Phylogeny, Prevalence, and Shiga Toxin (Stx) Production of Clinical Escherichia coli O157 Clade 2 Strains Isolated in Shimane Prefecture, Japan.

This study investigated the genetic and pathogenic variation of the subgroups of clade 2 strains of Shiga toxin (Stx)-producing Escherichia coli (STEC) O157. A total of 111 strains of STEC O157 isolated in Shimane prefecture, Japan, were classified in clade 2 (n = 39), clade 3 (n = 16), clade 4/5 (n = 3), clade 7 (n = 14), clade 8 (n = 17), and clade 12 (n = 22) by single-nucleotide polymorphism analysis and lineage-specific polymorphism assay-6. These results showed a distinct difference from our previous study in which clade 3 strains were the most prevalent strains in three other prefectures in Japan, indicating that the clade distribution of O157 strains was different in different geographic areas in Japan. Phylogenetic analysis using insertion sequence (IS) 629 distribution data showed that clade 2 strains formed two clusters, designated 2a and 2b. Stx2 production by cluster 2b strains was significantly higher than by cluster 2a strains (P < 0.01). In addition, population genetic analysis of the clade 2 strains showed significant linkage disequilibrium in the IS629 distribution of the strains in clusters 2a and 2b (P < 0.05). The ΦPT values calculated using the IS629 distribution data indicated that strains in clusters 2a and 2b were genetically different (P < 0.001). Cluster 2b strains are a highly pathogenic phylogenetic group and their geographic spread may be a serious public health concern.

Rapid and Accurate Diagnosis Based on Real-Time PCR Cycle Threshold Value for the Identification of Campylobacter jejuni, astA Gene-Positive Escherichia coli, and eae Gene-Positive E. coli.

We previously developed a multiplex real-time PCR assay (Rapid Foodborne Bacterial Screening 24 ver.5, [RFBS24 ver.5]) for simultaneous detection of 24 foodborne bacterial targets. Here, to overcome the discrepancy of the results from RFBS24 ver.5 and bacterial culture methods (BC), we analyzed 246 human clinical samples from 49 gastroenteritis outbreaks using RFBS24 ver.5 and evaluated the correlation between the cycle threshold (CT) value of RFBS24 ver.5 and the BC results. The results showed that the RFBS24 ver.5 was more sensitive than BC for Campylobacter jejuni and Escherichia coli harboring astA or eae, with positive predictive values (PPV) of 45.5-87.0% and a kappa coefficient (KC) of 0.60-0.92, respectively. The CTs were significantly different between BC-positive and -negative samples (p ＜ 0.01). All RFBS24 ver.5-positive samples were BC-positive under the lower confidence interval (CI) limit of 95% or 99% for the CT of the BC-negative samples. We set the 95% or 99% CI lower limit to the determination CT (d-CT) to discriminate for assured BC-positive results (d-CTs: 27.42-30.86), and subsequently the PPVs (94.7%-100.0%) and KCs (0.89-0.95) of the 3 targets were increased. Together, we concluded that the implication of a d-CT-based approach would be a valuable tool for rapid and accurate diagnoses using the RFBS24 ver.5 system.

Comparison of two methods of bacterial DNA extraction from human fecal samples contaminated with Clostridium perfringens, Staphylococcus aureus, Salmonella Typhimurium, and Campylobacter jejuni.

In this study, 2 methods of DNA extraction were evaluated for use in conjunction with the screening system Rapid Foodborne Bacterial Screening 24 (RFBS24), which employs multiplex real-time SYBR Green polymerase chain reaction (SG-PCR) and can simultaneously detect 24 target genes of foodborne pathogens in fecal DNA samples. The QIAamp DNA Stool mini kit (Qkit) and Ultra Clean Fecal DNA Isolation Kit (Ukit) were used for bacterial DNA extraction from fecal samples artificially inoculated with Clostridium perfringens, Staphylococcus aureus, Salmonella Typhimurium, and Campylobacter jejuni. SG-PCR and simplex real-time quantitative PCR (S-qPCR) analyses revealed higher copy numbers (8-234 times) of DNA in samples obtained using Ukit compared with those obtained using Qkit, resulting in lower cycle threshold values for the Ukit samples of the 4 bacteria on SG-PCR analysis. Fecal DNA samples from patients infected during foodborne outbreaks of Salmonella and Campylobacter were also prepared by Qkit and Ukit methods and subjected to RFBS24 analyses. Higher numbers of RFBS24 bacterial target genes were detected in DNA samples obtained using Ukit compared with those obtained using Qkit. Thus, the higher DNA extraction efficiency of the Ukit method compared with Qkit renders the former more useful in achieving improved detection rates of these 4 bacteria in fecal samples using SG-PCR.

Emergence of enterohemorrhagic Escherichia coli serovar O157 strains in clade 8 with highly similar pulsed-field gel electrophoresis patterns.

Enterohemorrhagic Escherichia coli serovar O157 (O157) strains with highly similar pulsed-field gel electrophoresis (PFGE) patterns were isolated in Japan during 2007 and 2008. Several genetic features related to O157 evolution were investigated to indicate whether homoplasy might have contributed to the highly similar PFGE patterns in these strains. The O157 strains were classified in lineage I/II, as defined by a lineage-specific polymorphism assay-6 with an atypical allele in Z5935 (code: 231111). Analysis of the insertion sites of stx(2) phage in these strains showed that the sites were "occupied" in yehV and "intact" in wrbA, indicating that the strains were derived from "Cluster 1" of "Subgroup C." When a specific single-nucleotide polymorphism in ECs2357 in clade 8 strains was investigated, all of the strains in the present study were confirmed to be clade 8 strains. These results indicated that the O157 strains in this study had common genetic features, suggesting that the highly similar PFGE patterns of these strains were not due to homoplasy. Because no common source of these strains could be identified in 2007 to 2008 in Japan, these strains may have emerged from a unique O157 clade 8 clone and then spread by dissemination in Japan.