Whole-Genome Draft Sequences of Six Commensal Fecal and Six Mastitis-Associated Escherichia coli Strains of Bovine Origin.

The bovine gastrointestinal tract is a natural reservoir for commensal and pathogenic Escherichia coli strains with the ability to cause mastitis. Here, we report the whole-genome sequences of six E. coli isolates from acute mastitis cases and six E. coli isolates from the feces of udder-healthy cows.

GENOME ANNOUNCEMENT

Although bovine intramammary infections with Escherichia coli mostly lead to an acute onset of mastitis, they can also result in a persistent infection of the udder with alternating subclinical or clinical periods (1). Additionally, no common virulence factor subset of mastitis-causing E. coli strains has been identified in previous studies (2).

To investigate the genomic potential of E. coli isolated from bovine mastitis, several draft genomes (3–5), as well as two complete genomes (6), have been published thus far. However, only two recent genomic E. coli mastitis studies included one commensal bovine isolate (7, 8). Because cows are a natural reservoir not only for pathogenic but also for commensal E. coli of high phylogenetic and genotypic diversity (2), we present here the draft genomes of six E. coli strains isolated from serous udder exudate of mastitis-afflicted cows and six E. coli strains isolated from the feces of udder-healthy cows (Table 1).

TABLE 1

Genome features and assembly metrics of the 12 E. coli whole-genome sequences

Strain	ECOR phylogroup(ST)	Source of isolation	Genome size (bp)	No. of contigs >500 bp	N50 (bp)	No. of CDSsa	Accession no.
131/07	A (ST10)	Udder acute mastitis	5,459,392	270	79,414	5,123	JXUH00000000
2772a	B1 (ST156)	Udder acute mastitis	4,949,901	93	163,837	4,621	LCVG00000000
3234/A	A (ST10)	Udder acute mastitis	5,482,981	290	95,923	5,211	LCVH00000000
MPEC4839	A (ST10)	Udder acute mastitis	4,866,885	124	133,521	4,502	JYHP00000000
MPEC4969	B1 (ST1125)	Udder acute mastitis	4,833,611	130	103,834	4,468	JYHQ00000000
RiKo 2299/09	B1 (ST448)	Healthy cow feces	4,954,750	129	114,991	4,587	JYKB00000000
RiKo 2305/09	B1 (ST410)	Healthy cow feces	4,806,931	123	129,952	4,429	JYPB00000000
RiKo 2308/09	A (ST167)	Healthy cow feces	5,112,873	186	83,735	4,685	LCVI00000000
RiKo 2331/09	B1 (ST1614)	Healthy cow feces	4,765,494	59	224,192	4,350	LCVJ00000000
RiKo 2340/09	A (ST167)	Healthy cow feces	5,024,854	204	82,522	4,568	LAGW00000000
RiKo 2351/09	B1 (ST88)	Healthy cow feces	5,297,190	252	102,610	4,931	LAUC00000000
UVM2	A (ST1091)	Udder acute mastitis	4,926,170	149	86,033	4,614	LAUD00000000

CDS, coding sequence.

All genomes were sequenced with an Illumina HiScan SQ sequencer with Nextera XT chemistry (Illumina, San Diego, CA, USA) for library preparation and a 101-bp paired-end sequencing run. Raw reads were quality controlled with FastQC version 0.11.2 (http://www.bioinformatics.bbsrc.ac.uk/projects/fastqc). Low-quality reads and adapter contaminations were trimmed with Cutadapt version 1.6 (9). All reads were randomly subsampled to an approximate 70-fold coverage for each strain with seqtk version 1.0-r32 (https://github.com/lh3/seqtk). Subsequently, the reads were de novo assembled with SPAdes version 3.1.1 (10). Assembly statistics were evaluated with QUAST version 2.3 (11), resulting in 59 to 290 contigs >500 bp and genome sizes ranging from 4,765,494 to 5,459,392 bp (Table 1).

The strains were classified evenly into phylogroups A or B1, regardless of isolation source, through the assignment of sequence types (ST) with e. coli_mlst version 0.3 (https://github.com/aleimba/bac-genomics-scripts/tree/master/ecoli_mlst) (12). The most prominent sequence type is ST10, but most of the strains were not closely phylogenetically related.

All genomes were annotated with Prokka version 1.9 (13) with E. coli 1303 (CP009166 to CP009169) or E. coli ECC-1470 (CP010344 to CP010345) as references for annotation for either the ECOR phylogroup A or B1 genomes, respectively. tRNAs were predicted with tRNAscan-SE version 1.3.1 (14). Additionally, the annotations were manually curated with Proteinortho version 5.11 (15), po2anno version 0.2 (https://github.com/aleimba/bac-genomics-scripts/tree/master/po2anno), ACT version 13.0.0 (16), and E. coli strains 1303 and ECC-1470 as references. Finally, tbl2tab version 0.2 (https://github.com/aleimba/bac-genomics-scripts/tree/master/tbl2tab) and Artemis version 16.0.0 (17) were used to refine the annotations after querying the Virulence Factors Database (18) and the ResFinder version 2.1 (19), VirulenceFinder version 1.2 (20), and SerotypeFinder version 1.0 (21) databases. In summary, between 4,350 and 5,211 coding DNA sequences were identified in the genomes with 3 to 7 rRNAs and 68 to 83 tRNAs.

The genome sequences in this study will serve as a useful resource for future comparative studies of E. coli strains associated with bovine mastitis in relationship to commensal strains and for the identification of potential virulence factors.

Nucleotide sequence accession numbers.

These whole-genome shotgun projects have been deposited at DDBJ/EMBL/GenBank under the accession numbers listed in Table 1. The versions described here are the first versions.