Clustered Regularly-interspaced Repeats Research Articles

Genetic Diversity of Non-O157 Shiga Toxin-Producing Escherichia coli Recovered From Patients in Michigan and Connecticut.

Shiga toxin-producing Escherichia coli (STEC) are important foodborne pathogens and non-O157 serotypes have been gradually increasing in frequency. The non-O157 STEC population is diverse and is often characterized using serotyping and/or multilocus sequence typing (MLST). Although spacers within clustered regularly interspaced repeat (CRISPR) regions were shown to comprise horizontally acquired DNA elements, this region does not actively acquire spacers in STEC. Hence, it is useful for further characterizing non-O157 STEC and examining relationships between strains. Our study goal was to evaluate the genetic relatedness of 41 clinical non-O157 isolates identified in Michigan between 2001 and 2005 while comparing to 114 isolates from Connecticut during an overlapping time period. Whole genome sequencing (WGS) was performed, and sequences were extracted for serotyping, MLST and CRISPR analysis. Phylogenetic analysis of MLST and CRISPR data was performed using the Neighbor joining and unweighted pair group method with arithmetic mean (UPGMA) algorithms, respectively. In all, 29 serogroups were identified; eight were unique to Michigan and 13 to Connecticut. “Big-six” serogroup frequencies were similar by state (Michigan: 73.2%, Connecticut: 81.6%), though STEC O121 was not found in Michigan. The distribution of sequence types (STs) and CRISPR profiles was also similar across states. Interestingly, big-six serogroups such as O103 and O26, grouped into different STs located on distinct branches of the phylogeny, further confirming that serotyping alone is not adequate for evaluating strain relatedness. Comparatively, the CRISPR analysis identified 361 unique spacers that grouped into 80 different CRISPR profiles. CRISPR spacers 231 and 317 were isolated from 79.2% (n = 118) and 59.1% (n = 88) of strains, respectively, regardless of serogroup and ST. Spacer profiles clustered according to the MLST analysis, though some discrepancies were noted. Indeed, use of both MLST and CRISPR typing enhanced the discriminatory power when compared to the use of each tool separately. These data highlight the genetic diversity of clinical STEC from different locations and show that CRISPR profiling can be used alongside MLST to discriminate related strains. Use of targeted sequencing approaches are particularly helpful for sites without WGS capabilities and can help define which strains require additional characterization using more discriminatory methods.

Genetic Variation in Shiga Toxin-producing Escherichia coli Recovered from Patients in Michigan and Connecticut

BackgroundShiga toxin-producing Escherichia coli (STEC) is a gram-negative foodborne pathogen that causes approximately 265,000 illnesses in the US annually. STEC O157 and six non-O157 STEC serotypes are most commonly associated with illness, though variation exists in the ability of different STEC types to cause disease. Consequently, we sought to examine genetic variation in virulence genes and clustered regularly interspaced repeat (CRISPR) loci among clinical strains of diverse lineages from different geographic locations.MethodsIsolates were collected from a sentinel surveillance in 2000–2006 by the Michigan Department of Health and Human Services (n = 44) and Connecticut Department of Public Health (n = 115). Whole genome sequencing was performed and genes for O-antigen (serotype), multilocus sequence typing (MLST) and virulence factors were extracted. CRISPRFinder and Geneious were used for CRISPR loci.ResultsA phylogenetic tree based on MLST found no geographic clustering of the strains. Similarly, no difference was observed for stx1 (MI: 89.6%, CT: 83.5%), stx2 (MI: 6.9%, CT: 7.1%), stx1/stx2 (MI: 3.4%, CT: 9.4%), ehxA (MI: 80.0%, CT: 81.2%) and eae (MI: 86.7%, CT: 94.1%) frequencies across geographical locations. Although the CRISPR loci were similar within related STs, which was independent of serotype, some variation was detected between locations.ConclusionThese data highlight the circulation of common non-O157 STEC lineages capable of causing disease in different populations. Strains had similar virulence gene profiles, though the diversity of the CRISPR loci varied across strains. The latter could be impacted by varying selective pressures that could affect disease frequencies and symptom severity. Continued surveillance of non-O157 STEC is needed to elucidate the genetic characteristics that are most important for disease severity.Disclosures All authors: No reported disclosures.