A Potential Virulence Gene, hylEfm Predominates in Enterococcus faecium of Clinical Origin
1Research and Medical Services, Louis Stokes Cleveland Veterans Affairs Medical Center and Department of Medicine, Case Western Reserve University, Cleveland, Ohio; 2Center for the Study of Emerging and Reemerging Pathogens, Division of Infectious Diseases, Departments of Internal Medicine and Microbiology and Molecular Genetics, University of Texas Medical School at Houston; 3Belgian Reference Centre for Enterococci, Department of Microbiology, University Hospital Antwerp, Antwerp, Belgium; 4Robert Koch Institute, Wernigerode Branch, Wernigerode, Germany
Received 9 July 2002; revised 8 October 2002; electronically published 8 January 2003.
An open reading frame (hylEfm) with homologies to previously described hyaluronidase genes has been identified in nonstool isolates of Enterococcus faecium. E. faecium isolates (n = 577) from diverse sources were screened for the presence of hylEfm and espEfm, a putative virulence gene associated with epidemic E. faecium strains. The presence of espEfm was roughly twice that of hylEfm, but both were found primarily in vancomycin-resistant E. faecium isolates in nonstool cultures obtained from patients hospitalized in the United States. These data suggest that specific E. faecium strains may be enriched in determinants that make them more likely to cause clinical infections. Differences in the prevalence of these strains may help explain variations in the clinical importance of multiresistant E. faecium across different continents.
Financial support: Research Service of the Department of Veterans Affairs (Merit Review to L.B.R.); National Institutes of Health, Division of Microbiology and Infectious Diseases (grants RO1 AI45626 [to L.B.R.] and RO1 AI42399 [to B.E.M]).
Reprints or correspondence: Dr. Louis B. Rice, Medical Service 111(W), Louis Stokes Cleveland Veterans Affairs Medical Center, 10701 East Blvd., Cleveland, OH ().
Enterococcus faecium generally is considered to be a species of limited virulence, at least in some measure, because it was found to be responsible for <10% of enterococcal infections. However, in the past decade, the relative percentage of enterococcal infections attributed to E. faecium has increased, in some instances to >30% of enterococcal infections [1]. Vancomycin-resistant enterococci, most of which are E. faecium, are now responsible for 25% of enterococcal infections that occur in intensive care units in the United States . Moreover, a recent report from China described a toxic shock–
like illness in both pigs and humans associated with isolation of E. faecium from blood and cerebrospinal fluid [2]. Taken together, these data suggest that some E. faecium strains may be accumulating determinants that increase their virulence.
Recently, a gene designated "espEfm" has been identified, whose presence is a marker for a clone of E. faecium that has spread across continents [3]. As part of an analysis of genomic DNA from multiresistant E. faecium C68 and its transconjugant CV133 [4], we encountered an open reading frame (ORF) with significant deduced amino acid identity to hyaluronidase genes described elsewhere [5]. We designated this ORF "hylEfm." Hyaluronidase has been proposed as a virulence factor in Streptococcus pyogenes, Staphylococcus aureus, and Streptococcus pneumoniae [6]. We screened 577 E. faecium strains obtained from diverse sources for the presence of hylEfm and espEfm. Both were present predominantly in multiresistant E. faecium isolated from nonstool cultures from patients in US hospitals.
Materials and methods. E. faecium C68, an ampicillin- and vancomycin-resistant clinical isolate, has been described elsewhere [4]. E. faecium CV133 is an ampicillin- and vancomycin-resistant transconjugant, which resulted from a mating between C68 and recipient strain E. faecium GE-1 [4]. For our analysis of the prevalence of espEfm and hylEfm, we studied 577 strains. Three hundred eighty-two strains were derived from various sources in the United States, including 24 different states. The remaining 195 strains were derived from locations outside the United States, including Argentina, Belgium, France, Germany, Norway, Saudi Arabia, Spain, and the United Kingdom. Strains were identified as E. faecium by use of standard techniques at each site of isolation. Isolates from humans were classified as nonstool or stool and as from hospitalized or community-based persons. Other isolates were either from animals, waste water, or probiotics. The site of isolation was characterized as either in the United States or outside the United States. Pulsed field gel electrophoresis (PFGE) analysis was used to determine clonal relationships among isolates derived from a common area in close temporal proximity. A single representative of each clone was included in these experiments.
hylEfm was first identified within the genome of E. faecium CV133 and was confirmed subsequently as present in C68. The hylEfm sequence was sought in the publicly available partial genome of E. faecium TX0016 . Two contigs in the reported sequence were connected using information derived from our analysis of CV133. Polymerase chain reaction (PCR), restriction digestion, and hybridization studies confirmed that the region present in the genome of CV133 was indistinguishable from that available in the database (data not shown). The sequence and map structure reported here represents a combination of sequence and structural analysis of the hylEfm region of CV133 and the database sequence.
PCR screening for espEfm in E. faecium isolates was accomplished by generating a 945-bp product using forward primer espEfm 5′
-TTGCTAATGCTAGTCCACGACC-3′
and reverse primer espEfm 5′
-GCGTCAACACTTGCATTGCCGA-3′
[7]. PCR screening for hylEfm in E. faecium isolates was accomplished by generating a 661-bp product using forward primer hylEfm 5′
-GAGTAGAGGAATATCTTAGC-3′
and reverse primer hylEfm 5′
-AGGCTCCAATTCTGT-3′
. PCRs were commenced by use of Taq polymerase after overnight cultures of test strains, were diluted to 2.5 × 107 cfu/mL in sterile water, and were heated to 95°C for 10 min. In one laboratory (Louis Stokes Cleveland Veterans Affairs, Cleveland) for a minority of cases (20 animal isolates and 15 vancomycin-susceptible nonstool E. faecium isolates from the United States), hylEfm was detected by Southern hybridization of digested genomic DNA, using a digoxigenin-labeled PCR product as a probe, and by a chemiluminescent detection method (Boehringer Mannheim).
In a second laboratory (Center for the Study of Emerging and Reemerging Pathogens, Houston), hylEfm and espEfm were detected by colony hybridization under high stringency conditions, as described elsewhere [8]. To check for the presence of hyl and esp genes in E. faecium isolates, respective intragenic fragments were amplified using forward primer hylEfm (5′
-GTT AGA AGA AGT CTG GAA ACC G-3′) and reverse primer hylEfm (5′
-TGC TAA GAT ATT CCT CTA CTC G-3′) or forward primer espEfm (5′
-TTG CTA ATG CTA GTC CAC GAC C-3′) and reverse primer espEfm (5′
-GCG TCA ACA CTT GCA TTG CCG A-3′) by PCR, verified by sequence, and used as probes for colony hybridization. The hylEfm sequence has been entered into under accession number AF544400.
Results. The hylEfm ORF is 1659 bp. Its start codon is preceded by a credible ribosome binding site (GGAGA). The G-C content of the ORF is 36%. Translation of the ORF yields a putative 65,051-kDa protein 553 aa in length. BLASTX search revealed 42% identity and 60% similarity over 533 aa to hyaluronidases from S. pyogenes [5] ( accession nos. NP269657 and NP607664). Comparable similarities were demonstrated to a putative hyaluronidase gene from a strain of Clostridium perfringens [9] ( accession no. NP562150).
hylEfm is located within an ∼
20-kb region that is flanked by inverted copies of putative insertion sequence IS1476. Upstream of hylEfm within the IS1476-flanked region exist ORFs with similarities to 2 component regulator systems and a β -galactosidase gene. Downstream of hylEfm lie an ORF with no significant homologies and an ORF encoding a putative protein with homology to previously described GMP synthases.
fig.ommitted
Figure 1. Map of the region within which hylEfm is located. Map was constructed by combining sequences determined with cloning fragments from Enterococcus faecium CV133 and aligning these sequences with those available in the E. faecium genome database. The entire sequence of the CV133 region was not determined, but the match with the database was precise in sequences that were determined. In addition, restriction maps of the cloned CV133 fragments matched precisely those predicted based on the available database. Putative genes functions were assigned based on the best "hits" resulting from BLASTP searches of translated open reading frames.
Results of PCR amplification and Southern hybridizations of DNA from E. faecium isolates of diverse origins are detailed in . Of 577 strains, espEfm was present in 193 (33.4%), whereas hylEfm was present in 102 (17.7%). Both determinants were found predominantly among nonstool isolates. One hundred sixty-three (84%) of 193 espEfm-positive strains and 93 (91%) of 102 hylEfm-positive strains were nonstool isolates obtained from humans. In addition, the presence of both determinants was closely associated with resistance to vancomycin, with 157 (81%) of 193 espEfm-positive strains and 85 (83%) of 102 hylEfm-positive strains expressing vancomycin resistance. Both determinants were found predominantly in strains derived from the United States, with 173 (90%) of 193 espEfm-positive strains and 100 (98%) of 102 hylEfm-positive strains derived from the United States. The 2 hylEfm-positive strains from outside the United States were isolated in Argentina, so all hylEfm-positive strains in this study were isolated in the Western hemisphere. Of the 220 vancomycin-resistant E. faecium isolates derived from the United States (all from humans), 147 (67%) were positive for espEfm and 83 (38%) were positive for hylEfm. This was in stark contrast to vancomycin-resistant strains isolated from humans outside the United States, where only 10.7% and 2.7% were positive for espEfm and hylEfm, respectively . Although we did identify a few strains that were positive for hylEfm and negative for espEfm, the large majority (>90%) of hylEfm-positive strains also were positive for espEfm.
fig.ommitted
Table 1. Characterization of Enterococcus faecium strains by origin, vancomycin resistance, and presence of espEfm and hylEfm.
Discussion. We have described a novel ORF from a clinical strain of multiresistant E. faecium with homology to previously described hyaluronidase genes from S. pyogenes and C. perfringens. Hyaluronidase has been proposed as a potential virulence factor in several gram-positive bacteria, including S. aureus, S. pneumoniae, and S. pyogenes [6]. Recent data from pneumococcal models of infection suggest that hyaluronidase may contribute to invasion of the nasopharynx that precedes central nervous system infection and contributue to an as yet undefined manner to pneumococcal pneumonia [10, 11].
Both hylEfm and espEfm were found overwhelmingly in E. faecium strains isolated from nonstool cultures from humans. Moreover, all stool samples yielding hylEfm-positive isolates were derived from patients hospitalized at least 8 days. In no instance were stool samples obtained from healthy community dwellers or from animals positive for hylEfm-positive E. faecium. E. faecium has traditionally been considered a bacterial species of limited virulence due to its involvement in a relatively low percentage of enterococcal infections (∼
10%) and the difficulty of establishing animal models of infection [10, 11]. The increase in the percentage of enterococcal infections caused by E. faecium over the past decade in the United States (30%–
40% in several surveys) suggests that these bacteria may have become more virulent. Although the increased resistance to antimicrobial agents in E. faecium over the same period of time should not be overlooked as a predisposing factor in the modern hospital, it is probably not the only explanation. After all, data from Europe in the 1990s suggested that fecal colonization by vancomycin-resistant E. faecium was quite frequent in some areas of Europe [12, 13] but was almost nonexistent in community dwellers from the United States. Yet the rates of infection and colonization by vancomycin-resistant enterococci in the hospital were far greater in the United States than in Europe over the same period.
The association between the presence of espEfm and hylEfm and both the expression of vancomycin resistance and origin in the United States are consistent with a scenario in which E. faecium isolates from the United States that are prevalent in the nosocomial setting had acquired resistance (e.g., to ampicillin) and virulence before the introduction of vancomycin resistance operons. Published reports document the rise of ampicillin resistance in E. faecium before the recognition of vancomycin resistance in this species [14]. It is worth noting that TX0016, the strain analyzed for the partial E. faecium database, is an ampicillin-resistant, vancomycin-sensitive isolate from a case of endocarditis (B.E.M., unpublished data). The introduction of the VanA and VanB operons into E. faecium strains rich in antimicrobial resistance and (presumably) virulence determinants, was then prompted by widespread use of orally administered vancomycin in the United States during the 1980s. This marriage of virulence determinants and multiresistance in the United States yielded a group of strains particularly suited to causing clinical infection in the modern hospital.
Why, then, is the situation different in Europe? Perhaps it is because the population of vancomycin-resistant E. faecium in Europe was created through use of the growth-promoting antibiotic avoparcin in animals during the 1970s through 1990s. These E. faecium strains, which as animal colonizers were devoid of virulence determinants selected for causing infections in humans, were sufficient to colonize the human population in Europe. The presence of substantial numbers of these less virulent vancomycin-resistant strains may have precluded the rapid emergence of the more pathogenic varieties by competing for one of the selective niches (growth in glycopeptide-rich environment) enjoyed exclusively by the more pathogenic strains in the United States.
We did not, as part of this study, rigorously exclude the possibility that the association of espEfm and hylEfm with vancomycin resistance in the United States was due in some part to clonal expansion of specific strains. It is worth noting that we identified espEfm from 21 of 24 states from which we derived strains, and hylEfm for 15 of those states. We were careful to use PFGE analysis to exclude duplicate clones within geographically localized clusters. If clonal expansion explained the movement of organisms between different geographic areas in our study, then we may not have recognized it. We would argue, however, that the clonal expansion of those specific strains across geographic regions suggests that they may be particularly suited to causing infection, thereby supporting our argument rather than diluting it. This scenario would be consistent with the observation that the variant esp found in E. faecium is a marker for clonal dissemination across geographic regions [3].
At present, we cannot say with certainty whether and to what extent E. faecium actually makes hyaluronidase and under what conditions this protein may be synthesized or exported. Northern hybridization experiments indicate that the hyaluronidase ORF is transcribed under nonselective growth conditions in vitro (data not shown). Therefore, we have compelling reason to believe that the protein is synthesized at least under some environmental conditions. Experiments to characterize the extent of hyaluronidase production by hylEfm-positive strains are ongoing.
Comparison of different S. aureus genome sequences indicates that substantial variation exists between strains, with much of the variability involving antimicrobial resistance determinants and pathogenicity islands [15]. Our data suggest that E. faecium should be added to the growing list of human pathogens with a variety of genomic inventories, some of which have evolved within and are particularly suited to surviving and causing infections in the modern hospital environment.
References
1.
Sahm DF, Marsilio MK, Piazza G. Antimicrobial resistance in key bloodstream bacterial isolates: electronic surveillance with the Surveillance Network Database–
USA. Clin Infect Dis 1999; 29:259–
63.
2.
Lu HZ, Weng XH, Li H, Yin YK, Pang MY, Tang YW. Enterococcus faecium–
related outbreak with molecular evidence of transmission from pigs to humans. J Clin Microbiol 2002; 40:913–
7.
3.
Willems RJ, Homan W, Top J, et al. Variant esp gene as a marker of a distinct genetic lineage of vancomycin-resistant Enterococcus faecium spreading in hospitals. Lancet 2001; 357:853–
5.
4.
Carias LL, Rudin SD, Donskey CJ, Rice LB. Genetic linkage and cotransfer of a novel, vanB-containing transposon (Tn5382) and a low-affinity penicillin-binding protein 5 gene in a clinical vancomycin-resistant Enterococcus faecium isolate. J Bacteriol 1998; 180:4426–
34.
5.
Ferretti JJ, McShan WM, Ajdic D, et al. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc Natl Acad Sci USA 2001; 98:4658–
63.
6.
Hynes WL, Walton SL. Hyaluronidases of gram-positive bacteria. FEMS Microbiol Lett 2000; 183:201–
7.
7.
Shankar V, Baghdayan AS, Huycke MM, Lindahl G, Gilmore MS. Infection-derived Enterococcus faecalis strains are enriched in esp, a gene encoding a novel surface protein. Infect Immun 1999; 67:193–
200.
8.
Singh KV, Coque TM, Weinstock GM, Murray BE. In vivo testing of an Enterococcus faecalis efaA mutant and use of efaA homologs for species identification. FEMS Immunol Med Microbiol 1998; 21:323–
31.
9.
Shimizu T, Ohtani K, Hirakawa H, et al. Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. Proc Natl Acad Sci USA 2002; 99:996–
1001.
10.
Polissi A, Pontiggia A, Feger G, et al. Large-scale identification of virulence genes from Streptococcus pneumoniae. Infect Immun 1998; 66:5620–
9.
11.
Berry AM, Paton JC. Additive attenuation of virulence of Streptococcus pneumoniae by mutation of the genes encoding pneumolysin and other putative pneumococcal virulence proteins. Infect Immun 2000; 68:133–
40.
12.
van der Auwera P, Pensart N, Korten V, Murray BE, Leclercq R. Influence of oral glycopeptides on the fecal flora of human volunteers: selection of highly glycopeptide-resistant enterococci. J Infect Dis 1996; 173:1129–
36.
13.
Klare I, Badstubner D, Konstabel C, Bohme G, Claus H, Witte W. Decreased incidence of VanA-type vancomycin-resistant enterococci isolated from poultry meat and from fecal samples of humans in the community after discontinuation of avoparcin usage in animal husbandry. Microb Drug Resist 1999; 5:45–
51.
14.
Chirurgi VA, Oster SE, Goldberg AA, McCabe RE. Nosocomial acquisition of -lactamase–
negative, ampicillin-resistant enterococcus. Arch Intern Med 1992; 152:1457–
61.
15.
Fitzgerald JR, Sturdevant DE, Mackie SM, Gill SR, Musser JM. Evolutionary genomics of Staphylococcus aureus: insights into the origin of methicillin-resistant strains and the toxic shock syndrome epidemic. Proc Natl Acad Sci USA 2001; 98:8821–
6., 百拇医药(Louis B. Rice Lenore Carias Susan Rudin Carl Vael Herman Goossens Carola Konstabel Ingo Klare Sreedh)
Received 9 July 2002; revised 8 October 2002; electronically published 8 January 2003.
An open reading frame (hylEfm) with homologies to previously described hyaluronidase genes has been identified in nonstool isolates of Enterococcus faecium. E. faecium isolates (n = 577) from diverse sources were screened for the presence of hylEfm and espEfm, a putative virulence gene associated with epidemic E. faecium strains. The presence of espEfm was roughly twice that of hylEfm, but both were found primarily in vancomycin-resistant E. faecium isolates in nonstool cultures obtained from patients hospitalized in the United States. These data suggest that specific E. faecium strains may be enriched in determinants that make them more likely to cause clinical infections. Differences in the prevalence of these strains may help explain variations in the clinical importance of multiresistant E. faecium across different continents.
Financial support: Research Service of the Department of Veterans Affairs (Merit Review to L.B.R.); National Institutes of Health, Division of Microbiology and Infectious Diseases (grants RO1 AI45626 [to L.B.R.] and RO1 AI42399 [to B.E.M]).
Reprints or correspondence: Dr. Louis B. Rice, Medical Service 111(W), Louis Stokes Cleveland Veterans Affairs Medical Center, 10701 East Blvd., Cleveland, OH ().
Enterococcus faecium generally is considered to be a species of limited virulence, at least in some measure, because it was found to be responsible for <10% of enterococcal infections. However, in the past decade, the relative percentage of enterococcal infections attributed to E. faecium has increased, in some instances to >30% of enterococcal infections [1]. Vancomycin-resistant enterococci, most of which are E. faecium, are now responsible for 25% of enterococcal infections that occur in intensive care units in the United States . Moreover, a recent report from China described a toxic shock–
like illness in both pigs and humans associated with isolation of E. faecium from blood and cerebrospinal fluid [2]. Taken together, these data suggest that some E. faecium strains may be accumulating determinants that increase their virulence.
Recently, a gene designated "espEfm" has been identified, whose presence is a marker for a clone of E. faecium that has spread across continents [3]. As part of an analysis of genomic DNA from multiresistant E. faecium C68 and its transconjugant CV133 [4], we encountered an open reading frame (ORF) with significant deduced amino acid identity to hyaluronidase genes described elsewhere [5]. We designated this ORF "hylEfm." Hyaluronidase has been proposed as a virulence factor in Streptococcus pyogenes, Staphylococcus aureus, and Streptococcus pneumoniae [6]. We screened 577 E. faecium strains obtained from diverse sources for the presence of hylEfm and espEfm. Both were present predominantly in multiresistant E. faecium isolated from nonstool cultures from patients in US hospitals.
Materials and methods. E. faecium C68, an ampicillin- and vancomycin-resistant clinical isolate, has been described elsewhere [4]. E. faecium CV133 is an ampicillin- and vancomycin-resistant transconjugant, which resulted from a mating between C68 and recipient strain E. faecium GE-1 [4]. For our analysis of the prevalence of espEfm and hylEfm, we studied 577 strains. Three hundred eighty-two strains were derived from various sources in the United States, including 24 different states. The remaining 195 strains were derived from locations outside the United States, including Argentina, Belgium, France, Germany, Norway, Saudi Arabia, Spain, and the United Kingdom. Strains were identified as E. faecium by use of standard techniques at each site of isolation. Isolates from humans were classified as nonstool or stool and as from hospitalized or community-based persons. Other isolates were either from animals, waste water, or probiotics. The site of isolation was characterized as either in the United States or outside the United States. Pulsed field gel electrophoresis (PFGE) analysis was used to determine clonal relationships among isolates derived from a common area in close temporal proximity. A single representative of each clone was included in these experiments.
hylEfm was first identified within the genome of E. faecium CV133 and was confirmed subsequently as present in C68. The hylEfm sequence was sought in the publicly available partial genome of E. faecium TX0016 . Two contigs in the reported sequence were connected using information derived from our analysis of CV133. Polymerase chain reaction (PCR), restriction digestion, and hybridization studies confirmed that the region present in the genome of CV133 was indistinguishable from that available in the database (data not shown). The sequence and map structure reported here represents a combination of sequence and structural analysis of the hylEfm region of CV133 and the database sequence.
PCR screening for espEfm in E. faecium isolates was accomplished by generating a 945-bp product using forward primer espEfm 5′
-TTGCTAATGCTAGTCCACGACC-3′
and reverse primer espEfm 5′
-GCGTCAACACTTGCATTGCCGA-3′
[7]. PCR screening for hylEfm in E. faecium isolates was accomplished by generating a 661-bp product using forward primer hylEfm 5′
-GAGTAGAGGAATATCTTAGC-3′
and reverse primer hylEfm 5′
-AGGCTCCAATTCTGT-3′
. PCRs were commenced by use of Taq polymerase after overnight cultures of test strains, were diluted to 2.5 × 107 cfu/mL in sterile water, and were heated to 95°C for 10 min. In one laboratory (Louis Stokes Cleveland Veterans Affairs, Cleveland) for a minority of cases (20 animal isolates and 15 vancomycin-susceptible nonstool E. faecium isolates from the United States), hylEfm was detected by Southern hybridization of digested genomic DNA, using a digoxigenin-labeled PCR product as a probe, and by a chemiluminescent detection method (Boehringer Mannheim).
In a second laboratory (Center for the Study of Emerging and Reemerging Pathogens, Houston), hylEfm and espEfm were detected by colony hybridization under high stringency conditions, as described elsewhere [8]. To check for the presence of hyl and esp genes in E. faecium isolates, respective intragenic fragments were amplified using forward primer hylEfm (5′
-GTT AGA AGA AGT CTG GAA ACC G-3′) and reverse primer hylEfm (5′
-TGC TAA GAT ATT CCT CTA CTC G-3′) or forward primer espEfm (5′
-TTG CTA ATG CTA GTC CAC GAC C-3′) and reverse primer espEfm (5′
-GCG TCA ACA CTT GCA TTG CCG A-3′) by PCR, verified by sequence, and used as probes for colony hybridization. The hylEfm sequence has been entered into under accession number AF544400.
Results. The hylEfm ORF is 1659 bp. Its start codon is preceded by a credible ribosome binding site (GGAGA). The G-C content of the ORF is 36%. Translation of the ORF yields a putative 65,051-kDa protein 553 aa in length. BLASTX search revealed 42% identity and 60% similarity over 533 aa to hyaluronidases from S. pyogenes [5] ( accession nos. NP269657 and NP607664). Comparable similarities were demonstrated to a putative hyaluronidase gene from a strain of Clostridium perfringens [9] ( accession no. NP562150).
hylEfm is located within an ∼
20-kb region that is flanked by inverted copies of putative insertion sequence IS1476. Upstream of hylEfm within the IS1476-flanked region exist ORFs with similarities to 2 component regulator systems and a β -galactosidase gene. Downstream of hylEfm lie an ORF with no significant homologies and an ORF encoding a putative protein with homology to previously described GMP synthases.
fig.ommitted
Figure 1. Map of the region within which hylEfm is located. Map was constructed by combining sequences determined with cloning fragments from Enterococcus faecium CV133 and aligning these sequences with those available in the E. faecium genome database. The entire sequence of the CV133 region was not determined, but the match with the database was precise in sequences that were determined. In addition, restriction maps of the cloned CV133 fragments matched precisely those predicted based on the available database. Putative genes functions were assigned based on the best "hits" resulting from BLASTP searches of translated open reading frames.
Results of PCR amplification and Southern hybridizations of DNA from E. faecium isolates of diverse origins are detailed in . Of 577 strains, espEfm was present in 193 (33.4%), whereas hylEfm was present in 102 (17.7%). Both determinants were found predominantly among nonstool isolates. One hundred sixty-three (84%) of 193 espEfm-positive strains and 93 (91%) of 102 hylEfm-positive strains were nonstool isolates obtained from humans. In addition, the presence of both determinants was closely associated with resistance to vancomycin, with 157 (81%) of 193 espEfm-positive strains and 85 (83%) of 102 hylEfm-positive strains expressing vancomycin resistance. Both determinants were found predominantly in strains derived from the United States, with 173 (90%) of 193 espEfm-positive strains and 100 (98%) of 102 hylEfm-positive strains derived from the United States. The 2 hylEfm-positive strains from outside the United States were isolated in Argentina, so all hylEfm-positive strains in this study were isolated in the Western hemisphere. Of the 220 vancomycin-resistant E. faecium isolates derived from the United States (all from humans), 147 (67%) were positive for espEfm and 83 (38%) were positive for hylEfm. This was in stark contrast to vancomycin-resistant strains isolated from humans outside the United States, where only 10.7% and 2.7% were positive for espEfm and hylEfm, respectively . Although we did identify a few strains that were positive for hylEfm and negative for espEfm, the large majority (>90%) of hylEfm-positive strains also were positive for espEfm.
fig.ommitted
Table 1. Characterization of Enterococcus faecium strains by origin, vancomycin resistance, and presence of espEfm and hylEfm.
Discussion. We have described a novel ORF from a clinical strain of multiresistant E. faecium with homology to previously described hyaluronidase genes from S. pyogenes and C. perfringens. Hyaluronidase has been proposed as a potential virulence factor in several gram-positive bacteria, including S. aureus, S. pneumoniae, and S. pyogenes [6]. Recent data from pneumococcal models of infection suggest that hyaluronidase may contribute to invasion of the nasopharynx that precedes central nervous system infection and contributue to an as yet undefined manner to pneumococcal pneumonia [10, 11].
Both hylEfm and espEfm were found overwhelmingly in E. faecium strains isolated from nonstool cultures from humans. Moreover, all stool samples yielding hylEfm-positive isolates were derived from patients hospitalized at least 8 days. In no instance were stool samples obtained from healthy community dwellers or from animals positive for hylEfm-positive E. faecium. E. faecium has traditionally been considered a bacterial species of limited virulence due to its involvement in a relatively low percentage of enterococcal infections (∼
10%) and the difficulty of establishing animal models of infection [10, 11]. The increase in the percentage of enterococcal infections caused by E. faecium over the past decade in the United States (30%–
40% in several surveys) suggests that these bacteria may have become more virulent. Although the increased resistance to antimicrobial agents in E. faecium over the same period of time should not be overlooked as a predisposing factor in the modern hospital, it is probably not the only explanation. After all, data from Europe in the 1990s suggested that fecal colonization by vancomycin-resistant E. faecium was quite frequent in some areas of Europe [12, 13] but was almost nonexistent in community dwellers from the United States. Yet the rates of infection and colonization by vancomycin-resistant enterococci in the hospital were far greater in the United States than in Europe over the same period.
The association between the presence of espEfm and hylEfm and both the expression of vancomycin resistance and origin in the United States are consistent with a scenario in which E. faecium isolates from the United States that are prevalent in the nosocomial setting had acquired resistance (e.g., to ampicillin) and virulence before the introduction of vancomycin resistance operons. Published reports document the rise of ampicillin resistance in E. faecium before the recognition of vancomycin resistance in this species [14]. It is worth noting that TX0016, the strain analyzed for the partial E. faecium database, is an ampicillin-resistant, vancomycin-sensitive isolate from a case of endocarditis (B.E.M., unpublished data). The introduction of the VanA and VanB operons into E. faecium strains rich in antimicrobial resistance and (presumably) virulence determinants, was then prompted by widespread use of orally administered vancomycin in the United States during the 1980s. This marriage of virulence determinants and multiresistance in the United States yielded a group of strains particularly suited to causing clinical infection in the modern hospital.
Why, then, is the situation different in Europe? Perhaps it is because the population of vancomycin-resistant E. faecium in Europe was created through use of the growth-promoting antibiotic avoparcin in animals during the 1970s through 1990s. These E. faecium strains, which as animal colonizers were devoid of virulence determinants selected for causing infections in humans, were sufficient to colonize the human population in Europe. The presence of substantial numbers of these less virulent vancomycin-resistant strains may have precluded the rapid emergence of the more pathogenic varieties by competing for one of the selective niches (growth in glycopeptide-rich environment) enjoyed exclusively by the more pathogenic strains in the United States.
We did not, as part of this study, rigorously exclude the possibility that the association of espEfm and hylEfm with vancomycin resistance in the United States was due in some part to clonal expansion of specific strains. It is worth noting that we identified espEfm from 21 of 24 states from which we derived strains, and hylEfm for 15 of those states. We were careful to use PFGE analysis to exclude duplicate clones within geographically localized clusters. If clonal expansion explained the movement of organisms between different geographic areas in our study, then we may not have recognized it. We would argue, however, that the clonal expansion of those specific strains across geographic regions suggests that they may be particularly suited to causing infection, thereby supporting our argument rather than diluting it. This scenario would be consistent with the observation that the variant esp found in E. faecium is a marker for clonal dissemination across geographic regions [3].
At present, we cannot say with certainty whether and to what extent E. faecium actually makes hyaluronidase and under what conditions this protein may be synthesized or exported. Northern hybridization experiments indicate that the hyaluronidase ORF is transcribed under nonselective growth conditions in vitro (data not shown). Therefore, we have compelling reason to believe that the protein is synthesized at least under some environmental conditions. Experiments to characterize the extent of hyaluronidase production by hylEfm-positive strains are ongoing.
Comparison of different S. aureus genome sequences indicates that substantial variation exists between strains, with much of the variability involving antimicrobial resistance determinants and pathogenicity islands [15]. Our data suggest that E. faecium should be added to the growing list of human pathogens with a variety of genomic inventories, some of which have evolved within and are particularly suited to surviving and causing infections in the modern hospital environment.
References
1.
Sahm DF, Marsilio MK, Piazza G. Antimicrobial resistance in key bloodstream bacterial isolates: electronic surveillance with the Surveillance Network Database–
USA. Clin Infect Dis 1999; 29:259–
63.
2.
Lu HZ, Weng XH, Li H, Yin YK, Pang MY, Tang YW. Enterococcus faecium–
related outbreak with molecular evidence of transmission from pigs to humans. J Clin Microbiol 2002; 40:913–
7.
3.
Willems RJ, Homan W, Top J, et al. Variant esp gene as a marker of a distinct genetic lineage of vancomycin-resistant Enterococcus faecium spreading in hospitals. Lancet 2001; 357:853–
5.
4.
Carias LL, Rudin SD, Donskey CJ, Rice LB. Genetic linkage and cotransfer of a novel, vanB-containing transposon (Tn5382) and a low-affinity penicillin-binding protein 5 gene in a clinical vancomycin-resistant Enterococcus faecium isolate. J Bacteriol 1998; 180:4426–
34.
5.
Ferretti JJ, McShan WM, Ajdic D, et al. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc Natl Acad Sci USA 2001; 98:4658–
63.
6.
Hynes WL, Walton SL. Hyaluronidases of gram-positive bacteria. FEMS Microbiol Lett 2000; 183:201–
7.
7.
Shankar V, Baghdayan AS, Huycke MM, Lindahl G, Gilmore MS. Infection-derived Enterococcus faecalis strains are enriched in esp, a gene encoding a novel surface protein. Infect Immun 1999; 67:193–
200.
8.
Singh KV, Coque TM, Weinstock GM, Murray BE. In vivo testing of an Enterococcus faecalis efaA mutant and use of efaA homologs for species identification. FEMS Immunol Med Microbiol 1998; 21:323–
31.
9.
Shimizu T, Ohtani K, Hirakawa H, et al. Complete genome sequence of Clostridium perfringens, an anaerobic flesh-eater. Proc Natl Acad Sci USA 2002; 99:996–
1001.
10.
Polissi A, Pontiggia A, Feger G, et al. Large-scale identification of virulence genes from Streptococcus pneumoniae. Infect Immun 1998; 66:5620–
9.
11.
Berry AM, Paton JC. Additive attenuation of virulence of Streptococcus pneumoniae by mutation of the genes encoding pneumolysin and other putative pneumococcal virulence proteins. Infect Immun 2000; 68:133–
40.
12.
van der Auwera P, Pensart N, Korten V, Murray BE, Leclercq R. Influence of oral glycopeptides on the fecal flora of human volunteers: selection of highly glycopeptide-resistant enterococci. J Infect Dis 1996; 173:1129–
36.
13.
Klare I, Badstubner D, Konstabel C, Bohme G, Claus H, Witte W. Decreased incidence of VanA-type vancomycin-resistant enterococci isolated from poultry meat and from fecal samples of humans in the community after discontinuation of avoparcin usage in animal husbandry. Microb Drug Resist 1999; 5:45–
51.
14.
Chirurgi VA, Oster SE, Goldberg AA, McCabe RE. Nosocomial acquisition of -lactamase–
negative, ampicillin-resistant enterococcus. Arch Intern Med 1992; 152:1457–
61.
15.
Fitzgerald JR, Sturdevant DE, Mackie SM, Gill SR, Musser JM. Evolutionary genomics of Staphylococcus aureus: insights into the origin of methicillin-resistant strains and the toxic shock syndrome epidemic. Proc Natl Acad Sci USA 2001; 98:8821–
6., 百拇医药(Louis B. Rice Lenore Carias Susan Rudin Carl Vael Herman Goossens Carola Konstabel Ingo Klare Sreedh)