Use of the espZ Gene Encoded in the Locus of Enter
http://www.100md.com
微生物临床杂志 2006年第2期
ABSTRACT
Infections with Shiga toxin-producing Escherichia coli (STEC) result in frequent cases of sporadic and outbreak-associated enteric bacterial disease in humans. Classification of STEC is by stx genotype (encoding the Shiga toxins), O and H antigen serotype, and seropathotype (subgroupings based upon the clinical relevance and virulence-related genotypes of individual serotypes). The espZ gene is encoded in the locus of enterocyte effacement (LEE) pathogenicity island responsible for the attaching and effacing (A/E) lesions caused by various E. coli pathogens (but not limited to STEC), and this individual gene (300 bp) has previously been identified as hypervariable among these A/E pathogens. Sequence analysis of the espZ locus encoded by additional STEC serotypes and strains (including O26:H11, O121:H19, O111:NM, O145:NM, O165:H25, O121:NM, O157:NM, O157:H7, and O5:NM) indicated that distinct sequence variants exist which correlate to subgroups among these serotypes. Allelic discrimination at the espZ locus was achieved using Light Upon eXtension real-time PCR and by liquid microsphere suspension arrays. The allele subtype of espZ did not correlate with STEC seropathotype classification; however, a correlation with the allele type of the LEE-encoded intimin (eae) gene was supported, and these sequence variations were conserved among individual serotypes. The study focused on the characterization of three clinically significant seropathotypes of LEE-positive STEC, and we have used the observed genetic variation at a pathogen-specific locus for detection and subtyping of STEC.
INTRODUCTION
Gastrointestinal infection by Shiga toxin-producing Escherichia coli (STEC) is largely due to serotype O157:H7 in North America, but infections with other serotypes also result in human disease (25). In Canada, 48 different serotypes of STEC (as identified on the basis of O-somatic and H-flagellar antigens) have been isolated from humans, and strains of serotypes O26:H11, O121:H19, O103:H2, O145:NM, and O111:NM have represented a large proportion of non-O157 isolates (38). STEC serotypes are classified into five seropathotypes (A through E) based upon both virulence gene content and clinical relevance, among which seropathotype A is solely comprised of O157:H7 and O157:NM strains; infections with these strains can result in serious disease symptoms and lead to outbreaks (15). Infection with STEC can result in diarrhea, hemorrhagic colitis, and hemolytic uremic syndrome, with possible disease outcomes including renal failure, neurological sequelae, and death (17, 33). Seropathotype B includes the non-O157 serotypes identified above, and while not known to cause large disease epidemics as frequently, infection with these pathogens can result in disease symptoms similar to those seen with seropathotype A strains (15). Notably, the differential capabilities for detection of O157 versus non-O157 serotypes in clinical laboratories may introduce reporting biases.
STEC disease manifestation correlates to the carriage of classical bacterial virulence determinants such as toxins and pathogenicity islands. The production of Shiga toxins, encoded by the stx1 and stx2 genes, is responsible for systemic disease symptoms, because necrotic and apoptotic cell death are induced after intracellular translocation (4). Stx1 is nearly identical to the cytotoxin produced by Shigella dysenteriae serotype 1 and is homogenous among E. coli carrying stx1, whereas several variants of stx2 have been identified, and the production of Stx2 is associated with hemolytic uremic syndrome (33). The carriage of stx genes is also variable among STEC serotypes, as the majority of seropathotype A strains encode both loci, whereas strains of the other seropathotypes typically encode a single toxin locus. A large subset of STEC strains (predominantly seropathotypes A and B) are also termed enterohemorrhagic E. coli (EHEC) strains and are partly characterized by attaching and effacing (A/E) lesions that they create on the intestinal epithelium, a histopathology resulting from the presence of the locus of enterocyte effacement (LEE) pathogenicity island (8, 19). While STEC strains are considered to be noninvasive, there is disruption of the brush border microvilli after actin rearrangements within epithelial cells. This process is induced by the LEE-encoded determinants that include 41 coding sequences (the majority of which are organized into five major operons) for type III secretion of effector and receptor proteins, including Tir, which allows the close association between the bacterial and the host epithelial cells via intimin (eae) present on the bacterial cell surface (6, 27).
Among the individual LEE-encoded genes present in different A/E pathotypes, including enteropathogenic E. coli (EPEC) and EHEC, a heterogeneous rate of genetic diversity has been observed (3, 27). It was our goal to select a LEE-encoded gene that would support the detection and subtyping of STEC, relying upon the ubiquity of the LEE among the most significant STEC serotypes and the large degree of genetic diversity observed at individual LEE genes. The loci that demonstrated the highest rates of diversity (measured as ) include those involved in interaction with the host cell (i.e., effectors and receptors), whereas the type III secretion apparatus had a lesser amount of diversity (3). Furthermore, the ratio of nonsynonymous mutation (dN) to synonymous mutation (dS) was markedly higher at those same loci involved in host interactions, and therefore these coding sequences may be undergoing positive selection (3), a process reflective of functional adaptation in different host environments and E. coli genetic backgrounds. The loci with the highest estimates included sepZ, tir, espA, espB, espF, espH, and eae, and some of these have previously been used for molecular subtyping of A/E pathogens (1, 2, 21, 23, 29, 40). Corresponding variability of SepZ protein primary sequences was also observed among EPEC and EHEC strains (9, 14, 27). Notably, the sepZ coding sequence was renamed espZ after determining that the gene product is secreted and translocated to eukaryotic cells (14). The espZ locus was also estimated as having the highest rate (3), and the resultant variability between EspZ proteins likely arose due to functional adaptation to host proteins, which are currently unknown (14). Our study was of the espZ nucleotide sequences encoded in clinically significant STEC. Molecular techniques were developed for the detection and subtyping of STEC at this locus, and phylogenetic analyses were performed to estimate the evolutionary history of the relationship between STEC serotypes and the LEE pathogenicity island.
MATERIALS AND METHODS
Bacterial strains. The panel of STEC strains collected for this study (Table 1) included representative isolates from each of the major serotypes observed in Canada (classified as seropathotype A, B, or C), and among individual serotypes we included strains with different stx genotypes when available. All isolates were from the National Microbiology Laboratory (NML) Bacteriology and Enteric Diseases Program culture collection, which originated from human sources at various Canadian Provincial Health laboratories during 1985 to 2005, and serotype and toxin genotype were confirmed at the NML (Table 1). Existing sepZ/espZ sequence data deposited in GenBank were used to initiate our characterization of these Canadian strains as follows: for O26:H11 (strain 6549), accession number AF035656; for O26:H- (413/89-1), AJ277443; for O157:H7 (EDL933), AAC31516; for O157:H7 (Sakai), BAB37994; for O157:H7 (86-24), AF035655; for O15:H- (RDEC-1), AF035651; for O15:H- (EPEC 83/89), AF453441; for O111:NM (EPEC B171), AF035653; for O103:H2 (RW1374), AJ303141; for O127:H6 (EPEC E2348/69), X94450; and for O55:H7 (EPEC), AF035652.
PCR and sequencing. Template DNA was prepared by centrifuging 1 ml of log-phase cultures grown in brain heart infusion broth, resuspending the pellet in 1 ml of TE buffer (Sigma, St. Louis, MO) (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) and boiling for 10 min. Boiled cells were pelleted, and the supernatant was removed and used as the template in real-time and standard PCRs. For quantitative determination of real-time PCR sensitivity, total genomic DNA was isolated from liquid cultures grown overnight in 8 ml brain heart infusion broth. After centrifugation, the bacterial pellet was resuspended in 2 ml of TE buffer with vortexing. Following the addition of lysozyme (Roche Diagnostics, Indianapolis, IN) (0.5 mg/ml), RNase (Roche Diagnostics) (1.5 μg/ml), and proteinase K (Sigma) (0.12 mg/ml), this mixture was incubated at 37°C for 1 h, and then sodium dodecyl sulfate (SDS) (Ambion, Austin, TX) was added to achieve a concentration of 0.1% (wt/vol) and the mixture was further incubated at 65°C until the suspension cleared. Organic extraction was performed using 15 ml Eppendorf Phase Lock tubes (Hamburg, Germany) with an equal volume of phenol:chloroform:isoamyl alcohol (Invitrogen, Burlington, ON) (25:24:1). Phenol-chloroform-isoamyl alcohol extraction was repeated until the aqueous layer was clear, and after a final extraction with 2 ml chloroform, the aqueous layer was transferred to a new 1.5 ml tube and 0.6 volumes of isopropanol and 0.1 volumes of 3 M sodium acetate (Ambion, pH 5.5) were added and DNA was precipitated at –20°C for 20 min. Following centrifugation, the DNA pellet was washed in 1 ml of 70% ethanol and resuspended in 200 μl of TE buffer. After quantification on a NanoDrop ND-1000 apparatus (NanoDrop Technologies, Rockland DE), the DNA was serially diluted from 33 ng/μl to 0.33 fg/μl.
Standard PCR was performed with Platinum Taq (Invitrogen), following the manufacturer's directions, and with the oligonucleotides described in Table 2. The thermocycling parameters for espZ required an initial denaturation at 94°C for 5 min and 35 cycles of denaturation at 94°C for 30 s, annealing at 50°C for 30 s, and extension at 68°C for 30 s, with a final extension at 68°C for 5 min. Conditions for PCR amplification of stx2 were initial denaturation at 94°C for 5 min and 30 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 45 s, and extension at 68°C for 30 s, with a final extension at 68°C for 7 min. PCR products for espZ and stx2 were purified using a QIAquick PCR purification kit (QIAGEN, Mississauga, ON) and sequenced using the same primers used to generate this template. Sequencing was performed on an ABI3730 apparatus (Applied Biosystems, Foster City, CA). Subtyping of stx2 variants was performed using oligonucleotides described by Wang et al. (36).
LUX (Light Upon eXtension) fluorogenic and unlabeled primer pairs (20) were designed using D-LUX designer software (Invitrogen) by targeting regions characteristic of each espZ allele (Fig. 1A) or for regions conserved in either stx1 or stx2 (see Results). LUX primers are self-quenching oligonucleotide primers that are labeled with a single fluorophore. Upon annealing to a specific DNA sequence, they become dequenched and the fluorescent signal increases. For LUX real-time PCR, Platinum quantitative PCR Supermix UDG (Invitrogen) was used for the amplification mixture with 200 nM of each LUX primer pair (Table 2) and 3 μl of template for a total reaction volume of 25 μl. Real-time PCRs were performed on the Cepheid SmartCycler 2.0 apparatus (Cepheid, Sunnyvale, CA), and samples were amplified after an initial denaturation at 95°C for 3 min and 40 cycles of denaturation at 95°C for 10 s, annealing at 55°C for 15 s, and an extension step at 72°C for 15 s. Fluorescence was detected at the annealing step, and the threshold level was set at 30 fluorescence units. A real-time PCR result was considered positive when the log fluorescent signal exceeded this threshold after background subtraction.
Microsphere liquid-suspension arrays. Allelic discrimination of espZ was achieved after PCR amplification of biotin-labeled espZ target DNA from group A and B STEC seropathotypes (O157:H7, O26:H11, O121:H19, and O111:NM) by use of a GeneAmp 9700 thermocycler (Applied Biosystems) and the thermocycling parameters detailed above. The 100 μl amplification mixture consisted of the following: 10 μl of 10x HiFi buffer (Invitrogen), 2 μl of deoxynucleoside triphosphates (Invitrogen) (10 μM each), 4 μl of MgSO4, 0.4 μl of Platinum Hi Fidelity Taq (Invitrogen), 61.6 μl of molecular biology grade water (Gibco, Grand Island, NY), and 10 μl each of primers GIL245 and GIL246-L (5' biotinylated; Table 2), for a final primer concentration of 1 μM each. The template (2 μl), prepared from a boiled cell resuspension as described above, was added to the reaction mixture. Successful PCR amplification of espZ was confirmed by agarose gel electrophoresis. PCRs were purified with Qiaquick DNA purification kits (QIAGEN) and eluted with 50 μl of EB buffer (QIAGEN). Oligonucleotide GIL246-L contains four bases with phosphorothioate linkages, as well as a biotin molecule, all at the 5' end. Of the two strands of the target DNA, the strand produced from GIL245 (espZ "sense" strand) is sensitive to T7 exonuclease digestion whereas the espZ antisense strand produced from GIL246-L is protected due to the phosphorothioate linkages (22). DNA digestion was performed by mixing 43 μl of purified PCR product with 5 μl of buffer 4 and 2 μl of T7 exonuclease (both from New England Biolabs, Ipswich, MA) (20 U total) and incubating at 37°C for 1 h. T7 exonuclease was inactivated by adding 2 μl of 0.5 M EDTA (Ambion). Selective degradation ensures elimination of the unlabeled target DNA strand, thereby preventing reannealing between the two target DNA strands during hybridization that, if left intact, would limit the intended hybridization between the biotin-labeled strand and the espZ allele-specific probes coupled to microspheres in subsequent steps.
Oligonucleotide probes for each espZ allele were designed matching the sense strand in highly variable regions characteristic of individual allele subtypes (Fig. 1A, Table 2). Oligonucleotides were screened using SBEprimer software (13) for potential secondary structures or cross-hybridization between probes. The oligonucleotide probes were synthesized with a 5' C-12 amine and coupled to xMAP carboxylated fluorescently coded microspheres (Luminex Corporation, Austin, TX). Microsphere sets 103, 105, 108, and 110 were coupled to oligonucleotides DOB70, DOB72, DOB73, and DOB74, respectively (Table 2). Microspheres (5.0 x 106) were transferred to a 1.5 ml microcentrifuge tube, centrifuged at 8,000 x g for 2 min, resuspended in 50 μl of 0.1 M MES buffer [2-(N-morpholino)ethanesulfonic acid; Sigma] (pH 4.5), and vortexed, and 6 μl of capture oligonucleotide (100 μM) was added to the respective bead set. A fresh solution of EDC [1-ethyl-3-(3-dimethylaminopropyl)carbodiimidehydrochloride; Pierce Biotechnology, Rockford, IL] (10 mg/ml) was prepared immediately before use, 2 μl was added to the bead-oligonucleotide mixture, and the mixture was vortexed and incubated at room temperature for 30 min in the dark. The EDC addition and incubation was repeated. Following incubation, beads were washed successively in 1 ml of 0.1% Tween and 1 ml of 0.1% SDS. Microspheres were resuspended in 100 μl of 0.1 M MES (pH 4.5) and enumerated on a hemocytometer.
For hybridization of biotin-labeled espZ target DNA strands to the capture probe-coupled microspheres, a reaction master mix was prepared in TE buffer at a concentration of 150 microspheres/μl for each of the four capture probe sets. Hybridizations were prepared in triplicate under low-light conditions as 50-μl reaction mixtures in Thermowell 96-well plates (Corning Incorporated, Corning, NY). Initially, 17 μl of the biotinylated PCR product was added to wells and denatured for 10 min at 95°C in a GeneAmp 9700 thermocycler, followed by addition of 33 μl of the reaction master mix and mixing by pipetting. Hybridizations were performed at 55°C for 20 min, after which 25 μl of the streptavidin R-phycoerythrin reporter dye (Molecular Probes, Eugene, OR) diluted to 10 μg/ml in 1.5x TMAC buffer (3 M tetramethylammonium chloride [Sigma], 0.1% SDS, 50 mM Tris-HCl [pH 8.0], 4 mM EDTA, pH 8.0) was added and mixed by pipetting. Plates were incubated at 55°C for an additional 10 min. Flow cytometry with a QIAGEN LiquiChip workstation was used to quantify hybridization events, with the following settings: reading gates set at 8,300 to 16,500; minimum of 100 events read; and microplate handler heater block maintained at 55°C during measurements. Mean fluorescence intensity signals for each espZ target DNA sample and the negative control (TE blank) were averaged among the triplicate wells.
Bioinformatics. Multiple sequence alignments were completed using ClustalW (www.ebi.ac.uk/clustalw/) and Boxshade (www.ch.embnet.org), neighbor-joining trees were constructed using MEGA3 (16), and genetic diversity statistics were calculated using DnaSP 4.10.3 (30). Pairwise global alignments were calculated using Align (www.ebi.ac.uk/emboss/align/#). Split decomposition analysis was performed using SplitsTree4 (12) and alignment inputs created by ClustalW, and calculations were performed using only parsimony-informative sites.
Nucleotide sequence accession numbers. The sequences determined in this study (see Results) have been deposited in GenBank under accession numbers DQ138070 to DQ138078 and DQ143180 to DQ143183.
RESULTS
Sequencing of the espZ hypervariable region. Molecular and phylogenetic characterization of the espZ locus was initiated by sequencing the central hypervariable region of this gene from seropathotype group A strains (O157:H7, O157:NM), group B strains (O26:H11, O121:H19, O111:NM, O145:NM) and group C strains which encode the LEE pathogenicity island (O5:NM, O121:NM, O165:NM). Oligonucleotides GIL245 and 246 (Table 2) were designed for the conserved C- and N-terminal encoding regions of espZ, observed in the sequence data from espZ (formerly sepZ) carried by various EPEC and EHEC strains (see Materials and Methods). These primers were able to produce an espZ product for each examined STEC serotype that was predicted to encode the LEE (Table 1), permitting sequencing of the intervening hypervariable region (Fig. 1A). The espZ PCR product was sequenced from four clinical strains of serotype O26:H11, three strains of serotype O121:H19, two strains of serotype O111:NM, and one strain for each of serotypes O5:NM, O145:NM, O157:NM, O157:NM, O121:NM, and O165:H25 (deposited in GenBank under accession numbers DQ138070 to DQ138078; strain numbers are indicated in Fig. 1B). Existing sequence data for serotypes O157:H7 and O26:H11 (see Materials and Methods) were identical to our data from Canadian strains; notably, the sequences of the amplified espZ product were identical for all strains of an individual serotype, but each examined serotype had a unique espZ allele (Fig. 1A), except for strains of serotype O121:H19 and O121:NM, in which the sequenced products were 100% identical.
Calculation of scores representing the distance between the espZ sequences for each STEC strain permitted construction of a phylogenetic tree (Fig. 1B), revealing four distinct clusters. Three clusters were constituted of multiple serotypes, and each of these clusters contained strains classified as representing different seropathotypes. The fourth STEC espZ cluster was constituted solely of O111:NM strains, and this group was most closely related to the O121-containing group and was 91% identical to O121:H19-encoded espZ. The overall pairwise sequence identity between espZ genes encoded by different STEC serotypes ranged from 67 to 100%. Sequence diversity existed in regions encoding both the predicted transmembrane domains as well as in the intervening loop region, in which the addition or deletion of up to two codons was observed (Fig. 1A). The genetic diversity at espZ, as calculated using data from 21 STEC strains identified in Fig. 1B, was = 0.23, and this was similar to previous measurements calculated using data from six EPEC and EHEC strains (3). This diversity index was higher than that observed for any other LEE-encoded locus (3), and we calculated a synonymous mutation rate (dS) of 0.33 and a nonsynonymous rate (dN) of 0.18 using espZ sequence data from these 21 STEC strains. The resulting dN/dS ratio was <1, and therefore espZ would be classified as undergoing purifying selection, but the relatively high dN value is still suggestive of adaptive evolution at this locus.
Intimin typing. One of the prototypical virulence factors encoded in the LEE is intimin (eae), and significant sequence variation between STEC serotypes has been observed at this locus (40). PCR-based allelic discrimination was performed at the eae locus for each of the 35 STEC strains in our panel (Table 3), and these results correspond to the intimin subtypes previously observed in other strains of the same serotypes (2, 23, 32). Complete congruence between intimin and espZ subtypes was also previously observed (14), and this observation extends to our data set; therefore, we propose that the four espZ lineages (Fig. 1A and 1B) should be named after the eae alleles they are coinherited with (?1, 1, 2, and ).
Real-time PCR allelic discrimination of espZ subtypes. Sequencing of espZ indicated distinct sequence variation between serotypes, and the divergent regions characteristic of each proposed espZ lineage were targeted for development of real-time LUX PCR probes, a novel real-time system requiring only one self-quenching fluorescently labeled primer and one unlabeled primer (20). LUX primer pairs were designed for ?1, 1, 2, and espZ alleles, and to provide additional strain characterization capabilities using the same platform, LUX primers for the Shiga toxin genes stx1 and stx2 were designed for conserved regions within each of these loci.
Real-time PCRs were performed with templates prepared from multiple clinical isolates for all LEE-positive serotypes classified in seropathotype group A, B, or C by use of each of the espZ-?1, -1, -2, and - LUX primer sets (Table 3). The specificity of each LUX primer set was 100% for each targeted serotype, based upon the known sequence data for other isolates of that same serotype. Notably, product was only detected with a single primer set for each examined template (Fig. 2A, showing data for O111:NM/espZ-2), and strains of the same serotype but of a different stx genotype had identical espZ alleles. STEC strains of serotype O91:H21 and O113:H21 that do not encode the LEE island were also tested with each espZ primer set (Table 3), and no products were detected, indicating that these LUX primers are only productive in the presence of espZ.
To determine the detection limits of the LUX real-time primer pairs, a dilution series of genomic DNA prepared from serotypes (O157:H7, O26:H11, O121:H19, and O111:NM) representing each of the four espZ allele subtypes was used as the template (Fig. 2B, showing data for O111:NM/espZ-2). The highest concentration of template examined was 100 ng/reaction, and a positive signal was produced for each sample. The lowest template concentrations that were productive (with fluorescence in excess of the threshold value) were 10 fg/reaction for O157:H7/espZ-1, 100 fg/reaction for O26:H11/espZ-?1, 100 pg/reaction for O121:H19/espZ-, and 10 pg/reaction for O111:NM/espZ-2; these lower detection limits correlated to 2 x 101, 2 x 102, 2 x 104, and 2 x 103 genome copies per reaction, respectively. Therefore, the largest dynamic range was achieved with LUX primers specific for espZ-1, where 8 orders of magnitude in DNA concentration were detectable.
Detection of stx1 and stx2. To allow for the concurrent detection of stx genes in STEC strains by use of LUX real-time technology, primer sets were developed for each of stx1 loci and the more variable stx2 locus. Numerous molecular reagents have already been developed for real-time PCR detection of Shiga toxin-encoding genes and other STEC-associated virulence factors (11, 21, 29, 36), and our method provides additional platform-specific reagents. To assist in a comparison between different stx2 alleles encoded by STEC required for primer design, the complete sequence of the stx2 locus (encoding Stx2a and Stx2b subunits) was determined for the lone O26:H11 strain carrying stx2 reported to the National Microbiology Laboratory (strain 02-6737), O111:NM strain 00-4748, and O121:H19 strains 03-2636 and 03-2642 (deposited in GenBank under accession numbers DQ143180 to DQ143183). These data identified a conserved region in the Stx2a subunit-encoding region that also matched LUX primer design parameters (data not shown), and all strains in our panel were examined for both stx1 and stx2 (Table 3). No discrepancies were observed between the LUX real-time results and the known stx genotypes and phenotypes of these strains. The sequence data for O26:H11, O121:H19, and O111:NM stx2 loci were 99.3, 99.9, and 100% identical, respectively, to that of O157:H7 strain EDL933 carrying stx2. Additionally, we subtyped the stx2 loci encoded by the O91:H21, O113:H21, and O165:H25 isolates in our panel as stx2c (data not shown), indicating that the stx2-specific LUX primers are minimally capable of detecting the vh-a and c-type variants.
Liquid microsphere suspension array discrimination of espZ subtypes. Genotyping at bacterial loci can also be achieved using liquid microsphere suspension array technology, and the espZ gene meets many of the requirements for such a genotyping assay. This target is small (300 bp) and contains a highly variable region (suitable for designing allele-specific probes) surrounded by highly conserved termini (suitable for designing universal primers for amplification of all alleles for that particular locus). Allelic discrimination of espZ was achieved for a representative strain of each of the four espZ lineages by addition of biotin-labeled, single-stranded espZ target DNA with four sets of differentially fluorescently coded microspheres covalently coupled with a probe specific for one of the four espZ alleles (Fig. 1A and Fig. 3). Single-stranded target DNA was selectively acquired after T7 exonuclease digestion of the unlabeled, exonuclease-sensitive strand in the GIL245/256-L espZ PCR product. After hybridizing individual targets with the full array of probe-coupled microspheres, the target was fluorescently labeled with streptavidin-R-phycoerythrin, and successful target-probe interactions were detected by Luminex flow cytometry. For each espZ target, only an interaction with the corresponding probe was detected.
DISCUSSION
To investigate genetic sequences that could potentially be used for molecular serotyping and characterization of STEC, we chose target loci common to the most frequently detected serotypes. The espZ gene is coinherited with other classical E. coli virulence determinants on the LEE, including sequences coding for intimin and effector proteins secreted by the type III secretion system. Our comparative DNA sequence analysis at the espZ locus revealed four distinct lineages among STEC strains, with heterogeneity observed between serotypes and conservation among strains of a single serotype. The observed sequence variation was developed into molecular tests that can accurately and rapidly identify toxin and espZ genotypes. These data also provide insight into the evolutionary history of the relationship between STEC serotypes and the LEE pathogenicity island.
Over 90% of STEC strains isolated in Canada are serotype O157:H7 or O157:NM (38), but it is estimated that between 20 and 50% of actual STEC infections result from non-O157 strains (17, 33). The bias towards O157 in clinical settings may be a result of the established methods for identification and the availability of selective media (39). To develop novel molecular methods specifically for non-O157 STEC, we had previously performed multilocus sequence typing on Canadian O26:H11 STEC isolates but were unable to identify significant genetic diversity to subtype this collection of strains (10). Whereas the goal of multilocus sequence typing is to identify polymorphisms between related strains, other typing methodologies such as serotyping can identify more global characteristics that are indicative of corresponding genetic traits (i.e., serotype O157:H7 strains produce Shiga toxins). The real-time PCR primers and microsphere-coupled probes described here cumulatively allow for the detection and genotyping of toxin and pathogenicity islands by targeting stx1, stx2, and LEE subtype-specific alleles of espZ. These methods were not capable of molecular serotyping (e.g., espZ-?1 does not exclusively correlate to O26:H11 strains) and do not provide additional discrimination of STEC isolates compared to typing of the eae locus, but genotyping of toxin and pathogenicity islands can be used to infer different STEC lineages. Previously developed detection methods for O serotypes have been developed using targets from the O-antigen cluster genes wzx and wzy carried by EHEC O157, O103, O26, and O113 serotypes (5, 11, 26), and examination of additional loci, including the genes described here, or of genes unique to individual serotypes such as fimbria-encoding determinants (31, 34) may cumulatively provide a means for molecular serotyping.
To our knowledge, this study is the first example of allelic discrimination at a bacterial virulence locus determined using LUX primers. The design of traditional PCR primers which would amplify all alleles of espZ was facilitated by sequence conservation at regions adjacent to the start and stop codon. The first 20 codons of espZ are sufficient to direct EspZ translocation (14), and this is a possible explanation for why this 5' segment is conserved among the different STEC serotypes. Alternatively, the sites within espZ selected for real-time LUX PCR and microsphere suspension array probe design occur principally in, or adjacent to, the segment encoding the first predicted transmembrane domain of EspZ (14). The intervening region between the two predicted transmembrane domains has been identified as the most divergent region of EspZ (14), and although significant nucleotide divergence exists in the region encoding the intervening loop between transmembrane domains, the probe design parameters for the microsphere suspension array and LUX technologies necessitated the design of primers and probes outside of this loop region (Fig. 1A) where subtle serotype-specific variations were available (e.g., O111:NM espZ versus O121:H19 espZ; Fig. 1A). The LUX real-time PCR technology offered quick resolution of espZ alleles (positive reactions within an hour after preparation of genomic template DNA); however, there are a greater number of espZ alleles than distinguishable fluorescent channels. Therefore, we did not multiplex this system to perform allele discrimination in a single reaction. Conversely, a single PCR using the espZ "universal" primer set was sufficient to initiate the microsphere-based allelic discrimination methodology, which was subsequently demultiplexed during flow cytometry of the probe-coupled microspheres. This latter method is therefore ideal for discrimination of loci that have a large number of alleles with definite sequence characteristics (for probe design), and additional targets could be incorporated into a single reaction mixture to provide additional typing capabilities.
The intimin-encoding gene eae was identified to exhibit a high degree of genetic diversity between A/E pathotypes of E. coli (3), and eae alleles are generally conserved between strains of the same STEC serotypes; therefore, this is also an appropriate target gene for molecular subtyping of E. coli. The diversity between eae alleles was calculated as = 0.14 for six A/E pathogens (3) and = 0.14 for the ?1, 1, 2, and alleles of eae (data not shown). Molecular subtyping of eae utilizes the 3' region encoding the Tir-binding domain (2, 40), and the diversity between the ?1, 1, 2, and subtypes at this region was = 0.31 (represented by 945 bp; data not shown). Although this region of eae is more diverse than espZ ( = 0.23), the espZ locus had ideal molecular characteristics for development of a liquid microsphere suspension assay, including conserved regions that surround the allele-specific sites (allowing universal amplification of all known espZ alleles and with a small fragment size suitable for this method). The allele-specific sites were also appropriate for LUX primer design. Furthermore, there was no evidence of recombination between espZ alleles (Fig. 4), whereas recombination was observed using split decomposition analysis for entire or partial coding sequences of eae subtypes (3, 40) or the 945-bp 3' terminus of ?1, 1, 2, and alleles (data not shown). The lack of observable recombination between espZ alleles is favorable for molecular typing, because the sites deemed characteristic for each allele are independently inherited in that lineage and are not recombined between unrelated lineages.
The LEE exhibits hallmark compositional traits that indicated that it was acquired through horizontal transfer (3), and our data suggest that it now appears to be stationary and is vertically transmitted by individual STEC lineages (i.e., serotypes). Congruence between STEC serotype and espZ allelic subtypes was observed, indicating that clonal dissemination of LEE variants (each categorized based upon eae and espZ allele carriage) occurred within individual serotypes. Between STEC strains of a single serotype, identical espZ alleles were observed, whereas between serotypes small to major variation occurred, ranging from 67 to 100% identity in pairwise sequence comparisons. The only observation of two STEC serotypes having identical espZ sequences was between O121:H19 and O121:NM strains. Furthermore, the congruence between serotype and espZ allelic variation does not support recent or frequent lateral transfer of the LEE pathogenicity island, and split decomposition analysis of STEC-encoded espZ did not indicate recombination between espZ alleles (Fig. 4). If the LEE is one of the founding genetic traits of STEC and if the espZ allele can be considered a marker for LEE evolution because it is highly polymorphic but not subject to recombination, then the lineage and serotype-specific variation observed at espZ could indicate the pattern of evolution for STEC serotypes, with a topology similar to that seen for espZ (Fig. 1B and 4). During the generation of three major lineages (?1, 1, and /2) through point mutation in the central region of espZ and insertion or deletion of whole codons in the loop-encoding region, these precursor LEE variants may have segregated to the progenitors of the currently observed serotypes, wherein additional serotype-specific variation arose, albeit subtle in some instances (e.g., serotypes O157:H7, O157:NM, and O145:NM, each encoding espZ-1, are 99% identical to each other, and O121:H19 and O103:H2 strains are 98% identical at espZ-). Additionally, each STEC serotype also would have concurrently evolved with the LEE (7, 37) by the gain or loss of stx genes (18), plasmids, antigenic determinants, O islands (28), and other virulence determinants that contribute to the differential pathogenicity observed between serotypes.
ACKNOWLEDGMENTS
Our gratitude goes to L. Chui, A. Paccagnella, J. Wylie, K. Forward, and G. Horsman of the Provincial Health Laboratories in Alberta, British Columbia, Manitoba, Nova Scotia, and Saskatchewan, respectively, for providing strains. Members of the NML DNA Core Facility performed sequencing reactions and oligonucleotide synthesis, and members of the NML Serotyping, Identification, and Molecular Typing Programs conducted the initial strain characterization. We also thank V. Goleski for assistance with the Luminex assays, J. McCrea for assistance with LUX assays, and S. Tyler for assistance with stx2 characterization.
This work was supported by Canadian Biotechnology Strategy Funds, administered by the Office of Biotechnology and Science, and Health Canada.
REFERENCES
Bertin, Y., K. Boukhors, V. Livrelli, and C. Martin. 2004. Localization of the insertion site and pathotype determination of the locus of enterocyte effacement of Shiga toxin-producing Escherichia coli strains. Appl. Environ. Microbiol. 70:61-68.
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.
Castillo, A., L. E. Eguiarte, and V. Souza. 2005. A genomic population genetics analysis of the pathogenic enterocyte effacement island in Escherichia coli: the search for the unit of selection. Proc. Natl. Acad. Sci. USA 102:1542-1547.
Cherla, R. P., S. Y. Lee, and V. L. Tesh. 2003. Shiga toxins and apoptosis. FEMS Microbiol. Lett. 228:159-166.
DebRoy, C., E. Roberts, J. Kundrat, M. A. Davis, C. E. Briggs, and P. M. Fratamico. 2004. Detection of Escherichia coli serogroups O26 and O113 by PCR amplification of the wzx and wzy genes. Appl. Environ. Microbiol. 70:1830-1832.
Donnenberg, M. S., and T. S. Whittam. 2001. Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J. Clin. Investig. 107:539-548.
Feng, P., K. A. Lampel, H. Karch, and T. S. Whittam. 1998. Genotypic and phenotypic changes in the emergence of Escherichia coli O157:H7. J. Infect. Dis. 177:1750-1753.
Frankel, G., A. D. Phillips, I. Rosenshine, G. Dougan, J. B. Kaper, and S. Knutton. 1998. Enteropathogenic and enterohaemorrhagic Escherichia coli: more subversive elements. Mol. Microbiol. 30:911-921.
Gartner, J. F., and M. A. Schmidt. 2004. Comparative analysis of locus of enterocyte effacement pathogenicity islands of atypical enteropathogenic Escherichia coli. Infect. Immun. 72:6722-6728.
Gilmour, M. W., T. D. Cote, J. Munro, L. Chui, J. Wylie, J. Isaac-Renton, G. Horsman, D. M. Tracz, A. Andrysiak, and L.-K. Ng. 2005. Multilocus sequence typing of Escherichia coli O26:H11 isolates carrying stx in Canada does not identify genetic diversity. J. Clin. Microbiol. 43:5319-5323.
Hsu, C. F., T. Y. Tsai, and T. M. Pan. 2005. Use of the duplex TaqMan PCR system for detection of Shiga-like toxin-producing Escherichia coli O157. J. Clin. Microbiol. 43:2668-2673.
Huson, D. H. 1998. SplitsTree: analyzing and visualizing evolutionary data. Bioinformatics 14:68-73.
Kaderali, L., A. Deshpande, J. P. Nolan, and P. S. White. 2003. Primer-design for multiplexed genotyping. Nucleic Acids Res. 31:1796-1802.
Kanack, K. J., J. A. Crawford, I. Tatsuno, M. A. Karmali, and J. B. Kaper. 2005. SepZ/EspZ is secreted and translocated into HeLa cells by the enteropathogenic Escherichia coli type III secretion system. Infect. Immun. 73:4327-4337.
Karmali, M. A., M. Mascarenhas, S. Shen, K. Ziebell, S. Johnson, R. Reid-Smith, J. Isaac-Renton, C. Clark, K. Rahn, and J. B. Kaper. 2003. Association of genomic O island 122 of Escherichia coli EDL 933 with verocytotoxin-producing Escherichia coli seropathotypes that are linked to epidemic and/or serious disease. J. Clin. Microbiol. 41:4930-4940.
Kumar, S., K. Tamura, and M. Nei. 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief. Bioinform. 5:150-163.
Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5:607-625.
Mellmann, A., M. Bielaszewska, L. B. Zimmerhackl, R. Prager, D. Harmsen, H. Tschape, and H. Karch. 2005. Enterohemorrhagic Escherichia coli in human infection: in vivo evolution of a bacterial pathogen. Clin. Infect. Dis. 41:785-792.
Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142-201.
Nazarenko, I., B. Lowe, M. Darfler, P. Ikonomi, D. Schuster, and A. Rashtchian. 27 February 2002, revision date. Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore. Nucleic Acids Res. 30:e37. http://nar.oxfordjournals.org/cgi/content/full/30/9/e37.
Nielsen, E. M., and M. T. Andersen. 2003. Detection and characterization of verocytotoxin-producing Escherichia coli by automated 5' nuclease PCR assay. J. Clin. Microbiol. 41:2884-2893.
Nikiforov, T. T., R. B. Rendle, M. L. Kotewicz, and Y. H. Rogers. 1994. The use of phosphorothioate primers and exonuclease hydrolysis for the preparation of single-stranded PCR products and their detection by solid-phase hybridization. PCR Methods Appl. 3:285-291.
Oswald, E., H. Schmidt, S. Morabito, H. Karch, O. Marches, and A. Caprioli. 2000. Typing of intimin genes in human and animal enterohemorrhagic and enteropathogenic Escherichia coli: characterization of a new intimin variant. Infect. Immun. 68:64-71.
Paton, A. W., and J. C. Paton. 1998. Detection and characterization of Shiga toxigenic Escherichia coli by using multiplex PCR assays for stx1, stx2, eaeA, enterohemorrhagic E. coli hlyA, rfbO111, and rfbO157. J. Clin. Microbiol. 36:598-602.
Paton, J. C., and A. W. Paton. 1998. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin. Microbiol. Rev. 11:450-479.
Perelle, S., F. Dilasser, J. Grout, and P. Fach. 2005. Detection of Escherichia coli serogroup O103 by real-time polymerase chain reaction. J. Appl. Microbiol. 98:1162-1168.
Perna, N. T., G. F. Mayhew, G. Posfai, S. Elliott, M. S. Donnenberg, J. B. Kaper, and F. R. Blattner. 1998. Molecular evolution of a pathogenicity island from enterohemorrhagic Escherichia coli O157:H7. Infect. Immun. 66:3810-3817.
Perna, N. T., G. Plunkett III, V. Burland, B. Mau, J. D. Glasner, D. J. Rose, G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J. Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis, A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Anantharaman, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner. 2001. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409:529-533.
Reischl, U., M. T. Youssef, J. Kilwinski, N. Lehn, W. L. Zhang, H. Karch, and N. A. Strockbine. 2002. Real-time fluorescence PCR assays for detection and characterization of Shiga toxin, intimin, and enterohemolysin genes from Shiga toxin-producing Escherichia coli. J. Clin. Microbiol. 40:2555-2565.
Rozas, J., J. C. Sanchez-DelBarrio, X. Messeguer, and R. Rozas. 2003. DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19:2496-2497.
Shen, S., M. Mascarenhas, R. Morgan, K. Rahn, and M. A. Karmali. 2005. Identification of four fimbria-encoding genomic islands that are highly specific for verocytotoxin-producing Escherichia coli serotype O157 strains. J. Clin. Microbiol. 43:3840-3850.
Tarr, C. L., T. M. Large, C. L. Moeller, D. W. Lacher, P. I. Tarr, D. W. Acheson, and T. S. Whittam. 2002. Molecular characterization of a serotype O121:H19 clone, a distinct Shiga toxin-producing clone of pathogenic Escherichia coli. Infect. Immun. 70:6853-6859.
Thorpe, C. M. 2004. Shiga toxin-producing Escherichia coli infection. Clin. Infect. Dis. 38:1298-1303.
Toma, C., E. Martinez Espinosa, T. Song, E. Miliwebsky, I. Chinen, S. Iyoda, M. Iwanaga, and M. Rivas. 2004. Distribution of putative adhesins in different seropathotypes of Shiga toxin-producing Escherichia coli. J. Clin. Microbiol. 42:4937-4946.
Tyler, S. D., W. M. Johnson, H. Lior, G. Wang, and K. R. Rozee. 1991. Identification of verotoxin type 2 variant B subunit genes in Escherichia coli by the polymerase chain reaction and restriction fragment length polymorphism analysis. J. Clin. Microbiol. 29:1339-1343.
Wang, G., C. G. Clark, and F. G. Rodgers. 2002. Detection in Escherichia coli of the genes encoding the major virulence factors, the genes defining the O157:H7 serotype, and components of the type 2 Shiga toxin family by multiplex PCR. J. Clin. Microbiol. 40:3613-3619.
Wick, L. M., W. Qi, D. W. Lacher, and T. S. Whittam. 2005.
Evolution of genomic content in the stepwise emergence of Escherichia coli O157:H7. J. Bacteriol. 187:1783-1791.
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.
Zadik, P. M., P. A. Chapman, and C. A. Siddons. 1993. Use of tellurite for the selection of verocytotoxigenic Escherichia coli O157. J. Med. Microbiol. 39:155-158.
Zhang, W. L., B. Kohler, E. Oswald, L. Beutin, H. Karch, S. Morabito, A. Caprioli, S. Suerbaum, and H. Schmidt. 2002. Genetic diversity of intimin genes of attaching and effacing Escherichia coli strains. J. Clin. Microbiol. 40:4486-4492.
National Microbiology Laboratory, Public Health Agency of Canada, Winnipeg, Manitoba, Canada(Matthew W. Gilmour, Dobry)
Infections with Shiga toxin-producing Escherichia coli (STEC) result in frequent cases of sporadic and outbreak-associated enteric bacterial disease in humans. Classification of STEC is by stx genotype (encoding the Shiga toxins), O and H antigen serotype, and seropathotype (subgroupings based upon the clinical relevance and virulence-related genotypes of individual serotypes). The espZ gene is encoded in the locus of enterocyte effacement (LEE) pathogenicity island responsible for the attaching and effacing (A/E) lesions caused by various E. coli pathogens (but not limited to STEC), and this individual gene (300 bp) has previously been identified as hypervariable among these A/E pathogens. Sequence analysis of the espZ locus encoded by additional STEC serotypes and strains (including O26:H11, O121:H19, O111:NM, O145:NM, O165:H25, O121:NM, O157:NM, O157:H7, and O5:NM) indicated that distinct sequence variants exist which correlate to subgroups among these serotypes. Allelic discrimination at the espZ locus was achieved using Light Upon eXtension real-time PCR and by liquid microsphere suspension arrays. The allele subtype of espZ did not correlate with STEC seropathotype classification; however, a correlation with the allele type of the LEE-encoded intimin (eae) gene was supported, and these sequence variations were conserved among individual serotypes. The study focused on the characterization of three clinically significant seropathotypes of LEE-positive STEC, and we have used the observed genetic variation at a pathogen-specific locus for detection and subtyping of STEC.
INTRODUCTION
Gastrointestinal infection by Shiga toxin-producing Escherichia coli (STEC) is largely due to serotype O157:H7 in North America, but infections with other serotypes also result in human disease (25). In Canada, 48 different serotypes of STEC (as identified on the basis of O-somatic and H-flagellar antigens) have been isolated from humans, and strains of serotypes O26:H11, O121:H19, O103:H2, O145:NM, and O111:NM have represented a large proportion of non-O157 isolates (38). STEC serotypes are classified into five seropathotypes (A through E) based upon both virulence gene content and clinical relevance, among which seropathotype A is solely comprised of O157:H7 and O157:NM strains; infections with these strains can result in serious disease symptoms and lead to outbreaks (15). Infection with STEC can result in diarrhea, hemorrhagic colitis, and hemolytic uremic syndrome, with possible disease outcomes including renal failure, neurological sequelae, and death (17, 33). Seropathotype B includes the non-O157 serotypes identified above, and while not known to cause large disease epidemics as frequently, infection with these pathogens can result in disease symptoms similar to those seen with seropathotype A strains (15). Notably, the differential capabilities for detection of O157 versus non-O157 serotypes in clinical laboratories may introduce reporting biases.
STEC disease manifestation correlates to the carriage of classical bacterial virulence determinants such as toxins and pathogenicity islands. The production of Shiga toxins, encoded by the stx1 and stx2 genes, is responsible for systemic disease symptoms, because necrotic and apoptotic cell death are induced after intracellular translocation (4). Stx1 is nearly identical to the cytotoxin produced by Shigella dysenteriae serotype 1 and is homogenous among E. coli carrying stx1, whereas several variants of stx2 have been identified, and the production of Stx2 is associated with hemolytic uremic syndrome (33). The carriage of stx genes is also variable among STEC serotypes, as the majority of seropathotype A strains encode both loci, whereas strains of the other seropathotypes typically encode a single toxin locus. A large subset of STEC strains (predominantly seropathotypes A and B) are also termed enterohemorrhagic E. coli (EHEC) strains and are partly characterized by attaching and effacing (A/E) lesions that they create on the intestinal epithelium, a histopathology resulting from the presence of the locus of enterocyte effacement (LEE) pathogenicity island (8, 19). While STEC strains are considered to be noninvasive, there is disruption of the brush border microvilli after actin rearrangements within epithelial cells. This process is induced by the LEE-encoded determinants that include 41 coding sequences (the majority of which are organized into five major operons) for type III secretion of effector and receptor proteins, including Tir, which allows the close association between the bacterial and the host epithelial cells via intimin (eae) present on the bacterial cell surface (6, 27).
Among the individual LEE-encoded genes present in different A/E pathotypes, including enteropathogenic E. coli (EPEC) and EHEC, a heterogeneous rate of genetic diversity has been observed (3, 27). It was our goal to select a LEE-encoded gene that would support the detection and subtyping of STEC, relying upon the ubiquity of the LEE among the most significant STEC serotypes and the large degree of genetic diversity observed at individual LEE genes. The loci that demonstrated the highest rates of diversity (measured as ) include those involved in interaction with the host cell (i.e., effectors and receptors), whereas the type III secretion apparatus had a lesser amount of diversity (3). Furthermore, the ratio of nonsynonymous mutation (dN) to synonymous mutation (dS) was markedly higher at those same loci involved in host interactions, and therefore these coding sequences may be undergoing positive selection (3), a process reflective of functional adaptation in different host environments and E. coli genetic backgrounds. The loci with the highest estimates included sepZ, tir, espA, espB, espF, espH, and eae, and some of these have previously been used for molecular subtyping of A/E pathogens (1, 2, 21, 23, 29, 40). Corresponding variability of SepZ protein primary sequences was also observed among EPEC and EHEC strains (9, 14, 27). Notably, the sepZ coding sequence was renamed espZ after determining that the gene product is secreted and translocated to eukaryotic cells (14). The espZ locus was also estimated as having the highest rate (3), and the resultant variability between EspZ proteins likely arose due to functional adaptation to host proteins, which are currently unknown (14). Our study was of the espZ nucleotide sequences encoded in clinically significant STEC. Molecular techniques were developed for the detection and subtyping of STEC at this locus, and phylogenetic analyses were performed to estimate the evolutionary history of the relationship between STEC serotypes and the LEE pathogenicity island.
MATERIALS AND METHODS
Bacterial strains. The panel of STEC strains collected for this study (Table 1) included representative isolates from each of the major serotypes observed in Canada (classified as seropathotype A, B, or C), and among individual serotypes we included strains with different stx genotypes when available. All isolates were from the National Microbiology Laboratory (NML) Bacteriology and Enteric Diseases Program culture collection, which originated from human sources at various Canadian Provincial Health laboratories during 1985 to 2005, and serotype and toxin genotype were confirmed at the NML (Table 1). Existing sepZ/espZ sequence data deposited in GenBank were used to initiate our characterization of these Canadian strains as follows: for O26:H11 (strain 6549), accession number AF035656; for O26:H- (413/89-1), AJ277443; for O157:H7 (EDL933), AAC31516; for O157:H7 (Sakai), BAB37994; for O157:H7 (86-24), AF035655; for O15:H- (RDEC-1), AF035651; for O15:H- (EPEC 83/89), AF453441; for O111:NM (EPEC B171), AF035653; for O103:H2 (RW1374), AJ303141; for O127:H6 (EPEC E2348/69), X94450; and for O55:H7 (EPEC), AF035652.
PCR and sequencing. Template DNA was prepared by centrifuging 1 ml of log-phase cultures grown in brain heart infusion broth, resuspending the pellet in 1 ml of TE buffer (Sigma, St. Louis, MO) (10 mM Tris-HCl, 1 mM EDTA, pH 8.0) and boiling for 10 min. Boiled cells were pelleted, and the supernatant was removed and used as the template in real-time and standard PCRs. For quantitative determination of real-time PCR sensitivity, total genomic DNA was isolated from liquid cultures grown overnight in 8 ml brain heart infusion broth. After centrifugation, the bacterial pellet was resuspended in 2 ml of TE buffer with vortexing. Following the addition of lysozyme (Roche Diagnostics, Indianapolis, IN) (0.5 mg/ml), RNase (Roche Diagnostics) (1.5 μg/ml), and proteinase K (Sigma) (0.12 mg/ml), this mixture was incubated at 37°C for 1 h, and then sodium dodecyl sulfate (SDS) (Ambion, Austin, TX) was added to achieve a concentration of 0.1% (wt/vol) and the mixture was further incubated at 65°C until the suspension cleared. Organic extraction was performed using 15 ml Eppendorf Phase Lock tubes (Hamburg, Germany) with an equal volume of phenol:chloroform:isoamyl alcohol (Invitrogen, Burlington, ON) (25:24:1). Phenol-chloroform-isoamyl alcohol extraction was repeated until the aqueous layer was clear, and after a final extraction with 2 ml chloroform, the aqueous layer was transferred to a new 1.5 ml tube and 0.6 volumes of isopropanol and 0.1 volumes of 3 M sodium acetate (Ambion, pH 5.5) were added and DNA was precipitated at –20°C for 20 min. Following centrifugation, the DNA pellet was washed in 1 ml of 70% ethanol and resuspended in 200 μl of TE buffer. After quantification on a NanoDrop ND-1000 apparatus (NanoDrop Technologies, Rockland DE), the DNA was serially diluted from 33 ng/μl to 0.33 fg/μl.
Standard PCR was performed with Platinum Taq (Invitrogen), following the manufacturer's directions, and with the oligonucleotides described in Table 2. The thermocycling parameters for espZ required an initial denaturation at 94°C for 5 min and 35 cycles of denaturation at 94°C for 30 s, annealing at 50°C for 30 s, and extension at 68°C for 30 s, with a final extension at 68°C for 5 min. Conditions for PCR amplification of stx2 were initial denaturation at 94°C for 5 min and 30 cycles of denaturation at 94°C for 30 s, annealing at 55°C for 45 s, and extension at 68°C for 30 s, with a final extension at 68°C for 7 min. PCR products for espZ and stx2 were purified using a QIAquick PCR purification kit (QIAGEN, Mississauga, ON) and sequenced using the same primers used to generate this template. Sequencing was performed on an ABI3730 apparatus (Applied Biosystems, Foster City, CA). Subtyping of stx2 variants was performed using oligonucleotides described by Wang et al. (36).
LUX (Light Upon eXtension) fluorogenic and unlabeled primer pairs (20) were designed using D-LUX designer software (Invitrogen) by targeting regions characteristic of each espZ allele (Fig. 1A) or for regions conserved in either stx1 or stx2 (see Results). LUX primers are self-quenching oligonucleotide primers that are labeled with a single fluorophore. Upon annealing to a specific DNA sequence, they become dequenched and the fluorescent signal increases. For LUX real-time PCR, Platinum quantitative PCR Supermix UDG (Invitrogen) was used for the amplification mixture with 200 nM of each LUX primer pair (Table 2) and 3 μl of template for a total reaction volume of 25 μl. Real-time PCRs were performed on the Cepheid SmartCycler 2.0 apparatus (Cepheid, Sunnyvale, CA), and samples were amplified after an initial denaturation at 95°C for 3 min and 40 cycles of denaturation at 95°C for 10 s, annealing at 55°C for 15 s, and an extension step at 72°C for 15 s. Fluorescence was detected at the annealing step, and the threshold level was set at 30 fluorescence units. A real-time PCR result was considered positive when the log fluorescent signal exceeded this threshold after background subtraction.
Microsphere liquid-suspension arrays. Allelic discrimination of espZ was achieved after PCR amplification of biotin-labeled espZ target DNA from group A and B STEC seropathotypes (O157:H7, O26:H11, O121:H19, and O111:NM) by use of a GeneAmp 9700 thermocycler (Applied Biosystems) and the thermocycling parameters detailed above. The 100 μl amplification mixture consisted of the following: 10 μl of 10x HiFi buffer (Invitrogen), 2 μl of deoxynucleoside triphosphates (Invitrogen) (10 μM each), 4 μl of MgSO4, 0.4 μl of Platinum Hi Fidelity Taq (Invitrogen), 61.6 μl of molecular biology grade water (Gibco, Grand Island, NY), and 10 μl each of primers GIL245 and GIL246-L (5' biotinylated; Table 2), for a final primer concentration of 1 μM each. The template (2 μl), prepared from a boiled cell resuspension as described above, was added to the reaction mixture. Successful PCR amplification of espZ was confirmed by agarose gel electrophoresis. PCRs were purified with Qiaquick DNA purification kits (QIAGEN) and eluted with 50 μl of EB buffer (QIAGEN). Oligonucleotide GIL246-L contains four bases with phosphorothioate linkages, as well as a biotin molecule, all at the 5' end. Of the two strands of the target DNA, the strand produced from GIL245 (espZ "sense" strand) is sensitive to T7 exonuclease digestion whereas the espZ antisense strand produced from GIL246-L is protected due to the phosphorothioate linkages (22). DNA digestion was performed by mixing 43 μl of purified PCR product with 5 μl of buffer 4 and 2 μl of T7 exonuclease (both from New England Biolabs, Ipswich, MA) (20 U total) and incubating at 37°C for 1 h. T7 exonuclease was inactivated by adding 2 μl of 0.5 M EDTA (Ambion). Selective degradation ensures elimination of the unlabeled target DNA strand, thereby preventing reannealing between the two target DNA strands during hybridization that, if left intact, would limit the intended hybridization between the biotin-labeled strand and the espZ allele-specific probes coupled to microspheres in subsequent steps.
Oligonucleotide probes for each espZ allele were designed matching the sense strand in highly variable regions characteristic of individual allele subtypes (Fig. 1A, Table 2). Oligonucleotides were screened using SBEprimer software (13) for potential secondary structures or cross-hybridization between probes. The oligonucleotide probes were synthesized with a 5' C-12 amine and coupled to xMAP carboxylated fluorescently coded microspheres (Luminex Corporation, Austin, TX). Microsphere sets 103, 105, 108, and 110 were coupled to oligonucleotides DOB70, DOB72, DOB73, and DOB74, respectively (Table 2). Microspheres (5.0 x 106) were transferred to a 1.5 ml microcentrifuge tube, centrifuged at 8,000 x g for 2 min, resuspended in 50 μl of 0.1 M MES buffer [2-(N-morpholino)ethanesulfonic acid; Sigma] (pH 4.5), and vortexed, and 6 μl of capture oligonucleotide (100 μM) was added to the respective bead set. A fresh solution of EDC [1-ethyl-3-(3-dimethylaminopropyl)carbodiimidehydrochloride; Pierce Biotechnology, Rockford, IL] (10 mg/ml) was prepared immediately before use, 2 μl was added to the bead-oligonucleotide mixture, and the mixture was vortexed and incubated at room temperature for 30 min in the dark. The EDC addition and incubation was repeated. Following incubation, beads were washed successively in 1 ml of 0.1% Tween and 1 ml of 0.1% SDS. Microspheres were resuspended in 100 μl of 0.1 M MES (pH 4.5) and enumerated on a hemocytometer.
For hybridization of biotin-labeled espZ target DNA strands to the capture probe-coupled microspheres, a reaction master mix was prepared in TE buffer at a concentration of 150 microspheres/μl for each of the four capture probe sets. Hybridizations were prepared in triplicate under low-light conditions as 50-μl reaction mixtures in Thermowell 96-well plates (Corning Incorporated, Corning, NY). Initially, 17 μl of the biotinylated PCR product was added to wells and denatured for 10 min at 95°C in a GeneAmp 9700 thermocycler, followed by addition of 33 μl of the reaction master mix and mixing by pipetting. Hybridizations were performed at 55°C for 20 min, after which 25 μl of the streptavidin R-phycoerythrin reporter dye (Molecular Probes, Eugene, OR) diluted to 10 μg/ml in 1.5x TMAC buffer (3 M tetramethylammonium chloride [Sigma], 0.1% SDS, 50 mM Tris-HCl [pH 8.0], 4 mM EDTA, pH 8.0) was added and mixed by pipetting. Plates were incubated at 55°C for an additional 10 min. Flow cytometry with a QIAGEN LiquiChip workstation was used to quantify hybridization events, with the following settings: reading gates set at 8,300 to 16,500; minimum of 100 events read; and microplate handler heater block maintained at 55°C during measurements. Mean fluorescence intensity signals for each espZ target DNA sample and the negative control (TE blank) were averaged among the triplicate wells.
Bioinformatics. Multiple sequence alignments were completed using ClustalW (www.ebi.ac.uk/clustalw/) and Boxshade (www.ch.embnet.org), neighbor-joining trees were constructed using MEGA3 (16), and genetic diversity statistics were calculated using DnaSP 4.10.3 (30). Pairwise global alignments were calculated using Align (www.ebi.ac.uk/emboss/align/#). Split decomposition analysis was performed using SplitsTree4 (12) and alignment inputs created by ClustalW, and calculations were performed using only parsimony-informative sites.
Nucleotide sequence accession numbers. The sequences determined in this study (see Results) have been deposited in GenBank under accession numbers DQ138070 to DQ138078 and DQ143180 to DQ143183.
RESULTS
Sequencing of the espZ hypervariable region. Molecular and phylogenetic characterization of the espZ locus was initiated by sequencing the central hypervariable region of this gene from seropathotype group A strains (O157:H7, O157:NM), group B strains (O26:H11, O121:H19, O111:NM, O145:NM) and group C strains which encode the LEE pathogenicity island (O5:NM, O121:NM, O165:NM). Oligonucleotides GIL245 and 246 (Table 2) were designed for the conserved C- and N-terminal encoding regions of espZ, observed in the sequence data from espZ (formerly sepZ) carried by various EPEC and EHEC strains (see Materials and Methods). These primers were able to produce an espZ product for each examined STEC serotype that was predicted to encode the LEE (Table 1), permitting sequencing of the intervening hypervariable region (Fig. 1A). The espZ PCR product was sequenced from four clinical strains of serotype O26:H11, three strains of serotype O121:H19, two strains of serotype O111:NM, and one strain for each of serotypes O5:NM, O145:NM, O157:NM, O157:NM, O121:NM, and O165:H25 (deposited in GenBank under accession numbers DQ138070 to DQ138078; strain numbers are indicated in Fig. 1B). Existing sequence data for serotypes O157:H7 and O26:H11 (see Materials and Methods) were identical to our data from Canadian strains; notably, the sequences of the amplified espZ product were identical for all strains of an individual serotype, but each examined serotype had a unique espZ allele (Fig. 1A), except for strains of serotype O121:H19 and O121:NM, in which the sequenced products were 100% identical.
Calculation of scores representing the distance between the espZ sequences for each STEC strain permitted construction of a phylogenetic tree (Fig. 1B), revealing four distinct clusters. Three clusters were constituted of multiple serotypes, and each of these clusters contained strains classified as representing different seropathotypes. The fourth STEC espZ cluster was constituted solely of O111:NM strains, and this group was most closely related to the O121-containing group and was 91% identical to O121:H19-encoded espZ. The overall pairwise sequence identity between espZ genes encoded by different STEC serotypes ranged from 67 to 100%. Sequence diversity existed in regions encoding both the predicted transmembrane domains as well as in the intervening loop region, in which the addition or deletion of up to two codons was observed (Fig. 1A). The genetic diversity at espZ, as calculated using data from 21 STEC strains identified in Fig. 1B, was = 0.23, and this was similar to previous measurements calculated using data from six EPEC and EHEC strains (3). This diversity index was higher than that observed for any other LEE-encoded locus (3), and we calculated a synonymous mutation rate (dS) of 0.33 and a nonsynonymous rate (dN) of 0.18 using espZ sequence data from these 21 STEC strains. The resulting dN/dS ratio was <1, and therefore espZ would be classified as undergoing purifying selection, but the relatively high dN value is still suggestive of adaptive evolution at this locus.
Intimin typing. One of the prototypical virulence factors encoded in the LEE is intimin (eae), and significant sequence variation between STEC serotypes has been observed at this locus (40). PCR-based allelic discrimination was performed at the eae locus for each of the 35 STEC strains in our panel (Table 3), and these results correspond to the intimin subtypes previously observed in other strains of the same serotypes (2, 23, 32). Complete congruence between intimin and espZ subtypes was also previously observed (14), and this observation extends to our data set; therefore, we propose that the four espZ lineages (Fig. 1A and 1B) should be named after the eae alleles they are coinherited with (?1, 1, 2, and ).
Real-time PCR allelic discrimination of espZ subtypes. Sequencing of espZ indicated distinct sequence variation between serotypes, and the divergent regions characteristic of each proposed espZ lineage were targeted for development of real-time LUX PCR probes, a novel real-time system requiring only one self-quenching fluorescently labeled primer and one unlabeled primer (20). LUX primer pairs were designed for ?1, 1, 2, and espZ alleles, and to provide additional strain characterization capabilities using the same platform, LUX primers for the Shiga toxin genes stx1 and stx2 were designed for conserved regions within each of these loci.
Real-time PCRs were performed with templates prepared from multiple clinical isolates for all LEE-positive serotypes classified in seropathotype group A, B, or C by use of each of the espZ-?1, -1, -2, and - LUX primer sets (Table 3). The specificity of each LUX primer set was 100% for each targeted serotype, based upon the known sequence data for other isolates of that same serotype. Notably, product was only detected with a single primer set for each examined template (Fig. 2A, showing data for O111:NM/espZ-2), and strains of the same serotype but of a different stx genotype had identical espZ alleles. STEC strains of serotype O91:H21 and O113:H21 that do not encode the LEE island were also tested with each espZ primer set (Table 3), and no products were detected, indicating that these LUX primers are only productive in the presence of espZ.
To determine the detection limits of the LUX real-time primer pairs, a dilution series of genomic DNA prepared from serotypes (O157:H7, O26:H11, O121:H19, and O111:NM) representing each of the four espZ allele subtypes was used as the template (Fig. 2B, showing data for O111:NM/espZ-2). The highest concentration of template examined was 100 ng/reaction, and a positive signal was produced for each sample. The lowest template concentrations that were productive (with fluorescence in excess of the threshold value) were 10 fg/reaction for O157:H7/espZ-1, 100 fg/reaction for O26:H11/espZ-?1, 100 pg/reaction for O121:H19/espZ-, and 10 pg/reaction for O111:NM/espZ-2; these lower detection limits correlated to 2 x 101, 2 x 102, 2 x 104, and 2 x 103 genome copies per reaction, respectively. Therefore, the largest dynamic range was achieved with LUX primers specific for espZ-1, where 8 orders of magnitude in DNA concentration were detectable.
Detection of stx1 and stx2. To allow for the concurrent detection of stx genes in STEC strains by use of LUX real-time technology, primer sets were developed for each of stx1 loci and the more variable stx2 locus. Numerous molecular reagents have already been developed for real-time PCR detection of Shiga toxin-encoding genes and other STEC-associated virulence factors (11, 21, 29, 36), and our method provides additional platform-specific reagents. To assist in a comparison between different stx2 alleles encoded by STEC required for primer design, the complete sequence of the stx2 locus (encoding Stx2a and Stx2b subunits) was determined for the lone O26:H11 strain carrying stx2 reported to the National Microbiology Laboratory (strain 02-6737), O111:NM strain 00-4748, and O121:H19 strains 03-2636 and 03-2642 (deposited in GenBank under accession numbers DQ143180 to DQ143183). These data identified a conserved region in the Stx2a subunit-encoding region that also matched LUX primer design parameters (data not shown), and all strains in our panel were examined for both stx1 and stx2 (Table 3). No discrepancies were observed between the LUX real-time results and the known stx genotypes and phenotypes of these strains. The sequence data for O26:H11, O121:H19, and O111:NM stx2 loci were 99.3, 99.9, and 100% identical, respectively, to that of O157:H7 strain EDL933 carrying stx2. Additionally, we subtyped the stx2 loci encoded by the O91:H21, O113:H21, and O165:H25 isolates in our panel as stx2c (data not shown), indicating that the stx2-specific LUX primers are minimally capable of detecting the vh-a and c-type variants.
Liquid microsphere suspension array discrimination of espZ subtypes. Genotyping at bacterial loci can also be achieved using liquid microsphere suspension array technology, and the espZ gene meets many of the requirements for such a genotyping assay. This target is small (300 bp) and contains a highly variable region (suitable for designing allele-specific probes) surrounded by highly conserved termini (suitable for designing universal primers for amplification of all alleles for that particular locus). Allelic discrimination of espZ was achieved for a representative strain of each of the four espZ lineages by addition of biotin-labeled, single-stranded espZ target DNA with four sets of differentially fluorescently coded microspheres covalently coupled with a probe specific for one of the four espZ alleles (Fig. 1A and Fig. 3). Single-stranded target DNA was selectively acquired after T7 exonuclease digestion of the unlabeled, exonuclease-sensitive strand in the GIL245/256-L espZ PCR product. After hybridizing individual targets with the full array of probe-coupled microspheres, the target was fluorescently labeled with streptavidin-R-phycoerythrin, and successful target-probe interactions were detected by Luminex flow cytometry. For each espZ target, only an interaction with the corresponding probe was detected.
DISCUSSION
To investigate genetic sequences that could potentially be used for molecular serotyping and characterization of STEC, we chose target loci common to the most frequently detected serotypes. The espZ gene is coinherited with other classical E. coli virulence determinants on the LEE, including sequences coding for intimin and effector proteins secreted by the type III secretion system. Our comparative DNA sequence analysis at the espZ locus revealed four distinct lineages among STEC strains, with heterogeneity observed between serotypes and conservation among strains of a single serotype. The observed sequence variation was developed into molecular tests that can accurately and rapidly identify toxin and espZ genotypes. These data also provide insight into the evolutionary history of the relationship between STEC serotypes and the LEE pathogenicity island.
Over 90% of STEC strains isolated in Canada are serotype O157:H7 or O157:NM (38), but it is estimated that between 20 and 50% of actual STEC infections result from non-O157 strains (17, 33). The bias towards O157 in clinical settings may be a result of the established methods for identification and the availability of selective media (39). To develop novel molecular methods specifically for non-O157 STEC, we had previously performed multilocus sequence typing on Canadian O26:H11 STEC isolates but were unable to identify significant genetic diversity to subtype this collection of strains (10). Whereas the goal of multilocus sequence typing is to identify polymorphisms between related strains, other typing methodologies such as serotyping can identify more global characteristics that are indicative of corresponding genetic traits (i.e., serotype O157:H7 strains produce Shiga toxins). The real-time PCR primers and microsphere-coupled probes described here cumulatively allow for the detection and genotyping of toxin and pathogenicity islands by targeting stx1, stx2, and LEE subtype-specific alleles of espZ. These methods were not capable of molecular serotyping (e.g., espZ-?1 does not exclusively correlate to O26:H11 strains) and do not provide additional discrimination of STEC isolates compared to typing of the eae locus, but genotyping of toxin and pathogenicity islands can be used to infer different STEC lineages. Previously developed detection methods for O serotypes have been developed using targets from the O-antigen cluster genes wzx and wzy carried by EHEC O157, O103, O26, and O113 serotypes (5, 11, 26), and examination of additional loci, including the genes described here, or of genes unique to individual serotypes such as fimbria-encoding determinants (31, 34) may cumulatively provide a means for molecular serotyping.
To our knowledge, this study is the first example of allelic discrimination at a bacterial virulence locus determined using LUX primers. The design of traditional PCR primers which would amplify all alleles of espZ was facilitated by sequence conservation at regions adjacent to the start and stop codon. The first 20 codons of espZ are sufficient to direct EspZ translocation (14), and this is a possible explanation for why this 5' segment is conserved among the different STEC serotypes. Alternatively, the sites within espZ selected for real-time LUX PCR and microsphere suspension array probe design occur principally in, or adjacent to, the segment encoding the first predicted transmembrane domain of EspZ (14). The intervening region between the two predicted transmembrane domains has been identified as the most divergent region of EspZ (14), and although significant nucleotide divergence exists in the region encoding the intervening loop between transmembrane domains, the probe design parameters for the microsphere suspension array and LUX technologies necessitated the design of primers and probes outside of this loop region (Fig. 1A) where subtle serotype-specific variations were available (e.g., O111:NM espZ versus O121:H19 espZ; Fig. 1A). The LUX real-time PCR technology offered quick resolution of espZ alleles (positive reactions within an hour after preparation of genomic template DNA); however, there are a greater number of espZ alleles than distinguishable fluorescent channels. Therefore, we did not multiplex this system to perform allele discrimination in a single reaction. Conversely, a single PCR using the espZ "universal" primer set was sufficient to initiate the microsphere-based allelic discrimination methodology, which was subsequently demultiplexed during flow cytometry of the probe-coupled microspheres. This latter method is therefore ideal for discrimination of loci that have a large number of alleles with definite sequence characteristics (for probe design), and additional targets could be incorporated into a single reaction mixture to provide additional typing capabilities.
The intimin-encoding gene eae was identified to exhibit a high degree of genetic diversity between A/E pathotypes of E. coli (3), and eae alleles are generally conserved between strains of the same STEC serotypes; therefore, this is also an appropriate target gene for molecular subtyping of E. coli. The diversity between eae alleles was calculated as = 0.14 for six A/E pathogens (3) and = 0.14 for the ?1, 1, 2, and alleles of eae (data not shown). Molecular subtyping of eae utilizes the 3' region encoding the Tir-binding domain (2, 40), and the diversity between the ?1, 1, 2, and subtypes at this region was = 0.31 (represented by 945 bp; data not shown). Although this region of eae is more diverse than espZ ( = 0.23), the espZ locus had ideal molecular characteristics for development of a liquid microsphere suspension assay, including conserved regions that surround the allele-specific sites (allowing universal amplification of all known espZ alleles and with a small fragment size suitable for this method). The allele-specific sites were also appropriate for LUX primer design. Furthermore, there was no evidence of recombination between espZ alleles (Fig. 4), whereas recombination was observed using split decomposition analysis for entire or partial coding sequences of eae subtypes (3, 40) or the 945-bp 3' terminus of ?1, 1, 2, and alleles (data not shown). The lack of observable recombination between espZ alleles is favorable for molecular typing, because the sites deemed characteristic for each allele are independently inherited in that lineage and are not recombined between unrelated lineages.
The LEE exhibits hallmark compositional traits that indicated that it was acquired through horizontal transfer (3), and our data suggest that it now appears to be stationary and is vertically transmitted by individual STEC lineages (i.e., serotypes). Congruence between STEC serotype and espZ allelic subtypes was observed, indicating that clonal dissemination of LEE variants (each categorized based upon eae and espZ allele carriage) occurred within individual serotypes. Between STEC strains of a single serotype, identical espZ alleles were observed, whereas between serotypes small to major variation occurred, ranging from 67 to 100% identity in pairwise sequence comparisons. The only observation of two STEC serotypes having identical espZ sequences was between O121:H19 and O121:NM strains. Furthermore, the congruence between serotype and espZ allelic variation does not support recent or frequent lateral transfer of the LEE pathogenicity island, and split decomposition analysis of STEC-encoded espZ did not indicate recombination between espZ alleles (Fig. 4). If the LEE is one of the founding genetic traits of STEC and if the espZ allele can be considered a marker for LEE evolution because it is highly polymorphic but not subject to recombination, then the lineage and serotype-specific variation observed at espZ could indicate the pattern of evolution for STEC serotypes, with a topology similar to that seen for espZ (Fig. 1B and 4). During the generation of three major lineages (?1, 1, and /2) through point mutation in the central region of espZ and insertion or deletion of whole codons in the loop-encoding region, these precursor LEE variants may have segregated to the progenitors of the currently observed serotypes, wherein additional serotype-specific variation arose, albeit subtle in some instances (e.g., serotypes O157:H7, O157:NM, and O145:NM, each encoding espZ-1, are 99% identical to each other, and O121:H19 and O103:H2 strains are 98% identical at espZ-). Additionally, each STEC serotype also would have concurrently evolved with the LEE (7, 37) by the gain or loss of stx genes (18), plasmids, antigenic determinants, O islands (28), and other virulence determinants that contribute to the differential pathogenicity observed between serotypes.
ACKNOWLEDGMENTS
Our gratitude goes to L. Chui, A. Paccagnella, J. Wylie, K. Forward, and G. Horsman of the Provincial Health Laboratories in Alberta, British Columbia, Manitoba, Nova Scotia, and Saskatchewan, respectively, for providing strains. Members of the NML DNA Core Facility performed sequencing reactions and oligonucleotide synthesis, and members of the NML Serotyping, Identification, and Molecular Typing Programs conducted the initial strain characterization. We also thank V. Goleski for assistance with the Luminex assays, J. McCrea for assistance with LUX assays, and S. Tyler for assistance with stx2 characterization.
This work was supported by Canadian Biotechnology Strategy Funds, administered by the Office of Biotechnology and Science, and Health Canada.
REFERENCES
Bertin, Y., K. Boukhors, V. Livrelli, and C. Martin. 2004. Localization of the insertion site and pathotype determination of the locus of enterocyte effacement of Shiga toxin-producing Escherichia coli strains. Appl. Environ. Microbiol. 70:61-68.
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.
Castillo, A., L. E. Eguiarte, and V. Souza. 2005. A genomic population genetics analysis of the pathogenic enterocyte effacement island in Escherichia coli: the search for the unit of selection. Proc. Natl. Acad. Sci. USA 102:1542-1547.
Cherla, R. P., S. Y. Lee, and V. L. Tesh. 2003. Shiga toxins and apoptosis. FEMS Microbiol. Lett. 228:159-166.
DebRoy, C., E. Roberts, J. Kundrat, M. A. Davis, C. E. Briggs, and P. M. Fratamico. 2004. Detection of Escherichia coli serogroups O26 and O113 by PCR amplification of the wzx and wzy genes. Appl. Environ. Microbiol. 70:1830-1832.
Donnenberg, M. S., and T. S. Whittam. 2001. Pathogenesis and evolution of virulence in enteropathogenic and enterohemorrhagic Escherichia coli. J. Clin. Investig. 107:539-548.
Feng, P., K. A. Lampel, H. Karch, and T. S. Whittam. 1998. Genotypic and phenotypic changes in the emergence of Escherichia coli O157:H7. J. Infect. Dis. 177:1750-1753.
Frankel, G., A. D. Phillips, I. Rosenshine, G. Dougan, J. B. Kaper, and S. Knutton. 1998. Enteropathogenic and enterohaemorrhagic Escherichia coli: more subversive elements. Mol. Microbiol. 30:911-921.
Gartner, J. F., and M. A. Schmidt. 2004. Comparative analysis of locus of enterocyte effacement pathogenicity islands of atypical enteropathogenic Escherichia coli. Infect. Immun. 72:6722-6728.
Gilmour, M. W., T. D. Cote, J. Munro, L. Chui, J. Wylie, J. Isaac-Renton, G. Horsman, D. M. Tracz, A. Andrysiak, and L.-K. Ng. 2005. Multilocus sequence typing of Escherichia coli O26:H11 isolates carrying stx in Canada does not identify genetic diversity. J. Clin. Microbiol. 43:5319-5323.
Hsu, C. F., T. Y. Tsai, and T. M. Pan. 2005. Use of the duplex TaqMan PCR system for detection of Shiga-like toxin-producing Escherichia coli O157. J. Clin. Microbiol. 43:2668-2673.
Huson, D. H. 1998. SplitsTree: analyzing and visualizing evolutionary data. Bioinformatics 14:68-73.
Kaderali, L., A. Deshpande, J. P. Nolan, and P. S. White. 2003. Primer-design for multiplexed genotyping. Nucleic Acids Res. 31:1796-1802.
Kanack, K. J., J. A. Crawford, I. Tatsuno, M. A. Karmali, and J. B. Kaper. 2005. SepZ/EspZ is secreted and translocated into HeLa cells by the enteropathogenic Escherichia coli type III secretion system. Infect. Immun. 73:4327-4337.
Karmali, M. A., M. Mascarenhas, S. Shen, K. Ziebell, S. Johnson, R. Reid-Smith, J. Isaac-Renton, C. Clark, K. Rahn, and J. B. Kaper. 2003. Association of genomic O island 122 of Escherichia coli EDL 933 with verocytotoxin-producing Escherichia coli seropathotypes that are linked to epidemic and/or serious disease. J. Clin. Microbiol. 41:4930-4940.
Kumar, S., K. Tamura, and M. Nei. 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Brief. Bioinform. 5:150-163.
Mead, P. S., L. Slutsker, V. Dietz, L. F. McCaig, J. S. Bresee, C. Shapiro, P. M. Griffin, and R. V. Tauxe. 1999. Food-related illness and death in the United States. Emerg. Infect. Dis. 5:607-625.
Mellmann, A., M. Bielaszewska, L. B. Zimmerhackl, R. Prager, D. Harmsen, H. Tschape, and H. Karch. 2005. Enterohemorrhagic Escherichia coli in human infection: in vivo evolution of a bacterial pathogen. Clin. Infect. Dis. 41:785-792.
Nataro, J. P., and J. B. Kaper. 1998. Diarrheagenic Escherichia coli. Clin. Microbiol. Rev. 11:142-201.
Nazarenko, I., B. Lowe, M. Darfler, P. Ikonomi, D. Schuster, and A. Rashtchian. 27 February 2002, revision date. Multiplex quantitative PCR using self-quenched primers labeled with a single fluorophore. Nucleic Acids Res. 30:e37. http://nar.oxfordjournals.org/cgi/content/full/30/9/e37.
Nielsen, E. M., and M. T. Andersen. 2003. Detection and characterization of verocytotoxin-producing Escherichia coli by automated 5' nuclease PCR assay. J. Clin. Microbiol. 41:2884-2893.
Nikiforov, T. T., R. B. Rendle, M. L. Kotewicz, and Y. H. Rogers. 1994. The use of phosphorothioate primers and exonuclease hydrolysis for the preparation of single-stranded PCR products and their detection by solid-phase hybridization. PCR Methods Appl. 3:285-291.
Oswald, E., H. Schmidt, S. Morabito, H. Karch, O. Marches, and A. Caprioli. 2000. Typing of intimin genes in human and animal enterohemorrhagic and enteropathogenic Escherichia coli: characterization of a new intimin variant. Infect. Immun. 68:64-71.
Paton, A. W., and J. C. Paton. 1998. Detection and characterization of Shiga toxigenic Escherichia coli by using multiplex PCR assays for stx1, stx2, eaeA, enterohemorrhagic E. coli hlyA, rfbO111, and rfbO157. J. Clin. Microbiol. 36:598-602.
Paton, J. C., and A. W. Paton. 1998. Pathogenesis and diagnosis of Shiga toxin-producing Escherichia coli infections. Clin. Microbiol. Rev. 11:450-479.
Perelle, S., F. Dilasser, J. Grout, and P. Fach. 2005. Detection of Escherichia coli serogroup O103 by real-time polymerase chain reaction. J. Appl. Microbiol. 98:1162-1168.
Perna, N. T., G. F. Mayhew, G. Posfai, S. Elliott, M. S. Donnenberg, J. B. Kaper, and F. R. Blattner. 1998. Molecular evolution of a pathogenicity island from enterohemorrhagic Escherichia coli O157:H7. Infect. Immun. 66:3810-3817.
Perna, N. T., G. Plunkett III, V. Burland, B. Mau, J. D. Glasner, D. J. Rose, G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J. Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis, A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Anantharaman, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch, and F. R. Blattner. 2001. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409:529-533.
Reischl, U., M. T. Youssef, J. Kilwinski, N. Lehn, W. L. Zhang, H. Karch, and N. A. Strockbine. 2002. Real-time fluorescence PCR assays for detection and characterization of Shiga toxin, intimin, and enterohemolysin genes from Shiga toxin-producing Escherichia coli. J. Clin. Microbiol. 40:2555-2565.
Rozas, J., J. C. Sanchez-DelBarrio, X. Messeguer, and R. Rozas. 2003. DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics 19:2496-2497.
Shen, S., M. Mascarenhas, R. Morgan, K. Rahn, and M. A. Karmali. 2005. Identification of four fimbria-encoding genomic islands that are highly specific for verocytotoxin-producing Escherichia coli serotype O157 strains. J. Clin. Microbiol. 43:3840-3850.
Tarr, C. L., T. M. Large, C. L. Moeller, D. W. Lacher, P. I. Tarr, D. W. Acheson, and T. S. Whittam. 2002. Molecular characterization of a serotype O121:H19 clone, a distinct Shiga toxin-producing clone of pathogenic Escherichia coli. Infect. Immun. 70:6853-6859.
Thorpe, C. M. 2004. Shiga toxin-producing Escherichia coli infection. Clin. Infect. Dis. 38:1298-1303.
Toma, C., E. Martinez Espinosa, T. Song, E. Miliwebsky, I. Chinen, S. Iyoda, M. Iwanaga, and M. Rivas. 2004. Distribution of putative adhesins in different seropathotypes of Shiga toxin-producing Escherichia coli. J. Clin. Microbiol. 42:4937-4946.
Tyler, S. D., W. M. Johnson, H. Lior, G. Wang, and K. R. Rozee. 1991. Identification of verotoxin type 2 variant B subunit genes in Escherichia coli by the polymerase chain reaction and restriction fragment length polymorphism analysis. J. Clin. Microbiol. 29:1339-1343.
Wang, G., C. G. Clark, and F. G. Rodgers. 2002. Detection in Escherichia coli of the genes encoding the major virulence factors, the genes defining the O157:H7 serotype, and components of the type 2 Shiga toxin family by multiplex PCR. J. Clin. Microbiol. 40:3613-3619.
Wick, L. M., W. Qi, D. W. Lacher, and T. S. Whittam. 2005.
Evolution of genomic content in the stepwise emergence of Escherichia coli O157:H7. J. Bacteriol. 187:1783-1791.
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.
Zadik, P. M., P. A. Chapman, and C. A. Siddons. 1993. Use of tellurite for the selection of verocytotoxigenic Escherichia coli O157. J. Med. Microbiol. 39:155-158.
Zhang, W. L., B. Kohler, E. Oswald, L. Beutin, H. Karch, S. Morabito, A. Caprioli, S. Suerbaum, and H. Schmidt. 2002. Genetic diversity of intimin genes of attaching and effacing Escherichia coli strains. J. Clin. Microbiol. 40:4486-4492.
National Microbiology Laboratory, Public Health Agency of Canada, Winnipeg, Manitoba, Canada(Matthew W. Gilmour, Dobry)