Whole-Genome Sequences of Agricultural, Host-Associated Campylobacter coli and Campylobacter jejuni Strains.

We report here the genome sequences of four agricultural, multidrug-resistant Campylobacter spp.: C. coli 11601 and C. jejuni 11601MD, isolated from turkey cecum and jejunum, respectively, and C. coli 6067 and C. coli 6461, isolated from turkey-house water and swine feces, respectively. The genomes provide insights on Campylobacter antimicrobial resistance and host adaptations.

GENOME ANNOUNCEMENT

Campylobacter jejuni and C. coli are leading agents of human foodborne disease that frequently colonize the mammalian and avian intestine (1–3). While certain Campylobacter lineages colonize diverse animal hosts, others exhibit preference for certain hosts or environments (3–7). Antimicrobial resistance (AMR) is frequently encountered among Campylobacter spp. (8). Much remains to be elucidated about the genomic basis of host preference and AMR in Campylobacter spp. of agricultural origin.

We determined draft genomes of four MDR strains from a longitudinal study of Campylobacter spp. from conventional turkey and swine farms in eastern North Carolina (9). C. coli 11601, resistant to tetracycline (T), streptomycin (S), erythromycin (E), kanamycin (K), nalidixic acid (N) and ciprofloxacin (C), was isolated in 2006 from the cecum of a turkey. On the other hand, C. jejuni 11601MD (resistant to T, K, N, C) was isolated from the jejunum of the same turkey. C. coli 6067 (resistant to T, N, C) was isolated in 2003 from drinker water in a turkey house and belonged to the turkey-associated "cluster II" lineage (10), while C. coli 6461 (resistant to T, S, E) was isolated in 2004 from swine feces and its DNA resisted digestion by MboI, an attribute found frequently in C. coli from swine, but not turkeys (11).

Genomic DNA was extracted using the DNeasy blood and tissue kit (QIAGEN, Valencia, CA, USA). DNA library preparation used the GS FLX Titanium rapid library preparation kit (Roche, Basel, Switzerland), and sequencing employed the 454 GS FLX high-throughput DNA sequencer (Roche). The raw reads were de novo assembled using Newbler version 1.1.03.24 with >30× estimated coverage, <77 contigs, N50 > 135,251 bp, providing draft genomes with total sizes of 1,695,750, 1,792,449, 1,957,712, and 1,736,776 bp for C. coli (6067, 6461, and 11601) and C. jejuni (11601MD) strains, respectively. Annotations were performed using the NCBI Prokaryotic Genome Annotation Pipeline (http://www.ncbi.nlm.nih.gov/genome/annotation_prok) and an updated version of the GAMOLA annotation suite (12). Genome annotation identified 1,618, 1,770, 1,939, 1,695 coding sequences, 76 to 123 pseudogenes, 1 to 4 rRNAs, and 35 to 36 tRNAs in C. coli 6067, 6461, and 11601 and C. jejuni 11601MD, respectively. In C. coli 11601 and C. jejuni 11601MD, the tetracycline resistance determinant tet(O) was plasmid-borne, with one contig of 11601MD harboring exclusively plasmid-associated sequences (13). As previously reported (14), tet(O) was found in the chromosome in C. coli 6067 and (in a different chromosomal locus) in C. coli 6461. Plasmids were also detected in C. coli 6067 and 6461.

Potential host association and AMR attributes of such agricultural strains remain poorly understood, with especially limited information from eastern North Carolina, a major turkey- and swine- producing region in the United States. The genome sequences in this announcement can be used to further elucidate population structure, adaptations, and evolutionary biology of agricultural C. coli and C. jejuni and inform analyses of Campylobacter transmission from agricultural reservoirs to the human food supply.

Accession number(s).

This whole-genome shotgun project has been deposited in DDBJ/ENA/GenBank under the accession numbers LKCQ00000000, LKCR00000000, LKCS00000000, and LKCT00000000. The versions described in this paper are the first versions.