Genome Sequences of 64 Non-O157:H7 Shiga Toxin-Producing Escherichia coli Strains

Shiga toxin-producing Escherichia coli (STEC) strains are human pathogens. Although >400 non-O157 serotypes have been involved in human disease, whole-genome sequencing information is missing for many serotypes. We sequenced 64 STEC strains comprising 38 serotypes, isolated from clinical sources, animals, and environmental samples, to improve the phylogenetic understanding of these important foodborne pathogens.

serotype, E. coli O157:H7, was associated with mild to severe human disease and outbreaks (1). Since then, its diversity has been widely studied (2). However, not only O157:H7 has been implicated with human diseases, as Ͼ400 STEC serotypes have also been found to be responsible for them (3). In the United States, six serogroups, O26, O45, O103, O111, O121, and O145, cause 70% of non-O157 STEC diseases (4). Although some of these serotypes have been described genetically (5), many more strains still have not been assigned to serogroups or sequenced.
Sixty-four STEC cultures were grown aerobically overnight in tryptic soy agar (TSA) at 37°C, and then DNA was extracted using the DNeasy blood and tissue kit (Qiagen, Valencia, CA). Libraries were prepared using the Nextera XT kit (Illumina, San Diego, CA) and 1 ng of genomic DNA. Sequences were obtained with the MiSeq Illumina V2 (2 ϫ 250 bp) or V3 kit (2 ϫ 300 bp), according to the manufacturer's instructions, and de novo-assembled sequences were generated using CLC Genomics Workbench version 7.6.1 (CLC bio, Germantown, MD, USA). Strains were sequenced to a coverage depth ranging from 34 to 118ϫ. We used Ridom SeqSphereϩ for in silico MLST analysis, and resulting sequences were annotated using the NCBI Prokaryotic Genomes Annotation Pipeline (PGAP) (http: //www.ncbi.nlm.nih.gov/genome/annotation_prok).
These genomes varied from 4.7 to 5.4 Mb; the number of contigs per assembly ranged from 52 to 303 (data not shown). Most strains represented 27 previously described sequence types (STs); seven strains were novel STs (Table 1). Using the CGE server at the Technical University of Denmark (10), we ran in silico analyses of sero-types and for the presence of virulence genes (stx, eae, and exhA). Forty-eight strains (75%) carried stx 1 , and 41 strains (64%) carried stx 2. A total of 25 strains (39%) carried both stx genes (Table 1). Twenty-five strains (39%) carried four different subtypes of eae: beta1, epsilon, gamma1, and theta2. The plasmid-borne exhA gene was detected in 52 strains (81%). The information we report here will help better understand the evolution of these emergent foodborne pathogens (non-O157 STECs) and improve the accuracy of foodrelated trace-back investigations outbreaks caused by these pathogens.
Nucleotide sequence accession numbers. The draft genome sequences for these 64 STEC strains are available in GenBank and are listed in Table 1.

ACKNOWLEDGMENTS
The study was supported by the FDA Foods Program Intramural Funds and the ORISE Fellowship Program.
The findings and conclusions in this report are those of the authors and do not necessary represent the official position of the Food and Drug Administration.
We thank Lili Fox Vélez for editorial support.