Complete Genome Sequences of Three Shiga Toxin-Producing Escherichia coli O111:H8 Strains Exhibiting an Aggregation Phenotype

Non-O157 Shiga toxin-producing Escherichia coli (STEC) strains are a common source of foodborne illness. STEC O111 is among the most prevalent non-O157 STEC serogroups.

ABSTRACT

Non-O157 Shiga toxin-producing Escherichia coli (STEC) strains are a common source of foodborne illness. STEC O111 is among the most prevalent non-O157 STEC serogroups. Few completed genomes of STEC O111 strains have been reported to date. We report here the complete genomic sequences of three O111:H8 strains that display a distinct aggregation phenotype.

ANNOUNCEMENT

Foodborne infections with Shiga toxin-producing Escherichia coli (STEC) strains are a global burden, being responsible worldwide for nearly 3 million cases of acute illness annually (1). Non-O157 STEC infections cause nearly 113,000 illnesses yearly in the United States (2), with STEC O111 infections accounting for about 19% of these illness cases between 2000 and 2010 (3). Many STEC O111 infections in the United States are sporadic; however, there have been 15 well-documented STEC O111 outbreaks in the United States since 1999 (4, 5), several of which were at least partially associated with produce (6–9). This announcement describes the complete genome sequences of three STEC O111 strains. Strains RM9131 and RM9975 were isolated in a major agricultural region for vegetable production in California from the intestinal contents of a cow and an American crow, respectively. Strain SD134209 was isolated from a patient with bloody diarrhea during an outbreak at a daycare facility in South Dakota (10). These three strains exhibited a strong aggregative phenotype characterized by the settlement of large bacterial assemblages in broth culture with agitation, a behavior commonly exhibited by environmental and outbreak STEC O111 isolates (11).

A single colony for each STEC O111 strain was grown overnight on LB agar plates at 37°C, and genomic DNA was extracted using a sucrose-Tris method, followed by phenol-chloroform cleanup, as described previously (12). Sequencing was performed using a Pacific Biosciences RS II platform (Menlo Park, CA), with 20-kb SMRTbell libraries, and assembled using PacBio HGAP version 3.0 in the SMRT Analysis package (version 2.3.0), as described previously (13). Additionally, sequencing was performed on a MiSeq platform (Illumina, Inc., San Diego, CA) with an LTP library preparation kit (Kapa Biosystems, Wilmington, MA), a MiSeq reagent kit version 2 (Illumina, Inc.), and 2 × 250-bp cycles, as described previously (13). A final base call validation of the PacBio contigs was performed using Illumina MiSeq reads trimmed using a quality score threshold of 20 or higher (≥Q20) and the reference assembler within Geneious (version 11.1; Biomatters, Ltd., Auckland, New Zealand). Protein-, rRNA-, and tRNA-coding genes were annotated using the NCBI Prokaryotic Genomes Automatic Annotation Pipeline (PGAAP) (14).

These three STEC O111:H8 strains exhibiting an aggregation phenotype were sequenced and shown to possess wild-type rpoS genes and curli operons, which are required for their aggregative behavior (15). The plasmid pAA and aggregative adherence fimbriae (AAF)-encoding genes, characteristic of enteroaggregative E. coli (16), were absent from all three STEC O111:H8 genomes. Each strain possessed a plasmid containing a hemolysin gene similar to that in other STEC strains (17). All three strains harbored a stx₁-containing bacteriophage located adjacent to the transfer-messenger RNA (tmRNA) gene ssrA. The clinical isolate SD134209 also possessed a stx₂-containing bacteriophage adjacent to the tRNA^Arg gene argW, while neither RM9131 nor RM9975 carried any bacteriophage adjacent to argW or stx₂ within the genome. The sequencing data of these STEC strains will add to our understanding of STEC O111 genomics, biology, and epidemiology and provide references for other STEC O111 sequencing projects.

Data availability.

The chromosomal and plasmid sequences for each strain have been deposited in GenBank, and the accession numbers are listed in Table 1.

TABLE 1.

Sequencing metrics and genomic data for Escherichia coli strains RM9131, RM9975, and SD134209

Feature	RM9131	RM9975	SD134209
Source	Cattle feces	American crow	Human
BioProject no.	PRJNA448153	PRJNA448155	PRJNA473579
SRA accession no.	SRP161702	SRP161706	SRP161696
Sequencing metrics
Illumina MiSeq platform
No. of reads	2,672,184	2,762,958	4,365,584
Avg length (bases)	223	224	200
Coverage (×)	113	119	154
PacBio platform
No. of reads	133,152	117,271	255,006
Avg length (bases)	7,246	7,335a	6,881
Coverage (×)	178	159	318
Genomic data
Size (bp)a
Chromosome	5,238,870	5,233,637	5,359,785
plasmid1	117,123	120,253	188,342
plasmid2	76,698	78,526	78,329
GenBank accession no.
Chromosome	CP028429	CP028432	CP029692
plasmid1	CP028430	CP028433	CP029690
plasmid2	CP028431	CP028434	CP029691
G+C content (%)	50.6	50.6	50.6
No. of CDSa	5,306	5,313	5,557
Plasmid-carried genes
plasmid1	Phage genes	Phage genes	aph6 str tet(B)b
plasmid2	Hemolysin	Hemolysin	Hemolysin

^aNumbers do not include pseudogenes; CDS, coding sequences.

^baph6, APH(6)-I family aminoglycoside O-phosphotransferase gene; str, streptomycin phosphotransferase gene; tet(B), tetracycline efflux transporter gene.