Genome Sequences of Four Yersinia enterocolitica Bioserotype 4/O:3 Isolates from Mammals

We report here the complete genome sequences of four European Yersinia enterocolitica mammalian isolates of bioserotype 4/O:3. The genomes have an average size of 4.50 Mb, a G+C content of 47%, and between 4,231 and 4,330 coding sequences (CDSs). No relevant differences were detected by genome comparison between mammalian and human isolates.

in mammals, with humans being mostly accidental hosts. In humans, after the ingestion of contaminated food or water, Y. enterocolitica colonizes the intestines, most frequently causing acute gastroenteritis with fever, vomiting, and diarrhea (1).
Y. enterocolitica consists of a biochemically and genetically heterogeneous group of organisms, divided into 6 biotypes (the nonvirulent 1A, the highly virulent 1B, and the low-virulence 2, 3, 4, and 5 biotypes) and Ͼ70 serotypes. Human infections with Y. enterocolitica are documented worldwide and are mostly caused by strains belonging to the pathogenic bioserotype 4/O:3 (2). Y. enterocolitica has been isolated from mammals, birds, and other animal species, as well as from the environment. Swine are the primary reservoir for food-borne illness associated with Y. enterocolitica, mainly of that of bioserotype 4/O:3 (3).
Genomes of Y. enterocolitica isolates of different bioserotypes have been sequenced, allowing the identification of serotypespecific features (4-9). Among these, four European clinical isolates (4, 10) and one Philippine isolate from swine (11), all belonging to the prevailing bioserotype 4/O:3, are currently available. Whole-genome sequences of Y. enterocolitica 4/O:3 isolates from animal sources would provide a comprehensive knowledge of the epidemiology and transmission of this frequently encountered bioserotype.
Four Y. enterocolitica strains of bioserotype 4/O:3 were selected, two isolates from pig (Y. enterocolitica YE-P1 and YE-P4), one dog isolate (Y. enterocolitica YE-149), and one strain isolated from calf (Y. enterocolitica YE-150). For each strain, a 150-bp paired-end library was constructed and used for whole-genome sequencing by the Illumina MiSeq technology (IMGM Laboratories, Martinsried, Germany). The run produced from 2.02 to 5.41 million reads, having an average length ranging from 142.44 to 147.34 bp and an average Phred quality score of 37.
Mapping assemblies using the published complete genome sequence of Y. enterocolitica strain Y11 (accession no. FR729477.2 for the chromosome and FR745874 for the plasmid) were performed by CLC Genomics Workbench version 6.0.2 (CLC bio, Aarhus, Denmark). On average, 94.8% of the reads were mapped to the bacterial chromosome, while 1.5% and 2.1% of the reads from strains YE-P1 and YE-149, respectively, were mapped against the plasmid. No mapping against the plasmid was performed in strains YE-P4 and YE-150, since they probably lost the pYV plasmid after subculturing. The average coverage ranged between 61.5ϫ and 159.3ϫ. We obtained 86 to 100 contigs (Ͼ200 bp in length) for each genome, with a total draft genome size of 4,464,171 to 4,550,830 bp and a GϩC content of 47%. Between 4,245 and 4,330 coding sequences (CDSs) and 62 to 64 tRNAs were predicted by genome annotation with Rapid Annotations using Subsystems Technology (RAST) (12).
Hypothetical proteins and prophages are the solely genetic differences identified by preliminary genomic comparison between pig and human Y. enterocolitica 4/O:3 isolates. Single nucleotide polymorphism analysis and detailed genome comparison will clarify whether there are different epidemiological origins between clinical and animal isolates.

ACKNOWLEDGMENT
This work was supported by the German Bundesministerium für Bildung und Forschung (BMBF) Network, grant FBI-Zoo.