Abstract
Vancomycin-resistant enterococci (VRE) are both of medical and public health importance associated with serious multidrug-resistant infections and persistent colonization. Enterococci are opportunistic environmental inhabitants with a remarkable adaptive capacity to evolve and transmit antimicrobial-resistant determinants. The VRE gene operons show distinct genetic variability and apparently continued evolution leading to a variety of antimicrobial resistance phenotypes and various environmental and livestock reservoirs for the most common van genes. Such complex diversity renders a number of important therapeutic options including “last resort antibiotics” ineffective and poses a particular challenge for clinical management. Enterococci resistance to glycopeptides and multidrug resistance warrants attention and continuous monitoring.
Introduction
E
Enterococci became recognized as important major nosocomial pathogens due to their natural-intrinsic resistance to several antimicrobials (e.g., penicillin, ampicillin, and most cephalosporins) and capacity to quickly acquire virulence and multidrug resistance determinants.13–18 In fact, enterococci can rapidly develop resistance following the clinical introduction of antimicrobial agents, including last resort antimicrobials used to treat glycopeptide and multidrug resistance (such as quinupristin-dalfopristin, linezolid, daptomycin, and tigecycline).13,18–20
Since the first reports of vancomycin-resistant enterococci (VRE) in the 1980 s,21,22 epidemiological studies have demonstrated serious health and economic impacts from VRE-associated infections and persistent colonization in human medicine.8,23–27 Alternatively, VRE are rare causes of infection in animals, and infrequently encountered in companion animals.2,28–30
The genetic and molecular basis of vancomycin resistance have been described with evidence that VREs may act as reservoirs and sources of other antimicrobial-resistant genes. In this article, glycopeptide and multidrug enterococcal resistance referred to by the historic term vancomycin-resistance enterococcus (VRE) and their associated clinical phenotypic/genotypic (Genotypes and phenotypes of VREs are referred to as Van. Genes, clusters, and operons of VREs are referred to as van) characteristics are reviewed and discussed. Also, the impact of the use and past-use of certain antimicrobial agents on the evolution and emergence of VRE are discussed.
Glycopeptides Resistance in Enterococci
The main mechanism of glycopeptide resistance (e.g., vancomycin) in enterococci involves the alteration of the peptidoglycan synthesis pathway, specifically the substitution of D-Alanine-D-Alanine (D-Ala-D-Ala), to either D-Alanine-D-Lactate (D-Ala-D-Lac) or D- Alanine-D-Serine (D-Ala-D-Ser).31,32 Such alterations can lead to variable expressions of glycopeptide resistance. For example, the respective altered D-Ala-D-Lac and D-Ala-D-Ser leads to less binding affinity of glycopeptide drugs compared to the normal cell wall precursors D-Ala-D-Ala; ∼1000-fold decreased binding affinity for D-Ala-D-lac and ∼7-fold for D-Ala-D-Ser.33–35 The ability to induce such alterations is related to several genes harbored on mobile genetic elements and/or chromosomally encoded regions of different Enterococcus species.18,32 The latter mechanisms appear to underlie most of the vancomycin-resistant phenotypes and the variable composition of Van-related forms of resistance; this distinction may provide insights into the different levels (low-level vs. high-level) of resistance to glycopeptides.34,36,37
The van Resistance Genetic Determinants of Enterococcal spp.
To date, operons related to vancomycin resistance for enterococci are described as vanA, -B, -C, -D, -E, -G, -L, -M, and N (Fig. 1). These are distinguishable by their degree of reduced susceptibility (i.e., resistance) to the glycopeptides, transferability, and inducibility (Table 1).12,18,31,32,34,37–50 An additional operon (vanF) has also been described but thus far only in Paenibacillus popilliae. 51 This vanF variant has a high similarity to the amino acid sequences of the vanA operon, and as such P. popilliae has been suggested as a possible origin for vancomycin resistance in enterococci.

Schematic diagram of van operons in Enterococci. Operons are aligned at the ligase genes and aligned according to the substitute peptidoglycan precursors as following:
Genes encode for proteins essential for expression of glycopeptides resistance (i.e., required genes for expressed VAN-resistance).
VRE, vancomycin resistance enterococci.
Most VRE outbreaks in human populations are attributed to the vanA and vanB gene clusters,52,53 both of which have also been identified in various colonized animals and environmental materials.54–56 These are complex distinct genetic determinants and the most globally widespread van-operons.57–60
The vanA operon is associated typically with transposons (Tn), such as Tn1546 involving two genes for the transposition of the element (orf1 and orf2), and one gene related to teicoplanin resistance (vanZ) (Fig. 1).61–63 The vanA gene cluster includes seven open reading frames transcribed from two separate promoters. The regulatory apparatus is encoded by the vanR (response regulator) and vanS (sensor kinase) two-component system, which are transcribed from a common promoter, while the remaining genes are transcribed from a second promoter.18,31 Gene products that specifically modify the production of peptidoglycan precursors include vanH (dehydrogenase that converts pyruvate to lactate) and vanA (ligase that forms D-Ala-D-Llac dipeptide).
An essential part of the vancomycin resistance phenomenon is that the production of the normal D-Ala-D-Ala end of the pentapeptide does not continue. This is resolved by the vanX and vanY genes where vanX (dipeptidase that cleaves D-Ala-D-Ala) hydrolyzes and thereby interrupts the production of the pentapeptides, and vanY (D, D-carboxypeptidase) cleaves the pentapeptides that might still be produced. There can be variations in the composition of this vancomycin resistance operon due to insertion of IS elements and these variants are described as vanA-like elements.
The typical vanB operon has a similar genetic backbone to the vanA. High-level vancomycin resistance has been reported within vanB subtypes and designated as vanB1-3 with vanB-2 as the most prevalent genotypes worldwide. 40 The transfer of vanB resistance alleles occurs through the acquisition and/or exchange of transposons such as Tn1547, Tn1549, and Tn5382.38,64,65 The conjugative vanB transposon, known as Tn1549 is widely prevalent among VanB-type enterococci and other gram-positive bacteria. 66 This is mainly a chromosomal transposon and less frequently found on plasmids.67,68
The genetic organization of vanB is similar to that of vanA, in that it contains two distinct promoters transcribing seven open reading frames, but there are some important differences (Fig. 1). For example, vanB encodes a two-component signaling system (named vanRB and vanSB) that is considerably different from that encoded in vanA. Furthermore, vanB encodes homologs of vanH and the D-Ala-D-Ala ligase (encoded by vanB), and the peptidases (vanX and vanY). However, vanB lacks a homolog of vanZ, and instead encodes a protein named vanW, whose role in resistance is not fully understood. 18
In contrast, the vanC operon is genetically different from vanA and vanB, and typically “less virulent” enterococci than those carrying inducible vanA and vanB gene clusters.34,41,42 The vanC-resistant subtypes, vanC-1, vanC-2, and vanC-3, are known to be intrinsically present in Enterococcus gallinarum, Enterococcus casseliflavus, and Enterococcus flavescens, respectively, 12 and sometimes seen as markers for these enterococcal species. The vanC-2 gene cluster in E. casseliflavus demonstrates a composition similar to the vanC-1 cluster in E. gallinarum.32,42 More recently, a vanC-4 subtype was also described with 93–95% nucleotide homology with vanC-2/-3. 41
The vanD operons are exclusively chromosomal and similar to vanA and vanB.43–45,69–72 This operon represents a number of different combinations of mutations and furthermore, different enterococcal strains demonstrate a constitutive resistance phenotype resulting from different mutations within the operon regulators. These various modifications have led to a wide range of subtypes and resistance phenotypes. 73 Nevertheless, the vanD is described sparsely among various enterococci species, and can also be carried by the vanC E. gallinarum. 74
The vanE gene cluster has been described in a few E. faecalis strains from North America and Australia.46,75,76 The vanE cluster resembles the vanC1, which occurs naturally in E. gallinarum (Fig. 1). Studies have suggested that a cluster within an integrase gene of the vanE- E. faecalis may have evolved with the acquisition of this operon; however, reports have not determined the transfer-ability of this gene cluster. 75 Enterococcus faecalis possessing a vanG cluster have been described and two subtypes were identified.47,77,78 In contrast to the other van D-Ala-D-Ser operons (i.e., vanE, vanC vanL and vanN), this operon has been shown to be transferable via a conjugative plasmid from E. faecalis.
The set of van genes was expanded recently after the discovery of vanL, vanM, and vanN operons. The vanM is genetically and phenotypically similar to vanA, vanB, and vanD, whereas both vanL and vanN are similar to vanC (Fig. 1). The vanL gene cluster exhibits 49–51% sequence identity to the vanE and vanC ligases. However, this VRE E. faecalis isolate did not demonstrate either a transfer or conjugation ability, suggesting that the vanL gene cluster is chromosomally encoded. 48 The vanM operon has been described in E. faecium VRE isolates and demonstrated a close gene arrangement to vanD and in vitro transferable resistance by conjugation. 49 The vanN operon is the most recently identified gene cluster described in E. faecium. 50 This is a similar operon to vanG, but more unique since it has also been reported to be transferable via mobile genetic elements, and only in E. faecium. In general, the D-Ala-D-Ser operons are frequently chromosomally encoded but some of the latest added operons (i.e., G and N) are transferable in E. faecalis and E. faecium respectively, and only exceptionally.
Phenotype Versus Genotype Expression of Vancomycin Resistance in Clinical Enterococcal Species
VRE phenotypes are defined by the expression of vancomycin-related resistance and virulence factors, regardless of the genes expressed but mainly related to their clinical characteristics. 79 The VanA and VanB phenotypes, and most recently the VanC, appear to be clinically the most important, and capable of conferring high-levels of resistance to vancomycin and teicoplanin; however, the VanA phenotype appears remarkably more virulent.
Motile enterococci, such as E. gallinarum, E. casseliflavus deserve special mention in phenotype designations, since the VanC phenotypes were previously less frequent causes of human clinical infections and confer a low-level of resistance to vancomycin, but not to teicoplanin.12,14 While the global incidence of VanC phenotypes remains relatively low, including a lower risk of mortality compared to VanA and VanB, 80 there has been a rise in VanC-type outbreaks reported in humans.81–85 For example, several human bacteremia cases have been identified as hospital-acquired VRE-C-type E. gallinarum and E. casseliflavus infections, of which a few cases failed to respond to therapeutic interventions with fatal consequences. 84 Moreover, a study involving 13 Greek human hospitals demonstrated that the VanC phenotype (i.e., C1, C2, and C3) was the most prevalent, representing 57% of VRE isolates followed respectively by VanA and VanB phenotypes. 86 A Swiss study identified the VanC genotype in 98% of 296 clinical VRE strains isolated from patients of which 290/296 VRE isolates were of fecal origin. 87 This is in contrast to the USA, where the VanC genotype is very rare compared to the dominant VanA and VanB genotypes. In addition, recent reports from other regions such as North Africa have revealed a low prevalence of clinical VanC enterococci in comparison to VanA type. 88
The VanC phenotype is known to be constitutively expressed; however, E. casseliflavus and E. gallinarum have been reported to harbor inducible resistance types as well. 89 Furthermore, E. gallinarum/casseliflavus have been found to carry other van gene clusters (i.e., vanA, vanB and vanD) and express high-level vancomycin resistance.90–94 VanC E. gallinarum have been detected in clinical isolates from humans and farm animals, carrying vanC-1 and vanA genes, and showing a high-level of resistance to glycopeptides and linezolid.81,82,92,95–98 E. gallinarum can capture the genetic determinants of high-level glycopeptide resistance and transfer these determinants to other enterococci (i.e., E. faecium).90,93,99 The emergence of high-level resistance to glycopeptides among less common motile enterococci, mainly E. gallinarum, suggests a need for increased awareness for detection and proper characterization of these microorganisms.100,101
Geno-Geographic Characteristics of the VRE and Enterococcal Clones
Both E. faecium and E. faecalis are distinct among enterococci and recognized as a major reservoir of acquired glycopeptide and multidrug-resistance (MDR) by carrying the most common reported van genes.1,8 A systematic review and meta-analysis involving Iran and few European countries revealed a close rate of prevalence between countries of VRE infections in humans and ranges between 8–13%, largely from E. faecalis carriers. 102 Whereas in the USA, E. faecium is the carrier of 95% of recovered VRE isolates 103 and the main clinical reservoir of the VanA and VanB genotypes in Europe, Northern and Latin America, and Southeast Asia.14,52,53,63,104 Alternatively, E. faecalis has been frequently associated with vanG, L and E genes (Table 1). The other van genes remain sparsely reported and neither responsible for most VRE clinical infections nor outbreaks in human medicine.
The VanA genotype is the most common among the reported E. faecium and E. faecalis isolates worldwide from humans and animals but less frequently reported in other enterococci species such as E. gallinarum, Enterococcus hira, and Enterococcus durans.18,105 For instance, a recent Iranian study describing a collection of 160 clinical enterococci reported the rate of VRE strains at 19%, largely E. faecium species dominantly carrying vanA genes. 104
On the other hand, the VanB genotype is reported more frequently in fewer countries such as Australia, Sweden, and Germany.106–108 Such variability in genotypes has been attributed to the different antimicrobial use patterns between different countries. For example, in Australia the VanB-VRE is the dominant E. faecium from human cases, but has neither been found in animals nor frequently in normal human feces. 109 However, indistinguishable vanB elements have been found in anaerobic commensal bacteria from human faeces, and might be a source of VanB resistance to enterococci. 110 In addition, vanB genetic determinants have been detected in VanC enterococcal species, 111 and vanC genes in E. faecalis and E. faecium from human infections.36,112–114 This supports the argument that glycopeptides resistance cannot be predicted only by the VRE genotype, particularly the motile species. 100
Elucidating further about VRE and enterococcal clinical outbreaks requires different molecular techniques to assess patterns and clonality of genomic DNA as well as subspecies phylogenetic relationships of enterococci. The application of such methods in recent years (i.e., amplified fragment length polymorphism, restriction endonuclease analysis, CRISPR analysis, multi-locus sequence typing (MLST) and whole genome datasets have separated E. faecium into two major clades designated as clades A and B.11,115–121 Clade A are mostly of clinical origin and associated with human-hospital infections and further divide into two subtypes: clade A1, which include epidemic hospital strains, and clade A2, which include animal and sporadic human infections. Whereas, clade B strains are of nonclinical origin and associated with community.117,120,122
The genome of clade A1 are characterized by harboring a greater abundance of virulence and antibiotic resistance genes compared to non-A1 lineages. 4 Clade A1 or clonal complex E. faecium 17 (CC17, a MLST designation) is a global polyclonal cluster of hospital adapted clones with the potential to cause invasive disease and adapted to the hospital environment and GIT colonization. This clone contains the well characterized global sequence type (ST) 17 and the descendant variants; ST16, ST78, ST63, ST64, and ST174. 122 The global clonal complex E. faecium CC17 is frequently characterized as multidrug resistant, including vancomycin, and typically carrying specific virulent gene markers, especially the hyaluronidase gene (hyl) and the enterococcal surface protein encoding gene (esp), as well as the IS16 insertion-element. These elements characterize CC17 clonal strains and associated with hospital acquired and MDR enterococcal infections. 88 Alternatively, E. faecalis are less restricted and some clones such as CC2, CC40, and CC87 are shared between hospitals and the community.
The global increase in MDR Enterococcus faecium strains has been linked to the emergence of lineage 78. Isolates belonging to sequence type 117 (ST117) within the lineage 78 are increasingly identified from clinical isolates in many European health institutions. 123 Other countries such as Australia have experienced the emergence of new and rapid spread of E. faecium clones within the clade A1 lineage replacing the common ST17 type by the ST203 type as the main cause of blood stream infections and a rising proportion of vanB VRE. 120 Genomic analysis of these particular new emergent clones revealed 40 unique genes including Tn1549 indicating a rapid change in the epidemiology and genomics of VRE in this country.
Genomic and epidemiological investigations of VRE are sparse from the under-developed regions, such as North African. Recent studies from such regions have reported novel information. For instance, a study from Tunisia has documented the prevalence of VanA - E. faecium from clinical and hospital sources at 5%, carrying purK and esp gene, and the IS16 element, but without the hyl gene. 88 Furthermore, e-BURST analysis of these strains has identified a new emergent clone ST910 as the dominant type and not belonging to the global CC17, suggesting a rapid evolution in this country. 88
A previous study from this North African country reported frequent sequence types ST18 and ST80 belonging to CC17 among clinical vanA E. faecium, all carrying the IS16 element, but lacking hyl and esp genes. 124 Similar variability was also reported from a Brazilian study describing VRE-E. faecium strains from colonized hospital patients showing ST412 as the most shared, belonging to CC17, harboring the IS16, Tn1546, esp, and acm genes, but lacking the hyl gene. 125 These results contrast previous reports from Brazil and suggesting interspecies horizontal transfer of genetic and resistant elements between the enterococci. 125
Global VRE and VSE isolates, both from human (i.e., hospital and community-acquired) and nonhuman (i.e., environment and animals) sources have revealed that hospital enterococcal isolates are related closely to/or descending from the MLST single clonal-complex (CC), CC17.23,126 Further analysis of E. faecium demonstrated that the major hospital clones ST78 and its single locus variants ST192 and ST203 group belong in a different BAPS group than ST17 and ST18.115,127,128 This indicates that hospital-associated E. faecium have not recently evolved from a single ancestor and consequently the designation of CC17 as a hospital-selected clonal complex may not be entirely accurate.
Also, the highly prevalent E. faecium hospital clones are almost absent in the community, 129 but this is not the case for E. faecalis. Based on MLST, there are seven most prevalent E. faecalis clones among clinical and outbreak-associated isolates (n = 355), including ST6, ST9, ST16, ST21, ST28, ST40, and ST87, accounting for only 37% of the hospital-derived isolates (http://efaecalis.mlst.net/). 130 These are found frequently in the community, farm animals and food products while the major hospital-derived E. faecium clones are found rarely in the community (8% of 513 nonhospital isolates).30,131,132
The complete genomic sequence of E. faecium was completed in 2012 and in-depth studies analyzing the genomics of the significant enterococci are still relatively small. 133 Most recent molecular epidemiological studies and genomic analysis using WGS have revealed discriminating information about the genomic characteristic and elements of enterococci involved in virulence and antimicrobial resistance. 121 However, knowledge is still incomplete, where future prospective analysis should rely on combining the current and future genomic insights with the subsequent functional and phenotypic change of the pathogen population and dynamics.
Insights into Animal-Associated VRE and Public Health Issues
Previously, the emergence of VREs in food-animal production systems has been largely attributed to the widespread use of sub-therapeutic avoparcin (a glycopeptide for animal use) for growth-promotion throughout Europe and other countries in the mid-1990 s.134–141 In fact, the isolation of different VRE genotypes, particularly VanA-VRE, from the community, animal faeces, sewage and raw meats, suggested that VREs colonized and spread as a result of the use of animal growth promoters (esp. avoparcin) and not human hospital-use alone.57,142–144
On the contrary, other countries like Canada and United States have never approved avoparcin and not reported VRE in animals until 2008.18,37 In North America, VREs are one of the major hospital acquired infections attributed largely to the clinical use of vancomycin in humans.145,146 In addition, the unique epidemiological distribution of VREs in Australia has been attributed to the combination of extensive use of vancomycin in humans and avoparcin as a growth promotor in the animal industry.4,147
Avoparcin, as a growth promoter, was banned in the EU in 1997 as detailed in the Commission Directive 97/6/EC. Despite the subsequent reduction of VRE prevalence from animal-derived products and fecal samples of humans, 148 VRE-related human infections and outbreaks have increased in the EU since 1999149–154 as well as the prevalence of VRE in asymptomatic human carriers in Europe.155,156 While the ban of antimicrobial growth promoters likely played an important role in reducing VRE carriers in animals, and in some cases a remarkable decrease in the incidence of VREs and glycopeptide-resistant Enterococcus species, the consequences of such previous livestock practices may not be reversible.143,148,157
In Norway, VREs were still easily found (96–98% proportion of positive samples) from poultry farms exposed previously to avoparcin after more than a year and a half after the ban, whereas none were isolated from the unexposed poultry and swine farms. 158 Moreover, 3 years after the ban of avoparcin, VRE were isolated from previously exposed and unexposed poultry farms (99% and 11% of the respective grouped-studies farms) as well as from farmers (18% and 1% from farmer samples from exposed and unexposed farms).159,160 Recent data from Europe revealed that VRE-VanA genotypes were detected in Danish poultry production 15 years after the ban of avoparcin and found in 47% of fecal samples with variable resistant phenotypes and diverse E. faecium clones. 161 In Italy and Germany, the frequency has been reduced, however, VREs can still be recovered from different farm animals, poultry meat and pork products, and fecal samples of healthy persons, 148 including clonally related vanA VREs. 105 Similar results have also been reported in Taiwan, where a ban of avoparcin in year 2000 resulted in a notable decreased VRE prevalence in chicken but persisted in the same populations. 162
Prior to the European avoparcin ban, high rates of VRE carriage were also reported in dogs. 163 A subsequent Dutch study, performed 5 years after the avoparcin ban, reported no VRE in a 100 dogs samples. 164 Other studies have found that healthy dogs and cats can be colonized by VRE.165–167 In addition, VRE isolates from dogs demonstrate similar genetic lineages to hospital-acquired infections in humans.168–170 Furthermore, VRE carrying vanA have been described in healthy horses in Italy, Poland, and Hungary. 171
When VRE contaminated meat was found in retail outlets as well as the fecal flora of humans, in countries with a heavy use of avoparcin,172,173 it was quickly concluded that the case against growth-promoting antibiotics, especially avoparcin, was proven.157,174,175 However, experiences gained after the ban have revealed some doubts as to the route of possible transmission of VRE determinants between animals and humans. For example, despite higher VRE carrier frequencies in animals in Europe, nosocomial VRE infections in humans were lower in Europe than in the United States 176 even though avoparcin was never used in animals in North America.
Initial reports from 10 to 20 years ago indicated that it has been difficult to prove a food-borne infection route to humans directly and the few examples published of identical VREs between humans and animals resulted from direct animal contact, rather than meat consumption. 173 Experimental infection of humans with animal VRE strains resulted in transient colonization only.177,178 Nevertheless, evidence obtained by using different molecular typing techniques, indicates strong correlations between VREs or genetic determinants in animals and humans (Table 2).23,65,130,131,166,177–199 These correlations between animal and human VRE carriage does not necessarily demonstrate causality, but that several reservoirs exist, including wild animals and plants. 200
VRE, vancomycin-resistant enterococci; PFGE, pulsed field gel electrophoresis; AFLP, amplified-fragment length polymorphism; MLST, multilocus sequence typing; PCR, polymerase chain reaction; CC, clonal complex; ST, sequence type.
Alternative theories have been proposed to explain VRE persistence among food producing animals, despite the avoparcin ban. 161 In Denmark, the use of the macrolide, tylosin, in pigs was suggested to co-select for VREs since the genes encoding the two resistances are located on the same plasmid.201,202 For example, in vitro studies have shown that ermB and vanA genes were located on the same transferable DNA elements, most likely large plasmids. A similar co-selection for VRE persistence has been proposed via tetracycline resistance coded by the tet(M), but transferability has not been easily demonstrated.135,201 Also, a conjugative plasmid conferring acquired copper resistance, via the tcrB gene, has been shown to be strongly correlated with macrolide [erm(B)] and glycopeptide resistance (vanA) genes in Enterococcus faecium isolates from pigs, whereby all genes can share the same conjugative plasmid.143,204
Another theory is that the plasmid addiction systems located on the same plasmid as the vanA gene would force the bacteria to retain the resistance. 204 Essentially, the plasmid addiction system is the strict natural selection of plasmid-encoded gene(s) that is required for the viability of the bacterium. 205
Nosocomial human-related VREs tend to be more clonal populations persisting in hospitals, and related to human consumption of antimicrobials or hospital environments.101,206 What appears to be important differences between the VRE subpopulations are other factors that determine spread and virulence. For instance, screening admitted patients for colonized VRE has significantly reduced nosocomial VRE outbreaks and distinguished the community-acquired VRE cases. 206
In addition, epidemic hospital-acquired strains of vancomycin-resistant E. faecium from the USA, Europe, and Australia often have certain virulence genes, esp 207 and hyl genes, 208 which are not found in nonepidemic or animal isolates, food or sewage. 209 These genes, are part of a putative pathogenicity island and considered to be a marker for nosocomial-epidemicity that contributes to the acquisition of antimicrobial resistant genes and spread of vancomycin-resistant E. faecium isolates in hospitals. 210 Thus, the administration of glycopeptides for humans and previous use as feed additive growth promoters in animals are important factors in the emergence of vancomycin resistance in enterococci,103,147 but other virulence factors appear necessary before these subpopulations are represented in human clinical infections. 211
Since year 2000, VRE rates in human clinical isolates have increased in numerous European countries despite glycopeptide resistance declining in nonhospital reservoirs. 212 The acquisition of vancomycin resistance by a distinct subpopulation of hospital-acquired ampicillin-resistant E. faecium strains has been the main factor responsible for the spread and dissemination. Clones belonging to the CC17 lineage are frequently characterized by ampicillin and fluoroquinolone resistance, and harboring the putative virulence genes esp and hyl.23,213–215 As late as 2016, there was a notable increase hospitalized human patients with VRE infections in Danish hospitals (DANMAP, 2016), mostly due to vanA E. faecium. In Europe, new reports of VRE are still being reported (e.g., West Balkan regions), as Enterococcus faecalis and E. faecium, both as VanA and VanB phenotypes. 216
It is possible that a proportion of European human hospitalized VRE cases stem from environmental or food sources either through bacterial contamination or indirectly, via gene transfer.217,218 This may in-part explain the different VRE subpopulations with variable levels of virulence associated with human clinical infections. For example, molecular typing of clinical VRE isolates from European hospitalized patients tend to be heterogeneous with respect to their genotypic fingerprints,176,207 whereas in the USA homogeneity is more common. 219 Other human VRE subpopulations have emerged, either as community-acquired VRE that are frequently susceptible to amoxicillin or gentamicin, or nosocomial VRE, which in contrast are also resistant to amoxicillin/gentamicin. 220
Emergence of MDR Among Enterococcus spp.
The emergence of multidrug resistant enterococci to the newer drugs including last resort antibiotics is worrisome and of global concern (www.who.int/mediacentre/news/releases/2017/bacteria-antibiotics-needed/en). These have variable susceptibility to even less antimicrobial agents and increasingly associated with significant morbidity and mortalities in human medicine. The concern is further complicated by the genetic intra-ability of enterococci to exchange resistance determinants and/or transfer to other Gram-positive organisms such as staphylococci and streptococci. 221 Generally, oxazolidinone-linezolid, daptomycin, quinupristin/dalfopristin (QD), and tigecycline are invaluable antibiotics frequently used to treat serious infections and therapeutic complications caused by multidrug resistant enterococcal infections, VRE and MRSA.222,223
Linezolid, daptomycin, and tigecycline have been increasingly utilized over the past decade as last-line therapeutics, to combat MDR enterococci and staphylococci, however, clinical isolates with reduced susceptibility have emerged.221,224–226 Oxazolidinone-linezolid was approved in 2000 and widely used, but the emergence of resistance is reportedly associated with its therapeutic and extended applications for VRE as well as vancomycin-susceptible enterococci cases. Linezolid-resistant enterococci (LRE) is of greater concern and increasingly encountered over the past decade although at a low prevalence of less than 1% among the enterococci and staphylococci.223,225 However, new generation oxazolidinones (i.e., Tedizolid phosphate) has been recently approved, which demonstrated better efficacy against clinical MDR Gram-positive pathogens such as MRSA, VRE, and LRE. 225
Resistance (or reduced susceptibility) to the oxazolidinone-linezolid is frequently reported to be mediated either by acquisition of the chloramphenicol–florfenicol resistance (cfr) gene or point mutations within the ribosomal complex system (23S rDNA or ribosomal protein genes). 225 Linezolid resistance is also linked to optrA gene, which recently was described in humans, food producing animals and products, from various global regions (i.e., China and Denmark). 218 Acquisition of the cfr gene is often carried on transferrable mobile genetic determinants, associated with resistant infections in livestock and reduced susceptibility to a broad range of drugs (i.e., Phenicols, Lincosamides, Oxazolidinones, Pleuromutilins, and Streptogramin A). Furthermore, a study from Germany has recently documented a cfr like-gene named cfr (B) variant gene locus in clinical strains of E. faecium with 99.9% identical sequence to a cfr-like gene from Clostridium difficile. 223 This new variant gene was found on both mobile and/or extra-chromosomal DNA. Recently, E. faecium was reported from food producing animals harboring a novel plasmid coding for multidrug resistance, including Linezolid. 227
In addition, tigecycline-resistant E. faecium is also emerging, but the precise mechanism is unclear. A recent genomic analysis of clinical enterococci expressing tigecycline-resistant revealed tetracycline-resistant determinants; tet(L)- encoded efflux pumps and tet(M)-encoded ribosomal protein in the expression of such resistance. 226 Another recent German investigation analyzing the spread of Tn1549-vanB-genotype resistance revealed the dissemination was attributed to the exchange of large chromosomal fragments between positive and negative vanB enterococci isolates thus providing a better understanding of the spread of multidrug-resistant pathogens. 108
In summary, VRE and enterococcal MDR infections are increasingly reported from humans with distinct geographic differences. The vanA-VRE are more commonly associated with poultry than other food animals (e.g., cattle, swine), with some geographic differences. On the other hand, VanC and VanN-VRE phenotypes have been detected frequently from farm and companion animals as well as meat products destined for human consumption (i.e., chicken, pigs, cow, and horses).228–230 Other reports have also shown the unusual presence of different van genes in E. casseliflavus isolated from farm animals. 231 Thus, animal microflora may play an important role in the development and emergence of novel and evolved geno/phenotypes of glycopeptide and multidrug resistances in different enterococcal strains as well as other bacterial strains.232,233 Nevertheless, the zoonotic risk of different resistant organisms including VREs from farms and livestock has not been fully defined. 234
Enterococci of foodborne origin are not known as a direct source of resistant enterococci in humans, but could pose a risk to transfer resistance determinants (e.g., van genes) to human-adapted enterococcal strains. For example, human colonization by VRE of animal origin has been documented including trans-conjugant transfer of van genes between enterococcal species within the human intestinal microflora.199,235,236 Also, the documentation of in vitro interspecies conjugal transfer of vancomycin resistance and MDR enterococcal phenotypes of environmental origin indicates the potential risk of these sources as well as the diverse ability to transfer such resistance determinants to human pathogens. 237 Antimicrobial-resistant E. faecalis and E. faecium of animal origin have demonstrated the ability to acquire resistance genes and act as donors of such genes.238,239 Such complex genetic exchanges and acquisitions might explain the continuing emergence and evolution of enterococcal antimicrobial resistance.
Conclusions and Prospective Challenges
Enterococcus species colonize and survive in widely different hosts and conditions. As part of the healthy intestinal flora and environment, enterococci are part of the cycle of continuous exposure to prophylactic/metaphylactic and therapeutic antimicrobial agents in human and animal hosts, which has likely contributed to their ability to acquire and develop unique profiles of virulence and antimicrobial resistance.
In contrast to vancomycin-resistant E. faecalis and E. faecium, other species such as E. casseliflavus or E. gallinarum have attracted less attention, despite associations with environmental and livestock sources as well as human infections and outbreaks. This suggests a need for continuous monitoring of a broader range of enterococcal species as well as other vancomycin-resistant genes (e.g., VanC genotypes).
The role of animals and food products in the human VRE population dynamics needs to be further defined. While many aspects of VRE are well known to human and public health, both veterinary and agricultural awareness of these ongoing issues is sparse. It may have been a premature conclusion that the animal involvement of VRE issues was resolved after the ban of avoparcin. The collection of evidence suggests that animal-associated vancomycin-resistant E. faecalis and E. faecium as direct sources of human infections is unlikely. The role of animal-associated vancomycin-resistant E. casseliflavus, E. gallinarum and others needs to be further defined.
Food animals can be reservoirs of van genes of clinical importance to human-related infections (e.g., vanA, vanC), including novel van genes. Further detailed sequencing of animal-associated van genes compared to clinical human VRE isolates could help to better define the significance of animal reservoirs. Transient colonisation of animal-associated VRE can lead to the transfer of van genes to human adapted enterococci strains with relevant virulence factors. The amplication of animal VRE reservoirs can be related to antimicrobial use patterns, as demonstrated by the previous massive consumption of avoparcin, with major spillover effects on human populations.
Persistence of van genes in animal populations has proven to be more challenging to resolve than previously thought. The potential co-selection of VREs in animals based on other antimicrobial consumptions (e.g., macrolides, tetracyclines) is worthy of further investigations. In contrast to humans, VRE infections in animals are rare, and infrequently seen in companion animals. Given the importance of animal reservoirs then VRE screening should be part of national antimicrobial resistance surveillance systems for indicator bacteria. Since the ban on avoparcin, veterinary medicine has demonstrated that the needs of animal health can be served without glycopeptide antimicrobials. As such, the class of glycopeptide antimicrobials should be reserved for human medicine.
Footnotes
Disclosure Statement
None of the authors of this article has a financial or personal relationship with other people or organizations that could inappropriately influence or bias the content of the article.
