10 Jul 2012: Debowski AW, Carnoy C, Verbrugghe P, Nilsson HO, Gauntlett JC, et al. (2012) Correction: Xer Recombinase and Genome Integrity in Helicobacter pylori, a Pathogen without Topoisomerase IV. PLOS ONE 7(7): 10.1371/annotation/f5629404-e0e6-4cc1-9d76-ab16238a48c0. doi: 10.1371/annotation/f5629404-e0e6-4cc1-9d76-ab16238a48c0 View correction
In the model organism E. coli, recombination mediated by the related XerC and XerD recombinases complexed with the FtsK translocase at specialized dif sites, resolves dimeric chromosomes into free monomers to allow efficient chromosome segregation at cell division. Computational genome analysis of Helicobacter pylori, a slow growing gastric pathogen, identified just one chromosomal xer gene (xerH) and its cognate dif site (difH). Here we show that recombination between directly repeated difH sites requires XerH, FtsK but not XerT, the TnPZ transposon associated recombinase. xerH inactivation was not lethal, but resulted in increased DNA per cell, suggesting defective chromosome segregation. The xerH mutant also failed to colonize mice, and was more susceptible to UV and ciprofloxacin, which induce DNA breakage, and thereby recombination and chromosome dimer formation. xerH inactivation and overexpression each led to a DNA segregation defect, suggesting a role for Xer recombination in regulation of replication. In addition to chromosome dimer resolution and based on the absence of genes for topoisomerase IV (parC, parE) in H. pylori, we speculate that XerH may contribute to chromosome decatenation, although possible involvement of H. pylori's DNA gyrase and topoisomerase III homologue are also considered. Further analyses of this system should contribute to general understanding of and possibly therapy development for H. pylori, which causes peptic ulcers and gastric cancer; for the closely related, diarrheagenic Campylobacter species; and for unrelated slow growing pathogens that lack topoisomerase IV, such as Mycobacterium tuberculosis.
Citation: Debowski AW, Carnoy C, Verbrugghe P, Nilsson H-O, Gauntlett JC, Fulurija A, et al. (2012) Xer Recombinase and Genome Integrity in Helicobacter pylori, a Pathogen without Topoisomerase IV. PLoS ONE 7(4): e33310. doi:10.1371/journal.pone.0033310
Editor: Niyaz Ahmed, University of Hyderabad, India
Received: December 15, 2011; Accepted: February 7, 2012; Published: April 12, 2012
Copyright: © 2012 Debowski et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study was supported by Ondek Pty Ltd, by research grants to DEB from the US National Institutes of Health R21 AI078237, R21 AI088337 and by a PhD scholarship from NSERC (Natural Sciences and Engineering Research Council of Canada) and the Australian Government to AWD and by a grant to MB from the University of Western Australia Research Grants Scheme RA/1/485/802. HN, JCG, AF, TC, BJM and MB are Employees of Ondek Pty Ltd and participated in the study design, data collection and analysis, and preparation of the manuscript. The other funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscripts.
Competing interests: The authors have read the journal's policy and have the following conflicts: Hans-Olof Nilsson, Jonathan C. Gauntlett, Alma Fulurija, Tania Camilleri, Barry J. Marshall and Mohammed Benghezal are Employees of Ondek Pty Ltd and participated in the study design, data collection and analysis, and preparation of the manuscript. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
Crossovers between circular monomeric chromosomes generate dimers and interlocked (catenated) structures that cannot segregate properly at cell division . Bacteria with circular chromosomes generally contain site-specific tyrosine Xer recombinases that act at cognate dif sites near where replication terminates and that resolve chromosome dimers to free monomers , . Deletion of dif or inactivation of a xer recombinase gene causes formation of abnormally partitioned nucleoids and cell filamentation in E. coli type model organisms . Cell filamentation results from SulA-mediated inhibition of cell division, is induced in the SOS response to DNA and chromosome breakage . Chromosome dimer resolution in most bacterial species, including E. coli, is mediated by two related recombinases, XerC and XerD, that function as a pair of heterodimers and that target a 28-bp dif site. The dif site is presented to the Xer complex by the FtsK DNA translocase protein. FtsK is anchored at incipient cell division septa, and interacts with a set of oriented and highly repeated 8 bp named KOPS sequences (FtsK-orienting polar sequences). The E. coli chromosome's dif region is rich in KOPS sites, which are in opposite orientation on each side of the dif site. Orientation-specific KOPS recognition and asymmetry in the KOPS distribution direct FtsK-chromosome interactions to effectively guide presentation of dif to the XerC-XerD complex for recombination. FtsK interacts specifically with the XerD component of the XerCD complex and with several other proteins including Topoisomerase IV, which are likely to be important for efficient, well-regulated chromosome separation and segregation , , . Many thousands of topological links arise as circular chromosomal DNAs are unwound during replication . In E. coli interlocked (catenated) chromosomes are resolved efficiently to monomers by topoisomerase IV , which is essential , primarily because of its high capacity to resolve interlocked chromosomes, and thereby allow efficient chromosome segregation, apace with rapid cell division . Of note, DNA gyrase (responsible for DNA supercoiling) and Xer recombination may play secondary roles in decatenation , ,. This is illustrated by the ability of E. coli's XerCD-dif-FtsK system can substitute for topoisomerase IV  to remove catenane links between circular DNAs in vitro without topoisomerase IV ; and the suppression of a temperature sensitive (conditional lethal) topoisomerase IV mutation by XerCD-dif recombination in E. coli producing an engineered FtsK protein with no septum anchor (FtsK50C) that is thus soluble in the cytoplasm . This Xer-mediated resolution of catenated DNAs entails multiple interconversions of catenated monomers and knotted dimers, removing a link at each step.
Xer/dif recombination systems have been detected computationally in many phyla ,  including Proteobacteria , Firmicutes ,  and Archae , . Our in silico analyses revealed that more than 85% proteobacterial species contain a conventional E. coli-type system in which related XerC and XerD recombinases act as heterodimers on cognate dif sites, with each Xer protein having a distinct role. However, Helicobacter pylori, the gastric pathogen implicated in peptic ulcer disease and gastric cancer, was inferred to contain just a single Xer recombinase, which was named XerH, as do the related Campylobacters, which cause diarrheal disease, and all other members of H. pylori's epsilon subgroup of Gram negative proteobacteria . Single Xer recombinase systems were also found in the Gram positive Streptococci and Lactococci (XerS)  and in Archaea (XerA) , . ftsK homologues are found in nearly all eubacterial species including the epsilon proteobacteria and Streptococci and Lactococci. Interestingly, several slow growing bacterial pathogenic genera, including Helicobacter, Campylobacter and Mycobacterium, lack parC and parE topoisomerase IV subunit genes . Further complicating inferences about XerH action in the case of H. pylori, many strains contain a second divergent xer recombinase gene, xerT, generally within a large TnPZ transposon , . Although its encoded XerT protein is needed for transposon excision and conjugative transfer, and probably also functions as a transposase, the possibility of XerT collaborating with XerH for chromosome resolution (as with XerC and XerD in E. coli) also merited testing.
The experiments presented here demonstrate site-specific recombination at H. pylori difH sites, and show that it requires XerH, FtsK and an intact difH sequence, but not XerT, and bring into focus the need to learn how catenanes are processed in the many other slow growing human pathogens that, like H. pylori, lack topoisomerase IV.
Materials and Methods
Bacterial strains and culture conditions
The H. pylori strains and plasmids used in this study are listed in Table S1. Streptomycin resistant rpsL-mutant strains were used for transformation with the difH repeat (rpsL-cat containing) cassette , . H. pylori strains were routinely grown at 37°C under microaerobic conditions on Columbia blood agar (CBA) plates containing 5% horse blood and Dent's antibiotic supplement (Oxoid). When appropriate, antibiotic selection in H. pylori was carried out by supplementing media with chloramphenicol, streptomycin, and/or kanamycin at final concentrations of 10 µg/ml. Escherichia coli DH5α was grown in Luria-Bertani broth. When necessary, antibiotics were added to the following final concentrations: ampicillin, 100 µg/ml; kanamycin, 50 µg/ml; and chloramphenicol, 20 µg/ml. H. pylori cultures were incubated at 37°C in sealed jars using the Anoxomat™ MarkII system (Mart Microbiology B.V., The Netherlands) after one atmosphere replacement using the following gas composition N2∶H2∶CO2, 85∶5∶10.
The oligonucleotides used in this study are listed in Table S2
Assays of difH site recombination
To test if the putative 40-bp H. pylori dif sequence (difH) ATTTAAAAGTTTGAAAAGTGCAGTTTTCATAACTAAATGA) was functional, a recombination assay was developed using a cassette containing both selectable (cat) and counterselectable (rpsL) genes, rpsL-cat (streptomycin susceptibility, chloramphenicol resistance, respectively), flanked by difH sites , , . This cassette was generated by PCR amplification from genomic DNA containing rpsL-cat using primers DifHPF and DifHPR, which contain direct repeats of difH and BamHI restriction sites near their 5′ ends. A control cassette containing 40 bp of sequence unrelated to difH was generated with primers NondifF and NondifR. These PCR products were cloned as described  to create plasmids pHInt_difH-RCAT-difH and pHInt_nondif-RCAT-nondif. Constructs were sequenced using primers SEQdifF and SEQdifR to ensure that difH and ‘nondif’ sequences were intact. Natural transformation of a derivative of H. pylori strain 26695 made resistant to streptomycin by mutation in its normal rspL gene (called 26695Str) with pHInt_difH-RCAT-difH and pHInt_nondif-RCAT- nondif  was used to place these cassettes in the H. pylori chromosome between genes HP0203 and HP0204 (strains 26695Str HP0203-4::difH-RCAT and 26695Str HP0203-4::nondif-RCAT, respectively). Chromosomal DNAs from the resulting chloramphenicol-resistant transformants were checked for streptomycin sensitivity (which is dominant to resistance in rpsLmut/rpsLWT partial diploids). These transformants and their descendants were also checked for correct insertion of the difH repeat cassette and for recombination at difH sites by PCR using primers 0203F and 0204R. Additionally, chloramphenicol-resistant transformants were subcultured on streptomycin-containing agar when needed to select for or to quantify rates of loss of the cassette. Chromosomal DNAs of representative streptomycin-resistant, chloramphenicol-sensitive derivatives were sequenced to confirm recombination at difH sites as diagrammed in Figure 1.
(A) Schematic depiction of XerH excision assay. The difH repeat cassette, consisting of streptomycin susceptibility-chloramphenicol resistance genes (rpsL-cat) flanked by difH sequence, was introduced into the H. pylori 26695Str (WT) genome by natural transformation and homologous recombination between genes HP0203 and HP0204 or in place of genes ureA and ureB. Recombination at difH sites leads to excision of rpsL-cat and one difH sequence, detectable by a PCR fragment 1.5 kb smaller than that from parental difH repeat cassette containing DNA. A cassette with nondif sequence flanking rpsL-cat served as a negative control. (B) Results of difH recombination assay in H. pylori 26695Str (WT) with 40 bp difH direct repeats or 40 bp nondif DNA direct repeats at the HP0203-HP0204 locus. With each of these cassettes, two independent clones were tested by diagnostic PCR. The 2.1 kb and 3.6 kb PCR products come difH recombinant and parental (difH repeat containing) DNAs, respectively. Lane C, control from wild-type H. pylori without difH repeat cassette. (C) difH recombination frequencies for difH repeat cassette located at HP0203-HP0204 or ureAB loci. Cells were grown on non-selective media for two or four days, re-streaked for single colonies, and ∼100–200 colonies were tested for retention or loss of rpsL and cat genes by replica plating to streptomycin and to chloramphenicol containing media. Experiments were performed in triplicates; horizontal bars indicate means and standard deviation.
Due to the low difH recombination rates in the HP0203-HP0204 intergenic region, the difH repeat cassette was also placed at the ureAB locus for further studies. Synthetic sequences containing direct repeats of wild-type or mutant difH sites separated by a BglII restriction site and flanked with BamHI sites were synthesized and cloned into plasmid pMA-RQ from Geneart (AG Regensburg, Germany) (mutant difH sequences shown in Table S1). A BamHI fragment containing the rpsL-cat cassette was cloned into BglII digested plasmids pDifWT, pDifM1, pDifM2, and pDifM3 (Table S1) to give rpsL-cat flanked with WT or mutated difH sequences in pDifWT-RC, pDifM1-RC, pDifM2-RC, pDifM3-RC respectively (Table S1). These difH flanked DNAs were excised with BamHI and cloned into plasmid pUreAB , a pBluescript-derived plasmid containing regions of homology for chromosomal replacement of [or insertion between] the ureA and ureB (urease) genes. This resulted in plasmids pUreAB_DifWT-RC, pUreAB_DifM1-RC, pUreAB_DifM2-RC and pUreAB_DifM3-RC, which were used for transformation of H. pylori strain 26695Str to create 26695Str ureAB::difHWT-RC, ureAB::difHM1-RC, ureAB::difHM2-RC and ureAB::difHM3-RC. Using primers UreABF and UreABR, chromosomal DNA of the resulting chloramphenicol-resistant transformants was checked for the correct insertion of the difH repeat cassette and for difH recombination at the ureAB locus.
H. pylori mutants used to assess H. pylori xerH and xerT roles in difH recombination
xerH is the only xer recombinase gene found in every H. pylori genome, although many strains, including 26695 (used here), also contain another xer family gene, xerT . Strain 26695Str derivatives with null mutations in these xer genes were constructed in order to test each gene's role in dif site recombination. xerH::rpsL-cat and xerT::rpsL-cat insertion mutant constructs were generated by PCR with overlapping primers , . Briefly, for xerH (HP0675), DNAs flanking and including much of this gene were amplified from 26695 genomic DNA using primers XerHrcat1 and XerHrcat2, and XerHrcat3 and XerHrcat4, respectively; and the rpsL-cat cassette was amplified using primers XerHrcat5 and XerHrcat6. Nested primers XerHrcat7 and XerHrcat8 were used for splicing overlap extension (SOE) PCR to generate a DNA segment containing rpsL-cat inserted within the HP0675 ORF, which would be suitable for transformation. The same strategy, using primers XerTrcat1 through XerTrcat8, was used to generate xerT::rpsL-cat mutant of the xerT gene (HP0995). Natural transformation of H. pylori strain 26695Str with these products yielded 26695Str xerH::rpsL-cat and 26695Str xerT::rpsL-cat. DNAs of transformants were checked for correct allelic insertion by PCR. Simple unmarked xerH and xerT deletion alleles (ΔxerH and ΔxerT) were then made by SOE PCR. Briefly, for xerH, 1-kb DNA segments flanking HP0675 were PCR amplified using primers XerHdel1 and XerHdel2, and XerHdel3 and XerHdel4, and fused in a second SOE PCR using nested primers XerHdel5 and XerHdel6. The ΔxerT allele was made similarly with primers XerTdel1 through XerTdel6. H. pylori strains containing xerH::rpsL-cat and xerT::rpsL-cat constructs were transformed with corresponding simple deletion DNAs, streptomycin resistance was selected, and transformants were screened for loss of chloramphenicol resistance and further checked by PCR. This resulted in strains 26695Str ΔxerH and 26695Str ΔxerT.
To assess the role of XerH and XerT in difH recombination, these ΔxerH and ΔxerT strains were transformed with pHInt_difH-RCAT-difH to give 26695Str ΔxerH; HP0203-4::difH-RCAT and 26695Str ΔxerT; HP0203-4::difH-RCAT; or with pUreAB_DifWT-RC to give 26695Str ΔxerH; ureAB::difHWT-RC and 26695Str ΔxerT; ureAB::difHWT-RC (difH flanking rpsL-cat at the ureAB locus). Chloramphenicol-resistant transformants were selected, and then assayed for difH site recombination by appearance of streptomycin resistance and loss of chloramphenicol resistance and by PCR.
Complementation of ΔxerH mutation
The xerH (HP0675) ORF was amplified from 26695 genomic DNA using primers XerHF and XerHR and the product was cloned downstream of the strong ureA promoter of pTrpA-up (Table S1) using NdeI and SalI restriction sites to create pTrpA-upXerH. Transformation of strain 26695Str ΔxerH with pTrpA-RC yielded 26695Str ΔxerHrecip, which was in turn transformed with pTrpA-upXerH to create strain 26695Str xerH comp, (ΔxerH complemented with highly expressed xerH gene at the chromosomal trpA locus, under ureA promoter control). Chromosomal DNAs of the resulting transformants were checked by PCR for correct allelic replacement at trpA and to verify that the cloned xerH gene was not at the normal chromosomal xerH locus. Two independently generated 26695Str xerH comp clones were then transformed with pUreAB_DifWT-RC to give 26695Str xerH comp; ureAB::difHWT-RC (difH flanking rpsL-cat at the ureAB locus). Chloramphenicol-resistant transformants were selected and assayed for recombination events at difH sites.
Construction of H. pylori mutants used to assess susceptibility to DNA damage
SOE PCR was used to generate a ruvC::rpsL-cat construct. For ruvC (HP0877), 1-kb DNA segments flanking HP0877 were amplified from 26695 genomic DNA using primers RuvCrcat1 and RuvCrcat2, and RuvCrcat3 and RuvCrcat4 respectively; and the rpsL-cat cassette was amplified using primers RuvCrcat5 and RuvCrcat6. Nested primers RuvCrcat7 and RuvCrcat8 were used to generate a ruvC::rpsL-cat containing SOE PCR product. Natural transformation of the H. pylori strain 26695 with the final SOE PCR products was performed to create 26695Str ΔruvC.
make a recG knockout mutant, two 1-kb DNA fragments upstream and downstream of the HP1523 ORF were amplified using primers RecGkan1 and RecGkan2, and RecGkan3 and RecGkan4 and fused by SOE PCR using primers RecGkan1 and RecGkan4, creating unique EcoRI and BamHI restriction sites. The final PCR product was blunt end cloned into pHSG576  to give pHRecG. The aphA cassette, conferring kanamycin resistance (KanR), was cloned into EcoRI and BamHI digested pHRecG, creating a nonpolar replacement of recG, pHRecG-Km. This plasmid was used for 26695Str transformation to create 26695Str ΔrecG.
Several attempts to construct the DNA segment to make a recA knockout by SOE PCR failed. Therefore, in vitro transposition was used to insert the rpsL-cat cassette into recA. The HP0153 ORF was amplified by PCR, using primers RecAF and RecAR, and cloned into pGEMT-Easy (Promega, Madison, WI) to create pRecA. In vitro transposition of a segment containing rpsL-cat flanked by ends of phage Mu into pRecA was done using MuA transposase (Finnzymes, Finland, F-750) to generate a library of pRecA-RC clones. H. pylori strain 26695Str was transformed to chloramphenicol resistance using this plasmid library to generate 26695Str recA::mu-rpsL-cat. Chromosomal DNA was isolated from transformants, and insertion of the mu-rpsL-cat-mu cassette into ORF HP0153 was confirmed by PCR. A similar strategy was used to truncate the H. pylori ftsK homologue. Primers FtsKF and FtsKR were used to amplify the HP1090 ORF and generate the ftsK containing plasmid pFtsK. In vitro transposition of mu-rpsL-cat into pFtsK generated a library of pFtsK-RC mutant DNAs which was used in transformation to make strain 26695 ftsK::mu-rpsL-cat. Transformants were characterised by PCR and DNA sequencing, and one with an insertion at the 454th codon from ftsK's 3′ end was identified.
Double and triple mutants were obtained by transforming 26695Str ΔxerH with PCR products containing ruvC::rpsL-cat to create 26695Str ΔxerHΔruvC; 26695Str ΔrecG with a recA::mu-rpsL-cat-containing PCR product to create 26695Str ΔrecG recA::mu-rpsL-cat; 26695Str ΔxerH, 26695Str ruvC::rpsL-cat and 26695Str ΔxerHΔruvC with pHRecG-Km to give 26695Str ΔxerHΔrecG, 26695Str ΔruvCΔrecG and 26695Str ΔxerHΔruvCΔrecG.
Construction of xerH and ruvC mutants in strain X47
To generate ΔxerH derivatives of X47 (which already contains an rpsL streptomycin resistance allele), this strain was transformed with genomic DNA from strain 26695str xerH::rpsL-cat. Genomic DNA isolated from the resulting transformants was used as template for PCR to confirm rpsL-cat insertion at xerH. A pool of X47 xerH::rpsL-cat clones was transformed with the PCR product obtained from 26695Str ΔxerH (unmarked deletion), with selection for streptomycin resistance and chloramphenicol susceptibility. Genomic DNA from the resulting X47 ΔxerH transformants was used as template for PCR to confirm clean deletion of xerH, as described for 26695.
To generate the ruvC mutant in X47 background, wild-type X47 was transformed with genomic DNA from strain 26695 ΔruvC::rpsL-cat. Genomic DNA isolated from the resulting transformants was used as template for PCR to confirm rpsL-cat insertion at the ruvC locus as described for 26695.
UV susceptibility assay
Fresh cultures of H. pylori, passaged the day before, were suspended in PBS (pH = 7.2) and standardized to an OD600 = 2. 50 µl aliquots of bacterial suspension was placed into a single well of a UV transparent 96 well plate (NUNC) and exposed to UV light at 312 nm using the TFX-35M transluminator (LifeTechnologies, Carlsbad, CA) at a distance of 45 cm for 0 to 75 sec. Serial dilutions of irradiated cells were plated onto CBA plates and incubated at 37°C for four days. Colonies were counted and percent survival was calculated. UV susceptibility experiments were repeated at least three times on two independent clones for each H. pylori strain.
DNA content analysis by Fluorescence Activated Cell Sorting (FACS)
The DNA content analysis was performed as described . Briefly, H. pylori cells were grown on blood agar plates and inoculated in BHI medium containing 10% Newborn Calf Serum (NCS) at OD600 = 2 and incubated in 100 µl in 96 well plate overnight (18 h) with shaking at 100 rpm. The bacteria were collected by centrifugation, the pellet was resuspended in 300 µl PBS and added to 900 µl PBS with 4% paraformaldehyde and samples were incubated 15 min at room temperature. The bacterial pellet was washed with PBS, resuspended in 100 µg/ml RNase A and 2 µg/ml Hoechst 333421 (Sigma) in PBS and incubated for 10 min at 37°C. The bacterial pellet was washed and resuspended in PBS. The analysis was performed with a Becton Dickinson FACSCanto II Flow cytometer.
Oxidative stress susceptibility assay
Susceptibility to oxidative stress using 2 mM and 20 mM paraquat was tested on Columbia agar plate using the disc method as described .
Antibiotic Sensitivity Testing
Wild-type, ΔxerH mutant and xerH complemented strains were inoculated on Columbia agar plates. A single E-test strip (AB Biodisk) was placed in the centre of each plate and the plates were incubated at 37°C for 6 days. The minimal inhibitory concentration (MIC) for ciprofloxacin was determined according to the manufacturer's instructions.
Fresh cultures, passaged the day before, were resuspended in BHI. 5 ml of BHI containing 10% NCS and Dent (Oxoid) was inoculated with 100 µl of a stock inoculum standardized to OD600 = 2. Growth studies were performed without any prior adaptation of H. pylori strains to liquid media. Growth was measured every 4 to 12 h for up to 40 h. Each experiment was done in duplicate and repeated at least twice.
H. pylori grown on Columbia agar were collected and washed in PBS (pH 7.4), prefixed in 2.5% glutaraldehyde in PBS buffer for 1 h, and then rinsed in PBS. After post-fixation in 1% osmium tetraoxide (in PBS), samples were dehydrated through ascending gradient of ethanol and then critical-point dried using carbon dioxide. Samples were sputter coated with palladium (4 nm) and examined using a SEM (Zeiss 1555 VP SEM) at 3 KV and a working distance of 6 mm.
Phylogeny of the single Xer recombinases
Experimental Infection of Mice
Helicobacter free C57BL/6J mice were purchased from the Animal Resource Centre (Perth, Western Australia). Studies were performed with approval from the UWA Animal Ethics Committee (approval no. 07/100/598). Each eight-week-old mouse was orogastrically inoculated with approximately 109 CFUs of H. pylori harvested from an overnight agar plate culture into BHI broth. Colonisation of mice inoculated with X47 wild-type or its ΔxerH mutant was evaluated 2 weeks after challenge as described .
The difH sequence undergoes site-specific recombination in H. pylori
An excision assay in H. pylori strain 26695 was designed to mimic chromosome dimer resolution (Figure 1A) and to test the ability of the recently identified difH site  to undergo site-specific recombination. This assay used a DNA segment with direct repeats of difH flanking rpsL and cat genes, which confer susceptibility to streptomycin and resistance to chloramphenicol, respectively. In our first tests this “difH repeat” cassette was inserted in the H. pylori chromosome between genes HP0203 and HP0204, which is about 525 kb from the normal difH site (Figure S1). A negative control strain contained an equivalent cassette, but with 40-bp of other DNA (‘nondif’) in place of difH at this same chromosomal location. In PCR tests, each of two bacterial clones containing the difH repeat cassette yielded two DNA fragments: 2.1 kb, expected of recombination between difH sites; and 3.6 kb, the full-length cassette (not recombinant at difH sites) (Figure 1B), whereas clones containing the nondif repeat cassette yielded only the 3.6 kb PCR product (Figure 1B). This outcome indicates that difH sites placed at a new chromosomal locus can undergo site-specific recombination.
difH recombination at the HP0203-HP0204 locus was also characterised by restoration of streptomycin resistance, which results from loss of the rpsL (susceptible) allele (difH recombination), or from rare gene conversion between the added rpsL gene in this cassette and the resistant allele at the normal chromosomal rpsL locus . Under our conditions, no (<0.1%) streptomycin resistant colonies were obtained when the 40 bp yeast DNA (‘nondif’) sequence flanked the rpsL-cat cassette. In contrast, many streptomycin resistant colonies were obtained in strains in which difH flanked rpsL-cat. All of these streptomycin resistant colonies were chloramphenicol sensitive. PCR confirmed that streptomycin sensitive clones still contained the full-length difH repeat cassette and that rpsL-cat was absent from streptomycin resistant clones. As expected, DNA sequencing confirmed that a single difH “scar” sequence had been retained in these streptomycin resistant, chloramphenicol sensitive excisants, (Figure 1 and data not shown). The frequency of XerH recombination at difH sites at the ureAB locus (about 647 kb from difH site; Figure S1) was also evaluated using this difH repeat cassette. Cells were grown for 2 and 4 days in non-selective medium (no chloramphenicol), streaked out for single colony isolates, and 100–200 colonies were then tested for streptomycin/chloramphenicol resistance/susceptibility phenotypes. Figure 1C shows that the difH recombination frequency is significantly higher for the cassette placed at ureAB than at HP0203-HP0204 after 4 days culture. We conclude that the chromosomal position of paired difH sites affects the frequency of recombination between them.
difH site recombination requires the single XerH recombinase
xerH was the only xer recombinase gene identified computationally in every fully sequenced genome of the Campylobacterales order, which includes the genus Helicobacter . However, many H. pylori strains also contain another xer-like recombinase gene named xerT (named for its association with the TnPZ transposon) , . In frame deletions were made in both xerH and xerT to test each of them for possible roles in difH site-specific recombination. Using the difH repeat cassette at the ureAB locus and our PCR assay, we found that xerH deletion blocked difH recombination, whereas xerT deletion did not (Figure 2A). In addition, no excision was detected in the ΔxerH mutant by testing ∼200 colonies for the emergence of streptomycin resistant, chloramphenicol sensitive colony phenotypes after 2 and 4 days growth on non-selective medium, whereas the isogenic ΔxerT mutant strain has undergone as much recombination as the xer-wild-type strain, or possibly more, after 2 and 4 days incubation (Figure 2B).
PCR-based difH recombination was assayed in H. pylori 26695Str (WT) and isogenic derivatives harbouring the difH repeat cassette after growth on non-selective media, essentially as in Figure 1. Two independent transformant clones of each strain were tested. The PCR product from control DNA (lane C), which did not have a difH repeat cassette, is slightly smaller than the product reflecting difH recombination because it does not contain the single copy of difH that remains after recombination (See Figure 1A). (A) Tests of difH recombination in WT and ΔxerH and ΔxerT derivatives, harbouring difH repeat cassette at the ureAB locus. (B) Tests of difH recombination in WT, ΔxerH, and ΔxerT derivatives, harbouring difH repeat cassette at the HP0203-HP0204 locus strains. (C) Tests of difH recombination in WT, ΔxerH and ΔxerH complemented with a highly expressed xerH gene using strains with the difH cassette at the ureAB locus.
Complementation of the ΔxerH allele with an intact xerH gene under a strong urease promoter (where it is probably over-expressed) restored difH site-specific recombination (Figure 2C). Indeed, the high level of the small PCR product in XerH complemented strains (difH recombination product; >80% of total) suggested that the amount of XerH protein per cell may be tightly controlled, and that increased XerH protein markedly increased the excision frequency, at least for difH at ectopic sites.
The plasmid used to introduce the difH repeat cassette into the H. pylori chromosome (pHInt_difH-RCAT-difH) did not undergo difH site recombination in E. coli (data not shown), indicating that E. coli's own XerC and XerD recombinases do not process difH; this was as expected, given difH and E. coli dif sequence divergence (Table 1 and ). However, expression of XerH in E. coli promoted excision at difH sites in 10% of this plasmid population (estimate based on plasmid restriction enzyme digest profile; data not shown). Taken together, our results indicate that recombination between difH sites is mediated by just one Xer recombinase without other species-specific factors and that XerH may be limiting, at least for difH sites far removed from their normal location. They further suggested that XerT might inhibit XerH, since more difH recombination was seen in the ΔxerT strain than in its wild-type parent (Figure 2B), although further testing is needed to learn if the stimulation seen in a ΔxerT strain reaches statistical and thus biological significance.
Point mutations in difH sequences that block XerH-mediated recombination
Before performing a mutational analysis of the difH, the nucleotide frequency at each position was calculated over 50 bp from 24 epsilon proteobacterial species to establish a consensus sequence (Table 2, Table S3). This revealed that difH consists of two highly conserved regions (position 17 to 23 and 29 to 34) separated by a variable region (position 24 to 26 and 28) and two other highly conserved positions, 10 and 14 (Figure 3A). In addition, positions 17 to 22 and 29 to 34 are always in a palindrome (inverted repeat), whereas positions 13, 23, 28, and 38 are not (Figure 3B). Thus, the difH sequence can be viewed as two matched domains flanking a unique 6 bp central region.
(A) difH consensus sequence and nucleotide variability for difH sequences from 24 epsilon- proteobacterial species (Table 2). If a given nucleotide is present in more than 50% of species it is written in upper case; if not, the most frequent nucleotide is in lower case. The nucleotide variability at each position was defined as 1–f, where f is the frequency of the most frequent nucleotide. (B) Palindromicity was analysed by comparing the 50-nt difH sequence with its inverted complementary counterpart in the 24 epsilon-proteobacterial species (Table 2). When a nucleotide was found both in difH and in the reverse complementary sequence, a value of 1 was given to the position. Next, the values for the 24 difH sequences for each position were added together to give the n value. The palindromicity frequency (fpal) was then estimated as: fpal = n/24, with 24 being the number of difH sequences analysed. A fpal value of 1 given to a nucleotide position means that the nucleotide is always part of a palindrome.
difH point mutations were made and tested for their ability to undergo difH recombination in H. pylori when at the ureAB locus. PCR tests showed that replacement of G by C in palindrome position 18 or of the four As by four Ts in palindrome positions 19 to 22 each abolished difH recombination in each of two independent clones (Figure 4A, lanes difH M1 and difH M2). difH recombination also seemed to be reduced by a C to G mutation at the non-palindromic but highly conserved position 23, although this mutation's effect seemed leaky (very weak excisant PCR product; Figure 4A, lane difH M3). In confirmation, tests for the appearance of streptomycin resistant clones showed that the difH M2 mutation abolished difH recombination, and that the difH M3 mutation caused a severe impairment of recombination (1±1% for difH M3 vs. 9±2% for difH–wild-type after 4 days incubation on non-selective medium; Figure 4B).
(A) The four A's in positions 19 to 22 of the difH sequence were changed to T's (empty stars), the G in position 18 was changed to C (black star) and the C in position 23 was changed to G (grey star). PCR tests were carried out as in Figure 1 on a clone harbouring WT difH repeats and two clones harbouring each mutant difH sequence (same sequence in each copy of difH) as indicated by the stars, difH M1, difH M2 and difH M3. No PCR product reflecting difH recombination was detected in difH M1, difH M2 mutants, whereas a weak band reflecting difH recombination was detected in the two clones with difH M3 sequences. Underlined nucleotides were found in a palindrome in all species studied. (B) difH recombination frequencies for WT difH and the difH M1, difH M2 and difH M3 mutant sequences carried out as in Figure 2B. Experiments were performed in triplicate; horizontal bars indicate means and standard deviations.
FtsK is required for XerH recombination
The presence of an ftsK homologue (HP1090) in H. pylori suggested that XerH action might require the FtsK DNA translocase, much as does XerC/D action in E. coli. To test the role of H. pylori's putative FtsK translocase protein, we deleted the 3′ terminal 1212 bp of the 2580 bp ftsK gene to generate a strain whose FtsK protein is truncated, and missing the γ regulatory domain that in the E. coli mediates FtsK interaction specifically with XerD, and not with the related XerC recombinase, for XerC/XerD mediated dif recombination . Our mutant protein retains the N-terminal membrane anchor domain, which is essential for viability . No difH recombination was detected by PCR in an H. pylori strain containing our mutant ftsK gene and the difH repeat cassette at the ureAB locus (Figure 5). We conclude that XerH mediated recombination between difH sites depends on FtsK in H. pylori, even for difH sites far from the normal difH locus (Figure S1), and despite H. pylori's use of a single Xer-type recombinase.
Recombination at difH sites was scored as in Figure 2A in H. pylori 26695Str (WT) and its derivative containing a C terminal deletion in ftsK, in each case with the difH repeat cassette at the ureAB locus.
In E. coli FtsK guides chromosomal translocation and presentation of dif to the Xer complex by a set of asymmetric KOPS sequences (5′-GGGNAGGG) whose distribution is skewed toward dif and the replication terminus and polarized (most on leading DNA strand during bidirectional chromosome replication). Although this octamer is also abundant in H. pylori genomes, we think it is unlikely to serve as KOPS for H. pylori's FtsK protein because its genomic distribution is neither skewed near difH nor highly polarized. The only H. pylori octamer mimicking E. coli's KOPS in its distribution is 5′-AGTAGGGG-3′ (Figure S1). This octamer was identified earlier by Hendrickson and Lawrence in their survey of many bacterial genomes  as a putative chromosome “architecture imparting sequence” for H. pylori. In accord with their view, we propose that it serves as H. pylori FtsK's guide for difH presentation, its KOPS sequence.
Slight growth defect, UV and ciprofloxacin susceptibility, and resistance to oxidative stress of ΔxerH mutant
Light and scanning electron microscopy indicated that loss of xerH did not cause filamentation in H. pylori equivalent to that caused by loss of xerC or xerD in E. coli (Figure 6A and 6B). The ΔxerH mutant grew less well than its wild-type parent did (Figure 6C), as did a derivative of this ΔxerH strain complemented by a highly expressed intact xerH gene (Figure 6C). These outcomes suggest that XerH levels are regulated – that either too much or too little can be deleterious.
(A) and (B) Electron micrographs. WT and ΔxerH mutant cells were fixed with glutaraldehyde and processed for scanning electron microscopy. Both WT and ΔxerH mutant cells displayed the characteristic curved rod morphology; in contrast to Δxer mutant E. coli, none were filamentous. (C) Growth curves of H. pylori in liquid medium. Cells were grown in BHI liquid medium supplemented with 10% NCS in microaerobic conditions with agitation. The optical densities (OD600 nm) of WT, ΔxerH and complemented strains were measured in triplicate for up to 40 h. (D) UV sensitivity of WT and recombination mutant H. pylori. Cells were exposed to UV light as described in the methods and viable colony forming units (survival) was determined. Each test was repeated at least three times; standard deviation is indicated. (E) UV sensitivity of WT and ftsK mutant H. pylori determined as in part D.
Studies in E. coli showing that xerC inactivation exacerbated the moderate UV susceptibility that is caused by a ruv-deficiency  prompted us to test if xerH inactivation affects H. pylori's UV susceptibility. Figure 6D shows that xerH inactivation caused UV sensitization, albeit less extreme than that caused by recA or ruvC inactivation, and that normal UV resistance was restored to a ΔxerH mutant by complementation with a functional xerH gene. Inactivation of xerH in a ΔruvC mutant did not affect this strain's normally very high UV sensitivity, whereas enhanced UV sensitivity was observed in a ΔrecG mutant with increased recombination  compared to either mutant alone (Figure 6D). The ΔrecA ΔrecG or ΔruvC ΔrecG double and ΔxerH ΔruvC ΔrecG triple deletion strains had the same phenotype as ΔrecA or ΔruvC single deletion strains (data not shown). Two failed attempts to obtain a ΔrecA derivative of a ΔxerH strain prevented assessment of UV sensitivity in the absence of both xerH and RecA-mediated generalized recombination. Deletion of sequences encoding the FtsK gamma (probable XerH interaction) domain caused moderate UV sensitization, almost as much as that caused by ΔxerH itself (Figure 6E). Ciprofloxacin induces DNA double-strand breaks that are repaired by RecA- and RuvABC-mediated homologous recombination .The ΔxerH mutant was more sensitive to ciprofloxacin than wild-type (Table 3) despite H. pylori's functional recA and ruvABC genes. This suggested a function other than DNA repair for XerH (e.g., dimeric and catenated chromosome resolution). The complemented xerH mutant (over-expressing XerH) was more resistant to ciprofloxacin (Table 3), again indicating a function other than DNA repair. Finally, the oxidative stress resistances of ΔxerH and ΔftsK mutant H. pylori were similar to that of wild- type, as were ΔrecA, ΔruvC and ΔrecG mutant strains (Figure S2). This outcome indicates that XerH recombination is not needed for base pair excision repair in H. pylori. Since the generalized recombination that UV, recG deletion and ciprofloxacin promote should result in formation of dimeric and catenated chromosomes, we suggest that failure to resolve such topological structures underlies the ΔxerH and ΔftsK mutant phenotypes. We propose that Xer recombination resolve chromosome dimers in H. pylori and speculate a role of XerH/difH in chromosome decatenation.
Impaired chromosome segregation in ΔxerH mutant
As noted above, an inability to resolve chromosome dimers and catenated chromosomes should block chromosome segregation at cell division. The lack in H. pylori of homologues of parC and parE, which in E. coli encode the two subunits topoisomerase IV, suggested that XerH-mediated recombination might also be used for chromosome decatenation. To test this idea, the DNA contents of wild-type, ΔxerH mutant and complemented ΔxerH mutant strains were analysed by flow cytometry after staining of DNA with Hoechst dye, much as in other studies of the hobA chromosome replication initiation gene . This showed that ΔxerH mutant cells contained more DNA on average than their wild-type parents did (Figure 7). In order to specifically test that XerH could perform decatenation, uncomplicated by chromosome dimers, which arise by generalized (RecA-mediated) recombination, we would have needed a ΔrecA ΔxerH double mutant. However, as noted above, we were unable to construct this double deletion strain.
We also note that the DNA contents of XerH complemented cells, which have increased XerH activity, was higher than those of isogenic wild-type cells (Figure 7). This suggests that excess XerH protein stimulates initiation of chromosome replication, or conceivably, that it interferes with chromosome segregation (reminiscent of that seen when XerH protein is absent). Taken together, these results suggest that ΔxerH mutants do not undergo efficient chromosome segregation, that they accumulate subpopulations, one with multiple (dimeric and perhaps entangled) chromosomes, and one without chromosomal DNA. A chromosome segregation defect is in line with a failure to resolve dimeric and catenated chromosomes, structures that Xer recombination can resolve in E. coli .
XerH is needed for gastric colonisation
The ΔxerH mutant's apparent defect in chromosome segregation and lack of severe growth phenotype in vitro prompted us to test if xerH is needed by H. pylori in its gastric mucosal environment. The ΔxerH allele, and for comparison, a ΔruvC allele, were transformed into strain X47, which colonizes mice robustly. C57BL/6J mice were inoculated orogastrically with mixtures of these mutant strains and their isogenic X47 wild-type parent; the mice were sacrificed two weeks later and gastric mucosal levels of H. pylori were assayed by bacterial culture. Compared to the robust stomach colonization observed for wild-type H. pylori (WT), the ΔxerH mutant did not colonize mice at all (Figure 8). Interestingly, the ΔruvC mutant, which exhibits a more severe DNA repair defect, did colonise mice, although with a 5-fold lower bacterial load than wild-type (Figure 8). The inability of the ΔxerH mutant to survive in the gastric niche contrasts with ΔruvC mutant colonization, and further supports the idea that XerH is not involved in DNA repair, but rather in chromosome maintenance such as chromosome dimer resolution and possibly in chromosome unlinking. This, in turn, suggests that the slow growing H. pylori depends on a unique chromosome replication and maintenance machinery to thrive in its special gastric niche.
The X and Y axes indicate the relative Hoechst fluorescence units and number of H. pylori cells, respectively. Dotted vertical lines indicate genome equivalents. The main fluorescent signal of wild-type H. pylori was considered as one genome equivalent, as described .
H. pylori X47 wild-type (WT) and isogenic ΔxerH and ΔruvC mutants were used to inoculate five to ten eight-week old C57BL/6J mice. Mice were sacrificed and colonisation levels in stomachs were measured as described in the methods. Data is presented as a scatter plot with each point representing the CFU count of one mouse stomach, and the solid line the geometric mean ± standard deviation for each group (WT, ΔxerH, and ΔruvC).
The present study confirmed our computational prediction  that H. pylori uses just one dedicated tyrosine recombinase, XerH, for site-specific recombination at a cognate chromosomal dif site (difH) – not a pair of distinct proteins akin to XerC and XerD tyrosine recombinases of E. coli and most other eubacterial species. For our experiments, we constructed a cassette with direct repeats of difH sites flanking counterselectable and selectable genes (rpsL and cat, respectively) and placed this “difH repeat” cassette at arbitrarily chosen H. pylori chromosomal locations. difH recombination was detected by loss of the rpsL-cat segment, as scored by bacterial phenotype or PCR. Deletion of xerH blocked recombination between difH sites in the H. pylori chromosome; and conversely, xerH expression promoted recombination between them in a plasmid in E. coli. The related XerT recombinase, present in many but not all H. pylori strains , was not needed for difH recombination. This fits with XerT's usually being associated with a widespread conjugative transposon, but not a fixed component of every H. pylori genome , . Single Xer proteins are also used for recombination at cognate dif sites in Lactococci , Streptococci  and related Gram positive genera , and in Archaea , , but they are distinct phylogenetically from H. pylori's XerH and XerT (Figure 9).
Species are representative of their respective taxonomic groups. XerC and XerD from E. coli and XerT from H. pylori were included in the study as reference. Amino acid sequence alignments were performed using Clustal W. The phylogenetic analyses, using the Neighbour-Joining method  were conducted in MEGA4 . The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) is shown next to the branches. The bootstrap consensus tree is taken to represent the evolutionary history of the taxa analysed. For each species, the accession number is indicated as well as the number of amino acid residues composing the recombinase.
Our ΔxerH mutant H. pylori exhibited a general DNA segregation defect. No typical filamentation was observed during normal growth (Figure 6), as in E. coli xerC or xerD or dif site mutant strains , or under UV stress (data not shown). This discrepancy may be explained by H. pylori's lack of an SOS response and E. coli-type cell division inhibitor (SulA), which is induced by the DNA breakage  that occurs when cell division proceeds without chromosome dimer resolution. Alternatively, the lack of filamentation in H. pylori ΔxerH mutant might be ascribed to H. pylori's much longer doubling time (3–4 hours) and small genome size (one-third E. coli's). Although E. coli chromosome replication takes some 40 min, rapidly growing E. coli (20 min generation time) can undergo multifork replication. In consequence, rounds of replication can initiate in one cell cycle, finish in the next cell cycle, and still allow segregation in which each daughter cell receives at least one complete genome. There should be no such need for multifork replication in H. pylori, with its small genome size, and leisurely growth rate. We propose that these features underlie the lack of filamentation in ΔxerH mutant H. pylori.
Deletion of xerH in H. pylori caused: (i) a slight growth defect in liquid culture (Figure 6C), as is typical of Δxer mutants of E. coli  (ii) a markedly increased sensitivity to DNA breakage inducing and homologous recombination stimulating UV irradiation (Figure 6, D and E), and ciprofloxacin (Table 3), (iii) an increased UV sensitivity of a ΔrecG mutant ; (iv) increased cellular DNA content (Figure 7), which we interpret as a defect in chromosome segregation; and (v) an inability to colonize mice (Figure 8). Overexpression of xerH in our complementation experiments also increased the level per cell. This unexpected finding suggests a role of XerH in regulation of DNA replication/segregation and merits further study. We also found that an intact FtsK DNA translocase protein was needed for difH recombination (Figure 5), presumably for effective difH site presentation to H. pylori's XerH recombination complex at the end of each DNA replication cycle, and much as expected based on E. coli results , , , , , , . Xer recombination likely depends on and is regulated by cognate FtsK proteins in all eubacterial species, although, curiously not in Archaea, since they lack obvious ftsK genes , , , . Deletion of H. pylori FtsK's cytoplasmic domain, including its putative γ segment, which probably interacts with XerH, led to a loss of difH site recombination. This implies FtsK-XerH interaction for coordinated control of chromosome dimer resolution and segregation to daughters at cell division. Our finding of XerH-dependent difH recombination in E. coli raises the possibility of low level FtsK-independent recombination in H. pylori as in Vibrios  or lack of FtsK-Xer interaction specificity as observed in Streptococci .
The importance of the wild-type difH sequence was confirmed by finding that each of our several difH sites mutations interfered with difH recombination (Figure 4). difH's short inverted repeats, which flank a small central unique sequence spacer, have a slight asymmetry (e.g., left arm positions 10 and 14) that is well conserved among epsilon proteobacteria (Figure 3A). We speculate that this asymmetry could be used to determine the time and place of the FtsK-XerH complex's DNA cleavages: first on one difH strand, and then the other – reminiscent of the sequential cleavages by the phylogenetically distinct single XerS recombinase , and formally equivalent to the different roles and timing of action of E. coli's XerC and XerD proteins in its heterodimeric recombinase . Our computational analysis further identified the sequence 5′-AGTAGGGG, whose polarized clustering near difH make it a prime candidate for H. pylori's KOPS; this octamer had also been noted earlier, and was formally proposed as a chromosome “architecture imparting sequence”, but without suggesting a molecular terms . H. pylori's putative KOPS diverges markedly from E. coli's KOPS octamer (Figure S1), a feature that should encourage comparison of FtsK-KOPS binding and associated interactions in these two species.
The reason for our inability to delete recA in a ΔxerH mutant strain is unclear, but might suggest a second important role for difH recombination. For example, many thousands of topological links created by unwinding and replication of double stranded circular DNAs must all be removed for proper chromosome segregation at cell division. The great majority of such links are removed in E. coli by topoisomerase (Topo) IV, which interacts with and is stimulated by FtsK. However, H. pylori seems to lack this essential enzyme: it has no close homologues of the Topo IV encoding parC and parE genes , and thus must use some other enzyme system. The possibility that XerH/difH recombination could allow H. pylori to avoid a chromosome decatenation dilemma is suggested by findings of E. coli XerC/D and dif-dependent (although inefficient) DNA decatenation in vitro; and by the in vivo XerC- and XerD-dependent suppression of a temperature sensitive (conditional lethal) Topo IV mutation when mutant soluble form of FtsK is overproduced . We think that the case for XerH-mediated decatenation in H. pylori would be strengthened if it were shown that this suppression reflects fulfilment of Topo IV's functions by a more effective XerC/XerD/FtsK complex, not just FtsK-mediated stabilization of an impaired (temperature sensitive) Topo IV . It is interesting in this context, that obvious parC and parE (Topo IV) homologues are found in Lactococci and Streptococci, which also use just a single Xer recombinase, these Gram positive species do not challenge our conventional understanding of how chromosomes are decatenated  in the way that H. pylori does. As a third case, Archaea, use just a single Xer recombinase (XerA) but lack obvious homologues of parC and parE and ftsK . Assuming that they will have developed yet another solution to the decatenation problem, valuable insights should emerge from detailed comparisons of daughter chromosome separation and chromosome integrity maintenance in diverse microbial species. One possible solution emerges from finding of higher decatenase activity in the DNA gyrases of M. tuberculosis and M. smegmatis than of E. coli , .
Although we can speculate that H. pylori chromosome decatenation is mediated by iterated round of XerH action on difH, we can also imagine DNA gyrase-mediated decatenation in H. pylori. This would be in accord with our ΔxerH strain's increased susceptibility to the gyrase inhibitor ciprofloxacin (Table 3), and E. coli DNA gyrase's low efficiency decatenation of linked circular DNAs in vitro (superimposed on its very efficient DNA negative supercoiling). As a final alternative, we can also imagine H. pylori's topoisomerase III (HP0116) mediating sufficient decatenation, by extrapolation from Topo III's activity in E. coli .
The analyses presented here suggest many valuable experiments for future studies, bringing into focus the need to learn how catenanes are processed in the many other slow growing human pathogens that, like H. pylori, lack topoisomerase IV. Particularly informative should be further molecular genetic and enzymologic analyses of H. pylori's XerH, DNA gyrase and TopoIII, in the context of this pathogen's small genome size and leisurely growth rate. The lessons learned should be applicable to the understanding, diagnosis and therapy for diverse pathogens and conditions: H. pylori itself, and peptic ulcer disease and gastric cancer; the closely related Campylobacters and associated diarrheal diseases; and equally, unrelated pathogens such as Mycobacterium tuberculosis, which chronically infects many millions of people worldwide, also without obvious genes for Topoisomerase IV.
Positions of putative KOPS sequences and other features in the H. pylori 26695 genome sequence. The H. pylori 26695 genome sequence was scanned for the octameric AGTAGGGG sequences that had been implicated computationally as likely to affect chromosome architecture  (A), and for the GGGNAGGG octamer that constitutes the KOPS sequence of E. coli  (B). The circular H. pylori genome is presented here as a linear structure, with ends corresponding to its origin of bidirectional replication. The AGTAGGGG and GGGNAGGG octamers are represented by red and blue plain diamonds, respectively. Also indicated are the locations of xerH and difH, and the HP0203-HP0204 and ureAB loci at which we had placed difH repeat cassette.
Sensitivity of H. pylori mutants to oxidative stress. Sensitivity to oxidative stress was evaluated in a disk assay using 2 mM or 20 mM of paraquat on blood agar plates that had previously been streaked for confluent growth with either mutant or wild-type cells as indicated. Following a 3–4 day incubation period, the clear zones surrounding the disks were measured. Experiments were repeated three times and standard deviation is indicated.
Plasmids and bacterial strains used in this study.
Oligonucleotide primers used in this study.
Nucleotide frequency (%) and consensus difH sequences from 24 epsilon-proteobacterial species chromosomes.
The SEM work was carried out using facilities at the Centre for Microscopy, Characterisation and Analysis, The University of Western Australia.
We thank Marie-Claude Serre for providing information on Archaea Xer recombinases.
Conceived and designed the experiments: AWD CC AF DEB BJM MB. Performed the experiments: AWD CC PV HN JCG TC MB. Analyzed the data: AWD CC PV HN DEB BJM MB. Contributed reagents/materials/analysis tools: AWD CC HN. Wrote the paper: AWD CC DEB MB.
- 1. Blakely G, May G, McCulloch R, Arciszewska LK, Burke M, et al. (1993) Two related recombinases are required for site-specific recombination at dif and cer in E. coli K12. Cell 75: 351–361.
- 2. Aussel L, Barre FX, Aroyo M, Stasiak A, Stasiak AZ, et al. (2002) FtsK Is a DNA motor protein that activates chromosome dimer resolution by switching the catalytic state of the XerC and XerD recombinases. Cell 108: 195–205.
- 3. Leslie NR, Sherratt DJ (1995) Site-specific recombination in the replication terminus region of Escherichia coli: functional replacement of dif. EMBO J 14: 1561–1570.
- 4. Huisman O, D'Ari R, Gottesman S (1984) Cell-division control in Escherichia coli: specific induction of the SOS function SfiA protein is sufficient to block septation. Proc Natl Acad Sci U S A 81: 4490–4494.
- 5. Crozat E, Grainge I (2010) FtsK DNA translocase: the fast motor that knows where it's going. Chembiochem 11: 2232–2243.
- 6. Hendrickson H, Lawrence JG (2006) Selection for chromosome architecture in bacteria. J Mol Evol 62: 615–629.
- 7. Grainge I, Bregu M, Vazquez M, Sivanathan V, Ip SC, et al. (2007) Unlinking chromosome catenanes in vivo by site-specific recombination. EMBO J 26: 4228–4238.
- 8. Zechiedrich EL, Cozzarelli NR (1995) Roles of topoisomerase IV and DNA gyrase in DNA unlinking during replication in Escherichia coli. Genes Dev 9: 2859–2869.
- 9. Kato J, Nishimura Y, Imamura R, Niki H, Hiraga S, et al. (1990) New topoisomerase essential for chromosome segregation in E. coli. Cell 63: 393–404.
- 10. Ip SC, Bregu M, Barre FX, Sherratt DJ (2003) Decatenation of DNA circles by FtsK-dependent Xer site-specific recombination. EMBO J 22: 6399–6407.
- 11. Espeli O, Lee C, Marians KJ (2003) A physical and functional interaction between Escherichia coli FtsK and topoisomerase IV. J Biol Chem 278: 44639–44644.
- 12. Carnoy C, Floquet S, Marceau M, Sebbane F, Haentjens-Herwegh S, et al. (2002) The superantigen gene ypm is located in an unstable chromosomal locus of Yersinia pseudotuberculosis. J Bacteriol 184: 4489–4499.
- 13. Kono N, Arakawa K, Tomita M (2011) Comprehensive prediction of chromosome dimer resolution sites in bacterial genomes. BMC Genomics 12: 19.
- 14. Carnoy C, Roten CA (2009) The dif/Xer recombination systems in proteobacteria. PLoS One 4: e6531.
- 15. Le Bourgeois P, Bugarel M, Campo N, Daveran-Mingot ML, Labonte J, et al. (2007) The unconventional Xer recombination machinery of Streptococci/Lactococci. PLoS Genet 3: e117.
- 16. Sciochetti SA, Piggot PJ, Blakely GW (2001) Identification and characterization of the dif Site from Bacillus subtilis. J Bacteriol 183: 1058–1068.
- 17. Cortez D, Quevillon-Cheruel S, Gribaldo S, Desnoues N, Sezonov G, et al. (2010) Evidence for a Xer/dif system for chromosome resolution in archaea. PLoS Genet 6: e1001166.
- 18. Duggin IG, Dubarry N, Bell SD (2011) Replication termination and chromosome dimer resolution in the archaeon Sulfolobus solfataricus. EMBO J 30: 145–153.
- 19. Ambur OH, Davidsen T, Frye SA, Balasingham SV, Lagesen K, et al. (2009) Genome dynamics in major bacterial pathogens. FEMS Microbiol Rev 33: 453–470.
- 20. Fischer W, Windhager L, Rohrer S, Zeiller M, Karnholz A, et al. (2010) Strain-specific genes of Helicobacter pylori: genome evolution driven by a novel type IV secretion system and genomic island transfer. Nucleic Acids Res.
- 21. Kersulyte D, Lee W, Subramaniam D, Anant S, Herrera P, et al. (2009) Helicobacter Pylori's plasticity zones are novel transposable elements. PLoS One 4: e6859.
- 22. Dailidiene D, Dailide G, Kersulyte D, Berg DE (2006) Contraselectable streptomycin susceptibility determinant for genetic manipulation and analysis of Helicobacter pylori. Appl Environ Microbiol 72: 5908–5914.
- 23. Dumrese C, Slomianka L, Ziegler U, Choi SS, Kalia A, et al. (2009) The secreted Helicobacter cysteine-rich protein A causes adherence of human monocytes and differentiation into a macrophage-like phenotype. FEBS Lett 583: 1637–1643.
- 24. Bloor AE, Cranenburgh RM (2006) An efficient method of selectable marker gene excision by Xer recombination for gene replacement in bacterial chromosomes. Appl Environ Microbiol 72: 2520–2525.
- 25. Styer CM, Hansen LM, Cooke CL, Gundersen AM, Choi SS, et al. (2010) Expression of the BabA adhesin during experimental infection with Helicobacter pylori. Infect Immun 78: 1593–1600.
- 26. Langford ML, Zabaleta J, Ochoa AC, Testerman TL, McGee DJ (2006) In vitro and in vivo complementation of the Helicobacter pylori arginase mutant using an intergenic chromosomal site. Helicobacter 11: 477–493.
- 27. Benghezal M, Dieye Y, Schwan C, Sehnal M, Fulurija A, et al. (2009) Synthetic operon. Australian Patent Application No 2009902990:
- 28. Shevchuk NA, Bryksin AV, Nusinovich YA, Cabello FC, Sutherland M, et al. (2004) Construction of long DNA molecules using long PCR-based fusion of several fragments simultaneously. Nucleic Acids Res 32: e19.
- 29. Schoep TD, Fulurija A, Good F, Lu W, Himbeck RP, et al. (2010) Surface properties of Helicobacter pylori urease complex are essential for persistence. PLoS One 5: e15042.
- 30. Takeshita S, Sato M, Toba M, Masahashi W, Hashimoto-Gotoh T (1987) High-copy-number and low-copy-number plasmid vectors for lacZ alpha-complementation and chloramphenicol- or kanamycin-resistance selection. Gene 61: 63–74.
- 31. Zakrzewska-Czerwinska J, Jakimowicz D, Zawilak-Pawlik A, Messer W (2007) Regulation of the initiation of chromosomal replication in bacteria. FEMS Microbiol Rev 31: 378–387.
- 32. Olczak AA, Olson JW, Maier RJ (2002) Oxidative-stress resistance mutants of Helicobacter pylori. J Bacteriol 184: 3186–3193.
- 33. Tamura K, Dudley J, Nei M, Kumar S (2007) MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) software version 4.0. Mol Biol Evol 24: 1596–1599.
- 34. Saitou N, Nei M (1987) The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4: 406–425.
- 35. Wang L, Lutkenhaus J (1998) FtsK is an essential cell division protein that is localized to the septum and induced as part of the SOS response. Mol Microbiol 29: 731–740.
- 36. Michel B, Recchia GD, Penel-Colin M, Ehrlich SD, Sherratt DJ (2000) Resolution of holliday junctions by RuvABC prevents dimer formation in rep mutants and UV-irradiated cells. Mol Microbiol 37: 180–191.
- 37. Kang J, Blaser MJ (2008) Repair and antirepair DNA helicases in Helicobacter pylori. J Bacteriol 190: 4218–4224.
- 38. Chen CR, Malik M, Snyder M, Drlica K (1996) DNA gyrase and topoisomerase IV on the bacterial chromosome: quinolone-induced DNA cleavage. J Mol Biol 258: 627–637.
- 39. Zawilak-Pawlik A, Kois A, Stingl K, Boneca IG, Skrobuk P, et al. (2007) HobA – a novel protein involved in initiation of chromosomal replication in Helicobacter pylori. Mol Microbiol 65: 979–994.
- 40. Kuempel PL, Henson JM, Dircks L, Tecklenburg M, Lim DF (1991) dif, a recA-independent recombination site in the terminus region of the chromosome of Escherichia coli. New Biol 3: 799–811.
- 41. Nolivos S, Pages C, Rousseau P, Le Bourgeois P, Cornet F (2010) Are two better than one? Analysis of an FtsK/Xer recombination system that uses a single recombinase. Nucleic Acids Res 38: 6477–6489.
- 42. Recchia GD, Aroyo M, Wolf D, Blakely G, Sherratt DJ (1999) FtsK-dependent and -independent pathways of Xer site-specific recombination. EMBO J 18: 5724–5734.
- 43. Val ME, Kennedy SP, El Karoui M, Bonne L, Chevalier F, et al. (2008) FtsK-dependent dimer resolution on multiple chromosomes in the pathogen Vibrio cholerae. PLoS Genet 4: e1000201.
- 44. Yates J, Aroyo M, Sherratt DJ, Barre FX (2003) Species specificity in the activation of Xer recombination at dif by FtsK. Mol Microbiol 49: 241–249.
- 45. Blakely GW, Davidson AO, Sherratt DJ (1997) Binding and cleavage of nicked substrates by site-specific recombinases XerC and XerD. J Mol Biol 265: 30–39.
- 46. Manjunatha UH, Dalal M, Chatterji M, Radha DR, Visweswariah SS, et al. (2002) Functional characterisation of mycobacterial DNA gyrase: an efficient decatenase. Nucleic Acids Res 30: 2144–2153.
- 47. Aubry A, Fisher LM, Jarlier V, Cambau E (2006) First functional characterization of a singly expressed bacterial type II topoisomerase: the enzyme from Mycobacterium tuberculosis. Biochem Biophys Res Commun 348: 158–165.
- 48. Nurse P, Levine C, Hassing H, Marians KJ (2003) Topoisomerase III can serve as the cellular decatenase in Escherichia coli. J Biol Chem 278: 8653–8660.
- 49. Bigot S, Saleh OA, Lesterlin C, Pages C, El Karoui M, et al. (2005) KOPS: DNA motifs that control E. coli chromosome segregation by orienting the FtsK translocase. EMBO J 24: 3770–3780.