Five genes encoding surface-exposed LPXTG proteins are enriched in hospital-adapted Enterococcus faecium clonal complex 17 isolates

Abstract

Most Enterococcus faecium isolates associated with hospital outbreaks and invasive infections belong to a distinct genetic subpopulation called clonal complex 17 (CC17). It has been postulated that the genetic evolution of CC17 involves the acquisition of various genes involved in antibiotic resistance, metabolic pathways, and virulence. To gain insight into additional genes that may have favored the rapid emergence of this nosocomial pathogen, we aimed to identify surface-exposed LPXTG cell wall-anchored proteins (CWAPs) specifically enriched in CC17 E. faecium. Using PCR and Southern and dot blot hybridizations, 131 E. faecium isolates (40 CC17 and 91 non-CC17) were screened for the presence of 22 putative CWAP genes identified from the E. faecium TX0016 genome. Five genes encoding LPXTG surface proteins were specifically enriched in E. faecium CC17 isolates. These five LPXTG surface protein genes were found in 28 to 40 (70 to 100%) of CC17 and in only 7 to 24 (8 to 26%) of non-CC17 isolates (P < 0.05). Three of these CWAP genes clustered together on the E. faecium TX0016 genome, which may comprise a novel enterococcal pathogenicity island covering E. faecium contig 609. Expression at the mRNA level was demonstrated, and immunotransmission electron microscopy revealed an association of the five LPXTG surface proteins with the cell wall. Minimal spanning tree analysis based on the presence and absence of 22 CWAP genes revealed grouping of all 40 CC17 strains together with 18 hospital-derived but evolutionary unrelated non-CC17 isolates in a distinct CWAP-enriched cluster, suggesting horizontal transfer of CWAP genes and a role of these CWAPs in hospital adaptation.

FIG. 1.

The genomic cluster (E. faecium contig 609) enriched in hospital-adapted E. faecium CC17 isolates. (A) The genomic organization of contig 609 as published at DDBJ/EMBL/GenBank. Direct repeats are indicated as dashed boxes in the scale at positions 2030 and 14751 of contig 609. The thymine-to-cytosine point mutation and the 19-bp oligonucleotide insertion are indicated with black arrows. The position and presumed direction of transcription of ORFs have been indicated by open arrows. (B) Alignment of the orf904-905 junction region. The TAA stop codon is depicted in bold for strain TX0016 at position 7688 (allele 1). In alleles 2, 3, and 4, a thymine-to-cytosine point mutation leads to a merged orf904.5. (C) Different alleles in the orf906-907 junction region. Allele 1 is found in 29 E. faecium isolates, including TX0016. Alleles 2, 3, and 4 resulted from three different 19-bp insertions (A, B, and C) and were found in 26, 3, and 2 isolates, respectively. The TAG stop codon is depicted in bold for strain TX0016 at position 9175 (allele 1) and is disrupted by the 19-bp insertion to yield orf906.7.

FIG. 2.

Western blots of protein extracts of E. faecium E135 and E380 isolates probed with mouse anti-904 (left) and mouse anti-906 (right) immune sera. The Western blots show expression of the 904.5 and 906.7 LPXTG proteins in E. faecium E380 (lanes 2 and 4), as depicted by arrows, and not in the negative control, E135 (lanes 1 and 3). Numbers depicted on the left indicate molecular masses in kDa.

FIG. 3.

mRNA expression of orf903, orf905, orf907, orf2351, and orf2430. Depicted is the mRNA of three outbreak-associated isolates, E155, E470, and E745 (lanes 2, 4, and 5), and of four clinical isolates, E380, E1165, E1172, and E1176 (lanes 3 and 6 to 8). Lane 1 contains mRNA of the E135 negative control, a community E. faecium isolate lacking all five CC17-enriched putative CWAP genes. Strains E380 and E1176 (lanes 3 and 8) are deficient for orf2430. (A to E) mRNA expression of the putative CWAP genes orf2351, orf2430, orf903, orf905, and orf907, respectively. (F) Control ddl PCRs on total mRNA preparations in which the reverse transcription reaction was omitted were all negative, demonstrating the absence of DNA contamination. (G) Control ddl reverse transcription-PCRs (internal housekeeping control) with E. faecium-specific ddl primers were all positive. The results are presented as amplified PCR products electrophoresed on the same ethidium bromide-stained 1.5% agarose gel. Lanes M show molecular mass markers.

FIG. 4.

Transmission electron microscope micrographs. E. faecium isolates E135, E380, E470, and E745 were negatively stained and labeled individually with five peptide immune sera directed against the five CC17-enriched LPXTG surface proteins followed by rabbit anti-mouse IgG and protein A-gold (15 nm). (A) E380 E. faecium cells (containing orf904.5 allele 2 and orf906.7 insertion A) incubated with anti-903 mouse immune serum. (B) E380 cells incubated with anti-904 mouse immune serum. (C) E380 cells incubated with anti-906 mouse immune serum. (D) E470 cells incubated with anti-2351 mouse immune serum. (E) E745 cells incubated with anti-2430 mouse immune serum. (F) E745 cells incubated with mouse preimmune serum. (G) E470 cells incubated with only rabbit anti-mouse IgG and protein A-gold (conjugate control). (H) E135 cells (negative control; deficient for orf903-906.7, orf2351, and orf2430) incubated with anti-903 immune serum (incubations with anti-904, -906, -2351, and -2430 immune sera, which also did not display gold labeling, are not shown). Bar, 200 nm; magnification, ×65,000.

FIG. 5.

Distribution of putative CWAP genes among E. faecium isolates. Each black circle represents an E. faecium isolate with a specific number of putative CWAP genes. Horizontal lines indicate the average number of putative CWAP genes in each epidemiological class of isolates. (A) Distribution among distinct epidemiological classes. The epidemiological classes are ordered with increasing clinical relevance. Annotation: Comm_surv, community surveillance isolates from human volunteers not connected to hospitals; Hosp_surv, hospital surveillance isolates from hospitalized patients not associated with enterococcal infection or outbreak; Clin_Isol, isolates from clinical sites from hospitalized patients; Outbreak, isolates from hospital outbreaks. (B) Distribution of putative CWAP genes among CC17 and non-CC17 E. faecium isolates. On average, CC17 isolates contain 21 of the 22 CWAP genes, while non-CC17 isolates contain 15 of the 22 CWAP genes in their genomes.

FIG. 6.

MST based on clustering on the presence and absence of the 22 putative CWAP genes. A categorical coefficient and the eBURST priority rule of the highest number of single-locus changes were used for the clustering. Circles represent putative CWAP gene profiles, and the sizes of the circles indicate the numbers of isolates. Thick short lines connecting two CWAP gene types denote types differing by a single CWAP gene, thin lines connect double CWAP gene variants, and dotted lines indicate the most likely connection between two types differing by more than two CWAP genes (black) or more than four CWAP genes (grey). (A) All 40 CC17 isolates are marked red and cluster together with 18 non-CC17 clinical isolates (marked white) in a CWAP-enriched cluster. (B) The presence of 0, 1 or 2, 3, or 4 or 5 CC17-enriched CWAP genes is indicated with colors depicted in the legend.