy by mixing 400 ␮l of spent medium with 200 ␮l of phosphate buffer (0.2 M Na2HPO4, 0.038 M NaH2PO4 [pH 7.4]) made up in 100% D2O and containing 0.06% sodium azide, 6 mM DFTMP [difluoro (trimethylsilyl) methylphosphonic acid], and 1.5 mM DSS (sodium 2,2-dimethyl-2-silapentane5-sulfonate) as a chemical shift reference. The sample was mixed, and 500 ␮l was transferred into a 5-mm NMR tube for spectral acquisition. The 1H NMR spectra were recorded at 600 MHz on a Bruker Avance spectrometer (Bruker BioSpin GmbH, Rheinstetten, Germany) running Topspin 2.0 software and fitted with a cryoprobe and a 60-slot autosampler. Each 1H NMR spectrum was acquired with 128 scans, a spectral width of 8,012.8 Hz, an acquisition time of 2.04 s, and a relaxation delay of 2.0 s. The “noesypr1d” presaturation sequence was used to suppress the residual water signal with a low-power selective irradiation at the water frequency during the recycle delay and a mixing time of 150 ms. Spectra were transformed with a 0.3-Hz line broadening, manually phased, baseline corrected, and referenced by setting the DSS methyl signal to 0 ppm. Absolute concentrations were obtained by using CHENOMX software (version 5.1) (43). Several metabolites were quantified from a possible library of 290, with quantification calculated relative to DSS.

RESULTS

The cj1500-cj1501 gene cluster is located in a conserved region of the C. jejuni genome. In the C. jejuni NCTC 11168 genome, the cj1501 gene is preceded by the cj1500 gene, which is annotated to encode an inner membrane protein orthologous to the E. coli YedE protein, and contains 10 transmembrane domains, suggesting a transporter function. Both genes are located divergently from the SeC-tRNA (selC, Fig. 1A) in an area of the genome (cj1429c-cj1519) that is highly conserved between the C. jejuni reference strains 81-176, 81116, and RM1221 (8, 16, 26). The cj1500 and cj1501 genes are present as two separate but adjacent genes in all sequenced C. jejuni genomes, with the exception of the C. jejuni 81116 genome (26) where these two genes are fused into a single open reading frame (C8J_1404), although the translational signals (ribosome binding site, GTG start codon) for translation of the Cj1501 ortholog are still present (Fig. 1A). The transcription start sites of the predicted cistrons in the cj1500-cj1514c region were determined using 5= RACE. The cj1500 and cj1501 genes are cotranscribed from a ␴70 promoter (gnTAnAAT [27]), with the RNA starting at the C residue 46 nucleotides upstream of the TTG start codon of cj1500 (Fig. 1B). Other transcription start sites were detected upstream of the cj1503c (putP), cj1505c, cj1506c (ccaA) (12), cj1508c (fdhD) (34), and cj1514c (fdhM) (14) genes, all with recognizable ␴70 ⫺10 promoter sequences (Fig. 1B). No transcription start site could be detected directly upstream of the fdhA gene, suggesting that transcription of the fdhABC genes originates from the promoter upstream of the fdhM gene (Fig. 1A and B). Inactivation of the cj1500 (fdhT) and cj1501 (fdhU) genes leads to the loss of FDH activity and expression. Insertional mutagenesis was used to investigate the function of the cj1500 and cj1501 genes in C. jejuni strain NCTC 11168, and the cj1500-cj1501 fusion gene C8J_1404 in strain 81116. The majority of the coding sequences of the cj1500, cj1501, and C8J_1404 genes were replaced by a kanamycin resistance gene, resulting in cj1500 and cj1501



FIG 1 Schematic representation of the region of the C. jejuni genome containing the fdhTU accessory genes and fdhABC structural genes of C. jejuni strain NCTC 11168 and 81116, as well as the promoters and transcription start sites driving transcription of the genes in this genomic region. (A) The region is shown from the selC gene (cjp26, SeC-tRNA) to the cj1514c gene (fdhM). In the strain 81116 genome this is the region from C8J_t0031 to C8J_1417 (26), and in strain 81-176 it is the region from CJJ81176_1753 (upstream of CJJ81176_1490) to CJJ81176_1506 (16). Gray arrows represent FDH subunit or accessory proteins, white arrows represent genes predicted to be involved in selenium metabolism, and black arrows represent genes not known to be involved in either FDH expression or selenium metabolism. Arrows indicate the presence of promoters in C. jejuni NCTC 11168, and the box shows the point mutation in strain 81116 leading to the fdhTU* fusion gene. (B) Transcription start sites and annotated promoters and 5= untranslated regions in the C. jejuni NCTC 11168 cj1500-cj1514c genomic region. The nucleotide position of the transcription start site on the NCTC 11168 genome is given on the left, whereas the sequence of the promoter and 5= UTR is shown with specific sections shown in capital letters, underlined from left to right: the ⫺10 promoter sequence, the transcription start site (indicated in boldface), the ribosome binding site, and the translational start codon. The use of “⬍Nx⬎” represents a stretch of x number nucleotides not shown.

single mutants and a cj1500-cj1501 double mutant in strain NCTC 11168, and a C8J_1404 mutant in strain 81116 (Table 1). We also complemented the different cj1500, cj1501, and C8J_1404 mutations by introducing the cj1501 gene in the cj0046 pseudogene, under the control of the C. jejuni fdxA promoter (36, 39) (Table 1). Initial characterization of the cj1500, cj1501, and cj1500-cj1501 mutants of C. jejuni strain NCTC 11168, and the C8J_1404 mutant of C. jejuni strain 81116 showed no difference between the wildtype strains and their isogenic mutants in terms of growth in Brucella media, auto-agglutination, motility, or resistance to acid shock (data not shown). Two-dimensional protein gel electrophoresis was used to identify proteins differentially expressed in the cj1501 mutant compared to the NCTC 11168 wild-type strain, using cells grown to mid-log-growth phase in standard brucella medium. Two proteins of approximately 100 and 26 kDa were consistently present in the wild-type strain (Fig. 2A) but absent in the cj1501 mutant. Subsequent identification of the 100- and 26-kDa proteins by MS demonstrated these proteins to be the Cj1511c (FdhA) FDH large subunit and the Cj1510c (FdhB) putative FDH iron-sulfur subunit, respectively. The absence of the FdhA and FdhB proteins resulted in the loss of FDH activity in the cj1500, cj1501, and C8J_1404 mutants, as measured by a benzylviologen-linked colorimetric assay (Fig. 3A), confirming the coupled function of the cj1500 and cj1501 genes and that the fused C8J_1404 gene in strain 81116 is functional. Genetic complementation with the cj1501 gene restored FDH activity in a cj1501 mutant but not in the cj1500 or C8J_1404 mutants (Fig. 3A), suggesting that both Cj1500 and


Cj1501 are required for FDH biosynthesis and enzyme activity. Hence, we have renamed the cj1500 and cj1501 genes as fdhT and fdhU, respectively, and the fusion gene C8J_1404 in strain 81116 as fdhTU* and will refer to these genes as such henceforth. The effect of the fdhU mutation on FDH protein expression was reflected at the transcriptional level, as measured by wholegenome microarray analysis. Transcriptomic comparison of the NCTC 11168 wild-type strain, fdhU mutant, complemented fdhU mutant, and wild-type strain overexpressing fdhU showed the successful complementation and overexpression of fdhU in the respective mutants (Fig. 2B). Subsequent analysis showed that transcription of the fdhA, fdhB, and fdhC genes was significantly reduced in the fdhU mutant but that the transcript levels were restored by genetic complementation (Fig. 2B) and that the overexpression of fdhU did not result in further increases in fdhABC transcription. Interestingly, the transcript levels of the upstream, cotranscribed fdhM (cj1514c) and cj1513c genes were not significantly altered in any of the mutants (Fig. 2B), suggesting that the effect on fdhABC transcription is likely to be posttranscriptional, possibly via RNA instability or degradation. Exogenous selenium complements the fdhT, fdhU, and fdhTU mutations. We hypothesized that the absence of FDH expression and activity in the fdhU mutant could be due to interference with selenium metabolism and investigated the effect of supplementation with exogenous selenium on FDH activity in the C. jejuni wild-type and the set of fdhT, fdhU, and fdhTU* mutants. We also generated strains with the fdhA (cj1511c), selA (cj1378), and selB (cj1379) genes inactivated and have tested the effect of

jb.asm.org 3817

Shaw et al.

FIG 2 Inactivation of the fdhU gene results in reduced expression of the FdhA and FdhB formate dehydrogenase subunit proteins in C. jejuni NCTC 11168 and is accompanied by reduced transcription of the fdhABC genes. (A) The wild-type strain and its isogenic fdhU mutant were grown to mid-log phase in brucella medium, and total protein was separated by two-dimensional (2D) protein gel electrophoresis. Two proteins were detected which showed differential expression between the wild-type strain and the fdhU mutant and were identified as the FdhA (⬃100-kDa) and FdhB (⬃26-kDa) proteins. A 2D gel is shown, with the regions containing the FdhA and FdhB proteins magnified, allowing comparison of protein expression. A representative example is shown of the three biological repeats examined. (B) Effect of inactivation, complementation, and overexpression of fdhU on RNA levels in C. jejuni NCTC 11168, as determined by microarray analysis. Black bars represent the fdhU mutant (11168 ⌬fdhU), white bars represent the complemented strain (11168 ⌬fdhU::fdhUC), gray bars represent the strain overexpressing fdhU (11168 wild type::fdhUC) (Table 1). RNA levels are expressed as the fold change of RNA levels in the mutant strain compared to the wild type (ratio of mutant to wild type) ⫾ the standard deviation. All values are averages from RNA samples isolated from three biological replicates, except for the fdhU overexpression strain, which is based on analysis two independent RNA samples. An asterisk indicates RNA levels significantly altered in either the fdhU mutant, complemented fdhU mutant, or the fdhU overexpression strain compared to the wildtype strain (P ⬍ 0.01).

3818

jb.asm.org

selenium supplementation on FDH activity. Inactivation of the selA and selB genes resulted in the absence of detectable FDH activity in the benzylviologen-linked assay, confirming the role of SeC in FDH expression and activity (Fig. 3B). As previously described (42), the fdhA mutant lacked any detectable FDH activity, thus confirming the absence of other formate-metabolizing enzymes in C. jejuni NCTC 11168 (Fig. 3B). Interestingly, supplementation of growth media with selenium dioxide resulted in partial restoration of FDH activity in the fdhT, fdhU, and fdhTU mutants to approximately one-third of the activity of the wildtype strain in C. jejuni NCTC 11168 and fully restored the FDH activity in the fdhTU* mutant of strain 81116 (Fig. 3B). Selenium supplementation failed to restore FDH activity in the selA, selB, and selAB mutants (Fig. 3B), suggesting that the absence of FDH expression and activity in the fdhT and fdhU mutants is due to a defect in selenium metabolism in these strains, which can be complemented by external supplementation of selenium. We hypothesize that a high external concentration of selenium allows the cell to bypass the FdhU function, allowing SeC synthesis and incorporation into the FdhA protein. To investigate whether the inactivation of fdhU and/or selenium supplementation of growth media affected transcriptional patterns in the cj1500-cj1514c region, 5= RACE was performed with primers located in fdhM, cj1513c, and both upstream and downstream of the SeC UGA codon of fdhA (Fig. 3C), using RNA isolated from the NCTC 11168 wild-type and fdhU mutant strains, grown with or without selenium dioxide supplementation. Only a single transcription start site was found with primers located in fdhM, cj1513c, and upstream of the fdhA SeC UGA codon, whereas no specific amplification product was found with the primer downstream of the fdhA SeC UGA codon (Fig. 3C). This was independent of selenium supplementation or inactivation of the fdhU gene, suggesting that any effect on the fdh transcripts is mediated at the posttranscriptional level. The other transcription start sites in the cj1503c-cj1514c region were not affected by fdhU inactivation or selenium supplementation (only shown for fdhD in Fig. 3C), whereas the fdhT amplification product was more present in the fdhU mutant samples (Fig. 3C). Inactivation of fdhU still allows very low levels of formate utilization. To further investigate the effect of the fdhU mutation on metabolite usage of C. jejuni, we performed 1H NMR-based metabolite analysis of spent growth media from cultures grown for 0, 2, 4, 8, and 24 h at 37°C. Figure 4A shows a representative example of such an analysis for the wild-type strain. The formate concentration in unsupplemented brucella broth was ⬃0.2 mM, and since an initial experiment demonstrated very rapid removal of formate by the wild-type strain (data not shown), brucella media were supplemented with formate to 1 mM to avoid the formate levels falling below the lower detection level too rapidly. As expected, the FDH-negative fdhA mutant was unable to utilize formate, and formate levels remained constant over an 8-h growth period (Fig. 4B). In contrast, the wild-type strain metabolized formate very rapidly, with formate levels dropping below detection levels within 2 h (Fig. 4B). Formate levels in media obtained from the fdhU mutant decreased between 8 h and 24 h, suggesting that there may be a very low level of FDH activity in the fdhU mutant, below detection levels of the benzylviologen-linked FDH assay (Fig. 3, 4B). In contrast, genetic complementation with the fdhU gene in the cj0046 pseudogene resulted in a formate utilization pattern identical to the wild-type strain. The positive effect of



exogenous selenium supplementation of growth media on FDH activity in the fdhU mutant was confirmed using 1H NMR-based metabolite analysis since the formate levels in spent medium were reduced formate levels to below detection level in 24 h, whereas selenium supplementation did not result in formate utilization in the fdhA mutant (Fig. 4B). Finally, the 1H NMR spectra for lactate, acetate, pyruvate, and succinate and other metabolites were analyzed to ascertain that there was no significant difference between the wild-type strain and the mutants tested for other metabolites (see Fig. 4B for lactate), thus confirming that the metabolite utilization phenotype of the fdhU mutant is limited to formate (data not shown). Presence of FdhT and FdhU orthologs in bacterial genomes coincides with the presence of SelA and SeC-containing FDH enzymes. To investigate the potential link between FdhTU orthologs and a SeC-containing FDH, we combined occurrence analysis using the EMBL String database (http://string.embl.de) (33) containing 943 bacterial taxa, 121 eukaryotic taxa and 69 archaeal taxa, with the information on selenium in the trace element database dbTEU (http://gladyshevlab.bwh.harvard.edu /trace_element) (46). FdhTU orthologs were detected in many genera throughout the bacterial kingdom, including the beta, gamma, and epsilon subdivisions of the Proteobacteria, in some members of the Firmicutes and Actinobacteridae, suggesting a joint function of the adjacent genes. When combined with searching for the presence of a SelA and FdhA orthologs, we did find a strong correlation between the predicted presence of adjacent genes encoding FdhTU orthologs and an SeC-containing FdhA in the gamma and epsilon subdivisions of the Proteobacteria (Table 3), whereas the reverse was not found, since SeC-containing FdhA were predicted in genomes lacking FdhTU orthologs (for example, in Wolinella succinogenes). It should be noted that the analysis only included SeC-containing FdhA, and not other possible selenoproteins. Despite this limitation, the bioinformatic analysis still suggests a link between FdhTU proteins and selenium metabolism in bacteria. DISCUSSION

Although C. jejuni is best known as a gastrointestinal pathogen of human and animals (20, 45), its natural niche is in the intestinal tracts of birds, where it can colonize to high densities in the cecum. Research on the metabolic features of C. jejuni has highlighted many adaptations of this bacterium to that cecal niche (22), such as the periplasmic fumarate reductase Mfr, which can also use

FIG 3 Selenium and genetic complementation restore formate dehydrogenase enzyme activity in C. jejuni but do not affect the transcription initiation of fdhABC and other genes in the cj1500-cj1514c genomic region. (A) Effect of inactivation of the fdhT and fdhU genes in C. jejuni NCTC 11168, and the fusion gene fdhTU* in C. jejuni strain 81116, and the effect of complementation with the NCTC 11168 fdhU gene under the control of the fdxA promoter,


on FDH activity. (B) Inactivation of the SeC biosynthesis pathway genes selA and selB results in the absence of FDH activity. Medium supplementation with selenium dioxide to a final concentration of 5 ␮M partially (NCTC 11168) or fully (strain 81116) restores FDH enzyme activity in fdhT, fdhU, and fdhTU* mutants but cannot restore FDH activity in selA and selB mutants. An fdhA mutant was included as a control. Black bars show the FDH activity in unsupplemented brucella broth, and white bars show the effect of medium supplementation with 5 ␮M selenium dioxide. Error bars represent the standard deviation from two (for FDH-negative samples) to five biological replicates. Asterisks indicate samples where activity was below detection level. ND, not determined; WT, wild-type strain. (C) Comparative 5= RACE determination of fdhA, cj1513c, fdhM, fdhT, and fdhD transcription start sites, using the NCTC 11168 wild-type strain and the fdhU mutant, grown in brucella broth with or without 5 ␮M selenium dioxide supplementation. Asterisks highlight TAP-specific amplification products representing transcription start sites shown in Fig. 1B, other bands are nonspecific amplification products.

jb.asm.org 3819

Shaw et al.

FIG 4 Inactivation of fdhU disrupts formate metabolism in C. jejuni, as determined by 1H NMR analysis of spent growth medium. (A) Representative example of the 1H NMR traces of spent medium, obtained from the wild-type strain, at time points of 0, 2, 4, 8, and 24 h. Peaks for which the metabolite has been identified are indicated as follows: asn, asparagine; asp, aspartate; glu, glutamate; lac, lactate; suc, succinate; pyr, pyruvate; and pyglu, pyroglutamate. (B) Comparison of formate concentrations in spent medium (supplemented with formate to 1 mM prior to incubation with cells) for wild-type C. jejuni NCTC 11168, the fdhU mutant, and the genetically complemented strain, and the fdhU mutant in selenium-supplemented medium. Formate levels go down very rapidly in the wild-type strain but go down partially in the mutant only after 24 h, showing the presence of very low levels of FDH activity in the fdhU mutant. The (partial) restoration of FDH activity in the fdhU mutant by genetic complementation results in very similar formate disappearance compared to the wild-type strain, whereas selenium supplementation results in reduction of formate levels, but more slowly than in the wild-type strain. A medium control is included, and an fdhA mutant with or without selenium supplementation, and none of these show removal of formate. The levels of lactate are shown for comparison and are independent of selenium supplementation or genetic inactivation and complementation. Bars show the average of data obtained from three biological replicates, and error bars denote standard deviation. Asterisks indicate samples where formate levels were below the detection level. ND, not determined; WT, wild-type strain.

3820

jb.asm.org



TABLE 3 Distribution of FdhTU orthologs, SelA, and selenoproteins in completed proteobacterial genomesa Genus (no. of species)

Presence of fdhTU operonb

SelA statusc

Selenoproteind

Epsilonproteobacteria Campylobacter (6) Arcobacter (1) Helicobacter (3) Sulfurimonas (1)

Yes (4/6 species) Yes Yes (1/3 species) Yes

All SelA⫹ SelA⫹ All SelA⫹ SelA⫹

FDH FDH FDH in one species FDH

No

SelA⫹

FDH

Gammaproteobacteria Escherichia (2) Salmonella (1) Yersinia (3) Serratia (1) Other Enterobacteriaceae (10)f Other Enterobacteriaceae (15)g Pseudomonas (9) Shewanella (14) Actinobacillus (2)

Yes (all) Yes Yes (all) Yes Yes (all) No Yes (2/9 species) Yes (5/14 species) Yes (1/2 species)

All SelA⫹ SelA⫹ All SelA⫹ SelA⫹ All SelA⫹ 4/15 SelA⫹* 4/9 SelA⫹* 5/14 SelA⫹* 2/2 SelA⫹*

FDH FDH FDH FDH FDH FDH in four species FDH in four species FDH in five species FDH in two species

Betaproteobacteria Nitrosomonas (2)

Yes (all)

Negative

None

Other Epsilonproteobacteria (2)e

a

Analysis based on version 9.0 of the String Database (http://string.embl.de, last accessed 27 April 2012) (33).The data set consists of 943 bacterial taxa, 121 eukaryotic taxa, and 69 archaeal taxa and only includes completed genome sequences. The modules used were neighborhood (for analysis of the presence of a fdhTU operon) and occurrence (with Cj1500 [FdhT], Cj1501 [FdhU], Cj1378 [SelA], and Cj1511c [FdhA]). b Since FdhU is homologous to TusA (19), only species with adjacent fdhTand fdhU orthologs were included in the analysis. c Determined based on the presence of an orthologous open reading frame and annotation, using String (33). *, For all categories, the SelA⫹ species are the same ones as the species with SeC-containing FDH enzyme(s). d Determined based on the trace element utilization database (dbTEU) (46) and the annotation of gene sequences. Some species have more than a single FDH enzyme. e “Other” here includes the Epsilonproteobacteria Wolinella (n ⫽ 1) and Nautilia (n ⫽ 1). f “Other” here includes the FdhTU-positive Enterobacteriaceae Shigella (n ⫽ 4), Klebsiella (n ⫽ 1), Cronobacter (n ⫽ 2), Citrobacter (n ⫽ 1), Proteus (n ⫽ 1), and Enterobacter (n ⫽ 1). g “Other” here includes FdhTU-negative Enterobacteriaceae Buchnera (n ⫽ 1), Dickeya (n ⫽ 2), Blochmania (n ⫽ 2), Hamiltonella (n ⫽ 1), Pectobacterium (n ⫽ 3), Edwardsiella (n ⫽ 2), Photorabdus (n ⫽ 2), Erwinia (n ⫽ 2), Sodalis (n ⫽ 1), and Wigglesworthia (n ⫽ 1).

fumarate analogs produced via fermentation by anaerobic bacteria in the cecal microbiota (9), and its versatility in using lactate as a carbon source (36). Another favored substrate for C. jejuni is formate, which is very rapidly used by C. jejuni when present in the growth medium (Fig. 4B) and to which C. jejuni displays chemotactic behavior (34). The inability to use formate results in lowered colonization in chickens when combined with the absence of hydrogenase activity (42), whereas absence of the formate chemoreceptor or the FDH accessory protein FdhD resulted in lowered immunopathology (but not reduced intestinal colonization) in gnotobiotic mice colonized with a humanized microflora (2). All of these studies highlight the importance of respiration and metabolism for C. jejuni in the intestinal environment. The metabolism of formate in C. jejuni is mediated by the formate dehydrogenase enzyme FDH. The C. jejuni FDH is a tungstoenzyme rather than a molybdoenzyme (31, 35) and is predicted to contain a selenocysteine (SeC) amino acid at position 181 (10); however, the requirement for selenium in FDH activity in C. jejuni was predicted but not demonstrated prior to the present study. We have shown here the need for selenium for FDH activity and that the Cj1500 (FdhT) and Cj1501 (FdhU) proteins are involved in FDH biogenesis, possibly by mediating selenium uptake or downstream processing. In an independent, concurrent study, Pryjma et al. (28) show that in C. jejuni strain 81-176, the fdhTU genes are required for optimal recovery following invasion of epithelial cells, and their data confirm the role of the fdhTU genes in FDH biosynthesis and activity in a third reference strain


of C. jejuni. Combined with earlier studies on formate metabolism and intestinal colonization studies and virulence of C. jejuni (2, 34, 42), this suggests an important contribution of FDH, and also FdhTU and potentially selenium in colonization and virulence properties of C. jejuni. Selenium is an essential trace element for many organisms and is used in proteins as the 21st amino acid selenocysteine in eukaryotes and prokaryotes (3). Next to FDH, SeC is mostly found in redox-active proteins such as peroxiredoxins, glycine reductases, glutaredoxin, and proline reductases (46, 47), but these proteins are absent in C. jejuni. Functional genomics approaches have greatly facilitated identification of the encoded selenoproteins, as exemplified by the database for trace element utilization (dbTEU) (46). In dbTEU, 220 species have an annotated FDH, and 120 of the 220 species listed have a SeC-containing FDH enzyme predicted. SeC is coded for by the UGA codon, which is commonly annotated as a stop codon, but in mRNA with a SECIS element, SeC insertion can occur at the UGA codon (11). The SECIS element binds the Sec-specific elongation factor (SelB) which forms a complex with tRNA-SeC (selC), which contains the anticodon for UGA. Synthesis of tRNA-SeC is dependent on the SeC synthase (SelA), which uses selenophosphate as the selenium donor, which is synthesized by the selenophosphate synthase SelD protein. SeC insertion is a unique example of cotranslational insertion of a nonstandard amino acid. The dual nature of the UGA codon (a stop codon and a SeC codon) raises an important question of the ability of the cell to distinguish between these two functions.

jb.asm.org 3821

Shaw et al.

The presence of a SECIS element is a marker for SeC UGA codon, but its sequence or secondary structure is not conserved between bacteria (11, 47), and hence predictions are unreliable. This has led to mis-annotation of genome sequences, as exemplified by the presence of two adjacent fdhA genes rather than a single fdhA gene (e.g., in Helicobacter hepaticus open reading frames HH0229 and HH0228 [32]). In addition to fdhA, the C. jejuni NCTC 11168 genome sequence contains two other candidates for SeC-containing proteins: a gene encoding a currently unannotated SelW orthologue in the intergenic region between cj0717 and cj0718 (11), and the SelD protein encoded by the cj1504c gene (Fig. 1A). Interestingly, translation of the cj1504c gene is probably initiated from a rare CTG start codon, a feature which is not apparent in related Campylobacter species and hence its significance is not known. Also, transcript levels of selD are not significantly affected by the inactivation of the fdhU gene, suggesting that not all selenoproteins are affected by the absence of fdhU (Fig. 2B). Analysis of RNA levels of the fdhABC genes in the fdhU mutant showed that the absence of FDH proteins in the fdhU mutant is linked to reduced RNA levels of the corresponding genes (Fig. 2B). Since the RNA levels of the fdhM and cj1513c genes were not affected in the fdhU mutant (Fig. 2B and 3C), we hypothesize that the reduced levels of fdhABC RNA is mediated at the posttranscriptional level in C. jejuni, possibly due to RNA instability in the absence of fdhABC translation beyond the UGA codon in fdhA when the SeC-tRNA is not available. An alternative explanation could be that this is due to increased transcription in the presence of selenium and binding of the SelB protein to the SECIS element (44), as has been suggested for an SeC-containing FDH in Treponema primitia, where sodium selenite-supplementation of growth media did not result in an increased transcript level upstream of the SeC UGA codon but did result in an increased transcript level downstream of the SeC UGA codon (23). In a previous bioinformatic study (47), the SirA-like domain was highlighted as one often found in the vicinity of the selD gene. This gene is present in a highly conserved region of the C. jejuni genome and contains several genes involved in FDH biogenesis and activity, and selenium metabolism (Fig. 1). The cj1501 and cj1505c genes both contain a SirA-like domain, whereas the cj1507c gene was identified as a ModE-like regulator controlling molybdenum and tungsten uptake (35). Selenocysteine biosynthesis is dependent on the presence of selenium, and selenium storage proteins have not been described to date. This implicates that selenium import will be required for biosynthesis of SeCcontaining enzymes but, surprisingly, little is known about selenium transport in bacteria (29). In E. coli, selenate [Se(VI)] is transported via the sulfate ABC transporter encoded by the cysAWTP operon (38), but orthologs of this system are absent in C. jejuni. Since FdhT has 10 predicted transmembrane regions, it is tempting to speculate that this inner membrane protein may function as a selenium transporter or channel, and this hypothesis is supported by the presence of a pfam04143 (sulfur transport) domain. In addition, FdhT orthologs are also annotated as containing a TIGR04112 (seleno_YedE) domain, which is commonly found in genomes also encoding a selenium trait. The homology of FdhU with the TusA sulfate relay protein (19) points to a possible role in intracellular movement of selenium or conversion of the selenium compound to SeC synthesis apparatus. Such a joint role is supported by the link of fdhTU operons with the presence of SeC-containing FDH enzymes in the Gamma- and Epsilon-sub-

3822

jb.asm.org

divisions of the Proteobacteria (Table 3). The role of FdhU is more difficult to predict, but the protein shares significant sequence homology and the N-terminal CPxP motif with the sulfurtransferase TusA which functions in a sulfur-relay system in E. coli (19). Preliminary evidence based on single replicate determination of the selenium content in C. jejuni wild-type and fdhU mutant suggests that the absence of FDH activity in the fdhU mutant is not due to a lack of selenium, since the cellular content of selenium was similar (⬃2 ⫻ 105 Se atoms/cell) between wild-type and fdhU mutant. Since it is unknown what selenium compound(s) are used, it is again tempting to speculate that FdhU is involved in interaction or modification with as-yet-unknown selenium compounds, transferring them to the selenocysteine-biosynthesis pathway. In summation, the present study is the first to provide direct evidence for the role of a SirA-like protein (FdhU) in selenium metabolism via its requirement for FDH biogenesis in C. jejuni. We also show here that inactivation of the fdhTU genes can be partially complemented by exogenous selenium, suggesting the presence of alternative, low-affinity pathways for selenium utilization or transport in C. jejuni, and this links FDH biogenesis and selenium metabolism in this important human bacterial pathogen. ACKNOWLEDGMENTS We thank Julea Butt and Sophie Marritt (University of East Anglia [UEA]) for assistance with the setup of the FDH assays, Graham Chilvers (UEA) for the selenium measurements, Elena Gomez and Tom Turner (Institute of Food Research [IFR]) for technical assistance, and Mark Reuter (IFR) and Dave Kelly and Andy Hitchcock (University of Sheffield) for helpful discussions, suggestions, and protocols. This research was supported by the Institute Strategic Programme Grant and a Doctoral Training Grant from the Biotechnology and Biological Research Council to the Institute of Food Research.

REFERENCES 1. Arner ES. 2002. Recombinant expression of mammalian selenocysteinecontaining thioredoxin reductase and other selenoproteins in Escherichia coli. Methods Enzymol. 347:226 –235. 2. Bereswill S, et al. 2011. Novel murine infection models provide deep insights into the “menage a trois” of Campylobacter jejuni, microbiota and host innate immunity. PLoS One 6:e20953. doi:10.1371/journal.pone.0020953. 3. Bock A. 2001. Selenium metabolism in bacteria, p 7–22. In Hatfield DL (ed), Selenium. Kluwer Academic Publishers, Hingham, MA. 4. Bock A, Forchhammer K, Heider J, Baron C. 1991. Selenoprotein synthesis: an expansion of the genetic code. Trends Biochem. Sci. 16:463– 467. 5. Bock A, et al. 1991. Selenocysteine: the 21st amino acid. Mol. Microbiol. 5:515–520. 6. Driscoll DM, Copeland PR. 2003. Mechanism and regulation of selenoprotein synthesis. Annu. Rev. Nutr. 23:17– 40. 7. EFSA. 2010. Scientific opinion on quantification of the risk posed by broiler meat to human campylobacteriosis in the EU. EFSA J. 8:1437. doi:10.2903/j.efsa.2010.1437. 8. Fouts DE, et al. 2005. Major structural differences and novel potential virulence mechanisms from the genomes of multiple Campylobacter species. PLoS Biol. 3:e15. doi:10.1371/journal.pbio.0030015. 9. Guccione E, et al. 2010. Reduction of fumarate, mesaconate and crotonate by Mfr, a novel oxygen-regulated periplasmic reductase in Campylobacter jejuni. Environ. Microbiol. 12:576 –591. 10. Gundogdu O, et al. 2007. Re-annotation and re-analysis of the Campylobacter jejuni NCTC11168 genome sequence. BMC Genomics 8:162. doi: 10.1186/1471-2164-8-162. 11. Gursinsky T, et al. 2008. Factors and selenocysteine insertion sequence



12. 13. 14. 15. 16. 17. 18.

19. 20. 21. 22. 23. 24. 25. 26. 27. 28.

29. 30.

requirements for the synthesis of selenoproteins from a gram-positive anaerobe in Escherichia coli. Appl. Environ. Microbiol. 74:1385–1393. Hartley-Tassell LE, et al. 2010. Identification and characterization of the aspartate chemosensory receptor of Campylobacter jejuni. Mol. Microbiol. 75:710 –730. Hatfield DL, Gladyshev VN. 2002. How selenium has altered our understanding of the genetic code. Mol. Cell. Biol. 22:3565–3576. Hitchcock A, et al. 2010. Roles of the twin-arginine translocase and associated chaperones in the biogenesis of the electron transport chains of the human pathogen Campylobacter jejuni. Microbiology 156:2994 –3010. Hoffman PS, Goodman TG. 1982. Respiratory physiology and energy conservation efficiency of Campylobacter jejuni. J. Bacteriol. 150:319 –326. Hofreuter D, et al. 2006. Unique features of a highly pathogenic Campylobacter jejuni strain. Infect. Immun. 74:4694 – 4707. Holmes K, et al. 2005. Campylobacter jejuni gene expression in response to iron limitation and the role of Fur. Microbiology 151:243–257. Holmes K, Tavender TJ, Winzer K, Wells JM, Hardie KR. 2009. AI-2 does not function as a quorum sensing molecule in Campylobacter jejuni during exponential growth in vitro. BMC Microbiol. 9:214. doi:10.1186/ 1471-2180-9-214. Ikeuchi Y, Shigi N, Kato J, Nishimura A, Suzuki T. 2006. Mechanistic insights into sulfur relay by multiple sulfur mediators involved in thiouridine biosynthesis at tRNA wobble positions. Mol. Cell 21:97–108. Janssen R, et al. 2008. Host-pathogen interactions in Campylobacter infections: the host perspective. Clin. Microbiol. Rev. 21:505–518. Jormakka M, Byrne B, Iwata S. 2003. Formate dehydrogenase–a versatile enzyme in changing environments. Curr. Opin. Struct. Biol. 13:418 – 423. Kelly DJ. 2008. Complexity and versatility in the physiology and metabolism of Campylobacter jejuni, p 41– 61. In Nachamkin I, Szymanski C, Blaser MJ (ed), Campylobacter, 3rd ed. ASM Press, Washington, DC. Matson EG, Zhang X, Leadbetter JR. 2010. Selenium controls transcription of paralogous formate dehydrogenase genes in the termite gut acetogen, Treponema primitia. Environ. Microbiol. 12:2245–2258. Mota CS, et al. 2011. The mechanism of formate oxidation by metaldependent formate dehydrogenases. J. Biol. Inorg. Chem. 16:1255–1268. Parkhill J, et al. 2000. The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403:665– 668. Pearson BM, et al. 2007. The complete genome sequence of Campylobacter jejuni strain 81116 (NCTC11828). J. Bacteriol. 189:8402– 8403. Petersen L, Larsen TS, Ussery DW, On SL, Krogh A. 2003. RpoD promoters in Campylobacter jejuni exhibit a strong periodic signal instead of a ⫺35 box. J. Mol. Biol. 326:1361–1372. Pryjma M, Apel D, Huynh S, Parker CT, Gaynor EC. 2012. FdhTUmodulated formate dehydrogenase expression and electron donor availability enhance recovery of Campylobacter jejuni following host cell infection. J. Bacteriol. 194:3803–3813. Rosen BP, Liu Z. 2009. Transport pathways for arsenic and selenium: a minireview. Environ. Int. 35:512–515. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.


31. Smart JP, Cliff MJ, Kelly DJ. 2009. A role for tungsten in the biology of Campylobacter jejuni: tungstate stimulates formate dehydrogenase activity and is transported via an ultra-high-affinity ABC system distinct from the molybdate transporter. Mol. Microbiol. 74:742–757. 32. Suerbaum S, et al. 2003. The complete genome sequence of the carcinogenic bacterium Helicobacter hepaticus. Proc. Natl. Acad. Sci. U. S. A. 100:7901–7906. 33. Szklarczyk D, et al. 2011. The STRING database in 2011: functional interaction networks of proteins, globally integrated and scored. Nucleic Acids Res. 39:D561–D568. 34. Tareen AM, Dasti JI, Zautner AE, Gross U, Lugert R. 2010. Campylobacter jejuni proteins Cj0952c and Cj0951c affect chemotactic behavior toward formic acid and are important for invasion of host cells. Microbiology 156:3123–3135. 35. Taveirne ME, Sikes ML, Olson JW. 2009. Molybdenum and tungsten in Campylobacter jejuni: their physiological role and identification of separate transporters regulated by a single ModE-like protein. Mol. Microbiol. 74:758 –771. 36. Thomas MT, et al. 2011. Two respiratory enzyme systems in Campylobacter jejuni NCTC 11168 contribute to growth on L-lactate. Environ. Microbiol. 13:48 – 61. 37. Thome R, et al. 2012. A sulfurtransferase is essential for activity of formate dehydrogenases in Escherichia coli. J. Biol. Chem. 287:4671– 4678. 38. Turner RJ, Weiner JH, Taylor DE. 1998. Selenium metabolism in Escherichia coli. Biometals 11:223–227. 39. van Vliet AHM, Baillon MA, Penn CW, Ketley JM. 2001. The ironinduced ferredoxin FdxA of Campylobacter jejuni is involved in aerotolerance. FEMS Microbiol. Lett. 196:189 –193. 40. van Vliet AHM, Wooldridge KG, Ketley JM. 1998. Iron-responsive gene regulation in a Campylobacter jejuni fur mutant. J. Bacteriol. 180:5291– 5298. 41. Walczak R, Hubert N, Carbon P, Krol A. 1997. Solution structure of SECIS, the mRNA element required for eukaryotic selenocysteine insertion-interaction studies with the SECIS-binding protein SBP. Biomed. Environ. Sci. 10:177–181. 42. Weerakoon DR, Borden NJ, Goodson CM, Grimes J, Olson JW. 2009. The role of respiratory donor enzymes in Campylobacter jejuni host colonization and physiology. Microb. Pathog. 47:8 –15. 43. Weljie AM, Newton J, Mercier P, Carlson E, Slupsky CM. 2006. Targeted profiling: quantitative analysis of 1H NMR metabolomics data. Anal. Chem. 78:4430 – 4442. 44. Yoshizawa S, et al. 2005. Structural basis for mRNA recognition by elongation factor SelB. Nat. Struct. Mol. Biol. 12:198 –203. 45. Young KT, Davis LM, Dirita VJ. 2007. Campylobacter jejuni: molecular biology and pathogenesis. Nat. Rev. Microbiol. 5:665– 679. 46. Zhang Y, Gladyshev VN. 2010. dbTEU: a protein database of trace element utilization. Bioinformatics 26:700 –702. 47. Zhang Y, Turanov AA, Hatfield DL, Gladyshev VN. 2008. In silico identification of genes involved in selenium metabolism: evidence for a third selenium utilization trait. BMC Genomics 9:251. doi:10.1186/14712164-9-251.

jb.asm.org 3823