Multilocus Sequence Typing of Escherichia coli O26:H11 Isolates Carrying stx in Canada Does Not Identify Genetic Diversity
http://www.100md.com
微生物临床杂志 2005年第10期
Emerging Bacterial Pathogens Program, National Microbiology Laboratory
Cadham Provincial Laboratory, Winnipeg, Manitoba
British Columbia Centre for Disease Control, Vancouver, British Columbia
Provincial Laboratory, Saskatchewan Health, Saskatoon, Saskatchewan
Provincial Laboratory for Public Health, Edmonton, Alberta, Canada
ABSTRACT
Multilocus sequence typing of 31 stx-carrying Escherichia coli O26:H11 strains isolated in Canada between 1999 and 2003 revealed a high degree of genetic relatedness at 10 loci, suggesting either that this is a clonal serotype (similar to O157:H7) or that additional genetic loci need to be examined.
TEXT
Shiga toxin-producing Escherichia coli (STEC) serotype O26:H11 is the second most common serotype of STEC identified in Canada, behind O157:H7 (12). STEC O26:H11 naturally occurs in cattle, diarrheic calves, and feedlots and has been reported globally in countries including, but not limited to, Japan, England, Argentina, and Canada (2, 5, 7, 9, 11). In humans, STEC O26:H11 causes sporadic cases and outbreaks of diarrhea, hemorrhagic colitis, and hemolytic-uremic syndrome; and the incidence of disease resulting from this serotype may outnumber that resulting from O157:H7 in particular regions (3, 10). The discrepancy between incidence and clinical reporting is partly due to a lack of accessible and standardized methods for differentiation of this pathogen. Nucleotide sequence data for this serotype are limited in comparison to those for O157:H7; however, the virulence determinants which have been consistently observed in O26:H11 isolates include the prophage-encoded stx1 and stx2 Shiga toxin genes, the hlyA-encoded hemolysin, high-molecular-weight plasmids, and the locus for enterocyte effacement pathogenicity island, which encodes a type III secretion system and the eae-1 intimin gene (3, 14). Multilocus sequencing typing (MLST) has previously been performed with strains of the O157:H7 and O78 serotypes (1, 8), and it was our objective to develop a MLST scheme to determine the genetic relatedness of stx1-carrying E. coli O26:H11 and potentially reveal epidemiological relationships.
The National Microbiology Laboratory (NML) in Winnipeg, Manitoba, Canada, received isolates of E. coli O26:H11 from provincial public health laboratories in Alberta, British Columbia, Saskatchewan, Manitoba, and Nova Scotia, a collection that represents a latitudinal cross-section of Canadian regions (Table 1). We sought to develop an alternative to pulsed-field gel electrophoresis (PFGE) in order to reveal epidemiological relationships, and MLST was a candidate methodology because of the ease of standardization and its potentially robust discriminatory power. Sequence analysis of the mdh, gnd, gcl, ppk, metA, fliC, ftsZ, relA, recN, and metG genes from strains isolated from 1999 to 2003 in the different Canadian provinces was completed. Included in this study were two non-H11 O26 isolates to identify serotype-specific genetic relatedness. Additionally, the sole stx1- and stx2-carrying O26:H11 isolate reported to NML was used to identify the genetic relatedness to stx1-carrying strains at the loci examined. The PFGE patterns obtained with XbaI digestion were also determined for each strain.
The oligonucleotide primers used to amplify mdh (580 bp), gnd (590 bp), gcl (758 bp), ppk (758 bp), and metA (601 bp) were described previously (1, 8). The ftsZ, relA, recN, and metG genes are identified as core bacterial genetic determinants (6) or are encoded in all E. coli strains, and each of the genes was examined for sequence polymorphisms. To identify target sites within the latter loci, multiple-sequence alignments were generated by using sequence data for E. coli O157:H7, CFT073, and K-12 (GenBank accession numbers NC_000913, NC_002655, and NC_004431, respectively) and ClustalW software (data not shown; http://www.ebi.ac.uk/clustalw/). Regions that demonstrated variability among these reference E. coli strains were selected; and primers designed for conserved regions surrounding these target sites were used for amplification and sequencing: ftsZ (450 bp), primers GIL213 (5'-GATCACTGAACTGTCCAAGCATG) and GIL214 (5'-TCAAGAGAAGTACCGATAACCAC), relA (470 bp), primers GIL215 (5' TCTGTTTCCTCCGAACAGGTCG) and GIL216 (5'-ACAATACGTACCGCACGCACATC), recN (467 bp), primers GIL217 (5'-ATTGCCACAGTATTACCAGTCGC) and GIL218 (5'-AGCAGTTTGCCGACAACTGC), and metG (503 bp), primers GIL219 (5'-TGGCTGACCCGCAGTTGTAC) and GIL220 (5'-GGTCAACTTTGGCGAAGTCGTC).
An additional target that we developed was the H11-specific region of the fliC gene. In E. coli, this locus contains conserved 5' and 3' regions and the interceding region is characteristic for individual H serotypes. The primers used to amplify the variable region of fliC (coordinates 361 to 1373; primers GIL211 [5'-GAMGAAATTGACCGYGTATCTG] and GIL212 [5'-ATGTTRGACACTTCGGTCGC]) were designed according to the fliC sequence of E. coli O157:H11 (GenBank accession number AY337472), with degenerate bases included to allow amplification of H19-encoding fliC (GenBank accession number AY337479); and these oligonucleotides, in addition to nested primers (GIL231 [5'-ACAGGGATTGATGCTACAGCAC] and GIL232 [5'-TTGTCGTATAAAGGCCAGCGC]), were used to sequence this subsection of fliC. The sequence data at individual loci were compared by using ClustalW, and any loci demonstrating variability from the consensus sequence developed were reamplified and sequenced to confirm genetic divergence. Allele sequence types were assigned in the order in which they were discovered (Table 1). Confirmation of the O and H serotypes was completed with antisera prepared at the National Microbiology Laboratory, and identification of the stx toxin genotype and phenotype was determined by PCR and cytotoxicity assays.
After sequence analysis of the first five test loci (mdh, gnd, gcl, ppk, and metA) in 30 stx1-carrying O26:H11 strains, only three single-nucleotide point mutations were identified; and only single-locus variants were observed (Table 1). The ppk locus of strain 03-4186 encoded a nonsynonymous GCG to ACG mutation (where the substituted bases are underlined), which resulted in the Ala271Thr substitution (the coordinates are in relation to those of O157:H7-encoded ppk), whereas the mdh locus of strain 01-6372 encoded a synonymous GTT-to-GTC mutation at Val93 and the mdh locus of strain 99-4610 encoded a synonymous CTG-to-TTG mutation at Leu186. The stx1- and stx2-carrying isolate 02-6737 had sequences identical to those of the predominant sequence type (sequence type 1) of the stx1-carrying strains. For each of the O26:H6 and O26:H32 isolates, sequences that diverged from the O26:H11 consensus sequence were observed at all five loci except gnd of strain 99-4328.
A supplementary set of five loci (fliC, ftsZ, relA, recN, and metG) was then examined for a subset of the O26 isolates (Table 1) in an attempt to identify additional genetic diversity. For all O26:H11 isolates examined (both stx1-carrying isolates and stx1- and stx2-carrying isolates), these targets each had identical sequences; and again, the O26:H6 and O26:H32 strains each had distinct sequence types (Table 1). The exception was the recN locus, which had an identical sequence type in all serotypes examined. Notably, no diversity was observed at the fliC locus; and strain 03-2830, which was typed as nonmotile for the H antigen, also contained the same sequence at the fliC locus as those strains serotyped as H11 (Table 1). To conduct a phylogenetic analysis between the six O26 multilocus sequence types observed in this study and with these same loci encoded by E. coli K-12, enterohemorrhagic E. coli O157:H7, and uropathogenic CFT073, we concatenated individual sequences for metA, mdh, gnd, gcl, ppk, ftsZ, relA, and metG to generate an artificial sequence representing each O26 sequence type and reference strain. These sequences were then aligned, and distance scores were generated with the ClustalW program to perform neighbor-joining clustering (Fig. 1). All four of the O26:H11 sequence types were present on a separate node, whereas O26:H6 was more similar to E. coli CFT073 and O26:H32 clustered with E. coli K-12.
PFGE of XbaI-digested samples was completed for each isolate; and cluster analysis (Fig. 2) confirmed the phylogenetic structure of the O26 isolates, as revealed by MLST, and identified additional, but modest genetic diversity within these strains. Both non-H11 strains (O26:H6 and O26:H32) clustered outside of the O26:H11 patterns, and only two identical patterns were observed within the O26:H11 cluster (strains 03-3099 and 03-3971, both of which were isolated in Alberta). This node also contained two other Albertan strains isolated in 2003. The majority of the O26:H11 strains in this study were isolated in Alberta in 2003; but these did not form a single cluster, nor did any other subgroup within the O26:H11 node correlate to the sequence types, the province of isolation, or the year of isolation.
Based upon the DNA sequences of the 10 loci examined, MLST did not reveal significant genetic diversity in the O26:H11 isolates, suggesting either that this is a clonal serotype or that additional genetic loci need to be examined. Similar results were previously observed after MLST of E. coli O157:H7, where Noller and colleagues examined seven housekeeping genes (and accumulated 311,000 total bases of sequence) without identifying a single mutation (8). The apparent clonality of the O26:H11 isolates may have resulted from the singular source of the strains (clinical samples from humans presenting with characteristic disease symptoms), whereas previous MLST analysis of O78 isolates revealed genetic diversity among strains isolated from different zoonotic origins and of different virulence phenotypes (1).
Of the individual loci that were analyzed, mdh had the most genetic variability. The majority of the loci (metA, gnd, gcl, fliC, ftsZ, relA, metG, and recN) were identical in all O26:H11 isolates. The complete homogeneity of the O26:H11 fliC locus has previously been demonstrated by PCR-restriction fragment length polymorphism analysis (13). Our sequencing typing method did, however, differentiate O26:H11 from other clones of E. coli, including the reference K-12, O157:H7, and CFT073 strains, as well as stx-negative O26:H6 and O26:H32 isolates. PFGE identified additional genetic diversity with the O26:H11 isolates sampled, indicating that the strains sampled are not truly clonal; however, no distinct epidemiological relationships were apparent in the PFGE data for the O26:H11 isolates.
The ideal molecular typing strategy would provide data that most accurately represent the genetic content of a strain and allow comparisons between strains to determine genetic variability and evolutionary lineages. There currently is no method for the quick or economical determination of such traits at the whole-genome level for individual strains; but MLST, even though it samples a very small proportion of the total genetic content, is hypothesized to provide such a representation by examining genetic loci that accumulate, transmit, and conserve mutations at a moderate level (4). Sequenced-based methods such as MLST have the advantage of determining exactly what contributes to the variability examined, but the rate of variability among natural populations (at the loci examined) may be insufficient to determine overall genetic relatedness, resulting in conclusions of clonality. Alternatively, the rate of variability may be too high and epidemiological relationships may be impossible to determine, even if they, in fact, exist. The MLST scheme described here for E. coli O26:H11 was unable to determine significant genetic variability in Canadian isolates, although PFGE distinctly indicated that such variability existed at the whole-genome level. Alternatively, with the genetic differentiation between STEC serotypes at individual loci revealed here (i.e., serotype-specific polymorphisms), MLST data have potential for use in the development of molecular serotyping tools for E. coli.
Nucleotide sequence accession numbers. The sequence data from this study were deposited in GenBank under accession numbers AY973395 to AY973421.
ACKNOWLEDGMENTS
Our gratitude goes to the DNA Core Facility, the Enteric Serotyping and Identification Unit, and the Molecular Typing Unit at the National Microbiology Laboratory for the synthesis of oligonucleotides and for performing DNA sequencing, bacterial serology, and cytotoxicity assays. We also thank the associated provincial health laboratories for providing strains.
This project was funded by the Office of Biotechnology and Science.
REFERENCES
Adiri, R. S., U. Gophna, and E. Z. Ron. 2003. Multilocus sequence typing (MLST) of Escherichia coli O78 strains. FEMS Microbiol. Lett. 222:199-203.
Aktan, I., K. A. Sprigings, R. M. La Ragione, L. M. Faulkner, G. A. Paiba, and M. J. Woodward. 2004. Characterisation of attaching-effacing Escherichia coli isolated from animals at slaughter in England and Wales. Vet. Microbiol. 102:43-53.
Blanco, J. E., M. Blanco, M. P. Alonso, A. Mora, G. Dahbi, M. A. Coira, and J. Blanco. 2004. Serotypes, virulence genes, and intimin types of Shiga toxin (verotoxin)-producing Escherichia coli isolates from human patients: prevalence in Lugo, Spain, from 1992 through 1999. J. Clin. Microbiol. 42:311-319.
Cooper, J. E., and E. J. Feil. 2004. Multilocus sequence typing—what is resolved Trends Microbiol. 12:373-377.
Fukushima, H., and R. Seki. 2004. High numbers of Shiga toxin-producing Escherichia coli found in bovine faeces collected at slaughter in Japan. FEMS Microbiol. Lett. 238:189-197.
Gil, R., F. J. Silva, J. Pereto, and A. Moya. 2004. Determination of the core of a minimal bacterial gene set. Microbiol. Mol. Biol. Rev. 68:518-537.
Mercado, E. C., A. Gioffre, S. M. Rodriguez, A. Cataldi, K. Irino, A. M. Elizondo, A. L. Cipolla, M. I. Romano, R. Malena, and M. A. Mendez. 2004. Non-O157 Shiga toxin-producing Escherichia coli isolated from diarrhoeic calves in Argentina. J. Vet. Med. B Infect. Dis. Vet. Public Health 51:82-88.
Noller, A. C., M. C. McEllistrem, O. C. Stine, J. G. Morris, Jr., D. J. Boxrud, B. Dixon, and L. H. Harrison. 2003. Multilocus sequence typing reveals a lack of diversity among Escherichia coli O157:H7 isolates that are distinct by pulsed-field gel electrophoresis. J. Clin. Microbiol. 41:675-679.
Renter, D. G., S. L. Checkley, J. Campbell, and R. King. 2004. Shiga toxin-producing Escherichia coli in the feces of Alberta feedlot cattle. Can. J. Vet. Res. 68:150-153.
Werber, D., A. Fruth, A. Liesegang, M. Littmann, U. Buchholz, R. Prager, H. Karch, T. Breuer, H. Tschape, and A. Ammon. 2002. A multistate outbreak of Shiga toxin-producing Escherichia coli O26:H11 infections in Germany, detected by molecular subtyping surveillance. J. Infect. Dis. 186:419-422.
Widiasih, D. A., N. Ido, K. Omoe, S. Sugii, and K. Shinagawa. 2004. Duration and magnitude of faecal shedding of Shiga toxin-producing Escherichia coli from naturally infected cattle. Epidemiol. Infect. 132:67-75.
Woodward, D. L., C. G. Clark, R. A. Caldeira, R. Ahmed, and F. G. Rodgers. 2002. Verotoxigenic Escherichia coli (VTEC): a major public health threat in Canada. Can. J. Infect. Dis. 13:321-330.
Zhang, W. L., M. Bielaszewska, J. Bockemuhl, H. Schmidt, F. Scheutz, and H. Karch. 2000. Molecular analysis of H antigens reveals that human diarrheagenic Escherichia coli O26 strains that carry the eae gene belong to the H11 clonal complex. J. Clin. Microbiol. 38:2989-2993.
Zhang, W. L., M. Bielaszewska, A. Liesegang, H. Tschape, H. Schmidt, M. Bitzan, and H. Karch. 2000. Molecular characteristics and epidemiological significance of Shiga toxin-producing Escherichia coli O26 strains. J. Clin. Microbiol. 38:2134-2140.(Matthew W. Gilmour, Tyler)
Cadham Provincial Laboratory, Winnipeg, Manitoba
British Columbia Centre for Disease Control, Vancouver, British Columbia
Provincial Laboratory, Saskatchewan Health, Saskatoon, Saskatchewan
Provincial Laboratory for Public Health, Edmonton, Alberta, Canada
ABSTRACT
Multilocus sequence typing of 31 stx-carrying Escherichia coli O26:H11 strains isolated in Canada between 1999 and 2003 revealed a high degree of genetic relatedness at 10 loci, suggesting either that this is a clonal serotype (similar to O157:H7) or that additional genetic loci need to be examined.
TEXT
Shiga toxin-producing Escherichia coli (STEC) serotype O26:H11 is the second most common serotype of STEC identified in Canada, behind O157:H7 (12). STEC O26:H11 naturally occurs in cattle, diarrheic calves, and feedlots and has been reported globally in countries including, but not limited to, Japan, England, Argentina, and Canada (2, 5, 7, 9, 11). In humans, STEC O26:H11 causes sporadic cases and outbreaks of diarrhea, hemorrhagic colitis, and hemolytic-uremic syndrome; and the incidence of disease resulting from this serotype may outnumber that resulting from O157:H7 in particular regions (3, 10). The discrepancy between incidence and clinical reporting is partly due to a lack of accessible and standardized methods for differentiation of this pathogen. Nucleotide sequence data for this serotype are limited in comparison to those for O157:H7; however, the virulence determinants which have been consistently observed in O26:H11 isolates include the prophage-encoded stx1 and stx2 Shiga toxin genes, the hlyA-encoded hemolysin, high-molecular-weight plasmids, and the locus for enterocyte effacement pathogenicity island, which encodes a type III secretion system and the eae-1 intimin gene (3, 14). Multilocus sequencing typing (MLST) has previously been performed with strains of the O157:H7 and O78 serotypes (1, 8), and it was our objective to develop a MLST scheme to determine the genetic relatedness of stx1-carrying E. coli O26:H11 and potentially reveal epidemiological relationships.
The National Microbiology Laboratory (NML) in Winnipeg, Manitoba, Canada, received isolates of E. coli O26:H11 from provincial public health laboratories in Alberta, British Columbia, Saskatchewan, Manitoba, and Nova Scotia, a collection that represents a latitudinal cross-section of Canadian regions (Table 1). We sought to develop an alternative to pulsed-field gel electrophoresis (PFGE) in order to reveal epidemiological relationships, and MLST was a candidate methodology because of the ease of standardization and its potentially robust discriminatory power. Sequence analysis of the mdh, gnd, gcl, ppk, metA, fliC, ftsZ, relA, recN, and metG genes from strains isolated from 1999 to 2003 in the different Canadian provinces was completed. Included in this study were two non-H11 O26 isolates to identify serotype-specific genetic relatedness. Additionally, the sole stx1- and stx2-carrying O26:H11 isolate reported to NML was used to identify the genetic relatedness to stx1-carrying strains at the loci examined. The PFGE patterns obtained with XbaI digestion were also determined for each strain.
The oligonucleotide primers used to amplify mdh (580 bp), gnd (590 bp), gcl (758 bp), ppk (758 bp), and metA (601 bp) were described previously (1, 8). The ftsZ, relA, recN, and metG genes are identified as core bacterial genetic determinants (6) or are encoded in all E. coli strains, and each of the genes was examined for sequence polymorphisms. To identify target sites within the latter loci, multiple-sequence alignments were generated by using sequence data for E. coli O157:H7, CFT073, and K-12 (GenBank accession numbers NC_000913, NC_002655, and NC_004431, respectively) and ClustalW software (data not shown; http://www.ebi.ac.uk/clustalw/). Regions that demonstrated variability among these reference E. coli strains were selected; and primers designed for conserved regions surrounding these target sites were used for amplification and sequencing: ftsZ (450 bp), primers GIL213 (5'-GATCACTGAACTGTCCAAGCATG) and GIL214 (5'-TCAAGAGAAGTACCGATAACCAC), relA (470 bp), primers GIL215 (5' TCTGTTTCCTCCGAACAGGTCG) and GIL216 (5'-ACAATACGTACCGCACGCACATC), recN (467 bp), primers GIL217 (5'-ATTGCCACAGTATTACCAGTCGC) and GIL218 (5'-AGCAGTTTGCCGACAACTGC), and metG (503 bp), primers GIL219 (5'-TGGCTGACCCGCAGTTGTAC) and GIL220 (5'-GGTCAACTTTGGCGAAGTCGTC).
An additional target that we developed was the H11-specific region of the fliC gene. In E. coli, this locus contains conserved 5' and 3' regions and the interceding region is characteristic for individual H serotypes. The primers used to amplify the variable region of fliC (coordinates 361 to 1373; primers GIL211 [5'-GAMGAAATTGACCGYGTATCTG] and GIL212 [5'-ATGTTRGACACTTCGGTCGC]) were designed according to the fliC sequence of E. coli O157:H11 (GenBank accession number AY337472), with degenerate bases included to allow amplification of H19-encoding fliC (GenBank accession number AY337479); and these oligonucleotides, in addition to nested primers (GIL231 [5'-ACAGGGATTGATGCTACAGCAC] and GIL232 [5'-TTGTCGTATAAAGGCCAGCGC]), were used to sequence this subsection of fliC. The sequence data at individual loci were compared by using ClustalW, and any loci demonstrating variability from the consensus sequence developed were reamplified and sequenced to confirm genetic divergence. Allele sequence types were assigned in the order in which they were discovered (Table 1). Confirmation of the O and H serotypes was completed with antisera prepared at the National Microbiology Laboratory, and identification of the stx toxin genotype and phenotype was determined by PCR and cytotoxicity assays.
After sequence analysis of the first five test loci (mdh, gnd, gcl, ppk, and metA) in 30 stx1-carrying O26:H11 strains, only three single-nucleotide point mutations were identified; and only single-locus variants were observed (Table 1). The ppk locus of strain 03-4186 encoded a nonsynonymous GCG to ACG mutation (where the substituted bases are underlined), which resulted in the Ala271Thr substitution (the coordinates are in relation to those of O157:H7-encoded ppk), whereas the mdh locus of strain 01-6372 encoded a synonymous GTT-to-GTC mutation at Val93 and the mdh locus of strain 99-4610 encoded a synonymous CTG-to-TTG mutation at Leu186. The stx1- and stx2-carrying isolate 02-6737 had sequences identical to those of the predominant sequence type (sequence type 1) of the stx1-carrying strains. For each of the O26:H6 and O26:H32 isolates, sequences that diverged from the O26:H11 consensus sequence were observed at all five loci except gnd of strain 99-4328.
A supplementary set of five loci (fliC, ftsZ, relA, recN, and metG) was then examined for a subset of the O26 isolates (Table 1) in an attempt to identify additional genetic diversity. For all O26:H11 isolates examined (both stx1-carrying isolates and stx1- and stx2-carrying isolates), these targets each had identical sequences; and again, the O26:H6 and O26:H32 strains each had distinct sequence types (Table 1). The exception was the recN locus, which had an identical sequence type in all serotypes examined. Notably, no diversity was observed at the fliC locus; and strain 03-2830, which was typed as nonmotile for the H antigen, also contained the same sequence at the fliC locus as those strains serotyped as H11 (Table 1). To conduct a phylogenetic analysis between the six O26 multilocus sequence types observed in this study and with these same loci encoded by E. coli K-12, enterohemorrhagic E. coli O157:H7, and uropathogenic CFT073, we concatenated individual sequences for metA, mdh, gnd, gcl, ppk, ftsZ, relA, and metG to generate an artificial sequence representing each O26 sequence type and reference strain. These sequences were then aligned, and distance scores were generated with the ClustalW program to perform neighbor-joining clustering (Fig. 1). All four of the O26:H11 sequence types were present on a separate node, whereas O26:H6 was more similar to E. coli CFT073 and O26:H32 clustered with E. coli K-12.
PFGE of XbaI-digested samples was completed for each isolate; and cluster analysis (Fig. 2) confirmed the phylogenetic structure of the O26 isolates, as revealed by MLST, and identified additional, but modest genetic diversity within these strains. Both non-H11 strains (O26:H6 and O26:H32) clustered outside of the O26:H11 patterns, and only two identical patterns were observed within the O26:H11 cluster (strains 03-3099 and 03-3971, both of which were isolated in Alberta). This node also contained two other Albertan strains isolated in 2003. The majority of the O26:H11 strains in this study were isolated in Alberta in 2003; but these did not form a single cluster, nor did any other subgroup within the O26:H11 node correlate to the sequence types, the province of isolation, or the year of isolation.
Based upon the DNA sequences of the 10 loci examined, MLST did not reveal significant genetic diversity in the O26:H11 isolates, suggesting either that this is a clonal serotype or that additional genetic loci need to be examined. Similar results were previously observed after MLST of E. coli O157:H7, where Noller and colleagues examined seven housekeeping genes (and accumulated 311,000 total bases of sequence) without identifying a single mutation (8). The apparent clonality of the O26:H11 isolates may have resulted from the singular source of the strains (clinical samples from humans presenting with characteristic disease symptoms), whereas previous MLST analysis of O78 isolates revealed genetic diversity among strains isolated from different zoonotic origins and of different virulence phenotypes (1).
Of the individual loci that were analyzed, mdh had the most genetic variability. The majority of the loci (metA, gnd, gcl, fliC, ftsZ, relA, metG, and recN) were identical in all O26:H11 isolates. The complete homogeneity of the O26:H11 fliC locus has previously been demonstrated by PCR-restriction fragment length polymorphism analysis (13). Our sequencing typing method did, however, differentiate O26:H11 from other clones of E. coli, including the reference K-12, O157:H7, and CFT073 strains, as well as stx-negative O26:H6 and O26:H32 isolates. PFGE identified additional genetic diversity with the O26:H11 isolates sampled, indicating that the strains sampled are not truly clonal; however, no distinct epidemiological relationships were apparent in the PFGE data for the O26:H11 isolates.
The ideal molecular typing strategy would provide data that most accurately represent the genetic content of a strain and allow comparisons between strains to determine genetic variability and evolutionary lineages. There currently is no method for the quick or economical determination of such traits at the whole-genome level for individual strains; but MLST, even though it samples a very small proportion of the total genetic content, is hypothesized to provide such a representation by examining genetic loci that accumulate, transmit, and conserve mutations at a moderate level (4). Sequenced-based methods such as MLST have the advantage of determining exactly what contributes to the variability examined, but the rate of variability among natural populations (at the loci examined) may be insufficient to determine overall genetic relatedness, resulting in conclusions of clonality. Alternatively, the rate of variability may be too high and epidemiological relationships may be impossible to determine, even if they, in fact, exist. The MLST scheme described here for E. coli O26:H11 was unable to determine significant genetic variability in Canadian isolates, although PFGE distinctly indicated that such variability existed at the whole-genome level. Alternatively, with the genetic differentiation between STEC serotypes at individual loci revealed here (i.e., serotype-specific polymorphisms), MLST data have potential for use in the development of molecular serotyping tools for E. coli.
Nucleotide sequence accession numbers. The sequence data from this study were deposited in GenBank under accession numbers AY973395 to AY973421.
ACKNOWLEDGMENTS
Our gratitude goes to the DNA Core Facility, the Enteric Serotyping and Identification Unit, and the Molecular Typing Unit at the National Microbiology Laboratory for the synthesis of oligonucleotides and for performing DNA sequencing, bacterial serology, and cytotoxicity assays. We also thank the associated provincial health laboratories for providing strains.
This project was funded by the Office of Biotechnology and Science.
REFERENCES
Adiri, R. S., U. Gophna, and E. Z. Ron. 2003. Multilocus sequence typing (MLST) of Escherichia coli O78 strains. FEMS Microbiol. Lett. 222:199-203.
Aktan, I., K. A. Sprigings, R. M. La Ragione, L. M. Faulkner, G. A. Paiba, and M. J. Woodward. 2004. Characterisation of attaching-effacing Escherichia coli isolated from animals at slaughter in England and Wales. Vet. Microbiol. 102:43-53.
Blanco, J. E., M. Blanco, M. P. Alonso, A. Mora, G. Dahbi, M. A. Coira, and J. Blanco. 2004. Serotypes, virulence genes, and intimin types of Shiga toxin (verotoxin)-producing Escherichia coli isolates from human patients: prevalence in Lugo, Spain, from 1992 through 1999. J. Clin. Microbiol. 42:311-319.
Cooper, J. E., and E. J. Feil. 2004. Multilocus sequence typing—what is resolved Trends Microbiol. 12:373-377.
Fukushima, H., and R. Seki. 2004. High numbers of Shiga toxin-producing Escherichia coli found in bovine faeces collected at slaughter in Japan. FEMS Microbiol. Lett. 238:189-197.
Gil, R., F. J. Silva, J. Pereto, and A. Moya. 2004. Determination of the core of a minimal bacterial gene set. Microbiol. Mol. Biol. Rev. 68:518-537.
Mercado, E. C., A. Gioffre, S. M. Rodriguez, A. Cataldi, K. Irino, A. M. Elizondo, A. L. Cipolla, M. I. Romano, R. Malena, and M. A. Mendez. 2004. Non-O157 Shiga toxin-producing Escherichia coli isolated from diarrhoeic calves in Argentina. J. Vet. Med. B Infect. Dis. Vet. Public Health 51:82-88.
Noller, A. C., M. C. McEllistrem, O. C. Stine, J. G. Morris, Jr., D. J. Boxrud, B. Dixon, and L. H. Harrison. 2003. Multilocus sequence typing reveals a lack of diversity among Escherichia coli O157:H7 isolates that are distinct by pulsed-field gel electrophoresis. J. Clin. Microbiol. 41:675-679.
Renter, D. G., S. L. Checkley, J. Campbell, and R. King. 2004. Shiga toxin-producing Escherichia coli in the feces of Alberta feedlot cattle. Can. J. Vet. Res. 68:150-153.
Werber, D., A. Fruth, A. Liesegang, M. Littmann, U. Buchholz, R. Prager, H. Karch, T. Breuer, H. Tschape, and A. Ammon. 2002. A multistate outbreak of Shiga toxin-producing Escherichia coli O26:H11 infections in Germany, detected by molecular subtyping surveillance. J. Infect. Dis. 186:419-422.
Widiasih, D. A., N. Ido, K. Omoe, S. Sugii, and K. Shinagawa. 2004. Duration and magnitude of faecal shedding of Shiga toxin-producing Escherichia coli from naturally infected cattle. Epidemiol. Infect. 132:67-75.
Woodward, D. L., C. G. Clark, R. A. Caldeira, R. Ahmed, and F. G. Rodgers. 2002. Verotoxigenic Escherichia coli (VTEC): a major public health threat in Canada. Can. J. Infect. Dis. 13:321-330.
Zhang, W. L., M. Bielaszewska, J. Bockemuhl, H. Schmidt, F. Scheutz, and H. Karch. 2000. Molecular analysis of H antigens reveals that human diarrheagenic Escherichia coli O26 strains that carry the eae gene belong to the H11 clonal complex. J. Clin. Microbiol. 38:2989-2993.
Zhang, W. L., M. Bielaszewska, A. Liesegang, H. Tschape, H. Schmidt, M. Bitzan, and H. Karch. 2000. Molecular characteristics and epidemiological significance of Shiga toxin-producing Escherichia coli O26 strains. J. Clin. Microbiol. 38:2134-2140.(Matthew W. Gilmour, Tyler)