Evolutionary History of hrgA Which Replaces the Restriction Gene hpyIIIR in the hpyIII Locus of Helicobacter pylori
First Department of Internal Medicine, Nagoya University School of Medicine, Nagoya, Japan,1 Departments of Medicine and Microbiology, New York University School of Medicine,2 Department of Veterans Affairs Medical Center, New York, New York,3 Molecular Microbiology and Genomics Consultants, Zotzenheim, Germany4n-zp%, 百拇医药
Received 24 June 2002/ Accepted 1 October 2002n-zp%, 百拇医药
ABSTRACTn-zp%, 百拇医药
A recently identified Helicobacter pylori gene, hrgA, was previously reported to be present in 70 (33%) of 208 strains examined (T. Ando, T. M. Wassenaar, R. M. Peek, R. A. Aras, A. I. Tschumi, L.-J. Van Doorn, K. Kusugami, and M. J. Blaser, Cancer Res. 62:2385-2389, 2002). Sequence analysis of nine such strains indicated that in each strain hrgA replaced hpyIIIR, which encodes a restriction endonuclease and which, together with the gene for its cognate methyltransferase, constitutes the hpyIII locus. As a consequence of either the hrgA insertion or independent mutations, hpyIIIM function was lost in 11 (5%) of the 208 strains examined, rendering chromosomal DNA sensitive to MboI digestion. The evolutionary history of the locus containing either hpyIII or hrgA was reconstructed. By homologous recombination involving flanking sequences, hrgA and hpyIIIR can replace one another in the hpyIII locus, and there is simultaneous replacement of several flanking genes. These findings, combined with the hpyIM/iceA2 locus discovered previously, suggest that the two most strongly conserved methylase genes of H. pylori, hpyIIIM and hpyIM, are both preceded by alternative genes that compete for presence at their loci.
INTRODUCTIONi1x[, http://www.100md.com
Type II restriction-modification (R-M) systems are comprised of paired enzymes, a restriction endonuclease that cleaves DNA within a specific 4- to 8-bp sequence and a methyltransferase that specifically methylates the DNA within the same sequence, protecting the sequence from cleavage (4, 5, 14). Helicobacter pylori, a gram-negative bacterium that colonizes the human stomach, affects the risk of getting upper gastrointestinal tract diseases, including gastric cancer (13). H. pylori strains are highly heterogeneous in terms of the number and nature of the R-M systems that they carry (2, 10, 11, 18, 19, 21, 23). During characterization of the hpyIIIR-hpyIIIM locus in Asian and Western strains, we found numerous strains with a novel gene that we designated hrgA in place of hpyIIIR (encoding an isoschizomer of Moraxella bovis MboI). The presence of hrgA appears to have predictive value for virulence in cagA-positive strains from Asia (3). Neither gene is essential, but since no strain that lacks or contains both genes has been identified thus far, it is hypothesized that there is selection for the presence of either gene. The work described here addressed the following questions. How conserved is hrgA? How did the hrgA-hpyIII locus evolve? And are hrgA and hpyIIIR functional, and can the two genes be exchanged by natural transformation? Mutants bearing antibiotic resistance cassettes were constructed to investigate exchange of the genes between strains by natural transformation. The results indicate that exchange of these genes is possible and may involve transfer of a DNA fragment containing substantial flanking sequences, thus increasing the potential for genomic plasticity in H. pylori.
MATERIALS AND METHODStd, http://www.100md.com
Bacterial strains and growth conditions. The H. pylori strains used in this study are summarized in . Escherichia coli was routinely grown at 37°C in Luria-Bertani broth or agar supplemented with ampicillin (100 µg/ml) and/or chloramphenicol (30 µg/ml), when appropriate. H. pylori strains were grown on Trypticase soy agar (TSA) with 5% sheep blood (BBL) or brucella serum (BBL) agar with 10% newborn calf serum (Intergen) at 37°C in an atmosphere containing 5% CO2.td, http://www.100md.com
fig.ommittedtd, http://www.100md.com
Strains used in this studytd, http://www.100md.com
DNA and protein techniques. Standard molecular techniques were used (15). H. pylori chromosomal DNA was prepared from cells of each strain after 48 h of growth on two agar plates, as described previously (3). Plasmid DNA was prepared from E. coli after overnight culture by using a midi-prep protocol (Qiagen Inc., Valencia, Calif.) according to the manufacturer's instructions. PCRs were performed in 50-µl mixtures containing 0.5 U of Taq polymerase (Qiagen), 1.5 mM MgCl2, and 200 ng of each primer. The PCR protocol (30 cycles) included a denaturation step at 94°C for 1 min, annealing at 5°C below the predicted melting temperature of the primers for 1 min, and extension at 72°C for 1 min/kb of amplification product. The primers reflecting conserved sequences in the hpyIII locus, hpRf, hpRr, hpMf, hpMr, hrgAf, hrgAr, locf, and locr, are described elsewhere (3). Other primers used in this study are listed in . Sequence analysis was performed as described elsewhere (3).
fig.ommittedun(^, http://www.100md.com
PCR primers used in this studyun(^, http://www.100md.com
Phylogenetic analysis and Ka/Ks ratios for hrgA and hpyIIIM. hrgA nucleotide sequences were aligned by using GCG Pileup (Wisconsin Package, version 9.1), and a phylogram was constructed by using Paup 4.0b2 (Sinauer Associates, Sunderland, Mass.) and was displayed by using midpoint rooting with Paup 3.1 (Illinois Natural History Survey, Champaign). To determine the ratios of the rate of nonsynonymous substitution (Ka) to the rate of synonymous substitution (Ks), multiple-sequence alignments were created with ClustalW and were analyzed by using SWAAP 1.0.0 (distributed by D. T. Pride and available at ).un(^, http://www.100md.com
Disruption of hpyIIIR or hrgA in H. pylori strain 26695 or JP26. Insertion of a chloramphenicol resistance gene (cat) in hrgA has been described previously (3). The same procedure was used to introduce cat into hpyIIIR of strain 26695 with primers NthpRf and XhhpMr . The product was cloned into pBluescript by using E. coli DH5{alpha} . A unique EcoRI site was created in hpyIIIR by performing inverse PCR with primers hpRinr and hpRinf. cat was amplified from pBSC103 (22) by using primers that added EcoRI restriction sites, as described previously (3). This cassette was ligated with the inverse PCR product, thereby disrupting hpyIIIR. H. pylori strains 26695, J99, J188, and B146 were transformed to chloramphenicol resistance with this construct to create 26695-hpyIIIR::cat, J99-hpyIIIR::cat, J188-hpyIIIR::cat, and B146-hpyIIIR::cat. An analogous procedure that was used to produce JP26-hrgA::cat (3) also was used to create J54-hrgA::cat and JP2-hrgA::cat. Chromosomal DNA was isolated from the transformants, and insertion of the cat cassette within hpyIIIR or hrgA was confirmed by PCR.
Transformation of strains with chromosomal DNA. H. pylori strains 26695, J99, J188, B146, JP26, J54, and JP2 were examined to determine whether they could be transformed by chromosomal DNA from the H. pylori hpyIIIR::cat or H. pylori hrgA::cat strain. Recipient H. pylori cells were harvested from a 48-h culture on a single agar plate in 1.0 ml of phosphate-buffered saline (PBS) (pH 7.4) and then centrifuged at 5,000 x g for 5 min, and the pellet was resuspended in 300 µl of PBS. Each transformation mixture, consisting of 25 µl of recipient cells and 30 ng of donor DNA, was spotted onto a TSA plate and then incubated overnight at 37°C in the presence of 5% CO2. The transformation mixture was harvested in 1 ml of PBS, and 100-µl aliquots of appropriate serial 10-fold dilutions were inoculated onto brucella serum agar containing chloramphenicol and onto TSA plates. All plates were incubated for 4 days at 37°C in an atmosphere containing 5% CO2. Transformation frequencies were determined by comparing the numbers of transformants and total viable cells.
Nucleotide sequence accession numbers. The GenBank accession numbers of sequences determined in this study are listed in .@23, http://www.100md.com
RESULTS AND DISCUSSION@23, http://www.100md.com
After the initial discovery in strain JP26 of hrgA (3), a gene of unknown function present in 33% of H. pylori strains, the sequence of this gene in eight other strains was determined. In two strains, JP2 (Japan) and 60190 (United States), multiple frameshifts were present in hrgA, which most likely eliminated expression in these strains. The translated hrgA sequences in the remaining seven H. pylori strains include three regions with multiple insertions or deletions (indels), all preserving frame, at positions 40 to 90, positions 255 to 266, and position 360 to the end (the amino acid positions include the gaps introduced in the alignment)@23, http://www.100md.com
fig.ommitted@23, http://www.100md.com
Alignment of deduced amino acid sequences encoded by hrgA from seven H. pylori strains. The JP26 sequence was used as the reference. Conserved amino acids are indicated by dots. Gaps introduced for optimal alignment are indicated by hyphens. The grey bars indicate three regions with frequent indels and substitutions. The stop codons are indicated by asterisks. The numbers at the top include introduced gaps.
Since hrgA was not significantly homologous to functionally annotated genes, nucleotide substitution rates were analyzed to obtain clues related to its function. Nonsynonymous substitutions result in a change in an amino acid in the sequence, so that the amino acid sequence is no longer conserved. Synonymous substitutions do not result in a change in the amino acid. Ka values are generally lower than Ks values; however, (segments of) genes under selective pressure for variation have higher Ka/Ks ratios than genes under selective pressure for conservation (24). For example, for hpyIIIM, the overall Ka is 0.04, the overall Ks is 0.18, and the Ka/Ks ratio is 0.24, values which are typical for H. pylori and suggest selection for a conserved amino acid sequence (1). However, for hrgA, a higher Ka/Ks ratio (0.74, based on a Ka value of 0.17 and a Ks value of 0.23) suggests that there is selective pressure for variation, which would be consistent with a role in virulence (16). When the Ka and Ks of hrgA were compared by using a window of 30 amino acids, a region (amino acid residues 50 to 90) was identified where Ka was higher than Ks. The finding that this region overlaps one of the regions where insertions and deletions are frequent suggests that this portion of the gene product may be exposed to host selection (for example, selection from the immune system). These data, in conjunction with the observed in-frame indels in hrgA, support the hypothesis that although not essential, hrgA is functional and not degenerate.
fig.ommitted0@, 百拇医药
Plot of Ka and Ks in hrgA sequences derived from seven strains, obtained by using sliding windows of 30 amino acids. The grey bars indicate the regions with frequent indels and substitutions identified in .0@, 百拇医药
Although certain indels or substitutions were present in both Japanese and Western strains, the overall phylogeny based on hrgA nucleotide acid sequences separates the Eastern and Western strains Although the number of strains analyzed is not large, the bootstrap values indicate that this separation is significant. To reconstruct the history of the hrgA/hpyIII locus, sequence analyses were extended upstream of hrgA or hpyIII and to the downstream hpyIIIM gene. Thus, the boundaries of the presumed replacement of hpyIIIR by hrgA could be determined, and a model for the history of the hpyIII locus of type I strains (containing hpyIIIR) and type II strains (containing hrgA) (3) could be constructed .0@, 百拇医药
fig.ommitted0@, 百拇医药
Phylogeny of hrgA based on nucleotide sequences. Bar = 10 nucleotide changes. The numbers at the nodes are bootstrap values per 100 runs.;zrz9p, 百拇医药
fig.ommitted;zrz9p, 百拇医药
(A) Reconstruction of the evolution of the hrgA/hpyIIIR locus in H. pylori strains, as deduced from the sequences of 13 strains. (B) Sequences of the regions upstream of hpyIIIR (strains 26695, 99517, 88-29, and J99) or hrgA (strains 99515, JP2, JP28, JP26, B128, 4602, 60190, 9627, and J54). Nucleotides identical to nucleotides in the 26695 sequence are indicated by dots, while gaps introduced for the alignment are indicated by hyphens. The positions where hrgA was introduced, replacing the upstream region of hpyIIIR, are indicated by arrows. Nucleotides identical to nucleotides in the JP2 sequence are indicated by colons. Translational start sites of both genes are underlined. (C) Sequences of the boundaries between hpyIIIR or hrgA and hpyIIIM of the strains shown in panel A. The start codon of hpyIIIM is underlined for each strain. An alternative start codon of hpyIIIM is shown for hrgA-containing strains. The 3' ends of hpyIIIR and hrgA are enclosed in boxes, and each stop codon is indicated by italics. Nucleotides identical to nucleotides in the 26695 sequence are indicated by dots, and nucleotides identical to nucleotides in the JP2 sequence are indicated by colons. Gaps introduced into the alignment are indicated by hyphens. The 3' end of the insertion containing hrgA for nine genes is indicated by an arrow. The insertion end could not be determined for strain 99515.
Type II restriction and modification genes are paired, and whereas cells with a modification gene can survive without the cognate restriction gene, cells with a functional restriction gene cannot survive without an intact and active modification gene. Thus, it must be assumed that hpyIIIR and hpyIIIM were once present together in the H. pylori chromosome and that in certain strains hpyIIIR was subsequently replaced by hrgA. Thus, in the most recent common ancestor of the H. pylori strains studied, an hpyIII R-M system likely was introduced downstream of fabD and tRNA Ser3, resulting in a strain similar to 26695 . Insertion of foreign DNA often occurs at tRNA loci (8). Strains with the insertion appear to have completely replaced the bacterial population lacking this R-M system, since no strains could be detected without the insertion. In several strains, subsequent mutations and deletions have inactivated hpyIIIR by premature termination of the open reading frame, without affecting the hpyIIIM function (for example, this occurred in strains J99 and 99517) (data not shown). Expression of hpyIIIM is indicated by resistance of the chromosomal DNA to MboI digestion (3). Among the hpyIIIR strains tested, only one (strain 88-29) is MboI sensitive , possibly due to a polar effect of a frameshift present in hpyIIIR. We interpret the presence of hrgA upstream of hpyIIIM in 33% of the strains to be the result of horizontal introduction in one or more ancestral strains, whereby hpyIIIR was replaced, after which hrgA spread by horizontal transformation in the H. pylori population.
The introduction of hrgA apparently included 59 to 65 nucleotides upstream of hrgA, which replaced 50 nucleotides upstream of hpyIIIR . The introduction of hrgA also resulted in replacement of approximately 30 nucleotides of hpyIIIM, after which all hpyIIIM sequences examined align. MboI-sensitive strains appear to be more common among hrgA-bearing strains; e.g., subsequent mutations in JP26, JP28, and 9627 resulted in inactivation of hpyIIIM.#, 百拇医药
The 3' insertion boundary is conserved in all strains except 99515 . The latter strain is an exception, in that the first part of hpyIIIM is identical to the sequence in strains bearing hpyIIIR, suggesting that during the hrgA insertion, less of the hpyIIIM sequence was replaced. Within the group of strains having identical hrgA 3' boundaries, one group (comprising JP26, B128, and 4602) also shared the 5' boundary; the other hrgA-containing strains have a different 5' insertion boundary, although the difference is only 12 nucleotides . This grouping does not correlate with geographical origin, since JP26, B128, and 4602 were isolated from patients from Japan, the United States, and Europe, respectively . Taken together, these observations suggest that multiple insertion events involving hrgA occurred before Eastern and Western human populations separated, which allowed conservation of insertion borders in strains from different geographical regions.
H. pylori strains possess either hpyIIIR or hrgA; thus, neither gene is essential. Nevertheless, in not one of the 208 strains tested were both genes missing or were both genes present (at any genomic locus, as detected by gene-specific PCRs) (unpublished results). Thus, the presence of either gene, whether functional or not, must be selected in H. pylori. Since natural competence augments the H. pylori repertoire for creating genomic diversity via homologous recombination (8, 17, 20), hrgA could have spread in the population by horizontal gene transfer after its original introduction into an H. pylori genome. Therefore, we hypothesized that strains can exchange hrgA for hpyIIIR, or vice versa, by natural transformation. This was confirmed experimentally, and a sequence analysis of the transformants obtained was performed to define the boundaries of the presumed replacement. Chromosomal DNA from mutants carrying a chloramphenicol resistance marker (cat) in hrgA (JP26-hrgA::cat) or hpyIIIR (26695-hpyIIIR::cat) was used to transform the reciprocal wild-type strains. The transformation efficiency varied between strains, and JP2 was an inefficient recipient . Transformation to a homologous strain (e.g., hpyIIIR DNA transformed into hpyIIIR-containing strains) was in general more efficient than the heterologous equivalent . However, recipient hrgA+ strains were more efficient in taking up heterologous hpyIII DNA than recipient hpyIIIR strains were in taking up hrgA DNA. The results suggest that replacement of hrgA by hpyIIIR by natural transformation may be more efficient than replacement of hpyIIIR by hrgA.
fig.ommittedih/s+v!, http://www.100md.com
Transformation frequencies of H. pylori strains with chromosomal DNA with a selectable marker in the hpyIIIR/hrgA locusih/s+v!, http://www.100md.com
The crossover locations involved in the homologous recombinations were determined by comparing the sequences of two transformants with the sequences of their parental strains. Transformation of strain 26695 with chromosomal DNA from JP26-hrgA::cat yielded a transformant, 26695::JP26-hrgA::cat, in which the crossovers occurred in the upstream fabD gene and in the downstream HP0093 gene, replacing a 4.0-kb fragment with a 5.2-kb fragment (including the cat cassette). In the inverse experiment, in which DNA from 26695-hpyIIIR::cat was transformed into strain JP26, the transformant studied (JP26::26695-hpyIIIR::cat) received an approximately 5.7-kb fragment (including the cat cassette) with crossovers in fabD and HP0095. These transformation results show that genetic exchange of hrgA and hpyIIIR can occur and that crossover events involve flanking genes, thus preserving the border of the original hrgA introduction. The ability to recombine DNA segments of various lengths around a locus with a (naturally) selectable phenotype is adaptive because it results in a population of variants available for survival under changing selective constraints. This property may in part help generate the enormous diversity present among H. pylori strains (7, 17).
fig.ommitted&30q:k, http://www.100md.com
Schematic diagram of transformation experiments exchanging hrgA and hpyIIIR. The donor DNA was derived from a mutant in which a selectable cat cassette was introduced into hrgA (in strain JP26) (top) or into hpyIIIR (in strain 26695) (bottom). The positions of the crossovers were determined by sequence analysis for each transformant, based on the polymorphisms between JP26 and 26695. The size of the transformed DNA is indicated. Sequences and lengths of the DNA segment with 100% identity around the crossover points are indicated by grey (for 26695 sequences) and black (for JP26 sequences).&30q:k, http://www.100md.com
Variation in the presence and expression of R-M systems or their constituents affects the ability of the host strain to further accept heterologous DNA fragments. After horizontal acquisition of the hpyIII R-M system, which may have occurred more than once in H. pylori (12), H. pylori may have coopted the methylase function, after which it could afford the loss of the hpyIIIR function. A parallel process may have occurred in another R-M locus, with iceA1 (encoding an isoschizomer of NlaIII) replaced by iceA2, in which the methylase gene was retained (6). No sequence similarity was detected, however, between hrgA and iceA2, and the presence of the hrgA gene and the presence of the iceA2 gene seem to be independent events.
Despite extensive diversity in most of the genes encoding methylases present in H. pylori strains (18), hpyIM (in the iceA locus) and hpyIIIM are the two genes that are universally (100%) and nearly universally (97%) expressed, respectively (18). Thus, selection for functional CATG and GATC methylases may be particularly strong and possibly stronger than selection for their functional restriction enzymes.${l, http://www.100md.com
It has been proposed that R-M systems parasitize bacterial genomes and are difficult to lose (9). In contrast, our data show that H. pylori strains are able to exchange hpyIIIR and hrgA by homologous recombination involving flanking sequences, suggesting that these genes can be in flux and that the presence of these genes is subject to selective constraints that remain to be determined.${l, http://www.100md.com
ACKNOWLEDGMENTS${l, http://www.100md.com
This study was supported in part by grant RO1GM63270 from the National Institutes of Health, by the Medical Research Service of the Department of Veterans Affairs, and by the Foundation for Bacteriology.
REFERENCES$df;gm, 百拇医药
Achtman, M., T. Azuma, D. E. Berg, Y. Ito, G. Morelli, Z. J. Pan, S. Suerbaum, S. A. Thompson, A. van der Ende, and L. J. van Doorn. 1999. Recombination and clonal groupings within Helicobacter pylori from different geographical regions. Mol. Microbiol. 32:459-470.$df;gm, 百拇医药
Alm, R. A., and T. J. Trust. 1999. Analysis of the genetic diversity of Helicobacter pylori: the role of two genomes. J. Mol. Med. 77:834-846.$df;gm, 百拇医药
Ando, T., T. M. Wassenaar, R. M. Peek, R. A. Aras, A. I. Tschumi, L.-J. Van Doorn, K. Kusugami, and M. J. Blaser. 2002. A Helicobacter pylori restriction endonuclease replacing gene, hrgA, associated with gastric cancer in Asian strains. Cancer Res. 62:2385-2389.$df;gm, 百拇医药
Bennett, S. P., and S. E. Halford. 1989. Recognition of DNA by type II restriction enzymes. Curr. Top. Cell. Regul. 30:57-104.$df;gm, 百拇医药
Bickle, T. A., and D. H. Kruger. 1993. Biology of DNA restriction. Microbiol. Rev. 57:434-450.$df;gm, 百拇医药
Figueiredo, C., W. G. Quint, R. Sanna, E. Sablon, J. P. Donahue, Q. Xu, G. G. Miller, R. M. Peek, M. J. Blaser, and L. J. van Doorn. 2000. Genetic organization and heterogeneity of the iceA locus of Helicobacter pylori. Gene 246:59-68.
Go, M. F., V. Kapur, D. Y. Graham, and J. M. Musser. 1996. Population genetic analysis of Helicobacter pylori by multilocus enzyme electrophoresis: extensive allelic diversity and recombinational population structure. J. Bacteriol. 178:3934-3938.665a#)?, 百拇医药
Israel, D. A., A. S. Lou, and M. J. Blaser. 2000. Characteristics of Helicobacter pylori natural transformation. FEMS Microbiol. Lett. 186:275-280.665a#)?, 百拇医药
Kobayashi, I., A. Nobusato, N. Kobayashi-Takahashi, and I. Uchiyama. 1999. Shaping the genome-restriction-modification systems as mobile genetic elements. Curr. Opin. Genet. Dev. 9:649-656.665a#)?, 百拇医药
Kong, H., L. F. Lin, N. Porter, S. Stickel, D. Byrd, J. Posfai, and R. J. Roberts. 2000. Functional analysis of putative restriction-modification system genes in the Helicobacter pylori J99 genome. Nucleic Acids Res. 28:3216-3223.665a#)?, 百拇医药
Lin, L. F., J. Posfai, R. J. Roberts, and H. Kong. 2001. Comparative genomics of the restriction-modification systems in Helicobacter pylori. Proc. Natl. Acad. Sci. USA 98:2740-2745.
Nobusato, A., I. Uchiyama, S. Ohashi, and I. Kobayashi. 2000. Insertion with long target duplication: a mechanism for gene mobility suggested from comparison of two related bacterial genomes. Gene 259:99-108.158]s-, 百拇医药
Peek, R. M., and M. J. Blaser. 2002. Helicobacter pylori and gastrointestinal tract adenomcarcinomas. Nat. Rev. Cancer 2:11-18.158]s-, 百拇医药
Pingoud, A., and A. Jeltsch. 1997. Recognition and cleavage of DNA by type-II restriction endonucleases. Eur. J. Biochem 246:1-22.158]s-, 百拇医药
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.158]s-, 百拇医药
Stockbauer, K. E., D. Grigsby, X. Pan, Y. X. Fu, L. M. Mejia, A. Cravioto, and J. M. Musser. 1998. Hypervariability generated by natural selection in an extracellular complement-inhibiting protein of serotype M1 strains of group A Streptococcus. Proc. Natl. Acad. Sci. USA 95:3128-3133.158]s-, 百拇医药
Suerbaum, S., J. Maynard Smith, K. Bapumia, G. Morelli, N. H. Smith, E. Kunstmann, I. Dyrek, and M. Achtman. 1998. Free recombination within Helicobacter pylori. Proc. Natl. Acad. Sci. USA 95:12619-12624.
Takata, T., R. Aras, D. Tavakoli, T. Ando, A. Z. Olivares, and M. J. Blaser. 2002. Phenotypic and genotypic variation in methylases involved in type II restriction-modification systems in Helicobacter pylori. Nucleic Acids Res. 30:2444-2452.0o}2jf., http://www.100md.com
Tomb, J. F., O. White, A. Kerlavage, R. A. Clayton, G. G. Sutton, R. D. Fleischmann, K. A. Ketchum, H. P. Gill, S. Klenk, B. A. Dougherty, et al. 1997. The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388:539-547.0o}2jf., http://www.100md.com
Tsuda, M., M. Karita, and T. Nakazawa. 1993. Genetic transformation in Helicobacter pylori. Microbiol. Immunol. 37:85-89.0o}2jf., http://www.100md.com
Vitkute, J., K. Stankevicius, G. Tamulaitiene, Z. Maneliene, A. Timinskas, D. E. Berg, and A. Janulaitis. 2001. Specificities of eleven different DNA methyltransferases of Helicobacter pylori strain 26695. J. Bacteriol. 183:443-450.0o}2jf., http://www.100md.com
Wang, Y., and D. E. Taylor. 1990. Chloramphenicol resistance in Campylobacter coli. Gene 94:23-28.0o}2jf., http://www.100md.com
Xu, Q., R. D. Morgan, R. J. Roberts, and M. J. Blaser. 2000. Identification of type II restriction and modification systems in Helicobacter pylori reveals their substantial diversity among strains. Proc. Natl. Acad. Sci. USA 97:9671-9676.0o}2jf., http://www.100md.com
Ziheng, Y., R. Nielsen, N. Goldman, and A.-M. Krabbe Pedersen. 2000. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics 155:431-449.(T. Ando R. A. Aras K. Kusugami M. J. Blaser and T. M. Wassenaar)
Received 24 June 2002/ Accepted 1 October 2002n-zp%, 百拇医药
ABSTRACTn-zp%, 百拇医药
A recently identified Helicobacter pylori gene, hrgA, was previously reported to be present in 70 (33%) of 208 strains examined (T. Ando, T. M. Wassenaar, R. M. Peek, R. A. Aras, A. I. Tschumi, L.-J. Van Doorn, K. Kusugami, and M. J. Blaser, Cancer Res. 62:2385-2389, 2002). Sequence analysis of nine such strains indicated that in each strain hrgA replaced hpyIIIR, which encodes a restriction endonuclease and which, together with the gene for its cognate methyltransferase, constitutes the hpyIII locus. As a consequence of either the hrgA insertion or independent mutations, hpyIIIM function was lost in 11 (5%) of the 208 strains examined, rendering chromosomal DNA sensitive to MboI digestion. The evolutionary history of the locus containing either hpyIII or hrgA was reconstructed. By homologous recombination involving flanking sequences, hrgA and hpyIIIR can replace one another in the hpyIII locus, and there is simultaneous replacement of several flanking genes. These findings, combined with the hpyIM/iceA2 locus discovered previously, suggest that the two most strongly conserved methylase genes of H. pylori, hpyIIIM and hpyIM, are both preceded by alternative genes that compete for presence at their loci.
INTRODUCTIONi1x[, http://www.100md.com
Type II restriction-modification (R-M) systems are comprised of paired enzymes, a restriction endonuclease that cleaves DNA within a specific 4- to 8-bp sequence and a methyltransferase that specifically methylates the DNA within the same sequence, protecting the sequence from cleavage (4, 5, 14). Helicobacter pylori, a gram-negative bacterium that colonizes the human stomach, affects the risk of getting upper gastrointestinal tract diseases, including gastric cancer (13). H. pylori strains are highly heterogeneous in terms of the number and nature of the R-M systems that they carry (2, 10, 11, 18, 19, 21, 23). During characterization of the hpyIIIR-hpyIIIM locus in Asian and Western strains, we found numerous strains with a novel gene that we designated hrgA in place of hpyIIIR (encoding an isoschizomer of Moraxella bovis MboI). The presence of hrgA appears to have predictive value for virulence in cagA-positive strains from Asia (3). Neither gene is essential, but since no strain that lacks or contains both genes has been identified thus far, it is hypothesized that there is selection for the presence of either gene. The work described here addressed the following questions. How conserved is hrgA? How did the hrgA-hpyIII locus evolve? And are hrgA and hpyIIIR functional, and can the two genes be exchanged by natural transformation? Mutants bearing antibiotic resistance cassettes were constructed to investigate exchange of the genes between strains by natural transformation. The results indicate that exchange of these genes is possible and may involve transfer of a DNA fragment containing substantial flanking sequences, thus increasing the potential for genomic plasticity in H. pylori.
MATERIALS AND METHODStd, http://www.100md.com
Bacterial strains and growth conditions. The H. pylori strains used in this study are summarized in . Escherichia coli was routinely grown at 37°C in Luria-Bertani broth or agar supplemented with ampicillin (100 µg/ml) and/or chloramphenicol (30 µg/ml), when appropriate. H. pylori strains were grown on Trypticase soy agar (TSA) with 5% sheep blood (BBL) or brucella serum (BBL) agar with 10% newborn calf serum (Intergen) at 37°C in an atmosphere containing 5% CO2.td, http://www.100md.com
fig.ommittedtd, http://www.100md.com
Strains used in this studytd, http://www.100md.com
DNA and protein techniques. Standard molecular techniques were used (15). H. pylori chromosomal DNA was prepared from cells of each strain after 48 h of growth on two agar plates, as described previously (3). Plasmid DNA was prepared from E. coli after overnight culture by using a midi-prep protocol (Qiagen Inc., Valencia, Calif.) according to the manufacturer's instructions. PCRs were performed in 50-µl mixtures containing 0.5 U of Taq polymerase (Qiagen), 1.5 mM MgCl2, and 200 ng of each primer. The PCR protocol (30 cycles) included a denaturation step at 94°C for 1 min, annealing at 5°C below the predicted melting temperature of the primers for 1 min, and extension at 72°C for 1 min/kb of amplification product. The primers reflecting conserved sequences in the hpyIII locus, hpRf, hpRr, hpMf, hpMr, hrgAf, hrgAr, locf, and locr, are described elsewhere (3). Other primers used in this study are listed in . Sequence analysis was performed as described elsewhere (3).
fig.ommittedun(^, http://www.100md.com
PCR primers used in this studyun(^, http://www.100md.com
Phylogenetic analysis and Ka/Ks ratios for hrgA and hpyIIIM. hrgA nucleotide sequences were aligned by using GCG Pileup (Wisconsin Package, version 9.1), and a phylogram was constructed by using Paup 4.0b2 (Sinauer Associates, Sunderland, Mass.) and was displayed by using midpoint rooting with Paup 3.1 (Illinois Natural History Survey, Champaign). To determine the ratios of the rate of nonsynonymous substitution (Ka) to the rate of synonymous substitution (Ks), multiple-sequence alignments were created with ClustalW and were analyzed by using SWAAP 1.0.0 (distributed by D. T. Pride and available at ).un(^, http://www.100md.com
Disruption of hpyIIIR or hrgA in H. pylori strain 26695 or JP26. Insertion of a chloramphenicol resistance gene (cat) in hrgA has been described previously (3). The same procedure was used to introduce cat into hpyIIIR of strain 26695 with primers NthpRf and XhhpMr . The product was cloned into pBluescript by using E. coli DH5{alpha} . A unique EcoRI site was created in hpyIIIR by performing inverse PCR with primers hpRinr and hpRinf. cat was amplified from pBSC103 (22) by using primers that added EcoRI restriction sites, as described previously (3). This cassette was ligated with the inverse PCR product, thereby disrupting hpyIIIR. H. pylori strains 26695, J99, J188, and B146 were transformed to chloramphenicol resistance with this construct to create 26695-hpyIIIR::cat, J99-hpyIIIR::cat, J188-hpyIIIR::cat, and B146-hpyIIIR::cat. An analogous procedure that was used to produce JP26-hrgA::cat (3) also was used to create J54-hrgA::cat and JP2-hrgA::cat. Chromosomal DNA was isolated from the transformants, and insertion of the cat cassette within hpyIIIR or hrgA was confirmed by PCR.
Transformation of strains with chromosomal DNA. H. pylori strains 26695, J99, J188, B146, JP26, J54, and JP2 were examined to determine whether they could be transformed by chromosomal DNA from the H. pylori hpyIIIR::cat or H. pylori hrgA::cat strain. Recipient H. pylori cells were harvested from a 48-h culture on a single agar plate in 1.0 ml of phosphate-buffered saline (PBS) (pH 7.4) and then centrifuged at 5,000 x g for 5 min, and the pellet was resuspended in 300 µl of PBS. Each transformation mixture, consisting of 25 µl of recipient cells and 30 ng of donor DNA, was spotted onto a TSA plate and then incubated overnight at 37°C in the presence of 5% CO2. The transformation mixture was harvested in 1 ml of PBS, and 100-µl aliquots of appropriate serial 10-fold dilutions were inoculated onto brucella serum agar containing chloramphenicol and onto TSA plates. All plates were incubated for 4 days at 37°C in an atmosphere containing 5% CO2. Transformation frequencies were determined by comparing the numbers of transformants and total viable cells.
Nucleotide sequence accession numbers. The GenBank accession numbers of sequences determined in this study are listed in .@23, http://www.100md.com
RESULTS AND DISCUSSION@23, http://www.100md.com
After the initial discovery in strain JP26 of hrgA (3), a gene of unknown function present in 33% of H. pylori strains, the sequence of this gene in eight other strains was determined. In two strains, JP2 (Japan) and 60190 (United States), multiple frameshifts were present in hrgA, which most likely eliminated expression in these strains. The translated hrgA sequences in the remaining seven H. pylori strains include three regions with multiple insertions or deletions (indels), all preserving frame, at positions 40 to 90, positions 255 to 266, and position 360 to the end (the amino acid positions include the gaps introduced in the alignment)@23, http://www.100md.com
fig.ommitted@23, http://www.100md.com
Alignment of deduced amino acid sequences encoded by hrgA from seven H. pylori strains. The JP26 sequence was used as the reference. Conserved amino acids are indicated by dots. Gaps introduced for optimal alignment are indicated by hyphens. The grey bars indicate three regions with frequent indels and substitutions. The stop codons are indicated by asterisks. The numbers at the top include introduced gaps.
Since hrgA was not significantly homologous to functionally annotated genes, nucleotide substitution rates were analyzed to obtain clues related to its function. Nonsynonymous substitutions result in a change in an amino acid in the sequence, so that the amino acid sequence is no longer conserved. Synonymous substitutions do not result in a change in the amino acid. Ka values are generally lower than Ks values; however, (segments of) genes under selective pressure for variation have higher Ka/Ks ratios than genes under selective pressure for conservation (24). For example, for hpyIIIM, the overall Ka is 0.04, the overall Ks is 0.18, and the Ka/Ks ratio is 0.24, values which are typical for H. pylori and suggest selection for a conserved amino acid sequence (1). However, for hrgA, a higher Ka/Ks ratio (0.74, based on a Ka value of 0.17 and a Ks value of 0.23) suggests that there is selective pressure for variation, which would be consistent with a role in virulence (16). When the Ka and Ks of hrgA were compared by using a window of 30 amino acids, a region (amino acid residues 50 to 90) was identified where Ka was higher than Ks. The finding that this region overlaps one of the regions where insertions and deletions are frequent suggests that this portion of the gene product may be exposed to host selection (for example, selection from the immune system). These data, in conjunction with the observed in-frame indels in hrgA, support the hypothesis that although not essential, hrgA is functional and not degenerate.
fig.ommitted0@, 百拇医药
Plot of Ka and Ks in hrgA sequences derived from seven strains, obtained by using sliding windows of 30 amino acids. The grey bars indicate the regions with frequent indels and substitutions identified in .0@, 百拇医药
Although certain indels or substitutions were present in both Japanese and Western strains, the overall phylogeny based on hrgA nucleotide acid sequences separates the Eastern and Western strains Although the number of strains analyzed is not large, the bootstrap values indicate that this separation is significant. To reconstruct the history of the hrgA/hpyIII locus, sequence analyses were extended upstream of hrgA or hpyIII and to the downstream hpyIIIM gene. Thus, the boundaries of the presumed replacement of hpyIIIR by hrgA could be determined, and a model for the history of the hpyIII locus of type I strains (containing hpyIIIR) and type II strains (containing hrgA) (3) could be constructed .0@, 百拇医药
fig.ommitted0@, 百拇医药
Phylogeny of hrgA based on nucleotide sequences. Bar = 10 nucleotide changes. The numbers at the nodes are bootstrap values per 100 runs.;zrz9p, 百拇医药
fig.ommitted;zrz9p, 百拇医药
(A) Reconstruction of the evolution of the hrgA/hpyIIIR locus in H. pylori strains, as deduced from the sequences of 13 strains. (B) Sequences of the regions upstream of hpyIIIR (strains 26695, 99517, 88-29, and J99) or hrgA (strains 99515, JP2, JP28, JP26, B128, 4602, 60190, 9627, and J54). Nucleotides identical to nucleotides in the 26695 sequence are indicated by dots, while gaps introduced for the alignment are indicated by hyphens. The positions where hrgA was introduced, replacing the upstream region of hpyIIIR, are indicated by arrows. Nucleotides identical to nucleotides in the JP2 sequence are indicated by colons. Translational start sites of both genes are underlined. (C) Sequences of the boundaries between hpyIIIR or hrgA and hpyIIIM of the strains shown in panel A. The start codon of hpyIIIM is underlined for each strain. An alternative start codon of hpyIIIM is shown for hrgA-containing strains. The 3' ends of hpyIIIR and hrgA are enclosed in boxes, and each stop codon is indicated by italics. Nucleotides identical to nucleotides in the 26695 sequence are indicated by dots, and nucleotides identical to nucleotides in the JP2 sequence are indicated by colons. Gaps introduced into the alignment are indicated by hyphens. The 3' end of the insertion containing hrgA for nine genes is indicated by an arrow. The insertion end could not be determined for strain 99515.
Type II restriction and modification genes are paired, and whereas cells with a modification gene can survive without the cognate restriction gene, cells with a functional restriction gene cannot survive without an intact and active modification gene. Thus, it must be assumed that hpyIIIR and hpyIIIM were once present together in the H. pylori chromosome and that in certain strains hpyIIIR was subsequently replaced by hrgA. Thus, in the most recent common ancestor of the H. pylori strains studied, an hpyIII R-M system likely was introduced downstream of fabD and tRNA Ser3, resulting in a strain similar to 26695 . Insertion of foreign DNA often occurs at tRNA loci (8). Strains with the insertion appear to have completely replaced the bacterial population lacking this R-M system, since no strains could be detected without the insertion. In several strains, subsequent mutations and deletions have inactivated hpyIIIR by premature termination of the open reading frame, without affecting the hpyIIIM function (for example, this occurred in strains J99 and 99517) (data not shown). Expression of hpyIIIM is indicated by resistance of the chromosomal DNA to MboI digestion (3). Among the hpyIIIR strains tested, only one (strain 88-29) is MboI sensitive , possibly due to a polar effect of a frameshift present in hpyIIIR. We interpret the presence of hrgA upstream of hpyIIIM in 33% of the strains to be the result of horizontal introduction in one or more ancestral strains, whereby hpyIIIR was replaced, after which hrgA spread by horizontal transformation in the H. pylori population.
The introduction of hrgA apparently included 59 to 65 nucleotides upstream of hrgA, which replaced 50 nucleotides upstream of hpyIIIR . The introduction of hrgA also resulted in replacement of approximately 30 nucleotides of hpyIIIM, after which all hpyIIIM sequences examined align. MboI-sensitive strains appear to be more common among hrgA-bearing strains; e.g., subsequent mutations in JP26, JP28, and 9627 resulted in inactivation of hpyIIIM.#, 百拇医药
The 3' insertion boundary is conserved in all strains except 99515 . The latter strain is an exception, in that the first part of hpyIIIM is identical to the sequence in strains bearing hpyIIIR, suggesting that during the hrgA insertion, less of the hpyIIIM sequence was replaced. Within the group of strains having identical hrgA 3' boundaries, one group (comprising JP26, B128, and 4602) also shared the 5' boundary; the other hrgA-containing strains have a different 5' insertion boundary, although the difference is only 12 nucleotides . This grouping does not correlate with geographical origin, since JP26, B128, and 4602 were isolated from patients from Japan, the United States, and Europe, respectively . Taken together, these observations suggest that multiple insertion events involving hrgA occurred before Eastern and Western human populations separated, which allowed conservation of insertion borders in strains from different geographical regions.
H. pylori strains possess either hpyIIIR or hrgA; thus, neither gene is essential. Nevertheless, in not one of the 208 strains tested were both genes missing or were both genes present (at any genomic locus, as detected by gene-specific PCRs) (unpublished results). Thus, the presence of either gene, whether functional or not, must be selected in H. pylori. Since natural competence augments the H. pylori repertoire for creating genomic diversity via homologous recombination (8, 17, 20), hrgA could have spread in the population by horizontal gene transfer after its original introduction into an H. pylori genome. Therefore, we hypothesized that strains can exchange hrgA for hpyIIIR, or vice versa, by natural transformation. This was confirmed experimentally, and a sequence analysis of the transformants obtained was performed to define the boundaries of the presumed replacement. Chromosomal DNA from mutants carrying a chloramphenicol resistance marker (cat) in hrgA (JP26-hrgA::cat) or hpyIIIR (26695-hpyIIIR::cat) was used to transform the reciprocal wild-type strains. The transformation efficiency varied between strains, and JP2 was an inefficient recipient . Transformation to a homologous strain (e.g., hpyIIIR DNA transformed into hpyIIIR-containing strains) was in general more efficient than the heterologous equivalent . However, recipient hrgA+ strains were more efficient in taking up heterologous hpyIII DNA than recipient hpyIIIR strains were in taking up hrgA DNA. The results suggest that replacement of hrgA by hpyIIIR by natural transformation may be more efficient than replacement of hpyIIIR by hrgA.
fig.ommittedih/s+v!, http://www.100md.com
Transformation frequencies of H. pylori strains with chromosomal DNA with a selectable marker in the hpyIIIR/hrgA locusih/s+v!, http://www.100md.com
The crossover locations involved in the homologous recombinations were determined by comparing the sequences of two transformants with the sequences of their parental strains. Transformation of strain 26695 with chromosomal DNA from JP26-hrgA::cat yielded a transformant, 26695::JP26-hrgA::cat, in which the crossovers occurred in the upstream fabD gene and in the downstream HP0093 gene, replacing a 4.0-kb fragment with a 5.2-kb fragment (including the cat cassette). In the inverse experiment, in which DNA from 26695-hpyIIIR::cat was transformed into strain JP26, the transformant studied (JP26::26695-hpyIIIR::cat) received an approximately 5.7-kb fragment (including the cat cassette) with crossovers in fabD and HP0095. These transformation results show that genetic exchange of hrgA and hpyIIIR can occur and that crossover events involve flanking genes, thus preserving the border of the original hrgA introduction. The ability to recombine DNA segments of various lengths around a locus with a (naturally) selectable phenotype is adaptive because it results in a population of variants available for survival under changing selective constraints. This property may in part help generate the enormous diversity present among H. pylori strains (7, 17).
fig.ommitted&30q:k, http://www.100md.com
Schematic diagram of transformation experiments exchanging hrgA and hpyIIIR. The donor DNA was derived from a mutant in which a selectable cat cassette was introduced into hrgA (in strain JP26) (top) or into hpyIIIR (in strain 26695) (bottom). The positions of the crossovers were determined by sequence analysis for each transformant, based on the polymorphisms between JP26 and 26695. The size of the transformed DNA is indicated. Sequences and lengths of the DNA segment with 100% identity around the crossover points are indicated by grey (for 26695 sequences) and black (for JP26 sequences).&30q:k, http://www.100md.com
Variation in the presence and expression of R-M systems or their constituents affects the ability of the host strain to further accept heterologous DNA fragments. After horizontal acquisition of the hpyIII R-M system, which may have occurred more than once in H. pylori (12), H. pylori may have coopted the methylase function, after which it could afford the loss of the hpyIIIR function. A parallel process may have occurred in another R-M locus, with iceA1 (encoding an isoschizomer of NlaIII) replaced by iceA2, in which the methylase gene was retained (6). No sequence similarity was detected, however, between hrgA and iceA2, and the presence of the hrgA gene and the presence of the iceA2 gene seem to be independent events.
Despite extensive diversity in most of the genes encoding methylases present in H. pylori strains (18), hpyIM (in the iceA locus) and hpyIIIM are the two genes that are universally (100%) and nearly universally (97%) expressed, respectively (18). Thus, selection for functional CATG and GATC methylases may be particularly strong and possibly stronger than selection for their functional restriction enzymes.${l, http://www.100md.com
It has been proposed that R-M systems parasitize bacterial genomes and are difficult to lose (9). In contrast, our data show that H. pylori strains are able to exchange hpyIIIR and hrgA by homologous recombination involving flanking sequences, suggesting that these genes can be in flux and that the presence of these genes is subject to selective constraints that remain to be determined.${l, http://www.100md.com
ACKNOWLEDGMENTS${l, http://www.100md.com
This study was supported in part by grant RO1GM63270 from the National Institutes of Health, by the Medical Research Service of the Department of Veterans Affairs, and by the Foundation for Bacteriology.
REFERENCES$df;gm, 百拇医药
Achtman, M., T. Azuma, D. E. Berg, Y. Ito, G. Morelli, Z. J. Pan, S. Suerbaum, S. A. Thompson, A. van der Ende, and L. J. van Doorn. 1999. Recombination and clonal groupings within Helicobacter pylori from different geographical regions. Mol. Microbiol. 32:459-470.$df;gm, 百拇医药
Alm, R. A., and T. J. Trust. 1999. Analysis of the genetic diversity of Helicobacter pylori: the role of two genomes. J. Mol. Med. 77:834-846.$df;gm, 百拇医药
Ando, T., T. M. Wassenaar, R. M. Peek, R. A. Aras, A. I. Tschumi, L.-J. Van Doorn, K. Kusugami, and M. J. Blaser. 2002. A Helicobacter pylori restriction endonuclease replacing gene, hrgA, associated with gastric cancer in Asian strains. Cancer Res. 62:2385-2389.$df;gm, 百拇医药
Bennett, S. P., and S. E. Halford. 1989. Recognition of DNA by type II restriction enzymes. Curr. Top. Cell. Regul. 30:57-104.$df;gm, 百拇医药
Bickle, T. A., and D. H. Kruger. 1993. Biology of DNA restriction. Microbiol. Rev. 57:434-450.$df;gm, 百拇医药
Figueiredo, C., W. G. Quint, R. Sanna, E. Sablon, J. P. Donahue, Q. Xu, G. G. Miller, R. M. Peek, M. J. Blaser, and L. J. van Doorn. 2000. Genetic organization and heterogeneity of the iceA locus of Helicobacter pylori. Gene 246:59-68.
Go, M. F., V. Kapur, D. Y. Graham, and J. M. Musser. 1996. Population genetic analysis of Helicobacter pylori by multilocus enzyme electrophoresis: extensive allelic diversity and recombinational population structure. J. Bacteriol. 178:3934-3938.665a#)?, 百拇医药
Israel, D. A., A. S. Lou, and M. J. Blaser. 2000. Characteristics of Helicobacter pylori natural transformation. FEMS Microbiol. Lett. 186:275-280.665a#)?, 百拇医药
Kobayashi, I., A. Nobusato, N. Kobayashi-Takahashi, and I. Uchiyama. 1999. Shaping the genome-restriction-modification systems as mobile genetic elements. Curr. Opin. Genet. Dev. 9:649-656.665a#)?, 百拇医药
Kong, H., L. F. Lin, N. Porter, S. Stickel, D. Byrd, J. Posfai, and R. J. Roberts. 2000. Functional analysis of putative restriction-modification system genes in the Helicobacter pylori J99 genome. Nucleic Acids Res. 28:3216-3223.665a#)?, 百拇医药
Lin, L. F., J. Posfai, R. J. Roberts, and H. Kong. 2001. Comparative genomics of the restriction-modification systems in Helicobacter pylori. Proc. Natl. Acad. Sci. USA 98:2740-2745.
Nobusato, A., I. Uchiyama, S. Ohashi, and I. Kobayashi. 2000. Insertion with long target duplication: a mechanism for gene mobility suggested from comparison of two related bacterial genomes. Gene 259:99-108.158]s-, 百拇医药
Peek, R. M., and M. J. Blaser. 2002. Helicobacter pylori and gastrointestinal tract adenomcarcinomas. Nat. Rev. Cancer 2:11-18.158]s-, 百拇医药
Pingoud, A., and A. Jeltsch. 1997. Recognition and cleavage of DNA by type-II restriction endonucleases. Eur. J. Biochem 246:1-22.158]s-, 百拇医药
Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.158]s-, 百拇医药
Stockbauer, K. E., D. Grigsby, X. Pan, Y. X. Fu, L. M. Mejia, A. Cravioto, and J. M. Musser. 1998. Hypervariability generated by natural selection in an extracellular complement-inhibiting protein of serotype M1 strains of group A Streptococcus. Proc. Natl. Acad. Sci. USA 95:3128-3133.158]s-, 百拇医药
Suerbaum, S., J. Maynard Smith, K. Bapumia, G. Morelli, N. H. Smith, E. Kunstmann, I. Dyrek, and M. Achtman. 1998. Free recombination within Helicobacter pylori. Proc. Natl. Acad. Sci. USA 95:12619-12624.
Takata, T., R. Aras, D. Tavakoli, T. Ando, A. Z. Olivares, and M. J. Blaser. 2002. Phenotypic and genotypic variation in methylases involved in type II restriction-modification systems in Helicobacter pylori. Nucleic Acids Res. 30:2444-2452.0o}2jf., http://www.100md.com
Tomb, J. F., O. White, A. Kerlavage, R. A. Clayton, G. G. Sutton, R. D. Fleischmann, K. A. Ketchum, H. P. Gill, S. Klenk, B. A. Dougherty, et al. 1997. The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388:539-547.0o}2jf., http://www.100md.com
Tsuda, M., M. Karita, and T. Nakazawa. 1993. Genetic transformation in Helicobacter pylori. Microbiol. Immunol. 37:85-89.0o}2jf., http://www.100md.com
Vitkute, J., K. Stankevicius, G. Tamulaitiene, Z. Maneliene, A. Timinskas, D. E. Berg, and A. Janulaitis. 2001. Specificities of eleven different DNA methyltransferases of Helicobacter pylori strain 26695. J. Bacteriol. 183:443-450.0o}2jf., http://www.100md.com
Wang, Y., and D. E. Taylor. 1990. Chloramphenicol resistance in Campylobacter coli. Gene 94:23-28.0o}2jf., http://www.100md.com
Xu, Q., R. D. Morgan, R. J. Roberts, and M. J. Blaser. 2000. Identification of type II restriction and modification systems in Helicobacter pylori reveals their substantial diversity among strains. Proc. Natl. Acad. Sci. USA 97:9671-9676.0o}2jf., http://www.100md.com
Ziheng, Y., R. Nielsen, N. Goldman, and A.-M. Krabbe Pedersen. 2000. Codon-substitution models for heterogeneous selection pressure at amino acid sites. Genetics 155:431-449.(T. Ando R. A. Aras K. Kusugami M. J. Blaser and T. M. Wassenaar)