Distribution of putative virulence markers in Enterococcus faecium: towards a safety profile review

Distribution of putative virulence markers in Enterococcus faecium: towards a safety profile review Abstract Objectives The criteria for identification of Enterococcus faecium (Efm) with the ability to cause human infections are currently being debated by the European Food Safety Authority (EFSA). Strains that have an MIC of ampicillin of ≤ 2 mg/L and lack IS16/esp/hyl genes should be regarded as safe for use as feed additives in animal nutrition, despite the lack of knowledge about putative virulence marker (PVM) distribution in community Efm. We analysed the distribution of major PVM and ampicillin phenotypes in large Efm collections to investigate further the safety of strains from a public health perspective. Methods Thirty-three PVM were assessed by PCR/sequencing among clonally disparate Efm (n = 328; 1986–2015) from different origins. We analysed ampicillin susceptibility (Etest/broth microdilution) according to EUCAST guidelines, clonal relationship (MLST) and genomic location of PVM (S1-PFGE/hybridization). Results Infection-derived Efm were more enriched in PVM and the increase in ampicillin MIC was positively correlated with an enrichment in different PVM. PVM coding for surface (esp/sgrA/ecbA/complete acm) and pili proteins, or others enhancing colonization (hyl/ptsD/orf1481) or plasticity (IS16), were strongly associated with clinical Efm (mostly clade A1), but also observed in clades A2/B at different rates. ptsD was a good marker of ampicillin-resistant Efm. ptsD, IS16, orf1481, sgrA and hospital variants of complete pili gene clusters are proposed as markers to assess the safety of Efm strains. Conclusions Our study expands on the distribution of PVM in diverse Efm lineages and demonstrates the enrichment in infection-derived strains of PVM not previously included in EFSA’s list of Efm safety criteria. The evidence of relevant Efm infection markers can impact the risk assessment of Efm strains in different public health contexts. Introduction Enterococcus faecium (Efm), a human and animal colonizer, is also a leading cause of hospital-acquired infections, mainly urinary tract and surgical infections, bacteraemia and endocarditis, the incidence of which continues to increase in Europe.1 Studies of the population structure of Efm have revealed that the species is split into two major clades comprising isolates principally associated with hospitalized humans (HA) and community-based individuals (CA), respectively.2 These groups have also been identified using different approaches and have alternatively been named clade A and clade B, epidemic hospital strains (A1) being distinguished from strains from animals and sporadic human infections (A2).3,4 Strains within the clade HA/clade A1 (formerly classified as CC17 by MLST), seem to be enriched in different adaptive features including metabolic capabilities, antibiotic resistance genes and traits enhancing colonization or pathogenicity in comparison with those of CA/clade B.3,5,6 The putative virulence genes most thoroughly studied at the functional level include those encoding a putative glycoside hydrolase (HylEfm) and several Efm surface proteins (Fms) involved in adhesion, biofilm formation and pili assembly, namely Esp (enterococcal surface protein), Acm (adhesin of collagen from Efm), Scm (second collagen adhesin from Efm), SgrA (serine-glutamate-repeat-containing-protein A) and EcbA (Efm-collagen-binding-protein A).6–9 In addition IS16, which is thought to enhance genomic plasticity, has been found to be a putative genetic marker of hospital strains.10 The occurrence of these and other putative virulence markers (PVM) that have recently arisen from genome sequencing projects (e.g. ptsD or orf1481 from a genomic island) and their association with infection-derived strains is lacking in comprehensive Efm collections.11,12 Indeed, the few gene-based studies suggesting that the enrichment of a number of Fms genes in infection-derived strains may favour hospital adaption included low numbers of animal isolates [n = 131 in one study (20 countries; 109 humans, 12 animals); n = 433 in another study (10 countries; 264 humans, 30 animals)].6,8,13 On the other hand, other surveillance studies related to outbreak reports or biased collections rarely include non-clinical strains. All the above-mentioned studies investigated only a few virulence markers.14–16 Most HA/clade A1 Efm isolates are also ampicillin resistant (>90%; MIC generally >128 mg/L) while most isolates associated with CA/clade B are susceptible to this antibiotic.3,17,18 Despite the available information related to the distribution of virulence traits in Efm isolates, there is a lack of evidence for a correlation between virulence profiles and an increased risk of human infection for Efm strains beyond a handful of well-known clones, and especially from the community setting. Indeed, most available information with respect to Efm PVM comes from clade A1, data from Efm in clades A2 and B remaining scarce. The need to evaluate the risk of infection and/or transmission of antibiotic resistance traits through the food chain and, more specifically, the risk of using Efm as a feed additive for animal nutrition is of utmost relevance to the European Food Safety Authority (EFSA) and currently under debate.19 Accordingly, any Efm strain that has an MIC of ampicillin of ≤2 mg/L and lacks all IS16, esp and hyl genes should be regarded as safe by potentially excluding high-risk strains of clade A1. To gain insights on this topic, we assessed the distribution of an extended set of PVM in a comprehensive collection of clonally diverse Efm and correlated the results with ampicillin-resistant phenotypes. These data will be helpful to support public health efforts in the containment of human infections caused by Efm, a bacterial species particularly prone to exhibit MDR. Materials and methods Epidemiology and characterization of the bacterial collection We selected a heterogeneous set of 328 Efm isolates (1986–2015) associated with infections of hospitalized patients (C, n = 193) and non-clinical sources (NC, n = 135) for this study (Table S1, available as Supplementary data at JAC Online). Non-clinical sources included food animals (n = 54, swine, chickens, aquaculture trout, calves), foodstuffs (n = 39, retail poultry, rainbow trout, ready-to-eat salads, cow’s milk, pork), environmental waters (n = 19), healthy volunteers (n = 16), wild birds (n = 6) and ambulatory patients (n = 1). The isolates were selected from a collection of ∼2000 Efm strains we have recovered over the last decades in different surveillance studies20–24 in order to obtain the maximum diversity in terms of origin/source, clonality and antibiotic resistance profiles. The clonal relationship of 284 isolates (representative of different sources and PVM profiles) was established by MLST, goeBURST and Bayesian Analysis Population Structure (BAPS) schemes, recently updated and representing a set of 1116 STs (http://pubmlst.org).20,25,26 Ampicillin susceptibility testing was performed by Etest (0.016–256 mg/L) and broth microdilution (0.0625–256 mg/L) methods following EUCAST guidelines.27 Cation-adjusted Mueller–Hinton II agar (bioMérieux, Marcy-l’Étoile, France) was always freshly prepared and Staphylococcus aureus ATCC 29213 and Enterococcus faecalis ATCC 29212 were used as controls. Extended screening of PVM PCR-based assays and sequencing of representative products were used to screen for the occurrence of 33 genes previously suggested as putative virulence or relevant genetic markers. They comprised: (i) IS16; (ii) hylEfm; (iii) ptsD; (iv) a marker of the putative genomic island mentioned previously and important in HA infections (orf1481); (v) genes coding for 24 cell-wall-anchored proteins (CWAP; also known as Efm surface proteins or Fms) that include esp, sgrA, fms3, fms6, fms7, fms12, fms22, orf773, orf2109 and 15 microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) [acm; scm; ecbA; fms15; fms20-fms21 (pili gene cluster 1, PGC-1; fms21 is also known as pilA); fms14-fms17-fms13 (PGC-2); ebpA-ebpB-ebpC (PGC-3; ebpC is also known as pilB); fms11-fms19-fms16 (PGC-4)] encoding collagen adhesins and pili; (vi) other adhesins (fnm and sagA); and (vii) genes coding for small WxL proteins (swpA, swpB, swpC).8–11,28–30 The acm (both the screening 1500 bp acm and the intact 2166 bp acm*), fms15, ecbA and fms21 were each tested by two sets of primers to cover false negatives obtained with primers targeting variable regions (ecbA, fms21) or potential pseudogenes (acm, fms15).14,28 Efm Aus0004, Aus0085 and C68 strains were used as positive controls for the PCR reactions. Efm strains JH2–2 and V583, both included as negative controls, did not amplify any of the 33 PVM genes. The categorization, nomenclatures and the primers used to search for PVM are given in Table 1.8,9,14,31 Table 1. Characteristics of the PVM used in this study, list of primers and conditions of PCR Category  Gene  Designation(s)  Description  Primers  Oligonucleotide sequence  Size (bp)  Ref.  CWAPa (surface-exposed LPXTG cell wall-anchored proteins) or Fms (Efm surface proteins)  esp  enterococcal surface protein  cell-wall-anchored surface protein with a role in biofilm formation and pathogenesis in endocarditis and UTI  esp-F  AGATTTCATCTTTGATTCTTGG  511  31  esp-R  AATTGATTCTTTAGCATCTGG  acm  adhesin of collagen from Efm fms8 (MSCRAMM)  first cell-wall-anchored collagen adhesin of Efm with increased collagen type I binding; Acm enhances initial adherence in vivo; role in endocarditis  acm-F2  CAGGCAGAGATATCAGCAG  1496  28  acm-R1  ATTCTCATTTGTAACGACTAGC  acm* (complete)  intact acm  acm-F1  GATTTTTGAGAGATGATATAGTAG  2166    acm-R3  ATTTTATTCTTTTGATTTCAGTC  scm  second collagen adhesin of Efm fms10 or orf418 (MSCRAMM)  collagen-binding protein that interacts with components of the extracellular matrix  scm-F  GAAACATACATTCCTAAAGACCCTGG  921  9  scm-R  ACCATTGTCGCTGATAAATATT  ecbA  Efm collagen-binding protein A fms18 or orf2430 (MSCRAMM)  cell surface protein implicated in adhesion to components of the extracellular matrix  ecbA-F  ATTGATGTGGAAACTGAGGG  1227  9  ecbA-R  TCCTTCTGGAGCCTCTACT  ecbA-2    ecbA2-F  GGTTGGACTGTCTTTGCGAATGGC  951  14  ecbA2-R  TGGCCGATTTACAATGAGTTCACTC  sgrA  serine–glutamate repeat containing protein A fms2 or orf2351  nidogen-binding LPXTG surface adhesin implicated in biofilm formation  fms2-F  AATGAACGGGCAAATGAG  671  8  fms2-R  CTTTTGTTCCTTAGTTGGTATGA  fms15  Efm surface protein 15 orf2514 (MSCRAMM)  pseudogene encoding a surface B-type Cna protein  fms15-F  GAGTCTTTAGCAGAACAGCCAGAGCG  1734  9  fms15-R  TCGATTGTCACTATTCACAAG  fms15′  orf2515  orf2515-F  GTGGTAGAATTGACGAAAGA  431  8  orf2515-R  AACAAGTAGCACACCCAATA  PGC-1 cluster  fms20  Efm surface protein 20 orf1901 (MSCRAMM)  PGC-1 is unique among Efm pilus loci as it also contains a housekeeping class A sortase and is located on a megaplasmid  fms20-F  CATTGTTTGGGTTAAAAACAG  803  9  fms20-R  TCAGCTTAGAGCTTCCTCCT  fms21  Efm surface protein 21 pilA or orf1904 (MSCRAMM)  major pilus protein of fms20-21 pili cluster similar to EbpC from E. faecalis  fms21-F  CTTATTGGAATGTTAGGAATCAT  757  9  fms21-R  TCAGTAGCAGTCAGCTTTCC  pilA2      pilA2-F  TGGTTGATCGGCAAATGTAA  211  14  pilA2-R  AGCAGATTATGGGGACGTTG  PGC-2 cluster  fms14  Efm surface protein 14 orf2010 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpA from Efm  fms14-F  CTGGCGAAATTACGACTATGCTACTA  926  9  fms14-R  GTCGCTTTATTATCTCCTTCTTTTAC  fms13  Efm surface protein 13 orf2008 (MSCRAMM)  major pilus protein of fms14-17-13 pili cluster similar to EbpC from Efm  fms13-F  GAAGAAGTCGCACAAAAAC  1493  9  fms13-R  TGACCCGCGTTTCATATTCG  fms17  Efm surface protein 17 orf2009 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpB from Efm  fms17-F  ATGAAAATGATGGCTTGGCT  827  9  fms17-R  GGATGATGACCTCGATTCTC  PGC-3 cluster  ebpAfm  endocarditis- and biofilm-associated pili A fms1 or orf2571 (MSCRAMM)  the pilus subunit accessory proteins A/B and the pilus subunit major protein C (PilB) of EmpABC cluster are orthologues of Efm Ebp subunit proteins associated with pathogenesis in experimental rat endocarditis and mouse UTI; PGC-1 and PGC-3 may play a role during colonization or pathogenesis in the mammalian host although their biological functions remain to be determined  ebpA-F  ACCAAAGCCAGACGAAATAGAAGAAG  848  9  ebpA-R  ATTGTTTTGGTCAGGTGCATCATAGA  ebpBfm  endocarditis- and biofilm-associated pili B fms5 or orf2570 (MSCRAMM)  ebpB-F  ATGTTTACCGCAGAAGCAAC  766  9  ebpB-R  ACTTGTATCCGTTGGCTGTT  ebpCfm  endocarditis- and biofilm-associated pili C pilB, fms9 or orf2569 (MSCRAMM)  ebpC-F  GAGGAGACAGCAGCTCAAG  1673  9  ebpC-R  TGTACCTTTGTGTTTATTTGGTA  PGC-4 cluster  fms11  Efm surface protein 11 orf903 (MSCRAMM)  genomic island with a putative role in biofilm formation of Efm  fms11-F  TGCTAACCAAACGACAGAGGAGAC  831  9  fms11-R  TTGTGCAAAAGAATAGCCCGTTAC  fms19  Efm surface protein 19 orf905 (MSCRAMM)  accessory protein of fms11-19-16 pili cluster similar to EbpB from Efm  fms19-F  GTGTGGAAGACGCACAAAGA  354  9  fms19-R  GGGACTTTATCCCCATCTGC  fms16  Efm surface protein 16 orf907 (MSCRAMM)  major pilus protein of fms11-19-16 pili cluster highly similar to the biofilm enhancer protein Bee3 of Efm  fms16-F  GCCGTATCACCTACACCAG  1204  9  fms16-R  GATTCCTTTAGGTTGGTTATC    fms3  Efm surface protein 3 orf371  cell surface protein Fms3 presumably of the Efm core genome  fms3-F  TGACTTCCAATGTACCGACA  669  8  fms3-R  GCTGCTGCGACTAACACAC  fms6  Efm surface protein 6  LPXTG family cell surface protein Fms6  fms6-F  ACCACGCAAGAAGCCATAAT  276  this study  fms6-R  CATCGATCTTATTGGGAGGGAT  fms7  Efm surface protein 7 orf2356  LPXTG family cell surface protein Fms7  fms7-F  GTTGGTCATCTTATTTGCTGTA  1340  8  fms7-R  TGCCCTGCTCCTTCTACTA  fms12  Efm surface protein 12 orf1996  LPXTG family cell surface protein Fms12  fms12-F  TGAAGTTGCTGCTAACGAAC  959  8  fms12-R  GCAACTTCGATTCCTTCTAC  fms22  Efm surface protein 22 orf884  LPXTG family cell surface protein Fms22  fms22-F  AATCAGACAGTCCACACAGAG  1168  8  fms22-R  ATGATTCCGCTCCACAGTA  orf773  cell cycle protein FtsW  cell surface protein, presumably of the Efm core genome  orf773-F  GCATCAGTCATTAACCAGAGTA  911  8  orf773-R  CCCTGTCAAAGGAATAACG  orf2109  hypothetical protein  cell surface protein, presumably of the Efm core genome  orf2109-F  GCAGGTGCAACTTATACATTAG  750  8  orf2109-R  CTTGATCCGTCGTAATATTGA  WxL proteins (cell surface proteins containing the WxL domain)  swpA  small WxL protein A  novel class of cell surface proteins found in most Efm isolates; functional role in virulence with involvement of WxL operons in bile salt stress and endocarditis pathogenesis was recently elucidated  swpA-F  TTCTTCTAGGCGGCTTGGCAA  625  30  swpA-R  AGCCACGAGGTTCCAAGTCAAA  swpB  small WxL protein B  swpB-F  GCATTAAGTACCGTTCTAGCTGG  729  30  swpB-R  CTGTAGGAGCATCTGTTAACGACC  swpC  small WxL protein C  swpC-F  ACATTTGAAGCAGGAGATGAGGG  474  30  swpC-R  AAGTCCCCACACCTGTTCCTGATTTAGC  Adhesins  sagA  secreted antigen A  broad-spectrum binding protein essential for growth and antigenic during infection; most abundant protein in biofilm-forming cells  sagA-F  CATGCTGACAGCAAAGTCA  816  29  sagA-R  AGAAGCACGCGAACAAGCA  fnm  fibronectin-binding protein of Efm (PavA-like class of adhesins)  Fnm affects Efm fibronectin adherence and has a role in the pathogenesis of experimental endocarditis  fnm-F  ATGAGCGTTCCGTCAGAAGA  887  this study  fnm-R  GTTGCTCCGCCAAAACAAGA  Other genetic markers  pstD  enzyme IID subunit of a putative phosphotransferase system  ptsD encodes a sugar-specific membrane-associated EIID subunit required for carbohydrate transport; it is the first gene contributing to intestinal colonization in Efm during antibiotic treatment  pstD-F  TATCAACGCGATCAAAACGA  241  11  pstD-R  CGTTCGCATACAGCTTTTCA  orf1481  sugar-binding protein encoded by a genomic island (8.5 kb)  orf1481 is one of the genes encoded by a putative 8.5 kb genomic island involved in carbohydrate transport and metabolism  orf1481-F  GTTTATCAACATGCTAGCCCA  437  29  orf1481-R  GCCAATGAGTTAGATGTAGCC    hylEfm  glycosyl-hydrolase  Hyl encodes a family 84 glycosyl-hydrolase variably detected in megaplasmids of clinical strains; its role in the pathogenesis of Efm remains unelucidated  hyl-F  ACAGAAGAGCTGCAGGAAATG  276  31  hyl-R  GACTGACGTCCAAGTTTCCAA  IS16    transposase enriched in hospital-associated strains; contributes to the genomic plasticity of Efm  IS16-F  CATGTTCCACGAACCAGAG  547  10  IS16-R  TCAAAAAGTGGGCTTGGC  Category  Gene  Designation(s)  Description  Primers  Oligonucleotide sequence  Size (bp)  Ref.  CWAPa (surface-exposed LPXTG cell wall-anchored proteins) or Fms (Efm surface proteins)  esp  enterococcal surface protein  cell-wall-anchored surface protein with a role in biofilm formation and pathogenesis in endocarditis and UTI  esp-F  AGATTTCATCTTTGATTCTTGG  511  31  esp-R  AATTGATTCTTTAGCATCTGG  acm  adhesin of collagen from Efm fms8 (MSCRAMM)  first cell-wall-anchored collagen adhesin of Efm with increased collagen type I binding; Acm enhances initial adherence in vivo; role in endocarditis  acm-F2  CAGGCAGAGATATCAGCAG  1496  28  acm-R1  ATTCTCATTTGTAACGACTAGC  acm* (complete)  intact acm  acm-F1  GATTTTTGAGAGATGATATAGTAG  2166    acm-R3  ATTTTATTCTTTTGATTTCAGTC  scm  second collagen adhesin of Efm fms10 or orf418 (MSCRAMM)  collagen-binding protein that interacts with components of the extracellular matrix  scm-F  GAAACATACATTCCTAAAGACCCTGG  921  9  scm-R  ACCATTGTCGCTGATAAATATT  ecbA  Efm collagen-binding protein A fms18 or orf2430 (MSCRAMM)  cell surface protein implicated in adhesion to components of the extracellular matrix  ecbA-F  ATTGATGTGGAAACTGAGGG  1227  9  ecbA-R  TCCTTCTGGAGCCTCTACT  ecbA-2    ecbA2-F  GGTTGGACTGTCTTTGCGAATGGC  951  14  ecbA2-R  TGGCCGATTTACAATGAGTTCACTC  sgrA  serine–glutamate repeat containing protein A fms2 or orf2351  nidogen-binding LPXTG surface adhesin implicated in biofilm formation  fms2-F  AATGAACGGGCAAATGAG  671  8  fms2-R  CTTTTGTTCCTTAGTTGGTATGA  fms15  Efm surface protein 15 orf2514 (MSCRAMM)  pseudogene encoding a surface B-type Cna protein  fms15-F  GAGTCTTTAGCAGAACAGCCAGAGCG  1734  9  fms15-R  TCGATTGTCACTATTCACAAG  fms15′  orf2515  orf2515-F  GTGGTAGAATTGACGAAAGA  431  8  orf2515-R  AACAAGTAGCACACCCAATA  PGC-1 cluster  fms20  Efm surface protein 20 orf1901 (MSCRAMM)  PGC-1 is unique among Efm pilus loci as it also contains a housekeeping class A sortase and is located on a megaplasmid  fms20-F  CATTGTTTGGGTTAAAAACAG  803  9  fms20-R  TCAGCTTAGAGCTTCCTCCT  fms21  Efm surface protein 21 pilA or orf1904 (MSCRAMM)  major pilus protein of fms20-21 pili cluster similar to EbpC from E. faecalis  fms21-F  CTTATTGGAATGTTAGGAATCAT  757  9  fms21-R  TCAGTAGCAGTCAGCTTTCC  pilA2      pilA2-F  TGGTTGATCGGCAAATGTAA  211  14  pilA2-R  AGCAGATTATGGGGACGTTG  PGC-2 cluster  fms14  Efm surface protein 14 orf2010 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpA from Efm  fms14-F  CTGGCGAAATTACGACTATGCTACTA  926  9  fms14-R  GTCGCTTTATTATCTCCTTCTTTTAC  fms13  Efm surface protein 13 orf2008 (MSCRAMM)  major pilus protein of fms14-17-13 pili cluster similar to EbpC from Efm  fms13-F  GAAGAAGTCGCACAAAAAC  1493  9  fms13-R  TGACCCGCGTTTCATATTCG  fms17  Efm surface protein 17 orf2009 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpB from Efm  fms17-F  ATGAAAATGATGGCTTGGCT  827  9  fms17-R  GGATGATGACCTCGATTCTC  PGC-3 cluster  ebpAfm  endocarditis- and biofilm-associated pili A fms1 or orf2571 (MSCRAMM)  the pilus subunit accessory proteins A/B and the pilus subunit major protein C (PilB) of EmpABC cluster are orthologues of Efm Ebp subunit proteins associated with pathogenesis in experimental rat endocarditis and mouse UTI; PGC-1 and PGC-3 may play a role during colonization or pathogenesis in the mammalian host although their biological functions remain to be determined  ebpA-F  ACCAAAGCCAGACGAAATAGAAGAAG  848  9  ebpA-R  ATTGTTTTGGTCAGGTGCATCATAGA  ebpBfm  endocarditis- and biofilm-associated pili B fms5 or orf2570 (MSCRAMM)  ebpB-F  ATGTTTACCGCAGAAGCAAC  766  9  ebpB-R  ACTTGTATCCGTTGGCTGTT  ebpCfm  endocarditis- and biofilm-associated pili C pilB, fms9 or orf2569 (MSCRAMM)  ebpC-F  GAGGAGACAGCAGCTCAAG  1673  9  ebpC-R  TGTACCTTTGTGTTTATTTGGTA  PGC-4 cluster  fms11  Efm surface protein 11 orf903 (MSCRAMM)  genomic island with a putative role in biofilm formation of Efm  fms11-F  TGCTAACCAAACGACAGAGGAGAC  831  9  fms11-R  TTGTGCAAAAGAATAGCCCGTTAC  fms19  Efm surface protein 19 orf905 (MSCRAMM)  accessory protein of fms11-19-16 pili cluster similar to EbpB from Efm  fms19-F  GTGTGGAAGACGCACAAAGA  354  9  fms19-R  GGGACTTTATCCCCATCTGC  fms16  Efm surface protein 16 orf907 (MSCRAMM)  major pilus protein of fms11-19-16 pili cluster highly similar to the biofilm enhancer protein Bee3 of Efm  fms16-F  GCCGTATCACCTACACCAG  1204  9  fms16-R  GATTCCTTTAGGTTGGTTATC    fms3  Efm surface protein 3 orf371  cell surface protein Fms3 presumably of the Efm core genome  fms3-F  TGACTTCCAATGTACCGACA  669  8  fms3-R  GCTGCTGCGACTAACACAC  fms6  Efm surface protein 6  LPXTG family cell surface protein Fms6  fms6-F  ACCACGCAAGAAGCCATAAT  276  this study  fms6-R  CATCGATCTTATTGGGAGGGAT  fms7  Efm surface protein 7 orf2356  LPXTG family cell surface protein Fms7  fms7-F  GTTGGTCATCTTATTTGCTGTA  1340  8  fms7-R  TGCCCTGCTCCTTCTACTA  fms12  Efm surface protein 12 orf1996  LPXTG family cell surface protein Fms12  fms12-F  TGAAGTTGCTGCTAACGAAC  959  8  fms12-R  GCAACTTCGATTCCTTCTAC  fms22  Efm surface protein 22 orf884  LPXTG family cell surface protein Fms22  fms22-F  AATCAGACAGTCCACACAGAG  1168  8  fms22-R  ATGATTCCGCTCCACAGTA  orf773  cell cycle protein FtsW  cell surface protein, presumably of the Efm core genome  orf773-F  GCATCAGTCATTAACCAGAGTA  911  8  orf773-R  CCCTGTCAAAGGAATAACG  orf2109  hypothetical protein  cell surface protein, presumably of the Efm core genome  orf2109-F  GCAGGTGCAACTTATACATTAG  750  8  orf2109-R  CTTGATCCGTCGTAATATTGA  WxL proteins (cell surface proteins containing the WxL domain)  swpA  small WxL protein A  novel class of cell surface proteins found in most Efm isolates; functional role in virulence with involvement of WxL operons in bile salt stress and endocarditis pathogenesis was recently elucidated  swpA-F  TTCTTCTAGGCGGCTTGGCAA  625  30  swpA-R  AGCCACGAGGTTCCAAGTCAAA  swpB  small WxL protein B  swpB-F  GCATTAAGTACCGTTCTAGCTGG  729  30  swpB-R  CTGTAGGAGCATCTGTTAACGACC  swpC  small WxL protein C  swpC-F  ACATTTGAAGCAGGAGATGAGGG  474  30  swpC-R  AAGTCCCCACACCTGTTCCTGATTTAGC  Adhesins  sagA  secreted antigen A  broad-spectrum binding protein essential for growth and antigenic during infection; most abundant protein in biofilm-forming cells  sagA-F  CATGCTGACAGCAAAGTCA  816  29  sagA-R  AGAAGCACGCGAACAAGCA  fnm  fibronectin-binding protein of Efm (PavA-like class of adhesins)  Fnm affects Efm fibronectin adherence and has a role in the pathogenesis of experimental endocarditis  fnm-F  ATGAGCGTTCCGTCAGAAGA  887  this study  fnm-R  GTTGCTCCGCCAAAACAAGA  Other genetic markers  pstD  enzyme IID subunit of a putative phosphotransferase system  ptsD encodes a sugar-specific membrane-associated EIID subunit required for carbohydrate transport; it is the first gene contributing to intestinal colonization in Efm during antibiotic treatment  pstD-F  TATCAACGCGATCAAAACGA  241  11  pstD-R  CGTTCGCATACAGCTTTTCA  orf1481  sugar-binding protein encoded by a genomic island (8.5 kb)  orf1481 is one of the genes encoded by a putative 8.5 kb genomic island involved in carbohydrate transport and metabolism  orf1481-F  GTTTATCAACATGCTAGCCCA  437  29  orf1481-R  GCCAATGAGTTAGATGTAGCC    hylEfm  glycosyl-hydrolase  Hyl encodes a family 84 glycosyl-hydrolase variably detected in megaplasmids of clinical strains; its role in the pathogenesis of Efm remains unelucidated  hyl-F  ACAGAAGAGCTGCAGGAAATG  276  31  hyl-R  GACTGACGTCCAAGTTTCCAA  IS16    transposase enriched in hospital-associated strains; contributes to the genomic plasticity of Efm  IS16-F  CATGTTCCACGAACCAGAG  547  10  IS16-R  TCAAAAAGTGGGCTTGGC  UTI, urinary tract infection. a CWAP typically contain an N-terminal signal sequence peptide and a C-terminal cell wall sorting signal (CWS). CWS consist of a conserved Leu–Pro–X–Thr–Gly (LPXTG) sortase substrate motif (where X denotes any amino acid) followed by a hydrophobic domain and positively charged amino acids. Twenty-four putative CWAP identified by the presence of a C-terminal CWS domain based on the analysis of the Efm TX0016 (DO) according to Hendrickx et al. and Sillanpaa et al.9 Location and transferability of specific virulence genes The plasmid location of fms21 (or pilA), hyl and IS16 has been extensively documented10,32,33 and was here assessed by hybridization of S1-digested genomic DNA using specific probes from Efm Aus0004 (pilA) and Efm C68 (IS16, hylEfm) as described previously.32 The transferability of specific PVM (IS16, hyl, ptsD, orf1481, esp, sgrA, ecbA and fms21), which were more commonly identified in clinical isolates and/or previously linked to plasmids, was assessed by testing these genes in representative transconjugants (TC) that were obtained in previous studies with vancomycin (n = 32) or ampicillin (n = 9) selective plates.17,21 The replication initiator protein (RIP) of pLG1 was tested (PCR/hybridization) in isolates carrying fms21 on megaplasmids, as described previously.21 Statistical analysis The profiles of PVM were visualized in a heatmap built with the ‘pheatmap’ package in ‘RStudio’. The differences between groups of isolates categorized according to sources, clades and resistance phenotypes (Table 2) in their distributions of PVM were compared using a two-tailed Fisher’s exact test. Table 2. Prevalence and distribution of all PVM genes according to sources, clades and antibiotic resistance phenotypes Genea  Distribution [n (%)] of virulence genes among different groups   total (N = 328)  clinical (N = 193)  non-clinical (N = 135)  A1b (N = 134)  A2 (N = 131)  B (N = 19)  clade A1c   ampicillin resistantd (N = 194)  ampicillin susceptible (N = 134)  MDR (N = 264)  non-MDR (N = 64)  ST17 (N = 40)  ST18 (N = 56)  ST78 (N = 38)  fnm  322 (98)  190 (98)  132 (98)  132 (98)  128 (98)  19 (100)  40 (100)  56 (100)  36 (95)  192 (99)  130 (97)  259 (98)  63 (98)  orf2109  322 (98)  189 (98)  133 (98)  133 (99)  129 (98)  17 (90)  39 (98)  56 (100)  38 (100)  194 (100)  128 (96)  260 (98)  62 (97)  swpA  323 (98)  190 (98)  133 (98)  133 (99)  129 (98)  19 (100)  39 (98)  56 (100)  38 (100)  194 (100)  129 (96)  260 (98)  63 (98)  fms6  318 (97)  188 (97)  130 (96)  133 (99)  128 (98)  18 (95)  39 (98)  56 (100)  38 (100)  191 (98)  127 (95)  257 (97)  61 (95)  fms3  310 (94)  186 (96)  124 (92)  131 (98)  120 (92)  17 (90)  39 (98)  54 (96)  38 (100)  190 (98)  120 (90)  251 (95)  59 (92)  swpB  310 (94)  188 (97)  130 (96)  130 (97)  123 (94)  18 (95)  39 (98)  55 (98)  36 (95)  188 (97)  122 (91)  249 (94)  61 (95)  fms7  308 (94)  181 (94)  127 (94)  132 (98)  120 (92)  15 (79)  39 (98)  55 (98)  38 (100)  192 (99)  116 (87)  251 (95)  57 (89)  orf773  302 (92)  179 (93)  123 (91)  131 (98)*  126 (96)*  6 (32)*  38 (95)  56 (100)  37 (97)  193 (100)  109 (81)  256 (97)  46 (72)  fms12  290 (88)  166 (86)  124 (92)  117 (87)  116 (88)  18 (95)  36 (90)  47 (84)  34 (90)  170 (88)  120 (90)  230 (87)  60 (94)  swpC  285 (87)  169 (88)  116 (86)  120 (90)  110 (84)  18 (95)  37 (92)  56 (100)  27 (71)  164 (84)  121 (90)  223 (84)  62 (97)  sagA  282 (86)  154 (80)  128 (95)  114 (85)  112 (86)  12 (63)  33 (82)  53 (95)  29 (76)  173 (89)  101 (75)  230 (87)  48 (75)  fms21  278 (85)  169 (88)  109 (81)  127 (95)  100 (76)  13 (68)  37 (92)  53 (95)  37 (97)  180 (93)*  98 (73)*  234 (89)*  44 (69)*  ebpB  270 (82)  169 (88)  101 (75)  124 (92)*  105 (80)*  9 (47)*  36 (90)  52 (93)  36 (95)  172 (89)  100 (75)  226 (86)  46 (72)  ebpC  256 (78)  152 (79)  104 (77)  115 (86)  102 (78)  9 (47)  32 (80)  50 (89)  33 (87)  162 (84)  94 (70)  216 (82)  40 (62)  fms17  261 (80)  157 (81)  104 (77)  121 (90)*  105 (80)*  4 (21)*  37 (92)  47 (84)  37 (97)  168 (87)  93 (70)  221 (84)  40 (62)  fms13  273 (83)  163 (84)  110 (82)  121 (90)*  112 (86)*  5 (26)*  38 (95)  48 (86)  35 (92)  172 (89)  101 (75)  228 (86)  45 (70)  fms22  259 (79)  164 (85)  95 (70)  116 (87)  91 (70)  17 (90)  37 (92)  42 (75)  37 (97)  170 (88)*  89 (66)*  209 (79)  46 (72)  fms15  260 (79)  168 (87)*  92 (68)*  122 (91)*  103 (79)*  1 (5)*  37 (92)  54 (96)  31 (82)  178 (92)*  82 (61)*  229 (87)*  31 (48)*  scm  244 (74)  142 (74)  102 (76)  104 (78)*  108 (82)*  4 (21)*  36 (90)  36 (64)  32 (84)  148 (76)  96 (72)  207 (78)  37 (58)  acme  231 (70)  139 (72)  92 (68)  97 (72)  79 (60)  18 (95)  28 (70)  42 (75)  27 (71)  147 (76)  84 (63)  188 (71)  43 (67)  ptsD  171 (52)  150 (78)*  21 (16)*  125 (93)*  29 (22)*  0*  37 (92)  54 (96)  34 (90)  167 (86)*  4 (3)*  169 (64)*  2 (3)*  IS16  173 (53)  147 (76)*  26 (19)*  126 (94)*  30 (23)*  0*  38 (95)  53 (95)  35 (92)  167 (86)*  6 (4)*  170 (64)*  3 (5)*  sgrA  177 (54)  146 (76)*  31 (23)*  124 (92)*  32 (24)*  1 (5)*  36 (90)  53 (95)  35 (92)  164 (84)*  13 (10)*  167 (63)*  10 (16)*  orf1481  169 (52)  146 (76)*  23 (17)*  125 (93)*  26 (20)*  1 (5)*  38 (95)  54 (96)  33 (87)  163 (84)*  6 (4)*  165 (62)*  4 (6)*  hyl  39 (12)  35 (18)*  4 (3)*  32 (24)*  6 (5)*  0*  16 (40)  9 (16)  7 (18)  37 (19)*  2 (2)*  39 (15)  0  acm* (complete)  90 (27)  70 (36)*  20 (15)*  53 (40)*  19 (14)*  3 (16)*  10 (25)  28 (50)  15 (40)  74 (38)*  16 (12)*  79 (30)  11 (17)  esp  108 (33)  95 (49)*  13 (10)*  80 (60)*  17 (13)*  2 (10)*  33 (82)  16 (29)  31 (82)*  103 (53)*  5 (4)*  106 (40)*  2 (3)*  ecbA  114 (35)  98 (51)*  16 (12)*  124 (92)*  32 (24)*  2 (10)*  33 (82)  20 (36)  35 (92)*  107 (55)*  7 (5)*  111 (42)*  3 (5)*  fms20  206 (63)  140 (72)*  66 (49)*  103 (77)*  64 (49)*  13 (68)*  32 (80)  35 (62)  36 (95)  143 (74)*  63 (47)*  173 (66)  33 (52)  fms20+fms21 (PGC-1)  200 (61)  137 (71)*  63 (47)*  102 (76)*  60 (46)*  13 (68)*  32 (80)  34 (61)  36 (95)  140 (72)*  60 (45)*  168 (64)  32 (50)  ebpA  172 (52)  132 (68)*  40 (30)*  109 (81)*  46 (35)*  0*  35 (88)  42 (75)  32 (84)  146 (75)*  26 (19)*  153 (58)  19 (30)  ebpA+ebpB+ebpC (PGC-3)  146 (44)  111 (58)*  35 (26)*  93 (69)*  39 (30)*  0*  25 (62)  39 (70)  29 (76)  125 (64)*  21 (15)*  130 (49)  16 (25)  fms14  194 (59)  147 (76)*  47 (35)*  114 (85)*  57 (44)*  3 (16)*  37 (92)  44 (79)  33 (87)  148 (76)*  46 (34)*  167 (63)  27 (42)  fms14-17-13 (PGC-2)  185 (56)  141 (73)*  44 (33)*  111 (83)  53 (40)*  3 (16)*  37 (92)  43 (77)  31 (82)  143 (74)*  42 (31)*  160 (61)  25 (39)  fms11  197 (60)  149 (77)*  48 (36)*  123 (92)*  43 (33)*  12 (63)*  38 (95)  54 (96)  31 (82)  169 (87)*  28 (21)*  183 (69)*  14 (22)*  fms19  205 (62)  151 (78)*  54 (40)*  125 (93)*  50 (38)*  10 (53)*  38 (95)  54 (96)  33 (87)  174 (90)*  31 (23)*  189 (72)*  16 (25)*  fms16  202 (62)  147 (76)*  55 (41)*  121 (90)*  50 (38)*  11 (58)*  35 (88)  54 (96)  32 (84)  170 (88)*  31 (23)*  184 (70)*  17 (27)*  fms11-19-16 (PGC-4)  187 (57)  143 (74)*  44 (33)*  117 (87)*  42 (32)*  10 (53)*  35 (88)  52 (93)  30 (79)  161 (83)*  26 (19)*  174 (66)*  13 (20)*  Genea  Distribution [n (%)] of virulence genes among different groups   total (N = 328)  clinical (N = 193)  non-clinical (N = 135)  A1b (N = 134)  A2 (N = 131)  B (N = 19)  clade A1c   ampicillin resistantd (N = 194)  ampicillin susceptible (N = 134)  MDR (N = 264)  non-MDR (N = 64)  ST17 (N = 40)  ST18 (N = 56)  ST78 (N = 38)  fnm  322 (98)  190 (98)  132 (98)  132 (98)  128 (98)  19 (100)  40 (100)  56 (100)  36 (95)  192 (99)  130 (97)  259 (98)  63 (98)  orf2109  322 (98)  189 (98)  133 (98)  133 (99)  129 (98)  17 (90)  39 (98)  56 (100)  38 (100)  194 (100)  128 (96)  260 (98)  62 (97)  swpA  323 (98)  190 (98)  133 (98)  133 (99)  129 (98)  19 (100)  39 (98)  56 (100)  38 (100)  194 (100)  129 (96)  260 (98)  63 (98)  fms6  318 (97)  188 (97)  130 (96)  133 (99)  128 (98)  18 (95)  39 (98)  56 (100)  38 (100)  191 (98)  127 (95)  257 (97)  61 (95)  fms3  310 (94)  186 (96)  124 (92)  131 (98)  120 (92)  17 (90)  39 (98)  54 (96)  38 (100)  190 (98)  120 (90)  251 (95)  59 (92)  swpB  310 (94)  188 (97)  130 (96)  130 (97)  123 (94)  18 (95)  39 (98)  55 (98)  36 (95)  188 (97)  122 (91)  249 (94)  61 (95)  fms7  308 (94)  181 (94)  127 (94)  132 (98)  120 (92)  15 (79)  39 (98)  55 (98)  38 (100)  192 (99)  116 (87)  251 (95)  57 (89)  orf773  302 (92)  179 (93)  123 (91)  131 (98)*  126 (96)*  6 (32)*  38 (95)  56 (100)  37 (97)  193 (100)  109 (81)  256 (97)  46 (72)  fms12  290 (88)  166 (86)  124 (92)  117 (87)  116 (88)  18 (95)  36 (90)  47 (84)  34 (90)  170 (88)  120 (90)  230 (87)  60 (94)  swpC  285 (87)  169 (88)  116 (86)  120 (90)  110 (84)  18 (95)  37 (92)  56 (100)  27 (71)  164 (84)  121 (90)  223 (84)  62 (97)  sagA  282 (86)  154 (80)  128 (95)  114 (85)  112 (86)  12 (63)  33 (82)  53 (95)  29 (76)  173 (89)  101 (75)  230 (87)  48 (75)  fms21  278 (85)  169 (88)  109 (81)  127 (95)  100 (76)  13 (68)  37 (92)  53 (95)  37 (97)  180 (93)*  98 (73)*  234 (89)*  44 (69)*  ebpB  270 (82)  169 (88)  101 (75)  124 (92)*  105 (80)*  9 (47)*  36 (90)  52 (93)  36 (95)  172 (89)  100 (75)  226 (86)  46 (72)  ebpC  256 (78)  152 (79)  104 (77)  115 (86)  102 (78)  9 (47)  32 (80)  50 (89)  33 (87)  162 (84)  94 (70)  216 (82)  40 (62)  fms17  261 (80)  157 (81)  104 (77)  121 (90)*  105 (80)*  4 (21)*  37 (92)  47 (84)  37 (97)  168 (87)  93 (70)  221 (84)  40 (62)  fms13  273 (83)  163 (84)  110 (82)  121 (90)*  112 (86)*  5 (26)*  38 (95)  48 (86)  35 (92)  172 (89)  101 (75)  228 (86)  45 (70)  fms22  259 (79)  164 (85)  95 (70)  116 (87)  91 (70)  17 (90)  37 (92)  42 (75)  37 (97)  170 (88)*  89 (66)*  209 (79)  46 (72)  fms15  260 (79)  168 (87)*  92 (68)*  122 (91)*  103 (79)*  1 (5)*  37 (92)  54 (96)  31 (82)  178 (92)*  82 (61)*  229 (87)*  31 (48)*  scm  244 (74)  142 (74)  102 (76)  104 (78)*  108 (82)*  4 (21)*  36 (90)  36 (64)  32 (84)  148 (76)  96 (72)  207 (78)  37 (58)  acme  231 (70)  139 (72)  92 (68)  97 (72)  79 (60)  18 (95)  28 (70)  42 (75)  27 (71)  147 (76)  84 (63)  188 (71)  43 (67)  ptsD  171 (52)  150 (78)*  21 (16)*  125 (93)*  29 (22)*  0*  37 (92)  54 (96)  34 (90)  167 (86)*  4 (3)*  169 (64)*  2 (3)*  IS16  173 (53)  147 (76)*  26 (19)*  126 (94)*  30 (23)*  0*  38 (95)  53 (95)  35 (92)  167 (86)*  6 (4)*  170 (64)*  3 (5)*  sgrA  177 (54)  146 (76)*  31 (23)*  124 (92)*  32 (24)*  1 (5)*  36 (90)  53 (95)  35 (92)  164 (84)*  13 (10)*  167 (63)*  10 (16)*  orf1481  169 (52)  146 (76)*  23 (17)*  125 (93)*  26 (20)*  1 (5)*  38 (95)  54 (96)  33 (87)  163 (84)*  6 (4)*  165 (62)*  4 (6)*  hyl  39 (12)  35 (18)*  4 (3)*  32 (24)*  6 (5)*  0*  16 (40)  9 (16)  7 (18)  37 (19)*  2 (2)*  39 (15)  0  acm* (complete)  90 (27)  70 (36)*  20 (15)*  53 (40)*  19 (14)*  3 (16)*  10 (25)  28 (50)  15 (40)  74 (38)*  16 (12)*  79 (30)  11 (17)  esp  108 (33)  95 (49)*  13 (10)*  80 (60)*  17 (13)*  2 (10)*  33 (82)  16 (29)  31 (82)*  103 (53)*  5 (4)*  106 (40)*  2 (3)*  ecbA  114 (35)  98 (51)*  16 (12)*  124 (92)*  32 (24)*  2 (10)*  33 (82)  20 (36)  35 (92)*  107 (55)*  7 (5)*  111 (42)*  3 (5)*  fms20  206 (63)  140 (72)*  66 (49)*  103 (77)*  64 (49)*  13 (68)*  32 (80)  35 (62)  36 (95)  143 (74)*  63 (47)*  173 (66)  33 (52)  fms20+fms21 (PGC-1)  200 (61)  137 (71)*  63 (47)*  102 (76)*  60 (46)*  13 (68)*  32 (80)  34 (61)  36 (95)  140 (72)*  60 (45)*  168 (64)  32 (50)  ebpA  172 (52)  132 (68)*  40 (30)*  109 (81)*  46 (35)*  0*  35 (88)  42 (75)  32 (84)  146 (75)*  26 (19)*  153 (58)  19 (30)  ebpA+ebpB+ebpC (PGC-3)  146 (44)  111 (58)*  35 (26)*  93 (69)*  39 (30)*  0*  25 (62)  39 (70)  29 (76)  125 (64)*  21 (15)*  130 (49)  16 (25)  fms14  194 (59)  147 (76)*  47 (35)*  114 (85)*  57 (44)*  3 (16)*  37 (92)  44 (79)  33 (87)  148 (76)*  46 (34)*  167 (63)  27 (42)  fms14-17-13 (PGC-2)  185 (56)  141 (73)*  44 (33)*  111 (83)  53 (40)*  3 (16)*  37 (92)  43 (77)  31 (82)  143 (74)*  42 (31)*  160 (61)  25 (39)  fms11  197 (60)  149 (77)*  48 (36)*  123 (92)*  43 (33)*  12 (63)*  38 (95)  54 (96)  31 (82)  169 (87)*  28 (21)*  183 (69)*  14 (22)*  fms19  205 (62)  151 (78)*  54 (40)*  125 (93)*  50 (38)*  10 (53)*  38 (95)  54 (96)  33 (87)  174 (90)*  31 (23)*  189 (72)*  16 (25)*  fms16  202 (62)  147 (76)*  55 (41)*  121 (90)*  50 (38)*  11 (58)*  35 (88)  54 (96)  32 (84)  170 (88)*  31 (23)*  184 (70)*  17 (27)*  fms11-19-16 (PGC-4)  187 (57)  143 (74)*  44 (33)*  117 (87)*  42 (32)*  10 (53)*  35 (88)  52 (93)  30 (79)  161 (83)*  26 (19)*  174 (66)*  13 (20)*  a PVM genes are grouped according to their occurrence and clustering in the heatmap (see two main clusters in Figure 2a). b Isolates were divided into three groups: clade A1 (BAPS 2.1a, 3.3a1 and 3.3a2), clade B (BAPS 1) and clade A2 (all remaining BAPS subgroups). c Clade A1 was subdivided into three main clonal lineages: ST17 (BAPS 3.3a2), ST18 (BAPS 3.3a1) and ST78 (BAPS 2.1a). d Ampicillin-resistant isolates were considered to have an MIC of ampicillin of ≥ 8 mg/L according to EUCAST guidelines. e Normal acm genes and pseudogenes were identified in both clinical and non-clinical strains, but a higher proportion of pseudogenes was identified in non-clinical settings. * P < 0.0001 [extremely significant according to Fisher’s exact test (α = 0.05) using GraphPad Prism software, version 7.0a]. Results and discussion Diversity of Efm clones Figure 1(a) illustrates the population structure of Efm isolates (125 STs) which was split into clades A and B [A1 (47%), A2 (46%) and B (7%)] and BAPS (7 BAPS groups and 17 BAPS subgroups). The BAPS groups 2 (36%) and 3 (52%) were overrepresented in comparison with BAPS groups 1, 5, 6, 7 or 9 (1%–7%). While isolates from hospitalized patients, hospital wastewater and swine were distributed among the three clades, isolates from wild birds were found only within clades A1 and A2, poultry isolates were confined to clade A2, and the remaining isolates (from cows, trout and healthy volunteers) belonged to clades A2 and B. Nevertheless, a strong relationship between BAPS 2.1a, 3.3a1 and 3.3a2 (forming the clade A1) and clinical isolates (85%), between BAPS 2.1b and animals (64%) and between BAPS 3.2 and pigs (69%) was observed (P < 0.05). Although clinical isolates frequently belonged to clade A1 (65%), a significant percentage belonged to clade A2 (31%) and more rarely to B (4%). Other studies have also recently linked a significant number of clinical isolates to clades A2 and B,3,4,20 suggesting the relevance of both endogenous and exogenous acquisition routes for infections caused by Efm.20,22 Figure 1. View largeDownload slide Snapshot of the population structure of Efm isolates tested by MLST (n = 284) according to ST (www.phyloviz.net) and BAPS subgroups. Circles represent different ST (ST in numbers within the circles) and the size of each circle is proportional to the number of isolates of each ST. (a) Isolates are differentiated by origin. BAPS grouping is the foundation for this phylogenetic tree, which was further split according to the grouping of isolates into clades. (b) Isolates are differentiated by the number of virulence genes. C, clinical; NC, non-clinical; H, hospital. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Snapshot of the population structure of Efm isolates tested by MLST (n = 284) according to ST (www.phyloviz.net) and BAPS subgroups. Circles represent different ST (ST in numbers within the circles) and the size of each circle is proportional to the number of isolates of each ST. (a) Isolates are differentiated by origin. BAPS grouping is the foundation for this phylogenetic tree, which was further split according to the grouping of isolates into clades. (b) Isolates are differentiated by the number of virulence genes. C, clinical; NC, non-clinical; H, hospital. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Distribution of PVM between Efm of clades A1, A2 and B demonstrates the repertoire of PVM among different Efm populations and the relevance of specific PVM as genetic markers of infection-derived strains Strains from clade A1, independent of their source, were the more frequently enriched in PVM (often >25 genes) and all isolates associated with hospital outbreak clones dating from ≥1992 carried 25–32 genes (Figure 1b). Within clade A2, a variable set of PVM genes was observed (<15 to >25 genes), even within isolates from the same source. Enriched A2 strains (>25 PVM genes) were mostly obtained from hospitals, hospital wastewater or wild birds and belonged to BAPS 2.3a, 3.1, 5 and 7. With the exception of two isolates enriched in PVM (ST108 and ST890 from a pig and a healthy volunteer), a reduced number of PVM were observed in clade B strains, independent of their source. Twenty PVM were carried by >70% of the isolates analysed, independent of their source or clade. They included surface-exposed genes [acm, scm, orf773, orf2109, fms3, fms6, fms7, fms12, fms15, fms22 and members of PGC such as PGC-1 (fms21), PGC-2 (ebpB, ebpC) and PGC-3 (fms13, fms17)], other adhesins (fnm, sagA) and all WxL genes (swpA, swpB, swpC) (Figure 2a and Table 2). This remarkable common backbone corresponding to a plethora of surface proteins with variable binding specificities may help Efm colonize different hosts, contributing both to its widespread occurrence and its virulence. It is possible that some of these predominant fms genes have differential expressions in different hosts or may be non-functional (pseudogenes), but that is beyond the scope of this study.9,28 Moreover, gene sequencing of representative isolates showed different variants of surface proteins and pilins in clinical and non-clinical Efm (hospital variant 1 and community variant 2, respectively, as described by Qin et al.34) (data not shown). Figure 2. View largeDownload slide (a) Heatmap of all virulence genes in the 328 Efm isolates from different sources. ABR, antibiotic resistance; E, environment, HP, hospitalized patients; HV, healthy volunteers. (b) Representation of the different PVM profiles (only major PVM) exhibited only by clinical Efm isolates causing different human infections according to their clades. *PVM patterns identified in different clades. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide (a) Heatmap of all virulence genes in the 328 Efm isolates from different sources. ABR, antibiotic resistance; E, environment, HP, hospitalized patients; HV, healthy volunteers. (b) Representation of the different PVM profiles (only major PVM) exhibited only by clinical Efm isolates causing different human infections according to their clades. *PVM patterns identified in different clades. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The other 13 genes analysed [either encoding surface-exposed molecules (esp, sgrA, ecbA, complete acm*), pili (fms11-fms19-fms16 (PGC-4), fms14, fms20 and ebpA) or other functions (IS16, hyl, ptsD, orf1481)], from now on designated major PVM, were more commonly identified in isolates from hospitalized patients (P < 0.0001), mostly within clade A1, and to a lesser extent clade A2 and/or B (Tables 2 and 3 and Figure 2a). ptsD, IS16, sgrA and orf1481 (group I) were the most common PVM genes found among clinical Efm, and were mostly from clade A1, less frequent in clade A2 and nearly absent in clade B (P < 0.0001; Tables 2 and 3 and Figure 2a). Among these four genes, ptsD was predominant among ampicillin-resistant Efm, both from clinical (all but four ampicillin resistant with an MIC of ampicillin of ≥8 mg/L; clades A1 and A2) and community (pigs, wild birds and healthy volunteers; all from clade A1) sources. Based on this observation of ptsD as a putative marker of relevant ampicillin-resistant Efm clones, we screened this gene in an additional 1077 Efm isolates from different sources and clonal backgrounds (317 ampicillin resistant and 760 ampicillin susceptible; Table S2). Only eight ampicillin-susceptible isolates carried ptsD (0.7%), four of which were associated with bloodstream infections, further confirming the strong positive association of ptsD with ampicillin-resistant Efm. IS16, orf1481 and sgrA were infrequent (5%–15%) but not as rare as ptsD in non-clinical ampicillin-susceptible Efm isolates (Table 2 and data not shown). This study corresponds to the first clear association of ptsD with diverse ampicillin-resistant Efm dating from 1986, and corroborates recent observations from other studies indicating the relevance of genes from phosphotransferase systems in individuals receiving β-lactam treatment.35 On the other hand, mutations related to ampicillin resistance seem to largely affect transport of carbohydrates.36 hyl, acm*, esp and ecbA genes (group II) were more often observed among clinical ampicillin-resistant Efm from clade A1 (P < 0.0001), although they were present at lower rates in all Efm studied, including clinical ones. esp and ecbA have been variably detected in different studies, which might depend on the clonal variability of the sample (strains of ST18 lineage were less enriched in esp and ecbA than ST17 and ST78 lineages in this collection; Table 2). Recent reports of local outbreaks or local collections describe high rates (>80%) of esp and ecbA in ST78 clones,20,37 while retrospective regional data show low rates (<30%) of these genes in clinical ST18-related clones20 (A. R. Freitas, C. Novais and L. Peixe, unpublished data). To a lesser extent, esp, ecbA and acm* were equally distributed between clades A2 and B. In contrast to pioneer studies, the functional role of hyl has been elucidated and appears not to be essential in the pathogenesis of Efm, possibly explaining its variable and often low frequency in infection-derived strains.38 Table 3. Patterns of major PVM among Efm isolates (n = 328) from different sources Patterns of major PVM   n  Hospital, n = 193  Ambulatory, n = 1  Healthy animals, n = 16  Farm animals, n = 54  Food, n = 39  Environment, n = 19  Wild birds, n = 6  group I   group II   group III   ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  PGC  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  —  1 PGC (1), 2 PGC (5), 3 PGC (18), 4 PGC (13)  ≥7  32a          4  1  ptsD  orf1481  IS16  sgrA  —  —  acm*  —  2 PGC (4), 3 PGC (7), 4 PGC (7)  ≥7  18a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  —  2 PGC (4), 3 PGC (4), 4 PGC (9)  ≥9  16a            1  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  hyl  3 PGC (5), 4 PGC (7)  ≥10  12a              ptsD  orf1481  IS16  sgrA  —  ecbA  acm*  —  3 PGC (7), 4 PGC (4)  ≥9  10a            1  ptsD  orf1481  IS16  sgrA  esp  —  —  —  2 PGC (1), 3 PGC (8), 4 PGC (2)  ≥7  9a          2    ptsD  orf1481  IS16  sgrA  —  ecbA  —  —  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥7  9a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥10  8a  1            ptsD  orf1481  IS16  sgrA  —  —  —    1 PGC (3), 2 PGC (2), 3 PGC (3), 4 PGC (2)  ≥5  7a          3    ptsD  orf1481  IS16  sgrA  —  —  acm*  hyl  3 PGC (4), 4 PGC (3)  ≥9  7a            1  ptsD  orf1481  IS16  sgrA  esp  —  acm*    2 PGC (2), 3 PGC (2), 4 PGC (2)  ≥8  6a              ptsD  orf1481  IS16  —  esp  ecbA  acm*  hyl  1 PGC  8  1              ptsD  orf1481  IS16  sgrA  esp  —  —  hyl  4 PGC  10  1              ptsD  orf1481  IS16  sgrA  esp  ecbA  —    —  6  1              ptsD  orf1481  IS16  —  esp  ecbA  acm*    4 PGC  10              1  ptsD  orf1481  IS16  —  —  —  acm*  hyl  2 PGC  7  1              ptsD  orf1481  IS16  sgrA  —  —  —  hyl  3 PGC (2), 4 PGC (2)  ≥8  2          2    ptsD  orf1481  IS16  sgrA  —  —  —    2 PGC  6      1          ptsD  orf1481  IS16  —  esp  ecbA  —  hyl  3 PGC  9  1              ptsD  orf1481  IS16  —  esp  ecbA  —    4 PGC (1)  9  1              ptsD  orf1481  IS16  —  —  —  —    3 PGC (1), 4 PGC (1)  ≥6            2    ptsD  orf1481  —  —  —  ecbA  —    —  3  1              ptsD  orf1481  —  sgrA  esp  ecbA  acm*    3 PGC  9        1        ptsD  orf1481  —  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥3  2              ptsD  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  9  2              ptsD  —  —  —  —  —  —  —  —  1  4              —  orf1481  IS16  —  esp  ecbA  —  —  3 PGC  7  1              —  orf1481  —  sgrA  —  —  —  —  2 PGC  4  1              —  orf1481  —  —  —  —  —  —  1 PGC (1), 2 PGC (1)  ≥2          2      —  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  8  1              —  —  IS16  sgrA  —  —  —  —  2 PGC (1), 3 PGC (2)  ≥4        3        —  —  IS16  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥2  1      1  1      —  —  IS16  —  esp  ecbA    hyl  —  4  1              —  —  IS16  —  esp  —  —  —  3 PGC (1)  5            1    —  —  —  sgrA  —  —  —  —  1 PGC (1), 2 PGC (2), 3 PGC (3)  ≥2  1    1  1  3      —  —  —  sgrA  —  —  acm*  —  1 PGC (1), 2 PGC (1)  ≥3        1  1      —  —  —  —  esp  —  acm*  —  1 PGC  3      1          —  —  —  —  —  ecbA  acm*  —  1 PGC  3          1      —  —  —  —  —  ecbA  —  —  1 PGC (2)  2        5        —  —  —  —  —  —  acm*  —  1 PGC (4), 2 PGC (6)  ≥2  3      4  3      —  —  —  —  —  —  —  esp  1 PGC (1)  2  2              —  —  —  —  —  —  —  —  1 PGC (37), 2 PGC (23), 3 PGC (11), 4 PGC (1)  ≥1  28b    5  17  17  4  1  —  —  —  —  —  —  —  —  —  0  3    8  21  11  1    Patterns of major PVM   n  Hospital, n = 193  Ambulatory, n = 1  Healthy animals, n = 16  Farm animals, n = 54  Food, n = 39  Environment, n = 19  Wild birds, n = 6  group I   group II   group III   ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  PGC  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  —  1 PGC (1), 2 PGC (5), 3 PGC (18), 4 PGC (13)  ≥7  32a          4  1  ptsD  orf1481  IS16  sgrA  —  —  acm*  —  2 PGC (4), 3 PGC (7), 4 PGC (7)  ≥7  18a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  —  2 PGC (4), 3 PGC (4), 4 PGC (9)  ≥9  16a            1  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  hyl  3 PGC (5), 4 PGC (7)  ≥10  12a              ptsD  orf1481  IS16  sgrA  —  ecbA  acm*  —  3 PGC (7), 4 PGC (4)  ≥9  10a            1  ptsD  orf1481  IS16  sgrA  esp  —  —  —  2 PGC (1), 3 PGC (8), 4 PGC (2)  ≥7  9a          2    ptsD  orf1481  IS16  sgrA  —  ecbA  —  —  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥7  9a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥10  8a  1            ptsD  orf1481  IS16  sgrA  —  —  —    1 PGC (3), 2 PGC (2), 3 PGC (3), 4 PGC (2)  ≥5  7a          3    ptsD  orf1481  IS16  sgrA  —  —  acm*  hyl  3 PGC (4), 4 PGC (3)  ≥9  7a            1  ptsD  orf1481  IS16  sgrA  esp  —  acm*    2 PGC (2), 3 PGC (2), 4 PGC (2)  ≥8  6a              ptsD  orf1481  IS16  —  esp  ecbA  acm*  hyl  1 PGC  8  1              ptsD  orf1481  IS16  sgrA  esp  —  —  hyl  4 PGC  10  1              ptsD  orf1481  IS16  sgrA  esp  ecbA  —    —  6  1              ptsD  orf1481  IS16  —  esp  ecbA  acm*    4 PGC  10              1  ptsD  orf1481  IS16  —  —  —  acm*  hyl  2 PGC  7  1              ptsD  orf1481  IS16  sgrA  —  —  —  hyl  3 PGC (2), 4 PGC (2)  ≥8  2          2    ptsD  orf1481  IS16  sgrA  —  —  —    2 PGC  6      1          ptsD  orf1481  IS16  —  esp  ecbA  —  hyl  3 PGC  9  1              ptsD  orf1481  IS16  —  esp  ecbA  —    4 PGC (1)  9  1              ptsD  orf1481  IS16  —  —  —  —    3 PGC (1), 4 PGC (1)  ≥6            2    ptsD  orf1481  —  —  —  ecbA  —    —  3  1              ptsD  orf1481  —  sgrA  esp  ecbA  acm*    3 PGC  9        1        ptsD  orf1481  —  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥3  2              ptsD  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  9  2              ptsD  —  —  —  —  —  —  —  —  1  4              —  orf1481  IS16  —  esp  ecbA  —  —  3 PGC  7  1              —  orf1481  —  sgrA  —  —  —  —  2 PGC  4  1              —  orf1481  —  —  —  —  —  —  1 PGC (1), 2 PGC (1)  ≥2          2      —  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  8  1              —  —  IS16  sgrA  —  —  —  —  2 PGC (1), 3 PGC (2)  ≥4        3        —  —  IS16  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥2  1      1  1      —  —  IS16  —  esp  ecbA    hyl  —  4  1              —  —  IS16  —  esp  —  —  —  3 PGC (1)  5            1    —  —  —  sgrA  —  —  —  —  1 PGC (1), 2 PGC (2), 3 PGC (3)  ≥2  1    1  1  3      —  —  —  sgrA  —  —  acm*  —  1 PGC (1), 2 PGC (1)  ≥3        1  1      —  —  —  —  esp  —  acm*  —  1 PGC  3      1          —  —  —  —  —  ecbA  acm*  —  1 PGC  3          1      —  —  —  —  —  ecbA  —  —  1 PGC (2)  2        5        —  —  —  —  —  —  acm*  —  1 PGC (4), 2 PGC (6)  ≥2  3      4  3      —  —  —  —  —  —  —  esp  1 PGC (1)  2  2              —  —  —  —  —  —  —  —  1 PGC (37), 2 PGC (23), 3 PGC (11), 4 PGC (1)  ≥1  28b    5  17  17  4  1  —  —  —  —  —  —  —  —  —  0  3    8  21  11  1    n, number of genes; PVM, putative virulence markers. a These isolates correspond to outbreak and epidemic Efm strains causing severe human infections. b Most of these isolates correspond to ampicillin-susceptible bloodstream Efm strains from clades A2 and B. Regarding pili (group III), the presence of the accessory fms20, fms14 and ebpA genes correlated well with the presence of complete PGC-1, PGC-2 and PGC-3, respectively. They were associated with both clinical and non-clinical Efm from different sources including foodstuffs and animals (Tables 2 and 3 and Figure 2a), but in this case too, sequencing of representatives demonstrated different protein variants between clinical (hospital variant 1) and non-clinical Efm isolates. Moreover, the co-occurrence of the four complete PGC was significantly more abundant in clinical ampicillin-resistant Efm from clade A1 (P < 0.0001; Table 2). In particular, the complete PGC-2 was present in all Efm from BAPS 3.1 (clade A2), a clonal subgroup that has been associated with miscellaneous isolates including limited clonal outbreaks (e.g. ST125, ST280).21 Its enrichment in specific pili might confer a selective advantage in different hosts. In contrast to the other PGC, both the accessory (fms11, fms19) and major (fms16) subunit pili genes of PGC-4 were strongly linked to ampicillin-resistant Efm strains. PGC-4 was identified in our work in different animal ampicillin-resistant Efm isolates from clade A2 (Figure 2a) and, in a previous study, were observed in most ampicillin-resistant isolates from pet animals.29 An association between specific PVM and particular types of infection could not be established. Current criteria for ‘safe Efm’ (MIC of ampicillin of ≤2 mg/L and absence of esp, hyl and IS16) need revising Figure 2(b) shows the distribution of the PVM profiles identified (n = 29) in the infection-derived isolates from the three clades, after including only major PVM from groups I–III. Some dominant PVM profiles were shared between different clades in different proportions. A significant number of Efm from clades A2 and B (including bloodstream isolates) only carried 1–3 complete PGC, a rare profile in clade A1. The seven bloodstream isolates from clade B contained esp + PGC-1 (n = 1), sgrA + PGC-4 (n = 1) and PGC-1 and/or PGC-4 (n = 5). When specifically analysing the Efm strains with an MIC of ampicillin of ≤2 mg/L, 29 ampicillin-susceptible Efm causing infections [clades: A1, n = 2; A2, n = 20; and B, n = 7; 28 bacteraemia and 1 urinary tract infection (UTI)] were identified with different PVM profiles (ptsD, sgrA, orf1481, acm* and/or PGC-1 to -4) but lacking esp, hyl and IS16 (Tables 4 and 5). For example, different strains from clade B (e.g. ST74, ST85; Table 4) corresponded to non-MDR ampicillin-susceptible bacteraemic isolates that only carried sgrA + PGC-4 or PGC. Twenty strains from clade A2 (15 STs) carried variable PVM profiles, mainly including PGC. Even though we cannot infer their function uniquely from their presence, which could be due, for example, to the high levels of recombination of Efm, our data clearly indicate that Efm strains with an MIC of ampicillin of ≤2 mg/L and which lack esp, hyl and IS16 genes can be associated with severe human infections, and more rarely with major human clones of clade A1. Table 4. Characteristics and major PVM of infection-derived Efm that had an MIC of ampicillin of ≤ 2 mg/L and lacked all IS16, esp and hyl genes (n = 29) Clade  BAPS  ST  ABR profile  Infection  Ward  MIC of ampicillin (mg/L)  Major PVM  Country  Year  A1  2.1a  ST442  CIP, ERY, STR, CHL, Q/D  bloodstream  infectious diseases  2  orf1481, sgrA, PGC-1, PGC-2  Spain  2005    3.3a1  ST18  CIP, ERY, STR  bloodstream  gastroenterology  1  PGC-1, PGC-2  Spain  1995  A2  3.1  ST22  ERY  bloodstream  emergency room  0.5  PGC-1, PGC-2, PGC-3  Spain  1995    3.1  ST22  CIP, ERY, TET  bloodstream  allergy (medical)  0.38  acm*, PGC-2  Spain  2010    3.1  ST22  ERY  bloodstream  emergency room  0.25  PGC-1, PGC-2, PGC-3  Spain  1999    3.1  ST32  ERY, TET  bloodstream  emergency room  1  acm*, PGC-2, PGC-3  Spain  1995    3.1  ST32  CIP, ERY  bloodstream  pneumology  0.5  ptsD, orf1481, PGC-3  Spain  2011    3.1  ST533  CIP, ERY  bloodstream  oncology  1  PGC-1, PGC-2  Spain  2006    3.1  ST214  CIP  bloodstream  paediatric cardiology  0.25  PGC-1, PGC-2, PGC-3  Spain  2006    3.1  ST214  CIP, ERY  bloodstream  gastroenterology  0.125  acm*, PGC-2  Spain  2006    3.2  ST29  CIP, ERY  bloodstream  emergency room  0.096  —  Spain  2007    3.3b  ST102  CIP, ERY  bloodstream  emergency room  0.25  PGC-1  Spain  2001    3.3b  ST888  CIP, ERY  bloodstream  emergency room  0.125  PGC-1  Spain  1999    3.3b  ST102  CIP, ERY  bloodstream  gastroenterology  0.25  ptsD  Spain  2008    2.1b  ST46  CIP, ERY  bloodstream  emergency room  0.25  sgrA, PGC-2, PGC-4  Spain  2001    2.1b  ST69  CIP  bloodstream  ICU  0.19  PGC-1, PGC-2  Spain  2005    2.1b  ST850  CIP, ERY  bloodstream  oncology  0.5  PGC-1, PGC-2  Spain  2011    2.3a  ST21  CIP, ERY  bloodstream  emergency room  2  PGC-2, PGC-3  Spain  1997    2.3a  ST1054  CIP  UTI  neurosurgery  0.25  PGC-2  Portugal  2001    2.3a  ST849  —  bloodstream  emergency room  1  ptsD, orf1481, PGC-2, PGC-3, PGC-4  Spain  2011    2.3b  ST247  —  bloodstream  emergency room  0.38  PGC-1, PGC-2, PGC-3  Spain  2000    7  ST675  CIP  bloodstream  emergency room  0.125  PGC-2  Spain  2008  B  1.2  ST74  ERY  bloodstream  emergency room  2  sgrA, PGC-4  Spain  2001    1.2  ST85  —  bloodstream  neurology  0.5  PGC-1, PGC-4  Spain  1995    1.2  ST85  ERY  bloodstream  emergency room  1  PGC-1  Spain  1997    1.2  ST96  CIP, ERY  bloodstream  emergency room  0.05  —  Spain  2002    1.2  ST178  —  bloodstream  surgery  1.5  PGC-1, PGC-4  Spain  1999    1.2  ST296  ERY, Q/D  bloodstream  urology  0.75  PGC-4  Spain  2013    1.5  ST678  ERY  bloodstream  surgery  1.5  PGC-1  Spain  1996  Clade  BAPS  ST  ABR profile  Infection  Ward  MIC of ampicillin (mg/L)  Major PVM  Country  Year  A1  2.1a  ST442  CIP, ERY, STR, CHL, Q/D  bloodstream  infectious diseases  2  orf1481, sgrA, PGC-1, PGC-2  Spain  2005    3.3a1  ST18  CIP, ERY, STR  bloodstream  gastroenterology  1  PGC-1, PGC-2  Spain  1995  A2  3.1  ST22  ERY  bloodstream  emergency room  0.5  PGC-1, PGC-2, PGC-3  Spain  1995    3.1  ST22  CIP, ERY, TET  bloodstream  allergy (medical)  0.38  acm*, PGC-2  Spain  2010    3.1  ST22  ERY  bloodstream  emergency room  0.25  PGC-1, PGC-2, PGC-3  Spain  1999    3.1  ST32  ERY, TET  bloodstream  emergency room  1  acm*, PGC-2, PGC-3  Spain  1995    3.1  ST32  CIP, ERY  bloodstream  pneumology  0.5  ptsD, orf1481, PGC-3  Spain  2011    3.1  ST533  CIP, ERY  bloodstream  oncology  1  PGC-1, PGC-2  Spain  2006    3.1  ST214  CIP  bloodstream  paediatric cardiology  0.25  PGC-1, PGC-2, PGC-3  Spain  2006    3.1  ST214  CIP, ERY  bloodstream  gastroenterology  0.125  acm*, PGC-2  Spain  2006    3.2  ST29  CIP, ERY  bloodstream  emergency room  0.096  —  Spain  2007    3.3b  ST102  CIP, ERY  bloodstream  emergency room  0.25  PGC-1  Spain  2001    3.3b  ST888  CIP, ERY  bloodstream  emergency room  0.125  PGC-1  Spain  1999    3.3b  ST102  CIP, ERY  bloodstream  gastroenterology  0.25  ptsD  Spain  2008    2.1b  ST46  CIP, ERY  bloodstream  emergency room  0.25  sgrA, PGC-2, PGC-4  Spain  2001    2.1b  ST69  CIP  bloodstream  ICU  0.19  PGC-1, PGC-2  Spain  2005    2.1b  ST850  CIP, ERY  bloodstream  oncology  0.5  PGC-1, PGC-2  Spain  2011    2.3a  ST21  CIP, ERY  bloodstream  emergency room  2  PGC-2, PGC-3  Spain  1997    2.3a  ST1054  CIP  UTI  neurosurgery  0.25  PGC-2  Portugal  2001    2.3a  ST849  —  bloodstream  emergency room  1  ptsD, orf1481, PGC-2, PGC-3, PGC-4  Spain  2011    2.3b  ST247  —  bloodstream  emergency room  0.38  PGC-1, PGC-2, PGC-3  Spain  2000    7  ST675  CIP  bloodstream  emergency room  0.125  PGC-2  Spain  2008  B  1.2  ST74  ERY  bloodstream  emergency room  2  sgrA, PGC-4  Spain  2001    1.2  ST85  —  bloodstream  neurology  0.5  PGC-1, PGC-4  Spain  1995    1.2  ST85  ERY  bloodstream  emergency room  1  PGC-1  Spain  1997    1.2  ST96  CIP, ERY  bloodstream  emergency room  0.05  —  Spain  2002    1.2  ST178  —  bloodstream  surgery  1.5  PGC-1, PGC-4  Spain  1999    1.2  ST296  ERY, Q/D  bloodstream  urology  0.75  PGC-4  Spain  2013    1.5  ST678  ERY  bloodstream  surgery  1.5  PGC-1  Spain  1996  ABR, antibiotic resistance; ERY, erythromycin; CIP, ciprofloxacin; CHL, chloramphenicol; Q/D, quinupristin/dalfopristin; STR, streptomycin; TET, tetracycline. acm*, complete acm. Only complete PGC were considered: PGC-1 (PilA), fms21-fms20; PGC-2, fms14-fms17-fms13; PGC-3 (PilB), ebpA-ebpB-ebpC; and PGC-4, fms11-fms19-fms16. Table 5. Distribution of major PVM [n (%)] among Efm isolates expressing different ampicillin MIC (mg/L) values     PGC-1 (PilA), fms21-fms20; PGC-2, fms14-fms17-fms13; PGC-3 (PilB), ebpA-ebpB-ebpC; PGC-4, fms11-fms19-fms16; PVM, putative virulence markers. Grey shading indicates groups of isolates that include at least one Efm clinical isolate involved in a human infection. Grey shading and bold formatting indicates groups of isolates that include at least one clinical isolate involved in a human infection with an MIC of ampicillin of ≤ 2 mg/L and which lacked all IS16/esp/hyl genes. Further studies are needed to clarify the pathogenic potential of different Efm hosts carrying individual or specific combinations of the PVM studied. In any case, our data clearly demonstrated that: (i) the PVM IS16, ptsD, sgrA and orf1481 were more frequent in clinical Efm and rare in community Efm, and might be considered as good markers of infection-derived Efm (group I); (ii) esp, hyl, ecbA and the complete acm were variably detected in clinical Efm, and never exclusively present in infection-derived Efm; (iii) even though more studies are necessary to better address the role of pili, our data suggest that particular variants (hospital variants in clinical Efm) might be important in infection-derived Efm [based on the fact that: (a) the protein variants of pili vary; (b) infection-derived Efm consistently carry at least one complete PGC; and (c) even infection-derived Efm expressing low levels of ampicillin resistance often exclusively carry PGC (Table 5)]; (iv) the occurrence of PVM from group II in infection-derived Efm strains was always associated with the detection of PVM from groups I and III; and (v) the cumulative presence of most of the PVM mentioned (groups I, II and III) is of concern since most outbreak, persistent and epidemic Efm strains were associated with the co-occurrence of several genes (>7) (Table 3). Not all PVM relevant to Efm pathogenicity were tested here and some are surely still unknown, but our data additionally supported the conclusion that a few genes promoting adherence to host tissues (such as PGC) and high-density colonization, which is probably enhanced in immunocompromised and/or elderly patients by multiple and prolonged antibiotic therapies, might favour human infection.20,39 PVM can be horizontally transferred along with antibiotic resistance fms21 (or pilA) was located in one or two plasmids of variable size (60–280 kb) carried since 1988 by unrelated Efm and occasionally co-located with vanA, IS16 and/or hylEfm, as observed in previous studies.32,33 Most fms21 (pilA) megaplasmids (>150 kb) hybridized with the RIP of pLG1 in isolates from different sources and backgrounds, which constitutes the first association of fms21 (pilA) with pLG1-like plasmids since its description.40 IS16 was either located in the chromosome and/or plasmids (50–220 kb), while hylEfm was only associated with megaplasmids (>180 kb). While fms21 (pilA) was identified in most of the 41 TC tested (85%, clinical and non-clinical donors), IS16, sgrA and hylEfm were transferred in 24%, 20% and 5% of the cases, respectively, all corresponding to donors of clinical origin. ptsD was not identified in any TC, despite its recent finding in a large non-pLG1 hyl-carrying plasmid41 and the descriptions of extensive transfer of sugar-related genes between clade A Efm strains,42 suggesting its location on the chromosome, on non-mobilizable plasmids, on plasmids that were not selected in the conjugation conditions used or on other mobile genetic elements (MGEs). It is noteworthy that sgrA, which was recently associated with a transferable pbp5 ampicillin resistant platform containing genes involved in different cell functions,17 was exclusively detected in clinical ampicillin-resistant donors (eight of nine TC). This is further supported by the data included in Table 5 showing a positive correlation for several PVM, including sgrA, and increasing ampicillin MIC values. Interestingly, within an ampicillin MIC range of 0.064–2 mg/L (mostly non-clinical isolates) the prevalence of PVM of groups I and II (see previous section) was quite low, and the near absence of isolates enriched in PVM for the lowest MIC tested (0.064–0.125 mg/L), compared with the high frequency of PVM for an ampicillin MIC of >256 mg/L, was remarkable. The MIC of ampicillin of ≤2 mg/L proposed by the EFSA is indeed associated with a lower number of PVM, but does not guarantee the absence of strains with virulence genes previously highly linked to human infections caused by Efm. The association between PVM and transferable MGEs means that the safety of particular Efm clonal lineages (e.g. clade B) cannot be assumed. Conclusions This is the first study to assess the distribution of an extended set of virulence genes in clonally diverse Efm from disparate hosts, correlating their occurrence with ampicillin-resistant phenotypes and proposing a consensus set of relevant PVM for the tracking of infection-derived Efm clones. One limitation of our study is the biased clinical collection of isolates (mostly VRE), though it still includes a large number of animal/food isolates compared with previous studies. Strains with higher ampicillin MIC values contained a higher number of PVM, independent of their clonal group or source, and might present a greater risk in terms of pathogenicity. PVM coding for surface (esp/sgrA/ecbA/complete acm) and pili proteins, or others enhancing colonization (hyl/ptsD/orf1481) or plasticity (IS16), were strongly associated with clinical Efm (mostly clade A1), but were also observed in clades A2/B at different rates. This study corroborates the association of particular pili variants with infection-derived Efm, stressing the need to sequence pili genes for strain risk analysis. Based on our data and in light of the current literature, we propose that ptsD, IS16, orf1481, sgrA and hospital variants of complete PGC should be considered when assessing the safety of Efm strains. The safety markers recently proposed by the EFSA as distinctive markers of safe Efm are not strict enough to avoid dissemination throughout microbial food additives of strains with the potential to cause human infections, thereby posing a risk to public health. Acknowledgements We are grateful for the technical assistance provided by Sara Aguiar, Elsa Martins, Houyem Elghaieb and Conception M. Rodríguez. We wish to thank Dr Ivan Literak and Dr Veronika Oravcová (University of Veterinary and Pharmaceutical Sciences Brno, Czech Republic), for providing the Efm strains of wild birds, and Dr Timothy P. Stinear (University of Melbourne, Australia), for the gift of Efm strains Aus0004 and Aus0085. Funding This work received financial support from the European Union (FEDER funds POCI/01/0145/FEDER/007728) and National Funds (FCT/MEC, Fundação para a Ciência e Tecnologia and Ministério da Educação e Ciência) under the Partnership Agreement PT2020 (UID/MULTI/04378/2013) and project NORTE-01-0145-FEDER-000011 [QREN – Qualidade e Segurança Alimentar – uma abordagem (nano)tecnológica]. This work was co-funded by a Research Grant (2013) from the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) (to A. R. F.). A. R. F. was supported by a fellowship (grant SFRH/BPD/96148/2013) from FCT through Programa Operacional Capital Humano (POCH). Work in the lab of T. M. C. is supported by the Instituto de Salud Carlos III (PI12-01581), and CIBER (CB06/02/0053), within the Plan Estatal de I+D+i 2013–2016 co-financed by the European Development Regional Fund ‘A way to achieve Europe’ (ERDF); also by the Regional Government of Madrid in Spain (PROMPT-S2010/BMD2414). Transparency declarations None to declare. Supplementary data Tables S1 and S2 are available as Supplementary data at JAC Online. References 1 ECDC. Annual Epidemiological Report 2014, Antimicrobial Resistance and Healthcare-Associated Infections. Stockholm, 2015. 2 Galloway-Peña J, Roh JH, Latorre M et al.   Genomic and SNP analyses demonstrate a distant separation of the hospital and community-associated clades of Enterococcus faecium. PLoS One  2012; 7: e30187. Google Scholar CrossRef Search ADS PubMed  3 Lebreton F, van Schaik W, McGuire AM et al.   Emergence of epidemic multidrug-resistant Enterococcus faecium from animal and commensal strains. MBio  2013; 4: e00534– 13. Google Scholar CrossRef Search ADS PubMed  4 Raven KE, Reuter S, Reynolds R et al.   A decade of genomic history for healthcare-associated Enterococcus faecium in the United Kingdom and Ireland. Genome Res  2016; 26: 1388– 96. Google Scholar CrossRef Search ADS PubMed  5 Guzman Prieto AM, van Schaik W, Rogers MRC et al.   Global emergence and dissemination of enterococci as nosocomial pathogens: attack of the clones? Front Microbiol  2016; 7: 788. Google Scholar CrossRef Search ADS PubMed  6 Sillanpää J, Prakash VP, Nallapareddy SR et al.   Distribution of genes encoding MSCRAMMs and pili in clinical and natural populations of Enterococcus faecium. J Clin Microbiol  2009; 47: 896– 901. Google Scholar CrossRef Search ADS PubMed  7 Rice LB, Carias L, Rudin S et al.   A potential virulence gene, hylEfm, predominates in Enterococcus faecium of clinical origin. J Infect Dis  2003; 187: 508– 12. Google Scholar CrossRef Search ADS PubMed  8 Hendrickx APA, van Wamel WJB, Posthuma G et al.   Five genes encoding surface-exposed LPXTG proteins are enriched in hospital-adapted Enterococcus faecium clonal complex 17 isolates. J Bacteriol  2007; 189: 8321– 32. Google Scholar CrossRef Search ADS PubMed  9 Sillanpaa J, Nallapareddy SR, Prakash VP et al.   Identification and phenotypic characterization of a second collagen adhesin, Scm, and genome-based identification and analysis of 13 other predicted MSCRAMMs, including four distinct pilus loci, in Enterococcus faecium. Microbiology  2008; 154: 3199– 211. Google Scholar CrossRef Search ADS PubMed  10 Werner G, Fleige C, Geringer U et al.   IS element IS16 as a molecular screening tool to identify hospital-associated strains of Enterococcus faecium. BMC Infect Dis  2011; 11: 80. Google Scholar CrossRef Search ADS PubMed  11 Zhang X, Top J, de Been M et al.   Identification of a genetic determinant in clinical Enterococcus faecium strains that contributes to intestinal colonization during antibiotic treatment. J Infect Dis  2013; 207: 1780– 6. Google Scholar CrossRef Search ADS PubMed  12 Heikens E, van Schaik W, Leavis HL et al.   Identification of a novel genomic island specific to hospital-acquired clonal complex 17 Enterococcus faecium isolates. Appl Environ Microbiol  2008; 74: 7094– 7. Google Scholar CrossRef Search ADS PubMed  13 Kim EB, Jin G-D, Lee J-Y et al.   Genomic features and niche-adaptation of Enterococcus faecium strains from Korean soybean-fermented foods. PLoS One  2016; 11: e0153279. Google Scholar CrossRef Search ADS PubMed  14 Sivertsen A, Billström H, Melefors Ö et al.   A multicentre hospital outbreak in Sweden caused by introduction of a vanB2 transposon into a stably maintained pRUM-plasmid in an Enterococcus faecium ST192 clone. PLoS One  2014; 9: e103274. Google Scholar CrossRef Search ADS PubMed  15 Almohamad S, Somarajan SR, Singh KV et al.   Influence of isolate origin and presence of various genes on biofilm formation by Enterococcus faecium. FEMS Microbiol Lett  2014; 353: 151– 6. Google Scholar CrossRef Search ADS PubMed  16 Sadowy E, Luczkiewicz A. Drug-resistant and hospital-associated Enterococcus faecium from wastewater, riverine estuary and anthropogenically impacted marine catchment basin. BMC Microbiol  2014; 14: 66. Google Scholar CrossRef Search ADS PubMed  17 Novais C, Tedim AP, Lanza VF et al.   Co-diversification of Enterococcus faecium core genomes and PBP5: evidences of pbp5 horizontal transfer. Front Microbiol  2016; 7: 1581. Google Scholar CrossRef Search ADS PubMed  18 Galloway-Peña JR, Rice LB, Murray BE. Analysis of PBP5 of early U.S. isolates of Enterococcus faecium: sequence variation alone does not explain increasing ampicillin resistance over time. Antimicrob Agents Chemother  2011; 55: 3272. Google Scholar CrossRef Search ADS PubMed  19 EFSA Panel on Additives and Products or Substances used in Animal Feed (FEEDAP). Guidance on the Safety Assessment of Enterococcus faecium in Animal Nutrition. 2012. 20 Tedim AP, Ruíz-Garbajosa P, Rodríguez MC et al.   Long-term clonal dynamics of Enterococcus faecium strains causing bloodstream infections (1995-2015) in Spain. J Antimicrob Chemother  2017; 72: 48– 55. Google Scholar CrossRef Search ADS PubMed  21 Freitas AR, Tedim AP, Francia MV et al.   Multilevel population genetic analysis of vanA and vanB Enterococcus faecium causing nosocomial outbreaks in 27 countries (1986-2012). J Antimicrob Chemother  2016; 71: 3351– 66. Google Scholar CrossRef Search ADS PubMed  22 Freitas AR, Coque TM, Novais C et al.   Human and swine hosts share vancomycin-resistant Enterococcus faecium CC17 and CC5 and Enterococcus faecalis CC2 clonal clusters harboring Tn1546 on indistinguishable plasmids. J Clin Microbiol  2011; 49: 925– 31. Google Scholar CrossRef Search ADS PubMed  23 Novais C, Coque TM, Costa MJ et al.   High occurrence and persistence of antibiotic-resistant enterococci in poultry food samples in Portugal. J Antimicrob Chemother  2005; 56: 1139– 43. Google Scholar CrossRef Search ADS PubMed  24 Novais C, Coque TM, Sousa JC et al.   Antimicrobial resistance among faecal enterococci from healthy individuals in Portugal. Clin Microbiol Infect  2006; 12: 1131– 4. Google Scholar CrossRef Search ADS PubMed  25 Willems RJL, Top J, van Schaik W et al.   Restricted gene flow among hospital subpopulations of Enterococcus faecium. MBio  2012; 3: e00151– 12. Google Scholar CrossRef Search ADS PubMed  26 Tedim AP, Ruiz-Garbajosa P, Corander J et al.   Population biology of intestinal Enterococcus isolates from hospitalized and nonhospitalized individuals in different age groups. Appl Environ Microbiol  2015; 81: 1820– 31. Google Scholar CrossRef Search ADS PubMed  27 EUCAST. Breakpoint Tables for Interpretation of MICs and Zone Diameters, Version 7.1, 2017. 28 Nallapareddy SR, Weinstock GM, Murray BE. Clinical isolates of Enterococcus faecium exhibit strain-specific collagen binding mediated by Acm, a new member of the MSCRAMM family. Mol Microbiol  2003; 47: 1733– 47. Google Scholar CrossRef Search ADS PubMed  29 de Regt MJA, van Schaik W, van Luit-Asbroek M et al.   Hospital and community ampicillin-resistant Enterococcus faecium are evolutionarily closely linked but have diversified through niche adaptation. PLoS One  2012; 7: e30319. Google Scholar CrossRef Search ADS PubMed  30 Galloway-Peña JR, Liang X, Singh KV et al.   The identification and functional characterization of WxL proteins from Enterococcus faecium reveal surface proteins involved in extracellular matrix interactions. J Bacteriol  2015; 197: 882– 92. Google Scholar CrossRef Search ADS PubMed  31 Vankerckhoven V, Van Autgaerden T, Vael C et al.   Development of a multiplex PCR for the detection of asa1, gelE, cylA, esp, and hyl genes in enterococci and survey for virulence determinants among European hospital isolates of Enterococcus faecium. J Clin Microbiol  2004; 42: 4473– 9. Google Scholar CrossRef Search ADS PubMed  32 Freitas AR, Tedim AP, Novais C et al.   Global spread of the hylEfm colonization-virulence gene in megaplasmids of the Enterococcus faecium CC17 polyclonal subcluster. Antimicrob Agents Chemother  2010; 54: 2660– 5. Google Scholar CrossRef Search ADS PubMed  33 Kim DS, Singh KV, Nallapareddy SR et al.   The fms21 (pilA)-fms20 locus encoding one of four distinct pili of Enterococcus faecium is harboured on a large transferable plasmid associated with gut colonization and virulence. J Med Microbiol  2010; 59: 505– 7. Google Scholar CrossRef Search ADS PubMed  34 Qin X, Galloway-Peña JR, Sillanpaa J et al.   Complete genome sequence of Enterococcus faecium strain TX16 and comparative genomic analysis of Enterococcus faecium genomes. BMC Microbiol  2012; 12: 135. Google Scholar CrossRef Search ADS PubMed  35 Pérez-Cobas AE, Artacho A, Knecht H et al.   Differential effects of antibiotic therapy on the structure and function of human gut microbiota. PLoS One  2013; 8: e80201. Google Scholar CrossRef Search ADS PubMed  36 Sacco E, Cortes M, Josseaume N et al.   Mutation landscape of acquired cross-resistance to glycopeptide and β-lactam antibiotics in Enterococcus faecium. Antimicrob Agents Chemother  2015; 59: 5306– 15. Google Scholar CrossRef Search ADS PubMed  37 Yang J, Jiang Y, Guo L et al.   Prevalence of diverse clones of vancomycin-resistant Enterococcus faecium ST78 in a Chinese hospital. Microb Drug Resist  2016; 22: 294– 300. Google Scholar CrossRef Search ADS PubMed  38 Panesso D, Montealegre MC, Rincón S et al.   The hylEfm gene in pHylEfm of Enterococcus faecium is not required in pathogenesis of murine peritonitis. BMC Microbiol  2011; 11: 20. Google Scholar CrossRef Search ADS PubMed  39 Flores-Mireles AL, Walker JN, Potretzke A et al.   Antibody-based therapy for enterococcal catheter-associated urinary tract infections. MBio  2016; 7: e01653-16. Google Scholar CrossRef Search ADS PubMed  40 Laverde Gomez JA, van Schaik W, Freitas AR et al.   A multiresistance megaplasmid pLG1 bearing a hylEfm genomic island in hospital Enterococcus faecium isolates. Int J Med Microbiol  2011; 301: 165– 75. Google Scholar CrossRef Search ADS PubMed  41 García-Solache M, Lebreton F, McLaughlin RE et al.   Homologous recombination within large chromosomal regions facilitates acquisition of β-lactam and vancomycin resistance in Enterococcus faecium. Antimicrob Agents Chemother  2016; 60: 5777– 86. Google Scholar CrossRef Search ADS PubMed  42 de Been M, van Schaik W, Cheng L et al.   Recent recombination events in the core genome are associated with adaptive evolution in Enterococcus faecium. Genome Biol Evol  2013; 5: 1524– 35. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Distribution of putative virulence markers in Enterococcus faecium: towards a safety profile review

Loading next page...
 
/lp/ou_press/distribution-of-putative-virulence-markers-in-enterococcus-faecium-34IG5pglry
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dkx387
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives The criteria for identification of Enterococcus faecium (Efm) with the ability to cause human infections are currently being debated by the European Food Safety Authority (EFSA). Strains that have an MIC of ampicillin of ≤ 2 mg/L and lack IS16/esp/hyl genes should be regarded as safe for use as feed additives in animal nutrition, despite the lack of knowledge about putative virulence marker (PVM) distribution in community Efm. We analysed the distribution of major PVM and ampicillin phenotypes in large Efm collections to investigate further the safety of strains from a public health perspective. Methods Thirty-three PVM were assessed by PCR/sequencing among clonally disparate Efm (n = 328; 1986–2015) from different origins. We analysed ampicillin susceptibility (Etest/broth microdilution) according to EUCAST guidelines, clonal relationship (MLST) and genomic location of PVM (S1-PFGE/hybridization). Results Infection-derived Efm were more enriched in PVM and the increase in ampicillin MIC was positively correlated with an enrichment in different PVM. PVM coding for surface (esp/sgrA/ecbA/complete acm) and pili proteins, or others enhancing colonization (hyl/ptsD/orf1481) or plasticity (IS16), were strongly associated with clinical Efm (mostly clade A1), but also observed in clades A2/B at different rates. ptsD was a good marker of ampicillin-resistant Efm. ptsD, IS16, orf1481, sgrA and hospital variants of complete pili gene clusters are proposed as markers to assess the safety of Efm strains. Conclusions Our study expands on the distribution of PVM in diverse Efm lineages and demonstrates the enrichment in infection-derived strains of PVM not previously included in EFSA’s list of Efm safety criteria. The evidence of relevant Efm infection markers can impact the risk assessment of Efm strains in different public health contexts. Introduction Enterococcus faecium (Efm), a human and animal colonizer, is also a leading cause of hospital-acquired infections, mainly urinary tract and surgical infections, bacteraemia and endocarditis, the incidence of which continues to increase in Europe.1 Studies of the population structure of Efm have revealed that the species is split into two major clades comprising isolates principally associated with hospitalized humans (HA) and community-based individuals (CA), respectively.2 These groups have also been identified using different approaches and have alternatively been named clade A and clade B, epidemic hospital strains (A1) being distinguished from strains from animals and sporadic human infections (A2).3,4 Strains within the clade HA/clade A1 (formerly classified as CC17 by MLST), seem to be enriched in different adaptive features including metabolic capabilities, antibiotic resistance genes and traits enhancing colonization or pathogenicity in comparison with those of CA/clade B.3,5,6 The putative virulence genes most thoroughly studied at the functional level include those encoding a putative glycoside hydrolase (HylEfm) and several Efm surface proteins (Fms) involved in adhesion, biofilm formation and pili assembly, namely Esp (enterococcal surface protein), Acm (adhesin of collagen from Efm), Scm (second collagen adhesin from Efm), SgrA (serine-glutamate-repeat-containing-protein A) and EcbA (Efm-collagen-binding-protein A).6–9 In addition IS16, which is thought to enhance genomic plasticity, has been found to be a putative genetic marker of hospital strains.10 The occurrence of these and other putative virulence markers (PVM) that have recently arisen from genome sequencing projects (e.g. ptsD or orf1481 from a genomic island) and their association with infection-derived strains is lacking in comprehensive Efm collections.11,12 Indeed, the few gene-based studies suggesting that the enrichment of a number of Fms genes in infection-derived strains may favour hospital adaption included low numbers of animal isolates [n = 131 in one study (20 countries; 109 humans, 12 animals); n = 433 in another study (10 countries; 264 humans, 30 animals)].6,8,13 On the other hand, other surveillance studies related to outbreak reports or biased collections rarely include non-clinical strains. All the above-mentioned studies investigated only a few virulence markers.14–16 Most HA/clade A1 Efm isolates are also ampicillin resistant (>90%; MIC generally >128 mg/L) while most isolates associated with CA/clade B are susceptible to this antibiotic.3,17,18 Despite the available information related to the distribution of virulence traits in Efm isolates, there is a lack of evidence for a correlation between virulence profiles and an increased risk of human infection for Efm strains beyond a handful of well-known clones, and especially from the community setting. Indeed, most available information with respect to Efm PVM comes from clade A1, data from Efm in clades A2 and B remaining scarce. The need to evaluate the risk of infection and/or transmission of antibiotic resistance traits through the food chain and, more specifically, the risk of using Efm as a feed additive for animal nutrition is of utmost relevance to the European Food Safety Authority (EFSA) and currently under debate.19 Accordingly, any Efm strain that has an MIC of ampicillin of ≤2 mg/L and lacks all IS16, esp and hyl genes should be regarded as safe by potentially excluding high-risk strains of clade A1. To gain insights on this topic, we assessed the distribution of an extended set of PVM in a comprehensive collection of clonally diverse Efm and correlated the results with ampicillin-resistant phenotypes. These data will be helpful to support public health efforts in the containment of human infections caused by Efm, a bacterial species particularly prone to exhibit MDR. Materials and methods Epidemiology and characterization of the bacterial collection We selected a heterogeneous set of 328 Efm isolates (1986–2015) associated with infections of hospitalized patients (C, n = 193) and non-clinical sources (NC, n = 135) for this study (Table S1, available as Supplementary data at JAC Online). Non-clinical sources included food animals (n = 54, swine, chickens, aquaculture trout, calves), foodstuffs (n = 39, retail poultry, rainbow trout, ready-to-eat salads, cow’s milk, pork), environmental waters (n = 19), healthy volunteers (n = 16), wild birds (n = 6) and ambulatory patients (n = 1). The isolates were selected from a collection of ∼2000 Efm strains we have recovered over the last decades in different surveillance studies20–24 in order to obtain the maximum diversity in terms of origin/source, clonality and antibiotic resistance profiles. The clonal relationship of 284 isolates (representative of different sources and PVM profiles) was established by MLST, goeBURST and Bayesian Analysis Population Structure (BAPS) schemes, recently updated and representing a set of 1116 STs (http://pubmlst.org).20,25,26 Ampicillin susceptibility testing was performed by Etest (0.016–256 mg/L) and broth microdilution (0.0625–256 mg/L) methods following EUCAST guidelines.27 Cation-adjusted Mueller–Hinton II agar (bioMérieux, Marcy-l’Étoile, France) was always freshly prepared and Staphylococcus aureus ATCC 29213 and Enterococcus faecalis ATCC 29212 were used as controls. Extended screening of PVM PCR-based assays and sequencing of representative products were used to screen for the occurrence of 33 genes previously suggested as putative virulence or relevant genetic markers. They comprised: (i) IS16; (ii) hylEfm; (iii) ptsD; (iv) a marker of the putative genomic island mentioned previously and important in HA infections (orf1481); (v) genes coding for 24 cell-wall-anchored proteins (CWAP; also known as Efm surface proteins or Fms) that include esp, sgrA, fms3, fms6, fms7, fms12, fms22, orf773, orf2109 and 15 microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) [acm; scm; ecbA; fms15; fms20-fms21 (pili gene cluster 1, PGC-1; fms21 is also known as pilA); fms14-fms17-fms13 (PGC-2); ebpA-ebpB-ebpC (PGC-3; ebpC is also known as pilB); fms11-fms19-fms16 (PGC-4)] encoding collagen adhesins and pili; (vi) other adhesins (fnm and sagA); and (vii) genes coding for small WxL proteins (swpA, swpB, swpC).8–11,28–30 The acm (both the screening 1500 bp acm and the intact 2166 bp acm*), fms15, ecbA and fms21 were each tested by two sets of primers to cover false negatives obtained with primers targeting variable regions (ecbA, fms21) or potential pseudogenes (acm, fms15).14,28 Efm Aus0004, Aus0085 and C68 strains were used as positive controls for the PCR reactions. Efm strains JH2–2 and V583, both included as negative controls, did not amplify any of the 33 PVM genes. The categorization, nomenclatures and the primers used to search for PVM are given in Table 1.8,9,14,31 Table 1. Characteristics of the PVM used in this study, list of primers and conditions of PCR Category  Gene  Designation(s)  Description  Primers  Oligonucleotide sequence  Size (bp)  Ref.  CWAPa (surface-exposed LPXTG cell wall-anchored proteins) or Fms (Efm surface proteins)  esp  enterococcal surface protein  cell-wall-anchored surface protein with a role in biofilm formation and pathogenesis in endocarditis and UTI  esp-F  AGATTTCATCTTTGATTCTTGG  511  31  esp-R  AATTGATTCTTTAGCATCTGG  acm  adhesin of collagen from Efm fms8 (MSCRAMM)  first cell-wall-anchored collagen adhesin of Efm with increased collagen type I binding; Acm enhances initial adherence in vivo; role in endocarditis  acm-F2  CAGGCAGAGATATCAGCAG  1496  28  acm-R1  ATTCTCATTTGTAACGACTAGC  acm* (complete)  intact acm  acm-F1  GATTTTTGAGAGATGATATAGTAG  2166    acm-R3  ATTTTATTCTTTTGATTTCAGTC  scm  second collagen adhesin of Efm fms10 or orf418 (MSCRAMM)  collagen-binding protein that interacts with components of the extracellular matrix  scm-F  GAAACATACATTCCTAAAGACCCTGG  921  9  scm-R  ACCATTGTCGCTGATAAATATT  ecbA  Efm collagen-binding protein A fms18 or orf2430 (MSCRAMM)  cell surface protein implicated in adhesion to components of the extracellular matrix  ecbA-F  ATTGATGTGGAAACTGAGGG  1227  9  ecbA-R  TCCTTCTGGAGCCTCTACT  ecbA-2    ecbA2-F  GGTTGGACTGTCTTTGCGAATGGC  951  14  ecbA2-R  TGGCCGATTTACAATGAGTTCACTC  sgrA  serine–glutamate repeat containing protein A fms2 or orf2351  nidogen-binding LPXTG surface adhesin implicated in biofilm formation  fms2-F  AATGAACGGGCAAATGAG  671  8  fms2-R  CTTTTGTTCCTTAGTTGGTATGA  fms15  Efm surface protein 15 orf2514 (MSCRAMM)  pseudogene encoding a surface B-type Cna protein  fms15-F  GAGTCTTTAGCAGAACAGCCAGAGCG  1734  9  fms15-R  TCGATTGTCACTATTCACAAG  fms15′  orf2515  orf2515-F  GTGGTAGAATTGACGAAAGA  431  8  orf2515-R  AACAAGTAGCACACCCAATA  PGC-1 cluster  fms20  Efm surface protein 20 orf1901 (MSCRAMM)  PGC-1 is unique among Efm pilus loci as it also contains a housekeeping class A sortase and is located on a megaplasmid  fms20-F  CATTGTTTGGGTTAAAAACAG  803  9  fms20-R  TCAGCTTAGAGCTTCCTCCT  fms21  Efm surface protein 21 pilA or orf1904 (MSCRAMM)  major pilus protein of fms20-21 pili cluster similar to EbpC from E. faecalis  fms21-F  CTTATTGGAATGTTAGGAATCAT  757  9  fms21-R  TCAGTAGCAGTCAGCTTTCC  pilA2      pilA2-F  TGGTTGATCGGCAAATGTAA  211  14  pilA2-R  AGCAGATTATGGGGACGTTG  PGC-2 cluster  fms14  Efm surface protein 14 orf2010 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpA from Efm  fms14-F  CTGGCGAAATTACGACTATGCTACTA  926  9  fms14-R  GTCGCTTTATTATCTCCTTCTTTTAC  fms13  Efm surface protein 13 orf2008 (MSCRAMM)  major pilus protein of fms14-17-13 pili cluster similar to EbpC from Efm  fms13-F  GAAGAAGTCGCACAAAAAC  1493  9  fms13-R  TGACCCGCGTTTCATATTCG  fms17  Efm surface protein 17 orf2009 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpB from Efm  fms17-F  ATGAAAATGATGGCTTGGCT  827  9  fms17-R  GGATGATGACCTCGATTCTC  PGC-3 cluster  ebpAfm  endocarditis- and biofilm-associated pili A fms1 or orf2571 (MSCRAMM)  the pilus subunit accessory proteins A/B and the pilus subunit major protein C (PilB) of EmpABC cluster are orthologues of Efm Ebp subunit proteins associated with pathogenesis in experimental rat endocarditis and mouse UTI; PGC-1 and PGC-3 may play a role during colonization or pathogenesis in the mammalian host although their biological functions remain to be determined  ebpA-F  ACCAAAGCCAGACGAAATAGAAGAAG  848  9  ebpA-R  ATTGTTTTGGTCAGGTGCATCATAGA  ebpBfm  endocarditis- and biofilm-associated pili B fms5 or orf2570 (MSCRAMM)  ebpB-F  ATGTTTACCGCAGAAGCAAC  766  9  ebpB-R  ACTTGTATCCGTTGGCTGTT  ebpCfm  endocarditis- and biofilm-associated pili C pilB, fms9 or orf2569 (MSCRAMM)  ebpC-F  GAGGAGACAGCAGCTCAAG  1673  9  ebpC-R  TGTACCTTTGTGTTTATTTGGTA  PGC-4 cluster  fms11  Efm surface protein 11 orf903 (MSCRAMM)  genomic island with a putative role in biofilm formation of Efm  fms11-F  TGCTAACCAAACGACAGAGGAGAC  831  9  fms11-R  TTGTGCAAAAGAATAGCCCGTTAC  fms19  Efm surface protein 19 orf905 (MSCRAMM)  accessory protein of fms11-19-16 pili cluster similar to EbpB from Efm  fms19-F  GTGTGGAAGACGCACAAAGA  354  9  fms19-R  GGGACTTTATCCCCATCTGC  fms16  Efm surface protein 16 orf907 (MSCRAMM)  major pilus protein of fms11-19-16 pili cluster highly similar to the biofilm enhancer protein Bee3 of Efm  fms16-F  GCCGTATCACCTACACCAG  1204  9  fms16-R  GATTCCTTTAGGTTGGTTATC    fms3  Efm surface protein 3 orf371  cell surface protein Fms3 presumably of the Efm core genome  fms3-F  TGACTTCCAATGTACCGACA  669  8  fms3-R  GCTGCTGCGACTAACACAC  fms6  Efm surface protein 6  LPXTG family cell surface protein Fms6  fms6-F  ACCACGCAAGAAGCCATAAT  276  this study  fms6-R  CATCGATCTTATTGGGAGGGAT  fms7  Efm surface protein 7 orf2356  LPXTG family cell surface protein Fms7  fms7-F  GTTGGTCATCTTATTTGCTGTA  1340  8  fms7-R  TGCCCTGCTCCTTCTACTA  fms12  Efm surface protein 12 orf1996  LPXTG family cell surface protein Fms12  fms12-F  TGAAGTTGCTGCTAACGAAC  959  8  fms12-R  GCAACTTCGATTCCTTCTAC  fms22  Efm surface protein 22 orf884  LPXTG family cell surface protein Fms22  fms22-F  AATCAGACAGTCCACACAGAG  1168  8  fms22-R  ATGATTCCGCTCCACAGTA  orf773  cell cycle protein FtsW  cell surface protein, presumably of the Efm core genome  orf773-F  GCATCAGTCATTAACCAGAGTA  911  8  orf773-R  CCCTGTCAAAGGAATAACG  orf2109  hypothetical protein  cell surface protein, presumably of the Efm core genome  orf2109-F  GCAGGTGCAACTTATACATTAG  750  8  orf2109-R  CTTGATCCGTCGTAATATTGA  WxL proteins (cell surface proteins containing the WxL domain)  swpA  small WxL protein A  novel class of cell surface proteins found in most Efm isolates; functional role in virulence with involvement of WxL operons in bile salt stress and endocarditis pathogenesis was recently elucidated  swpA-F  TTCTTCTAGGCGGCTTGGCAA  625  30  swpA-R  AGCCACGAGGTTCCAAGTCAAA  swpB  small WxL protein B  swpB-F  GCATTAAGTACCGTTCTAGCTGG  729  30  swpB-R  CTGTAGGAGCATCTGTTAACGACC  swpC  small WxL protein C  swpC-F  ACATTTGAAGCAGGAGATGAGGG  474  30  swpC-R  AAGTCCCCACACCTGTTCCTGATTTAGC  Adhesins  sagA  secreted antigen A  broad-spectrum binding protein essential for growth and antigenic during infection; most abundant protein in biofilm-forming cells  sagA-F  CATGCTGACAGCAAAGTCA  816  29  sagA-R  AGAAGCACGCGAACAAGCA  fnm  fibronectin-binding protein of Efm (PavA-like class of adhesins)  Fnm affects Efm fibronectin adherence and has a role in the pathogenesis of experimental endocarditis  fnm-F  ATGAGCGTTCCGTCAGAAGA  887  this study  fnm-R  GTTGCTCCGCCAAAACAAGA  Other genetic markers  pstD  enzyme IID subunit of a putative phosphotransferase system  ptsD encodes a sugar-specific membrane-associated EIID subunit required for carbohydrate transport; it is the first gene contributing to intestinal colonization in Efm during antibiotic treatment  pstD-F  TATCAACGCGATCAAAACGA  241  11  pstD-R  CGTTCGCATACAGCTTTTCA  orf1481  sugar-binding protein encoded by a genomic island (8.5 kb)  orf1481 is one of the genes encoded by a putative 8.5 kb genomic island involved in carbohydrate transport and metabolism  orf1481-F  GTTTATCAACATGCTAGCCCA  437  29  orf1481-R  GCCAATGAGTTAGATGTAGCC    hylEfm  glycosyl-hydrolase  Hyl encodes a family 84 glycosyl-hydrolase variably detected in megaplasmids of clinical strains; its role in the pathogenesis of Efm remains unelucidated  hyl-F  ACAGAAGAGCTGCAGGAAATG  276  31  hyl-R  GACTGACGTCCAAGTTTCCAA  IS16    transposase enriched in hospital-associated strains; contributes to the genomic plasticity of Efm  IS16-F  CATGTTCCACGAACCAGAG  547  10  IS16-R  TCAAAAAGTGGGCTTGGC  Category  Gene  Designation(s)  Description  Primers  Oligonucleotide sequence  Size (bp)  Ref.  CWAPa (surface-exposed LPXTG cell wall-anchored proteins) or Fms (Efm surface proteins)  esp  enterococcal surface protein  cell-wall-anchored surface protein with a role in biofilm formation and pathogenesis in endocarditis and UTI  esp-F  AGATTTCATCTTTGATTCTTGG  511  31  esp-R  AATTGATTCTTTAGCATCTGG  acm  adhesin of collagen from Efm fms8 (MSCRAMM)  first cell-wall-anchored collagen adhesin of Efm with increased collagen type I binding; Acm enhances initial adherence in vivo; role in endocarditis  acm-F2  CAGGCAGAGATATCAGCAG  1496  28  acm-R1  ATTCTCATTTGTAACGACTAGC  acm* (complete)  intact acm  acm-F1  GATTTTTGAGAGATGATATAGTAG  2166    acm-R3  ATTTTATTCTTTTGATTTCAGTC  scm  second collagen adhesin of Efm fms10 or orf418 (MSCRAMM)  collagen-binding protein that interacts with components of the extracellular matrix  scm-F  GAAACATACATTCCTAAAGACCCTGG  921  9  scm-R  ACCATTGTCGCTGATAAATATT  ecbA  Efm collagen-binding protein A fms18 or orf2430 (MSCRAMM)  cell surface protein implicated in adhesion to components of the extracellular matrix  ecbA-F  ATTGATGTGGAAACTGAGGG  1227  9  ecbA-R  TCCTTCTGGAGCCTCTACT  ecbA-2    ecbA2-F  GGTTGGACTGTCTTTGCGAATGGC  951  14  ecbA2-R  TGGCCGATTTACAATGAGTTCACTC  sgrA  serine–glutamate repeat containing protein A fms2 or orf2351  nidogen-binding LPXTG surface adhesin implicated in biofilm formation  fms2-F  AATGAACGGGCAAATGAG  671  8  fms2-R  CTTTTGTTCCTTAGTTGGTATGA  fms15  Efm surface protein 15 orf2514 (MSCRAMM)  pseudogene encoding a surface B-type Cna protein  fms15-F  GAGTCTTTAGCAGAACAGCCAGAGCG  1734  9  fms15-R  TCGATTGTCACTATTCACAAG  fms15′  orf2515  orf2515-F  GTGGTAGAATTGACGAAAGA  431  8  orf2515-R  AACAAGTAGCACACCCAATA  PGC-1 cluster  fms20  Efm surface protein 20 orf1901 (MSCRAMM)  PGC-1 is unique among Efm pilus loci as it also contains a housekeeping class A sortase and is located on a megaplasmid  fms20-F  CATTGTTTGGGTTAAAAACAG  803  9  fms20-R  TCAGCTTAGAGCTTCCTCCT  fms21  Efm surface protein 21 pilA or orf1904 (MSCRAMM)  major pilus protein of fms20-21 pili cluster similar to EbpC from E. faecalis  fms21-F  CTTATTGGAATGTTAGGAATCAT  757  9  fms21-R  TCAGTAGCAGTCAGCTTTCC  pilA2      pilA2-F  TGGTTGATCGGCAAATGTAA  211  14  pilA2-R  AGCAGATTATGGGGACGTTG  PGC-2 cluster  fms14  Efm surface protein 14 orf2010 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpA from Efm  fms14-F  CTGGCGAAATTACGACTATGCTACTA  926  9  fms14-R  GTCGCTTTATTATCTCCTTCTTTTAC  fms13  Efm surface protein 13 orf2008 (MSCRAMM)  major pilus protein of fms14-17-13 pili cluster similar to EbpC from Efm  fms13-F  GAAGAAGTCGCACAAAAAC  1493  9  fms13-R  TGACCCGCGTTTCATATTCG  fms17  Efm surface protein 17 orf2009 (MSCRAMM)  accessory protein of fms14-17-13 pili cluster similar to EbpB from Efm  fms17-F  ATGAAAATGATGGCTTGGCT  827  9  fms17-R  GGATGATGACCTCGATTCTC  PGC-3 cluster  ebpAfm  endocarditis- and biofilm-associated pili A fms1 or orf2571 (MSCRAMM)  the pilus subunit accessory proteins A/B and the pilus subunit major protein C (PilB) of EmpABC cluster are orthologues of Efm Ebp subunit proteins associated with pathogenesis in experimental rat endocarditis and mouse UTI; PGC-1 and PGC-3 may play a role during colonization or pathogenesis in the mammalian host although their biological functions remain to be determined  ebpA-F  ACCAAAGCCAGACGAAATAGAAGAAG  848  9  ebpA-R  ATTGTTTTGGTCAGGTGCATCATAGA  ebpBfm  endocarditis- and biofilm-associated pili B fms5 or orf2570 (MSCRAMM)  ebpB-F  ATGTTTACCGCAGAAGCAAC  766  9  ebpB-R  ACTTGTATCCGTTGGCTGTT  ebpCfm  endocarditis- and biofilm-associated pili C pilB, fms9 or orf2569 (MSCRAMM)  ebpC-F  GAGGAGACAGCAGCTCAAG  1673  9  ebpC-R  TGTACCTTTGTGTTTATTTGGTA  PGC-4 cluster  fms11  Efm surface protein 11 orf903 (MSCRAMM)  genomic island with a putative role in biofilm formation of Efm  fms11-F  TGCTAACCAAACGACAGAGGAGAC  831  9  fms11-R  TTGTGCAAAAGAATAGCCCGTTAC  fms19  Efm surface protein 19 orf905 (MSCRAMM)  accessory protein of fms11-19-16 pili cluster similar to EbpB from Efm  fms19-F  GTGTGGAAGACGCACAAAGA  354  9  fms19-R  GGGACTTTATCCCCATCTGC  fms16  Efm surface protein 16 orf907 (MSCRAMM)  major pilus protein of fms11-19-16 pili cluster highly similar to the biofilm enhancer protein Bee3 of Efm  fms16-F  GCCGTATCACCTACACCAG  1204  9  fms16-R  GATTCCTTTAGGTTGGTTATC    fms3  Efm surface protein 3 orf371  cell surface protein Fms3 presumably of the Efm core genome  fms3-F  TGACTTCCAATGTACCGACA  669  8  fms3-R  GCTGCTGCGACTAACACAC  fms6  Efm surface protein 6  LPXTG family cell surface protein Fms6  fms6-F  ACCACGCAAGAAGCCATAAT  276  this study  fms6-R  CATCGATCTTATTGGGAGGGAT  fms7  Efm surface protein 7 orf2356  LPXTG family cell surface protein Fms7  fms7-F  GTTGGTCATCTTATTTGCTGTA  1340  8  fms7-R  TGCCCTGCTCCTTCTACTA  fms12  Efm surface protein 12 orf1996  LPXTG family cell surface protein Fms12  fms12-F  TGAAGTTGCTGCTAACGAAC  959  8  fms12-R  GCAACTTCGATTCCTTCTAC  fms22  Efm surface protein 22 orf884  LPXTG family cell surface protein Fms22  fms22-F  AATCAGACAGTCCACACAGAG  1168  8  fms22-R  ATGATTCCGCTCCACAGTA  orf773  cell cycle protein FtsW  cell surface protein, presumably of the Efm core genome  orf773-F  GCATCAGTCATTAACCAGAGTA  911  8  orf773-R  CCCTGTCAAAGGAATAACG  orf2109  hypothetical protein  cell surface protein, presumably of the Efm core genome  orf2109-F  GCAGGTGCAACTTATACATTAG  750  8  orf2109-R  CTTGATCCGTCGTAATATTGA  WxL proteins (cell surface proteins containing the WxL domain)  swpA  small WxL protein A  novel class of cell surface proteins found in most Efm isolates; functional role in virulence with involvement of WxL operons in bile salt stress and endocarditis pathogenesis was recently elucidated  swpA-F  TTCTTCTAGGCGGCTTGGCAA  625  30  swpA-R  AGCCACGAGGTTCCAAGTCAAA  swpB  small WxL protein B  swpB-F  GCATTAAGTACCGTTCTAGCTGG  729  30  swpB-R  CTGTAGGAGCATCTGTTAACGACC  swpC  small WxL protein C  swpC-F  ACATTTGAAGCAGGAGATGAGGG  474  30  swpC-R  AAGTCCCCACACCTGTTCCTGATTTAGC  Adhesins  sagA  secreted antigen A  broad-spectrum binding protein essential for growth and antigenic during infection; most abundant protein in biofilm-forming cells  sagA-F  CATGCTGACAGCAAAGTCA  816  29  sagA-R  AGAAGCACGCGAACAAGCA  fnm  fibronectin-binding protein of Efm (PavA-like class of adhesins)  Fnm affects Efm fibronectin adherence and has a role in the pathogenesis of experimental endocarditis  fnm-F  ATGAGCGTTCCGTCAGAAGA  887  this study  fnm-R  GTTGCTCCGCCAAAACAAGA  Other genetic markers  pstD  enzyme IID subunit of a putative phosphotransferase system  ptsD encodes a sugar-specific membrane-associated EIID subunit required for carbohydrate transport; it is the first gene contributing to intestinal colonization in Efm during antibiotic treatment  pstD-F  TATCAACGCGATCAAAACGA  241  11  pstD-R  CGTTCGCATACAGCTTTTCA  orf1481  sugar-binding protein encoded by a genomic island (8.5 kb)  orf1481 is one of the genes encoded by a putative 8.5 kb genomic island involved in carbohydrate transport and metabolism  orf1481-F  GTTTATCAACATGCTAGCCCA  437  29  orf1481-R  GCCAATGAGTTAGATGTAGCC    hylEfm  glycosyl-hydrolase  Hyl encodes a family 84 glycosyl-hydrolase variably detected in megaplasmids of clinical strains; its role in the pathogenesis of Efm remains unelucidated  hyl-F  ACAGAAGAGCTGCAGGAAATG  276  31  hyl-R  GACTGACGTCCAAGTTTCCAA  IS16    transposase enriched in hospital-associated strains; contributes to the genomic plasticity of Efm  IS16-F  CATGTTCCACGAACCAGAG  547  10  IS16-R  TCAAAAAGTGGGCTTGGC  UTI, urinary tract infection. a CWAP typically contain an N-terminal signal sequence peptide and a C-terminal cell wall sorting signal (CWS). CWS consist of a conserved Leu–Pro–X–Thr–Gly (LPXTG) sortase substrate motif (where X denotes any amino acid) followed by a hydrophobic domain and positively charged amino acids. Twenty-four putative CWAP identified by the presence of a C-terminal CWS domain based on the analysis of the Efm TX0016 (DO) according to Hendrickx et al. and Sillanpaa et al.9 Location and transferability of specific virulence genes The plasmid location of fms21 (or pilA), hyl and IS16 has been extensively documented10,32,33 and was here assessed by hybridization of S1-digested genomic DNA using specific probes from Efm Aus0004 (pilA) and Efm C68 (IS16, hylEfm) as described previously.32 The transferability of specific PVM (IS16, hyl, ptsD, orf1481, esp, sgrA, ecbA and fms21), which were more commonly identified in clinical isolates and/or previously linked to plasmids, was assessed by testing these genes in representative transconjugants (TC) that were obtained in previous studies with vancomycin (n = 32) or ampicillin (n = 9) selective plates.17,21 The replication initiator protein (RIP) of pLG1 was tested (PCR/hybridization) in isolates carrying fms21 on megaplasmids, as described previously.21 Statistical analysis The profiles of PVM were visualized in a heatmap built with the ‘pheatmap’ package in ‘RStudio’. The differences between groups of isolates categorized according to sources, clades and resistance phenotypes (Table 2) in their distributions of PVM were compared using a two-tailed Fisher’s exact test. Table 2. Prevalence and distribution of all PVM genes according to sources, clades and antibiotic resistance phenotypes Genea  Distribution [n (%)] of virulence genes among different groups   total (N = 328)  clinical (N = 193)  non-clinical (N = 135)  A1b (N = 134)  A2 (N = 131)  B (N = 19)  clade A1c   ampicillin resistantd (N = 194)  ampicillin susceptible (N = 134)  MDR (N = 264)  non-MDR (N = 64)  ST17 (N = 40)  ST18 (N = 56)  ST78 (N = 38)  fnm  322 (98)  190 (98)  132 (98)  132 (98)  128 (98)  19 (100)  40 (100)  56 (100)  36 (95)  192 (99)  130 (97)  259 (98)  63 (98)  orf2109  322 (98)  189 (98)  133 (98)  133 (99)  129 (98)  17 (90)  39 (98)  56 (100)  38 (100)  194 (100)  128 (96)  260 (98)  62 (97)  swpA  323 (98)  190 (98)  133 (98)  133 (99)  129 (98)  19 (100)  39 (98)  56 (100)  38 (100)  194 (100)  129 (96)  260 (98)  63 (98)  fms6  318 (97)  188 (97)  130 (96)  133 (99)  128 (98)  18 (95)  39 (98)  56 (100)  38 (100)  191 (98)  127 (95)  257 (97)  61 (95)  fms3  310 (94)  186 (96)  124 (92)  131 (98)  120 (92)  17 (90)  39 (98)  54 (96)  38 (100)  190 (98)  120 (90)  251 (95)  59 (92)  swpB  310 (94)  188 (97)  130 (96)  130 (97)  123 (94)  18 (95)  39 (98)  55 (98)  36 (95)  188 (97)  122 (91)  249 (94)  61 (95)  fms7  308 (94)  181 (94)  127 (94)  132 (98)  120 (92)  15 (79)  39 (98)  55 (98)  38 (100)  192 (99)  116 (87)  251 (95)  57 (89)  orf773  302 (92)  179 (93)  123 (91)  131 (98)*  126 (96)*  6 (32)*  38 (95)  56 (100)  37 (97)  193 (100)  109 (81)  256 (97)  46 (72)  fms12  290 (88)  166 (86)  124 (92)  117 (87)  116 (88)  18 (95)  36 (90)  47 (84)  34 (90)  170 (88)  120 (90)  230 (87)  60 (94)  swpC  285 (87)  169 (88)  116 (86)  120 (90)  110 (84)  18 (95)  37 (92)  56 (100)  27 (71)  164 (84)  121 (90)  223 (84)  62 (97)  sagA  282 (86)  154 (80)  128 (95)  114 (85)  112 (86)  12 (63)  33 (82)  53 (95)  29 (76)  173 (89)  101 (75)  230 (87)  48 (75)  fms21  278 (85)  169 (88)  109 (81)  127 (95)  100 (76)  13 (68)  37 (92)  53 (95)  37 (97)  180 (93)*  98 (73)*  234 (89)*  44 (69)*  ebpB  270 (82)  169 (88)  101 (75)  124 (92)*  105 (80)*  9 (47)*  36 (90)  52 (93)  36 (95)  172 (89)  100 (75)  226 (86)  46 (72)  ebpC  256 (78)  152 (79)  104 (77)  115 (86)  102 (78)  9 (47)  32 (80)  50 (89)  33 (87)  162 (84)  94 (70)  216 (82)  40 (62)  fms17  261 (80)  157 (81)  104 (77)  121 (90)*  105 (80)*  4 (21)*  37 (92)  47 (84)  37 (97)  168 (87)  93 (70)  221 (84)  40 (62)  fms13  273 (83)  163 (84)  110 (82)  121 (90)*  112 (86)*  5 (26)*  38 (95)  48 (86)  35 (92)  172 (89)  101 (75)  228 (86)  45 (70)  fms22  259 (79)  164 (85)  95 (70)  116 (87)  91 (70)  17 (90)  37 (92)  42 (75)  37 (97)  170 (88)*  89 (66)*  209 (79)  46 (72)  fms15  260 (79)  168 (87)*  92 (68)*  122 (91)*  103 (79)*  1 (5)*  37 (92)  54 (96)  31 (82)  178 (92)*  82 (61)*  229 (87)*  31 (48)*  scm  244 (74)  142 (74)  102 (76)  104 (78)*  108 (82)*  4 (21)*  36 (90)  36 (64)  32 (84)  148 (76)  96 (72)  207 (78)  37 (58)  acme  231 (70)  139 (72)  92 (68)  97 (72)  79 (60)  18 (95)  28 (70)  42 (75)  27 (71)  147 (76)  84 (63)  188 (71)  43 (67)  ptsD  171 (52)  150 (78)*  21 (16)*  125 (93)*  29 (22)*  0*  37 (92)  54 (96)  34 (90)  167 (86)*  4 (3)*  169 (64)*  2 (3)*  IS16  173 (53)  147 (76)*  26 (19)*  126 (94)*  30 (23)*  0*  38 (95)  53 (95)  35 (92)  167 (86)*  6 (4)*  170 (64)*  3 (5)*  sgrA  177 (54)  146 (76)*  31 (23)*  124 (92)*  32 (24)*  1 (5)*  36 (90)  53 (95)  35 (92)  164 (84)*  13 (10)*  167 (63)*  10 (16)*  orf1481  169 (52)  146 (76)*  23 (17)*  125 (93)*  26 (20)*  1 (5)*  38 (95)  54 (96)  33 (87)  163 (84)*  6 (4)*  165 (62)*  4 (6)*  hyl  39 (12)  35 (18)*  4 (3)*  32 (24)*  6 (5)*  0*  16 (40)  9 (16)  7 (18)  37 (19)*  2 (2)*  39 (15)  0  acm* (complete)  90 (27)  70 (36)*  20 (15)*  53 (40)*  19 (14)*  3 (16)*  10 (25)  28 (50)  15 (40)  74 (38)*  16 (12)*  79 (30)  11 (17)  esp  108 (33)  95 (49)*  13 (10)*  80 (60)*  17 (13)*  2 (10)*  33 (82)  16 (29)  31 (82)*  103 (53)*  5 (4)*  106 (40)*  2 (3)*  ecbA  114 (35)  98 (51)*  16 (12)*  124 (92)*  32 (24)*  2 (10)*  33 (82)  20 (36)  35 (92)*  107 (55)*  7 (5)*  111 (42)*  3 (5)*  fms20  206 (63)  140 (72)*  66 (49)*  103 (77)*  64 (49)*  13 (68)*  32 (80)  35 (62)  36 (95)  143 (74)*  63 (47)*  173 (66)  33 (52)  fms20+fms21 (PGC-1)  200 (61)  137 (71)*  63 (47)*  102 (76)*  60 (46)*  13 (68)*  32 (80)  34 (61)  36 (95)  140 (72)*  60 (45)*  168 (64)  32 (50)  ebpA  172 (52)  132 (68)*  40 (30)*  109 (81)*  46 (35)*  0*  35 (88)  42 (75)  32 (84)  146 (75)*  26 (19)*  153 (58)  19 (30)  ebpA+ebpB+ebpC (PGC-3)  146 (44)  111 (58)*  35 (26)*  93 (69)*  39 (30)*  0*  25 (62)  39 (70)  29 (76)  125 (64)*  21 (15)*  130 (49)  16 (25)  fms14  194 (59)  147 (76)*  47 (35)*  114 (85)*  57 (44)*  3 (16)*  37 (92)  44 (79)  33 (87)  148 (76)*  46 (34)*  167 (63)  27 (42)  fms14-17-13 (PGC-2)  185 (56)  141 (73)*  44 (33)*  111 (83)  53 (40)*  3 (16)*  37 (92)  43 (77)  31 (82)  143 (74)*  42 (31)*  160 (61)  25 (39)  fms11  197 (60)  149 (77)*  48 (36)*  123 (92)*  43 (33)*  12 (63)*  38 (95)  54 (96)  31 (82)  169 (87)*  28 (21)*  183 (69)*  14 (22)*  fms19  205 (62)  151 (78)*  54 (40)*  125 (93)*  50 (38)*  10 (53)*  38 (95)  54 (96)  33 (87)  174 (90)*  31 (23)*  189 (72)*  16 (25)*  fms16  202 (62)  147 (76)*  55 (41)*  121 (90)*  50 (38)*  11 (58)*  35 (88)  54 (96)  32 (84)  170 (88)*  31 (23)*  184 (70)*  17 (27)*  fms11-19-16 (PGC-4)  187 (57)  143 (74)*  44 (33)*  117 (87)*  42 (32)*  10 (53)*  35 (88)  52 (93)  30 (79)  161 (83)*  26 (19)*  174 (66)*  13 (20)*  Genea  Distribution [n (%)] of virulence genes among different groups   total (N = 328)  clinical (N = 193)  non-clinical (N = 135)  A1b (N = 134)  A2 (N = 131)  B (N = 19)  clade A1c   ampicillin resistantd (N = 194)  ampicillin susceptible (N = 134)  MDR (N = 264)  non-MDR (N = 64)  ST17 (N = 40)  ST18 (N = 56)  ST78 (N = 38)  fnm  322 (98)  190 (98)  132 (98)  132 (98)  128 (98)  19 (100)  40 (100)  56 (100)  36 (95)  192 (99)  130 (97)  259 (98)  63 (98)  orf2109  322 (98)  189 (98)  133 (98)  133 (99)  129 (98)  17 (90)  39 (98)  56 (100)  38 (100)  194 (100)  128 (96)  260 (98)  62 (97)  swpA  323 (98)  190 (98)  133 (98)  133 (99)  129 (98)  19 (100)  39 (98)  56 (100)  38 (100)  194 (100)  129 (96)  260 (98)  63 (98)  fms6  318 (97)  188 (97)  130 (96)  133 (99)  128 (98)  18 (95)  39 (98)  56 (100)  38 (100)  191 (98)  127 (95)  257 (97)  61 (95)  fms3  310 (94)  186 (96)  124 (92)  131 (98)  120 (92)  17 (90)  39 (98)  54 (96)  38 (100)  190 (98)  120 (90)  251 (95)  59 (92)  swpB  310 (94)  188 (97)  130 (96)  130 (97)  123 (94)  18 (95)  39 (98)  55 (98)  36 (95)  188 (97)  122 (91)  249 (94)  61 (95)  fms7  308 (94)  181 (94)  127 (94)  132 (98)  120 (92)  15 (79)  39 (98)  55 (98)  38 (100)  192 (99)  116 (87)  251 (95)  57 (89)  orf773  302 (92)  179 (93)  123 (91)  131 (98)*  126 (96)*  6 (32)*  38 (95)  56 (100)  37 (97)  193 (100)  109 (81)  256 (97)  46 (72)  fms12  290 (88)  166 (86)  124 (92)  117 (87)  116 (88)  18 (95)  36 (90)  47 (84)  34 (90)  170 (88)  120 (90)  230 (87)  60 (94)  swpC  285 (87)  169 (88)  116 (86)  120 (90)  110 (84)  18 (95)  37 (92)  56 (100)  27 (71)  164 (84)  121 (90)  223 (84)  62 (97)  sagA  282 (86)  154 (80)  128 (95)  114 (85)  112 (86)  12 (63)  33 (82)  53 (95)  29 (76)  173 (89)  101 (75)  230 (87)  48 (75)  fms21  278 (85)  169 (88)  109 (81)  127 (95)  100 (76)  13 (68)  37 (92)  53 (95)  37 (97)  180 (93)*  98 (73)*  234 (89)*  44 (69)*  ebpB  270 (82)  169 (88)  101 (75)  124 (92)*  105 (80)*  9 (47)*  36 (90)  52 (93)  36 (95)  172 (89)  100 (75)  226 (86)  46 (72)  ebpC  256 (78)  152 (79)  104 (77)  115 (86)  102 (78)  9 (47)  32 (80)  50 (89)  33 (87)  162 (84)  94 (70)  216 (82)  40 (62)  fms17  261 (80)  157 (81)  104 (77)  121 (90)*  105 (80)*  4 (21)*  37 (92)  47 (84)  37 (97)  168 (87)  93 (70)  221 (84)  40 (62)  fms13  273 (83)  163 (84)  110 (82)  121 (90)*  112 (86)*  5 (26)*  38 (95)  48 (86)  35 (92)  172 (89)  101 (75)  228 (86)  45 (70)  fms22  259 (79)  164 (85)  95 (70)  116 (87)  91 (70)  17 (90)  37 (92)  42 (75)  37 (97)  170 (88)*  89 (66)*  209 (79)  46 (72)  fms15  260 (79)  168 (87)*  92 (68)*  122 (91)*  103 (79)*  1 (5)*  37 (92)  54 (96)  31 (82)  178 (92)*  82 (61)*  229 (87)*  31 (48)*  scm  244 (74)  142 (74)  102 (76)  104 (78)*  108 (82)*  4 (21)*  36 (90)  36 (64)  32 (84)  148 (76)  96 (72)  207 (78)  37 (58)  acme  231 (70)  139 (72)  92 (68)  97 (72)  79 (60)  18 (95)  28 (70)  42 (75)  27 (71)  147 (76)  84 (63)  188 (71)  43 (67)  ptsD  171 (52)  150 (78)*  21 (16)*  125 (93)*  29 (22)*  0*  37 (92)  54 (96)  34 (90)  167 (86)*  4 (3)*  169 (64)*  2 (3)*  IS16  173 (53)  147 (76)*  26 (19)*  126 (94)*  30 (23)*  0*  38 (95)  53 (95)  35 (92)  167 (86)*  6 (4)*  170 (64)*  3 (5)*  sgrA  177 (54)  146 (76)*  31 (23)*  124 (92)*  32 (24)*  1 (5)*  36 (90)  53 (95)  35 (92)  164 (84)*  13 (10)*  167 (63)*  10 (16)*  orf1481  169 (52)  146 (76)*  23 (17)*  125 (93)*  26 (20)*  1 (5)*  38 (95)  54 (96)  33 (87)  163 (84)*  6 (4)*  165 (62)*  4 (6)*  hyl  39 (12)  35 (18)*  4 (3)*  32 (24)*  6 (5)*  0*  16 (40)  9 (16)  7 (18)  37 (19)*  2 (2)*  39 (15)  0  acm* (complete)  90 (27)  70 (36)*  20 (15)*  53 (40)*  19 (14)*  3 (16)*  10 (25)  28 (50)  15 (40)  74 (38)*  16 (12)*  79 (30)  11 (17)  esp  108 (33)  95 (49)*  13 (10)*  80 (60)*  17 (13)*  2 (10)*  33 (82)  16 (29)  31 (82)*  103 (53)*  5 (4)*  106 (40)*  2 (3)*  ecbA  114 (35)  98 (51)*  16 (12)*  124 (92)*  32 (24)*  2 (10)*  33 (82)  20 (36)  35 (92)*  107 (55)*  7 (5)*  111 (42)*  3 (5)*  fms20  206 (63)  140 (72)*  66 (49)*  103 (77)*  64 (49)*  13 (68)*  32 (80)  35 (62)  36 (95)  143 (74)*  63 (47)*  173 (66)  33 (52)  fms20+fms21 (PGC-1)  200 (61)  137 (71)*  63 (47)*  102 (76)*  60 (46)*  13 (68)*  32 (80)  34 (61)  36 (95)  140 (72)*  60 (45)*  168 (64)  32 (50)  ebpA  172 (52)  132 (68)*  40 (30)*  109 (81)*  46 (35)*  0*  35 (88)  42 (75)  32 (84)  146 (75)*  26 (19)*  153 (58)  19 (30)  ebpA+ebpB+ebpC (PGC-3)  146 (44)  111 (58)*  35 (26)*  93 (69)*  39 (30)*  0*  25 (62)  39 (70)  29 (76)  125 (64)*  21 (15)*  130 (49)  16 (25)  fms14  194 (59)  147 (76)*  47 (35)*  114 (85)*  57 (44)*  3 (16)*  37 (92)  44 (79)  33 (87)  148 (76)*  46 (34)*  167 (63)  27 (42)  fms14-17-13 (PGC-2)  185 (56)  141 (73)*  44 (33)*  111 (83)  53 (40)*  3 (16)*  37 (92)  43 (77)  31 (82)  143 (74)*  42 (31)*  160 (61)  25 (39)  fms11  197 (60)  149 (77)*  48 (36)*  123 (92)*  43 (33)*  12 (63)*  38 (95)  54 (96)  31 (82)  169 (87)*  28 (21)*  183 (69)*  14 (22)*  fms19  205 (62)  151 (78)*  54 (40)*  125 (93)*  50 (38)*  10 (53)*  38 (95)  54 (96)  33 (87)  174 (90)*  31 (23)*  189 (72)*  16 (25)*  fms16  202 (62)  147 (76)*  55 (41)*  121 (90)*  50 (38)*  11 (58)*  35 (88)  54 (96)  32 (84)  170 (88)*  31 (23)*  184 (70)*  17 (27)*  fms11-19-16 (PGC-4)  187 (57)  143 (74)*  44 (33)*  117 (87)*  42 (32)*  10 (53)*  35 (88)  52 (93)  30 (79)  161 (83)*  26 (19)*  174 (66)*  13 (20)*  a PVM genes are grouped according to their occurrence and clustering in the heatmap (see two main clusters in Figure 2a). b Isolates were divided into three groups: clade A1 (BAPS 2.1a, 3.3a1 and 3.3a2), clade B (BAPS 1) and clade A2 (all remaining BAPS subgroups). c Clade A1 was subdivided into three main clonal lineages: ST17 (BAPS 3.3a2), ST18 (BAPS 3.3a1) and ST78 (BAPS 2.1a). d Ampicillin-resistant isolates were considered to have an MIC of ampicillin of ≥ 8 mg/L according to EUCAST guidelines. e Normal acm genes and pseudogenes were identified in both clinical and non-clinical strains, but a higher proportion of pseudogenes was identified in non-clinical settings. * P < 0.0001 [extremely significant according to Fisher’s exact test (α = 0.05) using GraphPad Prism software, version 7.0a]. Results and discussion Diversity of Efm clones Figure 1(a) illustrates the population structure of Efm isolates (125 STs) which was split into clades A and B [A1 (47%), A2 (46%) and B (7%)] and BAPS (7 BAPS groups and 17 BAPS subgroups). The BAPS groups 2 (36%) and 3 (52%) were overrepresented in comparison with BAPS groups 1, 5, 6, 7 or 9 (1%–7%). While isolates from hospitalized patients, hospital wastewater and swine were distributed among the three clades, isolates from wild birds were found only within clades A1 and A2, poultry isolates were confined to clade A2, and the remaining isolates (from cows, trout and healthy volunteers) belonged to clades A2 and B. Nevertheless, a strong relationship between BAPS 2.1a, 3.3a1 and 3.3a2 (forming the clade A1) and clinical isolates (85%), between BAPS 2.1b and animals (64%) and between BAPS 3.2 and pigs (69%) was observed (P < 0.05). Although clinical isolates frequently belonged to clade A1 (65%), a significant percentage belonged to clade A2 (31%) and more rarely to B (4%). Other studies have also recently linked a significant number of clinical isolates to clades A2 and B,3,4,20 suggesting the relevance of both endogenous and exogenous acquisition routes for infections caused by Efm.20,22 Figure 1. View largeDownload slide Snapshot of the population structure of Efm isolates tested by MLST (n = 284) according to ST (www.phyloviz.net) and BAPS subgroups. Circles represent different ST (ST in numbers within the circles) and the size of each circle is proportional to the number of isolates of each ST. (a) Isolates are differentiated by origin. BAPS grouping is the foundation for this phylogenetic tree, which was further split according to the grouping of isolates into clades. (b) Isolates are differentiated by the number of virulence genes. C, clinical; NC, non-clinical; H, hospital. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide Snapshot of the population structure of Efm isolates tested by MLST (n = 284) according to ST (www.phyloviz.net) and BAPS subgroups. Circles represent different ST (ST in numbers within the circles) and the size of each circle is proportional to the number of isolates of each ST. (a) Isolates are differentiated by origin. BAPS grouping is the foundation for this phylogenetic tree, which was further split according to the grouping of isolates into clades. (b) Isolates are differentiated by the number of virulence genes. C, clinical; NC, non-clinical; H, hospital. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Distribution of PVM between Efm of clades A1, A2 and B demonstrates the repertoire of PVM among different Efm populations and the relevance of specific PVM as genetic markers of infection-derived strains Strains from clade A1, independent of their source, were the more frequently enriched in PVM (often >25 genes) and all isolates associated with hospital outbreak clones dating from ≥1992 carried 25–32 genes (Figure 1b). Within clade A2, a variable set of PVM genes was observed (<15 to >25 genes), even within isolates from the same source. Enriched A2 strains (>25 PVM genes) were mostly obtained from hospitals, hospital wastewater or wild birds and belonged to BAPS 2.3a, 3.1, 5 and 7. With the exception of two isolates enriched in PVM (ST108 and ST890 from a pig and a healthy volunteer), a reduced number of PVM were observed in clade B strains, independent of their source. Twenty PVM were carried by >70% of the isolates analysed, independent of their source or clade. They included surface-exposed genes [acm, scm, orf773, orf2109, fms3, fms6, fms7, fms12, fms15, fms22 and members of PGC such as PGC-1 (fms21), PGC-2 (ebpB, ebpC) and PGC-3 (fms13, fms17)], other adhesins (fnm, sagA) and all WxL genes (swpA, swpB, swpC) (Figure 2a and Table 2). This remarkable common backbone corresponding to a plethora of surface proteins with variable binding specificities may help Efm colonize different hosts, contributing both to its widespread occurrence and its virulence. It is possible that some of these predominant fms genes have differential expressions in different hosts or may be non-functional (pseudogenes), but that is beyond the scope of this study.9,28 Moreover, gene sequencing of representative isolates showed different variants of surface proteins and pilins in clinical and non-clinical Efm (hospital variant 1 and community variant 2, respectively, as described by Qin et al.34) (data not shown). Figure 2. View largeDownload slide (a) Heatmap of all virulence genes in the 328 Efm isolates from different sources. ABR, antibiotic resistance; E, environment, HP, hospitalized patients; HV, healthy volunteers. (b) Representation of the different PVM profiles (only major PVM) exhibited only by clinical Efm isolates causing different human infections according to their clades. *PVM patterns identified in different clades. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide (a) Heatmap of all virulence genes in the 328 Efm isolates from different sources. ABR, antibiotic resistance; E, environment, HP, hospitalized patients; HV, healthy volunteers. (b) Representation of the different PVM profiles (only major PVM) exhibited only by clinical Efm isolates causing different human infections according to their clades. *PVM patterns identified in different clades. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The other 13 genes analysed [either encoding surface-exposed molecules (esp, sgrA, ecbA, complete acm*), pili (fms11-fms19-fms16 (PGC-4), fms14, fms20 and ebpA) or other functions (IS16, hyl, ptsD, orf1481)], from now on designated major PVM, were more commonly identified in isolates from hospitalized patients (P < 0.0001), mostly within clade A1, and to a lesser extent clade A2 and/or B (Tables 2 and 3 and Figure 2a). ptsD, IS16, sgrA and orf1481 (group I) were the most common PVM genes found among clinical Efm, and were mostly from clade A1, less frequent in clade A2 and nearly absent in clade B (P < 0.0001; Tables 2 and 3 and Figure 2a). Among these four genes, ptsD was predominant among ampicillin-resistant Efm, both from clinical (all but four ampicillin resistant with an MIC of ampicillin of ≥8 mg/L; clades A1 and A2) and community (pigs, wild birds and healthy volunteers; all from clade A1) sources. Based on this observation of ptsD as a putative marker of relevant ampicillin-resistant Efm clones, we screened this gene in an additional 1077 Efm isolates from different sources and clonal backgrounds (317 ampicillin resistant and 760 ampicillin susceptible; Table S2). Only eight ampicillin-susceptible isolates carried ptsD (0.7%), four of which were associated with bloodstream infections, further confirming the strong positive association of ptsD with ampicillin-resistant Efm. IS16, orf1481 and sgrA were infrequent (5%–15%) but not as rare as ptsD in non-clinical ampicillin-susceptible Efm isolates (Table 2 and data not shown). This study corresponds to the first clear association of ptsD with diverse ampicillin-resistant Efm dating from 1986, and corroborates recent observations from other studies indicating the relevance of genes from phosphotransferase systems in individuals receiving β-lactam treatment.35 On the other hand, mutations related to ampicillin resistance seem to largely affect transport of carbohydrates.36 hyl, acm*, esp and ecbA genes (group II) were more often observed among clinical ampicillin-resistant Efm from clade A1 (P < 0.0001), although they were present at lower rates in all Efm studied, including clinical ones. esp and ecbA have been variably detected in different studies, which might depend on the clonal variability of the sample (strains of ST18 lineage were less enriched in esp and ecbA than ST17 and ST78 lineages in this collection; Table 2). Recent reports of local outbreaks or local collections describe high rates (>80%) of esp and ecbA in ST78 clones,20,37 while retrospective regional data show low rates (<30%) of these genes in clinical ST18-related clones20 (A. R. Freitas, C. Novais and L. Peixe, unpublished data). To a lesser extent, esp, ecbA and acm* were equally distributed between clades A2 and B. In contrast to pioneer studies, the functional role of hyl has been elucidated and appears not to be essential in the pathogenesis of Efm, possibly explaining its variable and often low frequency in infection-derived strains.38 Table 3. Patterns of major PVM among Efm isolates (n = 328) from different sources Patterns of major PVM   n  Hospital, n = 193  Ambulatory, n = 1  Healthy animals, n = 16  Farm animals, n = 54  Food, n = 39  Environment, n = 19  Wild birds, n = 6  group I   group II   group III   ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  PGC  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  —  1 PGC (1), 2 PGC (5), 3 PGC (18), 4 PGC (13)  ≥7  32a          4  1  ptsD  orf1481  IS16  sgrA  —  —  acm*  —  2 PGC (4), 3 PGC (7), 4 PGC (7)  ≥7  18a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  —  2 PGC (4), 3 PGC (4), 4 PGC (9)  ≥9  16a            1  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  hyl  3 PGC (5), 4 PGC (7)  ≥10  12a              ptsD  orf1481  IS16  sgrA  —  ecbA  acm*  —  3 PGC (7), 4 PGC (4)  ≥9  10a            1  ptsD  orf1481  IS16  sgrA  esp  —  —  —  2 PGC (1), 3 PGC (8), 4 PGC (2)  ≥7  9a          2    ptsD  orf1481  IS16  sgrA  —  ecbA  —  —  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥7  9a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥10  8a  1            ptsD  orf1481  IS16  sgrA  —  —  —    1 PGC (3), 2 PGC (2), 3 PGC (3), 4 PGC (2)  ≥5  7a          3    ptsD  orf1481  IS16  sgrA  —  —  acm*  hyl  3 PGC (4), 4 PGC (3)  ≥9  7a            1  ptsD  orf1481  IS16  sgrA  esp  —  acm*    2 PGC (2), 3 PGC (2), 4 PGC (2)  ≥8  6a              ptsD  orf1481  IS16  —  esp  ecbA  acm*  hyl  1 PGC  8  1              ptsD  orf1481  IS16  sgrA  esp  —  —  hyl  4 PGC  10  1              ptsD  orf1481  IS16  sgrA  esp  ecbA  —    —  6  1              ptsD  orf1481  IS16  —  esp  ecbA  acm*    4 PGC  10              1  ptsD  orf1481  IS16  —  —  —  acm*  hyl  2 PGC  7  1              ptsD  orf1481  IS16  sgrA  —  —  —  hyl  3 PGC (2), 4 PGC (2)  ≥8  2          2    ptsD  orf1481  IS16  sgrA  —  —  —    2 PGC  6      1          ptsD  orf1481  IS16  —  esp  ecbA  —  hyl  3 PGC  9  1              ptsD  orf1481  IS16  —  esp  ecbA  —    4 PGC (1)  9  1              ptsD  orf1481  IS16  —  —  —  —    3 PGC (1), 4 PGC (1)  ≥6            2    ptsD  orf1481  —  —  —  ecbA  —    —  3  1              ptsD  orf1481  —  sgrA  esp  ecbA  acm*    3 PGC  9        1        ptsD  orf1481  —  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥3  2              ptsD  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  9  2              ptsD  —  —  —  —  —  —  —  —  1  4              —  orf1481  IS16  —  esp  ecbA  —  —  3 PGC  7  1              —  orf1481  —  sgrA  —  —  —  —  2 PGC  4  1              —  orf1481  —  —  —  —  —  —  1 PGC (1), 2 PGC (1)  ≥2          2      —  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  8  1              —  —  IS16  sgrA  —  —  —  —  2 PGC (1), 3 PGC (2)  ≥4        3        —  —  IS16  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥2  1      1  1      —  —  IS16  —  esp  ecbA    hyl  —  4  1              —  —  IS16  —  esp  —  —  —  3 PGC (1)  5            1    —  —  —  sgrA  —  —  —  —  1 PGC (1), 2 PGC (2), 3 PGC (3)  ≥2  1    1  1  3      —  —  —  sgrA  —  —  acm*  —  1 PGC (1), 2 PGC (1)  ≥3        1  1      —  —  —  —  esp  —  acm*  —  1 PGC  3      1          —  —  —  —  —  ecbA  acm*  —  1 PGC  3          1      —  —  —  —  —  ecbA  —  —  1 PGC (2)  2        5        —  —  —  —  —  —  acm*  —  1 PGC (4), 2 PGC (6)  ≥2  3      4  3      —  —  —  —  —  —  —  esp  1 PGC (1)  2  2              —  —  —  —  —  —  —  —  1 PGC (37), 2 PGC (23), 3 PGC (11), 4 PGC (1)  ≥1  28b    5  17  17  4  1  —  —  —  —  —  —  —  —  —  0  3    8  21  11  1    Patterns of major PVM   n  Hospital, n = 193  Ambulatory, n = 1  Healthy animals, n = 16  Farm animals, n = 54  Food, n = 39  Environment, n = 19  Wild birds, n = 6  group I   group II   group III   ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  PGC  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  —  1 PGC (1), 2 PGC (5), 3 PGC (18), 4 PGC (13)  ≥7  32a          4  1  ptsD  orf1481  IS16  sgrA  —  —  acm*  —  2 PGC (4), 3 PGC (7), 4 PGC (7)  ≥7  18a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  —  2 PGC (4), 3 PGC (4), 4 PGC (9)  ≥9  16a            1  ptsD  orf1481  IS16  sgrA  esp  ecbA  —  hyl  3 PGC (5), 4 PGC (7)  ≥10  12a              ptsD  orf1481  IS16  sgrA  —  ecbA  acm*  —  3 PGC (7), 4 PGC (4)  ≥9  10a            1  ptsD  orf1481  IS16  sgrA  esp  —  —  —  2 PGC (1), 3 PGC (8), 4 PGC (2)  ≥7  9a          2    ptsD  orf1481  IS16  sgrA  —  ecbA  —  —  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥7  9a              ptsD  orf1481  IS16  sgrA  esp  ecbA  acm*  hyl  2 PGC (1), 3 PGC (4), 4 PGC (4)  ≥10  8a  1            ptsD  orf1481  IS16  sgrA  —  —  —    1 PGC (3), 2 PGC (2), 3 PGC (3), 4 PGC (2)  ≥5  7a          3    ptsD  orf1481  IS16  sgrA  —  —  acm*  hyl  3 PGC (4), 4 PGC (3)  ≥9  7a            1  ptsD  orf1481  IS16  sgrA  esp  —  acm*    2 PGC (2), 3 PGC (2), 4 PGC (2)  ≥8  6a              ptsD  orf1481  IS16  —  esp  ecbA  acm*  hyl  1 PGC  8  1              ptsD  orf1481  IS16  sgrA  esp  —  —  hyl  4 PGC  10  1              ptsD  orf1481  IS16  sgrA  esp  ecbA  —    —  6  1              ptsD  orf1481  IS16  —  esp  ecbA  acm*    4 PGC  10              1  ptsD  orf1481  IS16  —  —  —  acm*  hyl  2 PGC  7  1              ptsD  orf1481  IS16  sgrA  —  —  —  hyl  3 PGC (2), 4 PGC (2)  ≥8  2          2    ptsD  orf1481  IS16  sgrA  —  —  —    2 PGC  6      1          ptsD  orf1481  IS16  —  esp  ecbA  —  hyl  3 PGC  9  1              ptsD  orf1481  IS16  —  esp  ecbA  —    4 PGC (1)  9  1              ptsD  orf1481  IS16  —  —  —  —    3 PGC (1), 4 PGC (1)  ≥6            2    ptsD  orf1481  —  —  —  ecbA  —    —  3  1              ptsD  orf1481  —  sgrA  esp  ecbA  acm*    3 PGC  9        1        ptsD  orf1481  —  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥3  2              ptsD  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  9  2              ptsD  —  —  —  —  —  —  —  —  1  4              —  orf1481  IS16  —  esp  ecbA  —  —  3 PGC  7  1              —  orf1481  —  sgrA  —  —  —  —  2 PGC  4  1              —  orf1481  —  —  —  —  —  —  1 PGC (1), 2 PGC (1)  ≥2          2      —  —  IS16  sgrA  esp  ecbA  —  —  4 PGC  8  1              —  —  IS16  sgrA  —  —  —  —  2 PGC (1), 3 PGC (2)  ≥4        3        —  —  IS16  —  —  —  —  —  1 PGC (1), 3 PGC (1)  ≥2  1      1  1      —  —  IS16  —  esp  ecbA    hyl  —  4  1              —  —  IS16  —  esp  —  —  —  3 PGC (1)  5            1    —  —  —  sgrA  —  —  —  —  1 PGC (1), 2 PGC (2), 3 PGC (3)  ≥2  1    1  1  3      —  —  —  sgrA  —  —  acm*  —  1 PGC (1), 2 PGC (1)  ≥3        1  1      —  —  —  —  esp  —  acm*  —  1 PGC  3      1          —  —  —  —  —  ecbA  acm*  —  1 PGC  3          1      —  —  —  —  —  ecbA  —  —  1 PGC (2)  2        5        —  —  —  —  —  —  acm*  —  1 PGC (4), 2 PGC (6)  ≥2  3      4  3      —  —  —  —  —  —  —  esp  1 PGC (1)  2  2              —  —  —  —  —  —  —  —  1 PGC (37), 2 PGC (23), 3 PGC (11), 4 PGC (1)  ≥1  28b    5  17  17  4  1  —  —  —  —  —  —  —  —  —  0  3    8  21  11  1    n, number of genes; PVM, putative virulence markers. a These isolates correspond to outbreak and epidemic Efm strains causing severe human infections. b Most of these isolates correspond to ampicillin-susceptible bloodstream Efm strains from clades A2 and B. Regarding pili (group III), the presence of the accessory fms20, fms14 and ebpA genes correlated well with the presence of complete PGC-1, PGC-2 and PGC-3, respectively. They were associated with both clinical and non-clinical Efm from different sources including foodstuffs and animals (Tables 2 and 3 and Figure 2a), but in this case too, sequencing of representatives demonstrated different protein variants between clinical (hospital variant 1) and non-clinical Efm isolates. Moreover, the co-occurrence of the four complete PGC was significantly more abundant in clinical ampicillin-resistant Efm from clade A1 (P < 0.0001; Table 2). In particular, the complete PGC-2 was present in all Efm from BAPS 3.1 (clade A2), a clonal subgroup that has been associated with miscellaneous isolates including limited clonal outbreaks (e.g. ST125, ST280).21 Its enrichment in specific pili might confer a selective advantage in different hosts. In contrast to the other PGC, both the accessory (fms11, fms19) and major (fms16) subunit pili genes of PGC-4 were strongly linked to ampicillin-resistant Efm strains. PGC-4 was identified in our work in different animal ampicillin-resistant Efm isolates from clade A2 (Figure 2a) and, in a previous study, were observed in most ampicillin-resistant isolates from pet animals.29 An association between specific PVM and particular types of infection could not be established. Current criteria for ‘safe Efm’ (MIC of ampicillin of ≤2 mg/L and absence of esp, hyl and IS16) need revising Figure 2(b) shows the distribution of the PVM profiles identified (n = 29) in the infection-derived isolates from the three clades, after including only major PVM from groups I–III. Some dominant PVM profiles were shared between different clades in different proportions. A significant number of Efm from clades A2 and B (including bloodstream isolates) only carried 1–3 complete PGC, a rare profile in clade A1. The seven bloodstream isolates from clade B contained esp + PGC-1 (n = 1), sgrA + PGC-4 (n = 1) and PGC-1 and/or PGC-4 (n = 5). When specifically analysing the Efm strains with an MIC of ampicillin of ≤2 mg/L, 29 ampicillin-susceptible Efm causing infections [clades: A1, n = 2; A2, n = 20; and B, n = 7; 28 bacteraemia and 1 urinary tract infection (UTI)] were identified with different PVM profiles (ptsD, sgrA, orf1481, acm* and/or PGC-1 to -4) but lacking esp, hyl and IS16 (Tables 4 and 5). For example, different strains from clade B (e.g. ST74, ST85; Table 4) corresponded to non-MDR ampicillin-susceptible bacteraemic isolates that only carried sgrA + PGC-4 or PGC. Twenty strains from clade A2 (15 STs) carried variable PVM profiles, mainly including PGC. Even though we cannot infer their function uniquely from their presence, which could be due, for example, to the high levels of recombination of Efm, our data clearly indicate that Efm strains with an MIC of ampicillin of ≤2 mg/L and which lack esp, hyl and IS16 genes can be associated with severe human infections, and more rarely with major human clones of clade A1. Table 4. Characteristics and major PVM of infection-derived Efm that had an MIC of ampicillin of ≤ 2 mg/L and lacked all IS16, esp and hyl genes (n = 29) Clade  BAPS  ST  ABR profile  Infection  Ward  MIC of ampicillin (mg/L)  Major PVM  Country  Year  A1  2.1a  ST442  CIP, ERY, STR, CHL, Q/D  bloodstream  infectious diseases  2  orf1481, sgrA, PGC-1, PGC-2  Spain  2005    3.3a1  ST18  CIP, ERY, STR  bloodstream  gastroenterology  1  PGC-1, PGC-2  Spain  1995  A2  3.1  ST22  ERY  bloodstream  emergency room  0.5  PGC-1, PGC-2, PGC-3  Spain  1995    3.1  ST22  CIP, ERY, TET  bloodstream  allergy (medical)  0.38  acm*, PGC-2  Spain  2010    3.1  ST22  ERY  bloodstream  emergency room  0.25  PGC-1, PGC-2, PGC-3  Spain  1999    3.1  ST32  ERY, TET  bloodstream  emergency room  1  acm*, PGC-2, PGC-3  Spain  1995    3.1  ST32  CIP, ERY  bloodstream  pneumology  0.5  ptsD, orf1481, PGC-3  Spain  2011    3.1  ST533  CIP, ERY  bloodstream  oncology  1  PGC-1, PGC-2  Spain  2006    3.1  ST214  CIP  bloodstream  paediatric cardiology  0.25  PGC-1, PGC-2, PGC-3  Spain  2006    3.1  ST214  CIP, ERY  bloodstream  gastroenterology  0.125  acm*, PGC-2  Spain  2006    3.2  ST29  CIP, ERY  bloodstream  emergency room  0.096  —  Spain  2007    3.3b  ST102  CIP, ERY  bloodstream  emergency room  0.25  PGC-1  Spain  2001    3.3b  ST888  CIP, ERY  bloodstream  emergency room  0.125  PGC-1  Spain  1999    3.3b  ST102  CIP, ERY  bloodstream  gastroenterology  0.25  ptsD  Spain  2008    2.1b  ST46  CIP, ERY  bloodstream  emergency room  0.25  sgrA, PGC-2, PGC-4  Spain  2001    2.1b  ST69  CIP  bloodstream  ICU  0.19  PGC-1, PGC-2  Spain  2005    2.1b  ST850  CIP, ERY  bloodstream  oncology  0.5  PGC-1, PGC-2  Spain  2011    2.3a  ST21  CIP, ERY  bloodstream  emergency room  2  PGC-2, PGC-3  Spain  1997    2.3a  ST1054  CIP  UTI  neurosurgery  0.25  PGC-2  Portugal  2001    2.3a  ST849  —  bloodstream  emergency room  1  ptsD, orf1481, PGC-2, PGC-3, PGC-4  Spain  2011    2.3b  ST247  —  bloodstream  emergency room  0.38  PGC-1, PGC-2, PGC-3  Spain  2000    7  ST675  CIP  bloodstream  emergency room  0.125  PGC-2  Spain  2008  B  1.2  ST74  ERY  bloodstream  emergency room  2  sgrA, PGC-4  Spain  2001    1.2  ST85  —  bloodstream  neurology  0.5  PGC-1, PGC-4  Spain  1995    1.2  ST85  ERY  bloodstream  emergency room  1  PGC-1  Spain  1997    1.2  ST96  CIP, ERY  bloodstream  emergency room  0.05  —  Spain  2002    1.2  ST178  —  bloodstream  surgery  1.5  PGC-1, PGC-4  Spain  1999    1.2  ST296  ERY, Q/D  bloodstream  urology  0.75  PGC-4  Spain  2013    1.5  ST678  ERY  bloodstream  surgery  1.5  PGC-1  Spain  1996  Clade  BAPS  ST  ABR profile  Infection  Ward  MIC of ampicillin (mg/L)  Major PVM  Country  Year  A1  2.1a  ST442  CIP, ERY, STR, CHL, Q/D  bloodstream  infectious diseases  2  orf1481, sgrA, PGC-1, PGC-2  Spain  2005    3.3a1  ST18  CIP, ERY, STR  bloodstream  gastroenterology  1  PGC-1, PGC-2  Spain  1995  A2  3.1  ST22  ERY  bloodstream  emergency room  0.5  PGC-1, PGC-2, PGC-3  Spain  1995    3.1  ST22  CIP, ERY, TET  bloodstream  allergy (medical)  0.38  acm*, PGC-2  Spain  2010    3.1  ST22  ERY  bloodstream  emergency room  0.25  PGC-1, PGC-2, PGC-3  Spain  1999    3.1  ST32  ERY, TET  bloodstream  emergency room  1  acm*, PGC-2, PGC-3  Spain  1995    3.1  ST32  CIP, ERY  bloodstream  pneumology  0.5  ptsD, orf1481, PGC-3  Spain  2011    3.1  ST533  CIP, ERY  bloodstream  oncology  1  PGC-1, PGC-2  Spain  2006    3.1  ST214  CIP  bloodstream  paediatric cardiology  0.25  PGC-1, PGC-2, PGC-3  Spain  2006    3.1  ST214  CIP, ERY  bloodstream  gastroenterology  0.125  acm*, PGC-2  Spain  2006    3.2  ST29  CIP, ERY  bloodstream  emergency room  0.096  —  Spain  2007    3.3b  ST102  CIP, ERY  bloodstream  emergency room  0.25  PGC-1  Spain  2001    3.3b  ST888  CIP, ERY  bloodstream  emergency room  0.125  PGC-1  Spain  1999    3.3b  ST102  CIP, ERY  bloodstream  gastroenterology  0.25  ptsD  Spain  2008    2.1b  ST46  CIP, ERY  bloodstream  emergency room  0.25  sgrA, PGC-2, PGC-4  Spain  2001    2.1b  ST69  CIP  bloodstream  ICU  0.19  PGC-1, PGC-2  Spain  2005    2.1b  ST850  CIP, ERY  bloodstream  oncology  0.5  PGC-1, PGC-2  Spain  2011    2.3a  ST21  CIP, ERY  bloodstream  emergency room  2  PGC-2, PGC-3  Spain  1997    2.3a  ST1054  CIP  UTI  neurosurgery  0.25  PGC-2  Portugal  2001    2.3a  ST849  —  bloodstream  emergency room  1  ptsD, orf1481, PGC-2, PGC-3, PGC-4  Spain  2011    2.3b  ST247  —  bloodstream  emergency room  0.38  PGC-1, PGC-2, PGC-3  Spain  2000    7  ST675  CIP  bloodstream  emergency room  0.125  PGC-2  Spain  2008  B  1.2  ST74  ERY  bloodstream  emergency room  2  sgrA, PGC-4  Spain  2001    1.2  ST85  —  bloodstream  neurology  0.5  PGC-1, PGC-4  Spain  1995    1.2  ST85  ERY  bloodstream  emergency room  1  PGC-1  Spain  1997    1.2  ST96  CIP, ERY  bloodstream  emergency room  0.05  —  Spain  2002    1.2  ST178  —  bloodstream  surgery  1.5  PGC-1, PGC-4  Spain  1999    1.2  ST296  ERY, Q/D  bloodstream  urology  0.75  PGC-4  Spain  2013    1.5  ST678  ERY  bloodstream  surgery  1.5  PGC-1  Spain  1996  ABR, antibiotic resistance; ERY, erythromycin; CIP, ciprofloxacin; CHL, chloramphenicol; Q/D, quinupristin/dalfopristin; STR, streptomycin; TET, tetracycline. acm*, complete acm. Only complete PGC were considered: PGC-1 (PilA), fms21-fms20; PGC-2, fms14-fms17-fms13; PGC-3 (PilB), ebpA-ebpB-ebpC; and PGC-4, fms11-fms19-fms16. Table 5. Distribution of major PVM [n (%)] among Efm isolates expressing different ampicillin MIC (mg/L) values     PGC-1 (PilA), fms21-fms20; PGC-2, fms14-fms17-fms13; PGC-3 (PilB), ebpA-ebpB-ebpC; PGC-4, fms11-fms19-fms16; PVM, putative virulence markers. Grey shading indicates groups of isolates that include at least one Efm clinical isolate involved in a human infection. Grey shading and bold formatting indicates groups of isolates that include at least one clinical isolate involved in a human infection with an MIC of ampicillin of ≤ 2 mg/L and which lacked all IS16/esp/hyl genes. Further studies are needed to clarify the pathogenic potential of different Efm hosts carrying individual or specific combinations of the PVM studied. In any case, our data clearly demonstrated that: (i) the PVM IS16, ptsD, sgrA and orf1481 were more frequent in clinical Efm and rare in community Efm, and might be considered as good markers of infection-derived Efm (group I); (ii) esp, hyl, ecbA and the complete acm were variably detected in clinical Efm, and never exclusively present in infection-derived Efm; (iii) even though more studies are necessary to better address the role of pili, our data suggest that particular variants (hospital variants in clinical Efm) might be important in infection-derived Efm [based on the fact that: (a) the protein variants of pili vary; (b) infection-derived Efm consistently carry at least one complete PGC; and (c) even infection-derived Efm expressing low levels of ampicillin resistance often exclusively carry PGC (Table 5)]; (iv) the occurrence of PVM from group II in infection-derived Efm strains was always associated with the detection of PVM from groups I and III; and (v) the cumulative presence of most of the PVM mentioned (groups I, II and III) is of concern since most outbreak, persistent and epidemic Efm strains were associated with the co-occurrence of several genes (>7) (Table 3). Not all PVM relevant to Efm pathogenicity were tested here and some are surely still unknown, but our data additionally supported the conclusion that a few genes promoting adherence to host tissues (such as PGC) and high-density colonization, which is probably enhanced in immunocompromised and/or elderly patients by multiple and prolonged antibiotic therapies, might favour human infection.20,39 PVM can be horizontally transferred along with antibiotic resistance fms21 (or pilA) was located in one or two plasmids of variable size (60–280 kb) carried since 1988 by unrelated Efm and occasionally co-located with vanA, IS16 and/or hylEfm, as observed in previous studies.32,33 Most fms21 (pilA) megaplasmids (>150 kb) hybridized with the RIP of pLG1 in isolates from different sources and backgrounds, which constitutes the first association of fms21 (pilA) with pLG1-like plasmids since its description.40 IS16 was either located in the chromosome and/or plasmids (50–220 kb), while hylEfm was only associated with megaplasmids (>180 kb). While fms21 (pilA) was identified in most of the 41 TC tested (85%, clinical and non-clinical donors), IS16, sgrA and hylEfm were transferred in 24%, 20% and 5% of the cases, respectively, all corresponding to donors of clinical origin. ptsD was not identified in any TC, despite its recent finding in a large non-pLG1 hyl-carrying plasmid41 and the descriptions of extensive transfer of sugar-related genes between clade A Efm strains,42 suggesting its location on the chromosome, on non-mobilizable plasmids, on plasmids that were not selected in the conjugation conditions used or on other mobile genetic elements (MGEs). It is noteworthy that sgrA, which was recently associated with a transferable pbp5 ampicillin resistant platform containing genes involved in different cell functions,17 was exclusively detected in clinical ampicillin-resistant donors (eight of nine TC). This is further supported by the data included in Table 5 showing a positive correlation for several PVM, including sgrA, and increasing ampicillin MIC values. Interestingly, within an ampicillin MIC range of 0.064–2 mg/L (mostly non-clinical isolates) the prevalence of PVM of groups I and II (see previous section) was quite low, and the near absence of isolates enriched in PVM for the lowest MIC tested (0.064–0.125 mg/L), compared with the high frequency of PVM for an ampicillin MIC of >256 mg/L, was remarkable. The MIC of ampicillin of ≤2 mg/L proposed by the EFSA is indeed associated with a lower number of PVM, but does not guarantee the absence of strains with virulence genes previously highly linked to human infections caused by Efm. The association between PVM and transferable MGEs means that the safety of particular Efm clonal lineages (e.g. clade B) cannot be assumed. Conclusions This is the first study to assess the distribution of an extended set of virulence genes in clonally diverse Efm from disparate hosts, correlating their occurrence with ampicillin-resistant phenotypes and proposing a consensus set of relevant PVM for the tracking of infection-derived Efm clones. One limitation of our study is the biased clinical collection of isolates (mostly VRE), though it still includes a large number of animal/food isolates compared with previous studies. Strains with higher ampicillin MIC values contained a higher number of PVM, independent of their clonal group or source, and might present a greater risk in terms of pathogenicity. PVM coding for surface (esp/sgrA/ecbA/complete acm) and pili proteins, or others enhancing colonization (hyl/ptsD/orf1481) or plasticity (IS16), were strongly associated with clinical Efm (mostly clade A1), but were also observed in clades A2/B at different rates. This study corroborates the association of particular pili variants with infection-derived Efm, stressing the need to sequence pili genes for strain risk analysis. Based on our data and in light of the current literature, we propose that ptsD, IS16, orf1481, sgrA and hospital variants of complete PGC should be considered when assessing the safety of Efm strains. The safety markers recently proposed by the EFSA as distinctive markers of safe Efm are not strict enough to avoid dissemination throughout microbial food additives of strains with the potential to cause human infections, thereby posing a risk to public health. Acknowledgements We are grateful for the technical assistance provided by Sara Aguiar, Elsa Martins, Houyem Elghaieb and Conception M. Rodríguez. We wish to thank Dr Ivan Literak and Dr Veronika Oravcová (University of Veterinary and Pharmaceutical Sciences Brno, Czech Republic), for providing the Efm strains of wild birds, and Dr Timothy P. Stinear (University of Melbourne, Australia), for the gift of Efm strains Aus0004 and Aus0085. Funding This work received financial support from the European Union (FEDER funds POCI/01/0145/FEDER/007728) and National Funds (FCT/MEC, Fundação para a Ciência e Tecnologia and Ministério da Educação e Ciência) under the Partnership Agreement PT2020 (UID/MULTI/04378/2013) and project NORTE-01-0145-FEDER-000011 [QREN – Qualidade e Segurança Alimentar – uma abordagem (nano)tecnológica]. This work was co-funded by a Research Grant (2013) from the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) (to A. R. F.). A. R. F. was supported by a fellowship (grant SFRH/BPD/96148/2013) from FCT through Programa Operacional Capital Humano (POCH). Work in the lab of T. M. C. is supported by the Instituto de Salud Carlos III (PI12-01581), and CIBER (CB06/02/0053), within the Plan Estatal de I+D+i 2013–2016 co-financed by the European Development Regional Fund ‘A way to achieve Europe’ (ERDF); also by the Regional Government of Madrid in Spain (PROMPT-S2010/BMD2414). Transparency declarations None to declare. Supplementary data Tables S1 and S2 are available as Supplementary data at JAC Online. References 1 ECDC. Annual Epidemiological Report 2014, Antimicrobial Resistance and Healthcare-Associated Infections. Stockholm, 2015. 2 Galloway-Peña J, Roh JH, Latorre M et al.   Genomic and SNP analyses demonstrate a distant separation of the hospital and community-associated clades of Enterococcus faecium. PLoS One  2012; 7: e30187. Google Scholar CrossRef Search ADS PubMed  3 Lebreton F, van Schaik W, McGuire AM et al.   Emergence of epidemic multidrug-resistant Enterococcus faecium from animal and commensal strains. MBio  2013; 4: e00534– 13. Google Scholar CrossRef Search ADS PubMed  4 Raven KE, Reuter S, Reynolds R et al.   A decade of genomic history for healthcare-associated Enterococcus faecium in the United Kingdom and Ireland. Genome Res  2016; 26: 1388– 96. Google Scholar CrossRef Search ADS PubMed  5 Guzman Prieto AM, van Schaik W, Rogers MRC et al.   Global emergence and dissemination of enterococci as nosocomial pathogens: attack of the clones? Front Microbiol  2016; 7: 788. Google Scholar CrossRef Search ADS PubMed  6 Sillanpää J, Prakash VP, Nallapareddy SR et al.   Distribution of genes encoding MSCRAMMs and pili in clinical and natural populations of Enterococcus faecium. J Clin Microbiol  2009; 47: 896– 901. Google Scholar CrossRef Search ADS PubMed  7 Rice LB, Carias L, Rudin S et al.   A potential virulence gene, hylEfm, predominates in Enterococcus faecium of clinical origin. J Infect Dis  2003; 187: 508– 12. Google Scholar CrossRef Search ADS PubMed  8 Hendrickx APA, van Wamel WJB, Posthuma G et al.   Five genes encoding surface-exposed LPXTG proteins are enriched in hospital-adapted Enterococcus faecium clonal complex 17 isolates. J Bacteriol  2007; 189: 8321– 32. Google Scholar CrossRef Search ADS PubMed  9 Sillanpaa J, Nallapareddy SR, Prakash VP et al.   Identification and phenotypic characterization of a second collagen adhesin, Scm, and genome-based identification and analysis of 13 other predicted MSCRAMMs, including four distinct pilus loci, in Enterococcus faecium. Microbiology  2008; 154: 3199– 211. Google Scholar CrossRef Search ADS PubMed  10 Werner G, Fleige C, Geringer U et al.   IS element IS16 as a molecular screening tool to identify hospital-associated strains of Enterococcus faecium. BMC Infect Dis  2011; 11: 80. Google Scholar CrossRef Search ADS PubMed  11 Zhang X, Top J, de Been M et al.   Identification of a genetic determinant in clinical Enterococcus faecium strains that contributes to intestinal colonization during antibiotic treatment. J Infect Dis  2013; 207: 1780– 6. Google Scholar CrossRef Search ADS PubMed  12 Heikens E, van Schaik W, Leavis HL et al.   Identification of a novel genomic island specific to hospital-acquired clonal complex 17 Enterococcus faecium isolates. Appl Environ Microbiol  2008; 74: 7094– 7. Google Scholar CrossRef Search ADS PubMed  13 Kim EB, Jin G-D, Lee J-Y et al.   Genomic features and niche-adaptation of Enterococcus faecium strains from Korean soybean-fermented foods. PLoS One  2016; 11: e0153279. Google Scholar CrossRef Search ADS PubMed  14 Sivertsen A, Billström H, Melefors Ö et al.   A multicentre hospital outbreak in Sweden caused by introduction of a vanB2 transposon into a stably maintained pRUM-plasmid in an Enterococcus faecium ST192 clone. PLoS One  2014; 9: e103274. Google Scholar CrossRef Search ADS PubMed  15 Almohamad S, Somarajan SR, Singh KV et al.   Influence of isolate origin and presence of various genes on biofilm formation by Enterococcus faecium. FEMS Microbiol Lett  2014; 353: 151– 6. Google Scholar CrossRef Search ADS PubMed  16 Sadowy E, Luczkiewicz A. Drug-resistant and hospital-associated Enterococcus faecium from wastewater, riverine estuary and anthropogenically impacted marine catchment basin. BMC Microbiol  2014; 14: 66. Google Scholar CrossRef Search ADS PubMed  17 Novais C, Tedim AP, Lanza VF et al.   Co-diversification of Enterococcus faecium core genomes and PBP5: evidences of pbp5 horizontal transfer. Front Microbiol  2016; 7: 1581. Google Scholar CrossRef Search ADS PubMed  18 Galloway-Peña JR, Rice LB, Murray BE. Analysis of PBP5 of early U.S. isolates of Enterococcus faecium: sequence variation alone does not explain increasing ampicillin resistance over time. Antimicrob Agents Chemother  2011; 55: 3272. Google Scholar CrossRef Search ADS PubMed  19 EFSA Panel on Additives and Products or Substances used in Animal Feed (FEEDAP). Guidance on the Safety Assessment of Enterococcus faecium in Animal Nutrition. 2012. 20 Tedim AP, Ruíz-Garbajosa P, Rodríguez MC et al.   Long-term clonal dynamics of Enterococcus faecium strains causing bloodstream infections (1995-2015) in Spain. J Antimicrob Chemother  2017; 72: 48– 55. Google Scholar CrossRef Search ADS PubMed  21 Freitas AR, Tedim AP, Francia MV et al.   Multilevel population genetic analysis of vanA and vanB Enterococcus faecium causing nosocomial outbreaks in 27 countries (1986-2012). J Antimicrob Chemother  2016; 71: 3351– 66. Google Scholar CrossRef Search ADS PubMed  22 Freitas AR, Coque TM, Novais C et al.   Human and swine hosts share vancomycin-resistant Enterococcus faecium CC17 and CC5 and Enterococcus faecalis CC2 clonal clusters harboring Tn1546 on indistinguishable plasmids. J Clin Microbiol  2011; 49: 925– 31. Google Scholar CrossRef Search ADS PubMed  23 Novais C, Coque TM, Costa MJ et al.   High occurrence and persistence of antibiotic-resistant enterococci in poultry food samples in Portugal. J Antimicrob Chemother  2005; 56: 1139– 43. Google Scholar CrossRef Search ADS PubMed  24 Novais C, Coque TM, Sousa JC et al.   Antimicrobial resistance among faecal enterococci from healthy individuals in Portugal. Clin Microbiol Infect  2006; 12: 1131– 4. Google Scholar CrossRef Search ADS PubMed  25 Willems RJL, Top J, van Schaik W et al.   Restricted gene flow among hospital subpopulations of Enterococcus faecium. MBio  2012; 3: e00151– 12. Google Scholar CrossRef Search ADS PubMed  26 Tedim AP, Ruiz-Garbajosa P, Corander J et al.   Population biology of intestinal Enterococcus isolates from hospitalized and nonhospitalized individuals in different age groups. Appl Environ Microbiol  2015; 81: 1820– 31. Google Scholar CrossRef Search ADS PubMed  27 EUCAST. Breakpoint Tables for Interpretation of MICs and Zone Diameters, Version 7.1, 2017. 28 Nallapareddy SR, Weinstock GM, Murray BE. Clinical isolates of Enterococcus faecium exhibit strain-specific collagen binding mediated by Acm, a new member of the MSCRAMM family. Mol Microbiol  2003; 47: 1733– 47. Google Scholar CrossRef Search ADS PubMed  29 de Regt MJA, van Schaik W, van Luit-Asbroek M et al.   Hospital and community ampicillin-resistant Enterococcus faecium are evolutionarily closely linked but have diversified through niche adaptation. PLoS One  2012; 7: e30319. Google Scholar CrossRef Search ADS PubMed  30 Galloway-Peña JR, Liang X, Singh KV et al.   The identification and functional characterization of WxL proteins from Enterococcus faecium reveal surface proteins involved in extracellular matrix interactions. J Bacteriol  2015; 197: 882– 92. Google Scholar CrossRef Search ADS PubMed  31 Vankerckhoven V, Van Autgaerden T, Vael C et al.   Development of a multiplex PCR for the detection of asa1, gelE, cylA, esp, and hyl genes in enterococci and survey for virulence determinants among European hospital isolates of Enterococcus faecium. J Clin Microbiol  2004; 42: 4473– 9. Google Scholar CrossRef Search ADS PubMed  32 Freitas AR, Tedim AP, Novais C et al.   Global spread of the hylEfm colonization-virulence gene in megaplasmids of the Enterococcus faecium CC17 polyclonal subcluster. Antimicrob Agents Chemother  2010; 54: 2660– 5. Google Scholar CrossRef Search ADS PubMed  33 Kim DS, Singh KV, Nallapareddy SR et al.   The fms21 (pilA)-fms20 locus encoding one of four distinct pili of Enterococcus faecium is harboured on a large transferable plasmid associated with gut colonization and virulence. J Med Microbiol  2010; 59: 505– 7. Google Scholar CrossRef Search ADS PubMed  34 Qin X, Galloway-Peña JR, Sillanpaa J et al.   Complete genome sequence of Enterococcus faecium strain TX16 and comparative genomic analysis of Enterococcus faecium genomes. BMC Microbiol  2012; 12: 135. Google Scholar CrossRef Search ADS PubMed  35 Pérez-Cobas AE, Artacho A, Knecht H et al.   Differential effects of antibiotic therapy on the structure and function of human gut microbiota. PLoS One  2013; 8: e80201. Google Scholar CrossRef Search ADS PubMed  36 Sacco E, Cortes M, Josseaume N et al.   Mutation landscape of acquired cross-resistance to glycopeptide and β-lactam antibiotics in Enterococcus faecium. Antimicrob Agents Chemother  2015; 59: 5306– 15. Google Scholar CrossRef Search ADS PubMed  37 Yang J, Jiang Y, Guo L et al.   Prevalence of diverse clones of vancomycin-resistant Enterococcus faecium ST78 in a Chinese hospital. Microb Drug Resist  2016; 22: 294– 300. Google Scholar CrossRef Search ADS PubMed  38 Panesso D, Montealegre MC, Rincón S et al.   The hylEfm gene in pHylEfm of Enterococcus faecium is not required in pathogenesis of murine peritonitis. BMC Microbiol  2011; 11: 20. Google Scholar CrossRef Search ADS PubMed  39 Flores-Mireles AL, Walker JN, Potretzke A et al.   Antibody-based therapy for enterococcal catheter-associated urinary tract infections. MBio  2016; 7: e01653-16. Google Scholar CrossRef Search ADS PubMed  40 Laverde Gomez JA, van Schaik W, Freitas AR et al.   A multiresistance megaplasmid pLG1 bearing a hylEfm genomic island in hospital Enterococcus faecium isolates. Int J Med Microbiol  2011; 301: 165– 75. Google Scholar CrossRef Search ADS PubMed  41 García-Solache M, Lebreton F, McLaughlin RE et al.   Homologous recombination within large chromosomal regions facilitates acquisition of β-lactam and vancomycin resistance in Enterococcus faecium. Antimicrob Agents Chemother  2016; 60: 5777– 86. Google Scholar CrossRef Search ADS PubMed  42 de Been M, van Schaik W, Cheng L et al.   Recent recombination events in the core genome are associated with adaptive evolution in Enterococcus faecium. Genome Biol Evol  2013; 5: 1524– 35. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please email: journals.permissions@oup.com.

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve Freelancer

DeepDyve Pro

Price
FREE
$49/month

$360/year
Save searches from
Google Scholar,
PubMed
Create lists to
organize your research
Export lists, citations
Read DeepDyve articles
Abstract access only
Unlimited access to over
18 million full-text articles
Print
20 pages/month
PDF Discount
20% off