A commensal gone bad: complete genome sequence of the prototypical enterotoxigenic Escherichia coli strain H10407

Abstract

In most cases, Escherichia coli exists as a harmless commensal organism, but it may on occasion cause intestinal and/or extraintestinal disease. Enterotoxigenic E. coli (ETEC) is the predominant cause of E. coli-mediated diarrhea in the developing world and is responsible for a significant portion of pediatric deaths. In this study, we determined the complete genomic sequence of E. coli H10407, a prototypical strain of enterotoxigenic E. coli, which reproducibly elicits diarrhea in human volunteer studies. We performed genomic and phylogenetic comparisons with other E. coli strains, revealing that the chromosome is closely related to that of the nonpathogenic commensal strain E. coli HS and to those of the laboratory strains E. coli K-12 and C. Furthermore, these analyses demonstrated that there were no chromosomally encoded factors unique to any sequenced ETEC strains. Comparison of the E. coli H10407 plasmids with those from several ETEC strains revealed that the plasmids had a mosaic structure but that several loci were conserved among ETEC strains. This study provides a genetic context for the vast amount of experimental and epidemiological data that have been published.

FIG. 1.

Circular representation of the E. coli H10407 chromosome. From the outside in, the outer circle 1 marks the positions of regions of difference (mentioned in the text), including prophage (light pink), as well as regions differentially present in other E. coli strains (blue) (see table S1 in the supplemental material). Circle 2 shows the sizes in bp. Circles 3 and 4 show the positions of CDSs transcribed in clockwise and counterclockwise directions, respectively. Genes in circles 3 and 4 are color coded according to the functions of their gene products: dark green, membrane or surface structures; yellow, central or intermediary metabolism; cyan, degradation of macromolecules; red, information transfer/cell division; cerise, degradation of small molecules; pale blue, regulators; salmon pink, pathogenicity or adaptation; black, energy metabolism; orange, conserved hypothetical; pale green, unknown; brown, pseudogenes. Circles 5 and 6 and circles 9 and10 show the positions of E. coli H10407 genes that have orthologues (by reciprocal FASTA analysis) in E. coli K-12 MG1655 (blue) or E. coli 042 (green), respectively. Circles 7 and 8 and circles 11 and 12 show the positions of genes unique to E. coli H10407 compared to E. coli K-12 MG1655 (red) or E. coli 042 (gray), respectively. Circle 13 shows a plot of G+C contents (in a 10-kb window). Circle 14 shows a plot of GC skew ([G − C]/[G + C], in a 10-kb window).

FIG. 2.

Comparison of the genetic contents of the E. coli H10407 chromosome with those of the chromosomes of other sequenced strains of E. coli. (A) Comparison of E. coli H10407 with the three nonpathogenic E. coli strains HS, C, and K-12 revealed that the four strains share a large proportion of common genes. Only 599 E. coli H10407-specific genes were identified. The E. coli H10407-specific CDSs are not thought to be associated with virulence (see the text for details). (B) Comparison of E. coli H10407 with the genome-sequenced ETEC strains E24377A and B7A. The four strains possess 3,553 genes in common; however, the ETEC strains share only 188 genes not present in the commensal strain E. coli HS. The latter genes are not unique to ETEC; they are widely distributed among E. coli strains and are largely present among nonpathogenic strains of E. coli, such as E. coli K-12.

FIG. 3.

Nucleotide sequence comparison of large conjugative-like plasmids from ETEC strains. Plasmid sequences from each strain were concatenated and compared using BLASTn. BLAST matches longer than 250 bp are shown as gray blocks in a comparison between plasmids from E24377A (pETEC_80, pETEC_74, pETEC_73, and pETEC_35), H10407 (pETEC948 and pETEC666), E1392/75 (pETEC1018, pETEC746, and pETEC557), and C921b-1 (pCoo). The shading of the gray blocks is proportional to the BLAST match (minimum, 80% nucleotide identity; maximum, 100% nucleotide identity). Each plasmid is denoted as a black line; the identity of each plasmid is noted above the line, and the source ETEC strain from which the plasmids are derived is given on the left side of the diagram. Coding sequences are depicted by arrows and are colored according to known or predicted functions: blue, virulence related; red, plasmid-related protein; green, outer membrane related (includes conjugal transfer loci); pink, transposase/insertion element related; light blue, regulatory protein; orange, conserved hypothetical protein; uncolored, hypothetical protein. The positions of genes encoding known or predicted virulence-related proteins are denoted by white boxes containing the gene names. In addition, the locus encoding the R64 conjugative pilus and the variant PilV tips is also depicted. The putative origin of replication associated with each of the plasmids is highlighted within yellow-shaded boxes. The chimeric nature of the plasmids is clearly visible, with recombination between plasmids a frequent occurrence. The unlabeled figure was prepared using a custom script (M. J. Sullivan and S. A. Beatson, unpublished data).

FIG. 4.

Comparison of the EAEC aat-aap locus with the aat-cexE loci of ETEC strains. (A) The genetic organizations of the aat and cexE loci are depicted. The level of amino acid identity for each component of the aat-cexE system is shown; the percentages represent comparison with the E. coli H10407 orthologues. Orthologues are colored coded for ease of identification. Genes that are not juxtaposed are depicted with a blue line separating them. (B) Amino acid sequence alignment of ETEC CexE proteins with the EAEC 042 dispersin. All three proteins possess a signal sequence that is cleaved after the amino acid at position 21 in the alignment. There is limited conservation in the sequences; however, two cysteine residues that are disulfide bonded in dispersin are conserved. Based on the structure of dispersin, the remainder of the conserved residues appear to represent hydrophobic core residues required for structural integrity of the molecule. Asterisks indicate positions of amino acid identity; periods and colons show positions of low and high amino acid similarity.

FIG. 5.

Arrangement of the pilV shufflon region of E. coli E1392/75 pETEC746. Annotation of the pilV region is shown using the Artemis sequence viewer (1). Sequence blocks encoding C-terminal fragments of PilV are found in both orientations between pilV and the rci recombinase gene. Identical 13-bp repeats (GTGCCAATCCGGT) are shown as miscellaneous features and mark the predicted sites of recombination between the C-terminal fragments and the pilV gene.