Version May 2024
INTRODUCTION — Enterococci, formerly called group D streptococci, come well equipped with a variety of intrinsic (ie, naturally occurring) antibiotic resistances; they are also capable of acquiring new resistance genes and/or mutations. The combination of high-level resistance to ampicillin, vancomycin, and aminoglycosides continues to be common among hospital-acquired Enterococcus faecium in the United States and has a major impact on therapeutic options. (See "Vancomycin-resistant enterococci: Epidemiology, prevention, and control".)
Problems in the treatment of enterococcal infections were noticed as early as the 1950s with the observation that enterococcal endocarditis was not cured nearly as often as streptococcal endocarditis with penicillin [1]. The reason for the poorer response appears to be that penicillin is not as bactericidal against enterococci as it is (or was in the 1950s) against most viridans-group streptococci. This phenomenon, which has been commonly described as tolerance, is characteristic of many enterococcal strains and, even in those that do not initially display tolerance, tolerance can be rapidly elicited by pulsed (intermittent) penicillin exposure. The latter observation led to some support for the use of continuous-infusion penicillin or ampicillin in an attempt to avoid eliciting tolerance [2].
GENETICS OF RESISTANCE IN ENTEROCOCCI — Until the last few decades, enterococci could be treated with penicillin, ampicillin, or vancomycin with or without an aminoglycoside. Some enterococci have now acquired resistance to these and many other agents as a result of mutations (eg, causing high-level resistance to streptomycin or to fluoroquinolones) or the acquisition of new gene(s).
With respect to the acquisition of new genes, enterococci have several different ways of transferring DNA by conjugation (bacterial mating):
●One mechanism, involving pheromone-responsive plasmids, causes plasmid transfer between Enterococcus faecalis isolates at a very high frequency [3].
●Another mechanism involves other plasmids that can transfer among a broad range of species and genera, although usually at a moderately low frequency [4].
●A third mechanism (conjugative transposition) involves transfer of specialized transposons at low frequency but to a very broad range of different kinds of bacteria [5]. Conjugative transposons are relatively nonselective in their host range and are one of the few types of elements known to have crossed the gram-positive/gram-negative barrier in naturally occurring clinical isolates and to then cause resistance in these various hosts [6].
●A fourth mechanism involves the transfer of large fragments of chromosomal DNA directly from one cell to the other via conjugation. This has been described with conjugative transposons and, in E. faecalis, with pheromone-responsive plasmids; with the latter, the transfer of chromosomal DNA appears to be dependent upon recombination occurring between homologous sequences present on these plasmids and the chromosome [7].
Many of the acquired resistances of enterococci involve antibiotics infrequently used to treat enterococcal infections. These include tetracyclines, macrolides, and high levels of clindamycin, rifampin, and fluoroquinolones. It seems likely that these resistances have emerged among enterococci that were colonizing humans or animals to whom antibiotics were given for other reasons.
This observation serves as a reminder that the colonizing flora is at least as likely to be influenced by the use of antibiotics as is the infecting organism. Colonizing bacteria may actually be more capable of developing resistance because they coexist with multiple other bacterial species in nonsterile sites like the gastrointestinal tract and therefore have access to their resistance genes.
BETA-LACTAM RESISTANCE — Unlike many streptococci, an important intrinsic property of enterococci is their resistance to many beta-lactam compounds. The greatest degree of resistance is to aztreonam, which lacks activity against all gram-positive cocci, and the cephalosporins (although new generation cephalosporins, such as ceftobiprole and ceftaroline, exhibit some activity against E. faecalis but not most E. faecium), followed by the antistaphylococcal penicillins such as methicillin and the carboxypenicillins such as ticarcillin.
Even with the more active beta-lactams (eg, penicillin, ampicillin, piperacillin), it takes 10 to 1000 times more drug to inhibit an average Enterococcus than an average Streptococcus. E. faecalis, the more susceptible of the two predominant enterococcal species, is usually inhibited by 1 to 4 mcg/mL of ampicillin and 2 to 8 mcg/mL of penicillin; the comparable minimal inhibitory concentrations (MICs) for E. faecium are typically 8 to 32 mcg/mL. However, E. faecium strains that are much more highly resistant (MICs >32 mcg/mL) to ampicillin have emerged in the past several decades.
Ampicillin resistance — A gene coding for penicillinase was first reported in staphylococci and later found, although very rarely, in enterococci [8]. Most penicillinase-producing enterococci have been E. faecalis, but eight isolates of E. faecium from a polyclonal outbreak in Italy were reported to possess the beta-lactamase gene [9].
Relatively little penicillinase is produced by enterococci in comparison with most Staphylococcus aureus, which can result in a problem of detection for the clinical laboratory. At the standard inoculum used for susceptibility testing, penicillinase-producing enterococci appear no more resistant to penicillin or ampicillin than a nonpenicillinase-producing organism; however, these organisms will show resistance at a high inoculum when more enzyme is present (table 1).
An in vivo effect of penicillinase has been demonstrated in animal models with a better response to penicillins in the presence of beta-lactamase inhibitors. The Clinical Laboratory Standards Institute (CLSI) currently recommends that, because of the rarity of penicillinase-positive enterococci, a test for penicillinase need not be performed routinely but can be used in selected cases, such as endocarditis and other serious infections [10]. If recognized, these organisms pose no real therapeutic dilemma since they are susceptible to beta-lactamase inhibitor combinations, such as ampicillin-sulbactam or piperacillin-tazobactam, and to vancomycin.
A more difficult problem is posed by enterococci with high-level resistance to ampicillin due to a nonpenicillinase mechanism, since this resistance is not overcome by adding a beta-lactamase inhibitor. This phenotype is primarily seen in E. faecium. Isolates of this species recovered prior to the mid- to late 1980s, although more resistant to ampicillin than E. faecalis, were usually inhibited by <8 to 32 mcg/mL of ampicillin. More recently, a trend toward much higher levels of resistance has been observed among nosocomial isolates, with some strains failing to be inhibited by 256 mcg/mL of ampicillin or more [11].
The intrinsic resistance of E. faecium appears to be due to the presence of a cell wall synthesis enzyme (encoded by a gene that is part of the core genome) that is relatively resistant to inhibition by penicillin. This low-affinity penicillin-binding protein (PBP) is called PBP5 [12,13]. There are two versions of PBP5: PBP5-R and PBP5-S. PBP5-R is typically observed in strains of hospital-associated E. faecium with high MICs for ampicillin. PBP5-S is typically found in nonclinical or community-associated strains and confers lower MICs of ampicillin. Higher levels of resistance to beta-lactam antibiotics appear to involve one or more of the following mechanisms: increased expression of PBP5-R, alterations in the PBP5 protein around the active site, and/or utilization of a beta-lactam–insensitive transpeptidase for cell wall synthesis [14-16].
Some strains of E. faecalis have been reported with higher than usual MICs of penicillin and/or imipenem; this has been associated with amino acid changes in PBP4, the low-affinity PBP of this species. These strains may still test susceptible to ampicillin (albeit with higher than usual MICs) and raise caution about using ampicillin susceptibility to predict susceptibility of E. faecalis to penicillin and carbapenems [17-19]. A similar discrepancy has been noted between ampicillin and piperacillin, with some suggesting that ampicillin susceptibility may also not imply piperacillin susceptibility (particularly when piperacillin-tazobactam is used as an antienterococcal agent in intra-abdominal infections) [20]. Similar to E. faecium, isolates of E. faecalis with elevated penicillin MICs have been associated with increased expression of the pbp4 gene (encoding a PBP homologous to PBP5 in E. faecium) and changes in the PBP4 amino acid sequence [21,22].
Enterococci are also intrinsically resistant to cephalosporins with a high MIC for the majority of these compounds; lower MICs are seen with newer cephalosporins, such as ceftobiprole and ceftaroline. Intrinsic resistance to cephalosporins in E. faecalis is complex and has been tied to a number of proteins that impact cell wall synthesis. First, the presence of a putative serine-threonine kinase (designated IreK) is reciprocally regulated by IreP, a protein phosphatase whose gene is encoded immediately upstream of IreK [23]. A similar serine-threonine kinase system (designated Stk) has also been implicated in cephalosporin resistance in E. faecium [24]. Second, the two-component regulatory system CroRS (consisting of the histidine kinase CroS and the response regulator CroR) modulates cephalosporin resistance and has been tied to PBP4 expression in E. faecalis [25]. Finally, the low molecular weight PBP2a (a monofunctional transpeptidase) is also required for resistance to cephalosporins in both E. faecalis and E. faecium [26].
AMINOGLYCOSIDE RESISTANCE — Enterococci are intrinsically resistant to low to moderate levels of aminoglycosides. The minimum inhibitory concentration (MIC) of gentamicin is usually 8 to 64 mcg/mL and that of streptomycin is 64 to 512 mcg/mL, which makes aminoglycosides not active in vivo as monotherapy. However, synergism is generally seen with these two compounds when enterococci are exposed to a combination of the aminoglycoside with a cell wall active agent, such as penicillin or vancomycin. In this setting, there is usually a marked increase in killing (>100 colony-forming units [CFU]/mL, more killing than with penicillin alone), and a bactericidal effect (defined as ≥1000 CFU/mL or a 99.9 percent decrease from the starting inoculum) is achieved by 24 hours.
This synergistic effect presumably explains the clinical observation in the 1950s that a combination of penicillin with streptomycin was much more effective for treatment of enterococcal endocarditis than penicillin alone [1]. This regimen subsequently became the standard of care but has more recently been replaced by the combination of ampicillin plus ceftriaxone [27,28]. (See "Antimicrobial therapy of left-sided native valve endocarditis", section on 'Enterococci'.)
In addition to the usual increased MICs of aminoglycosides for all enterococci, a naturally occurring characteristic of E. faecium is higher MICs of tobramycin (MICs 64 to 1000 mcg/mL) and resistance to synergism with this aminoglycoside. This is due to the presence of an aminoglycoside-modifying enzyme, a tobramycin 6'-acetyltransferase [AAC(6')-Ii], which modifies tobramycin but not gentamicin [29]. This enzyme eliminates synergism between cell wall–active agents and tobramycin, kanamycin, netilmicin, and sisomicin. Similarly, many enterococci possess the aph(3')-IIIa gene, which confers high-level resistance to kanamycin and abolishes the synergistic effect of amikacin. Therefore, with minor exceptions, gentamicin and streptomycin are the only two aminoglycosides that should be considered to achieve synergistic therapy when treating enterococcal infections.
High-level resistance to streptomycin and gentamicin — One of the more serious of the acquired resistances of enterococci is high-level resistance to both streptomycin and gentamicin. The significance of high-level resistance to these aminoglycosides is that it eliminates the synergism expected between gentamicin or streptomycin and a cell wall–active agent such as penicillin or vancomycin. However, this significance has been lessened by reports of success of ampicillin plus ceftriaxone for therapy of E. faecalis endocarditis.
A striking report of this phenomenon in 1959 described an enterococcus from a patient with endocarditis that was not killed by the combination of penicillin and streptomycin [30]. The patient had relapsing disease and was subsequently treated by penicillin plus neomycin (the only other aminoglycoside available at that time). While this combination produced a clinical cure, it also produced deafness and resulted in an article entitled "Deaf or dead" [30]. By 1970, a number of enterococcal isolates had acquired high-level resistance to streptomycin and to kanamycin.
●Streptomycin resistance can be caused either by a ribosomal mutation, which leads to very high levels of resistance (MICs >64,000 mcg/mL), or by the acquisition of a streptomycin-modifying enzyme (an adenylyltransferase), which usually produces MICs between 4000 and 16,000 mcg/mL (table 1) [31].
●Gentamicin was found to be active against highly streptomycin- and kanamycin-resistant organisms and, by the mid-1970s, was being widely used for enterococcal endocarditis. However, since 1979, enterococci with high-level resistance to gentamicin have been increasingly reported [32]. This resistance is most often due to the production of an enzyme with two functional domains: one with 2''-phosphotransferase activity and one with 6'-acetyltransferase activity. These combined enzymatic activities result in high-level resistance and/or resistance to synergism to all commercially available aminoglycosides except streptomycin. The gene encoding this bifunctional enzyme, aph(2'')-Ia/aac(6')-Ie, is identical to a gene previously found in staphylococci (table 1).
●The co-occurrence of high-level resistance to gentamicin and high-level resistance to streptomycin was reported in 1983, which documented, for the first time, strains of enterococci for which there was no known synergistic bactericidal regimen [33]. The addition of any aminoglycoside to ampicillin or vancomycin for such strains results in no more killing than ampicillin alone.
The Clinical Laboratory Standards Institute recommends screening of enterococci for high-level resistance to both gentamicin and streptomycin [10]. This is done by testing for growth at concentrations of 2000 mcg/mL and 500 mcg/mL of streptomycin and gentamicin, respectively, on brain heart infusion agar (BHI); when using BHI broth, the recommended concentration for streptomycin is 1000 mcg/mL. Alternatively, disks of gentamicin 120 mcg (correlates with 500 mcg/mL) and streptomycin 300 mcg (correlates with 1000 mcg/mL) on Mueller-Hinton agar can be used. If the results with the disk diffusion test are not conclusive, high-level resistance should be confirmed by other methods. Detection of high-level resistance to both these compounds precludes the use of available aminoglycosides for synergism in any clinical situation.
Four other gentamicin-resistance genes have been found that remain uncommon (table 1) [34].
●One, designated aph(2")-Ic, was found in a veterinary isolate of Enterococcus gallinarum with a gentamicin MIC of 256 mcg/mL [35] and has been found in an occasional E. faecium and E. faecalis with intermediate-level resistance to gentamicin. The resulting enzyme is active against gentamicin and tobramycin but is not very active against netilmicin or amikacin.
●The second gene, aph(2")-Id, found in a few isolates of Enterococcus casseliflavus and E. faecium, confers high-level resistance to gentamicin, but the encoded enzyme is also not active against amikacin.
●The third gentamicin resistance gene, aph(2")-Ib, found in E. faecium, was associated with MICs >500 mcg/mL of gentamicin [34]. It also conferred resistance to synergism with most aminoglycosides other than amikacin and streptomycin.
●A fourth gene encodes a 16S ribosomal RNA methyltransferase enzyme (designated EfmM) that appears to contribute to intrinsic aminoglycoside resistance in E. faecium [36].
The combination of ceftriaxone (or cefotaxime) with ampicillin is effective in the treatment of endocarditis due to isolates of E. faecalis exhibiting high-level resistance to all aminoglycosides [27,28]. (See "Antimicrobial therapy of left-sided native valve endocarditis", section on 'Enterococci'.)
TRIMETHOPRIM-SULFAMETHOXAZOLE AND CLINDAMYCIN RESISTANCE — Other resistances that appear to be naturally occurring traits of E. faecalis include low-level resistance to clindamycin due to the presence of lsa found in almost all strains [37,38] and in vivo resistance to trimethoprim-sulfamethoxazole, despite in vitro susceptibility to this combination [39,40].
Trimethoprim and sulfamethoxazole are both inhibitors of the synthesis of folic acid, a necessary component for bacterial growth and survival. In vivo resistance of enterococci to trimethoprim-sulfamethoxazole in animal models appears to be due to the fact that these organisms, unlike most other bacteria, can use preformed folic acid, that is, folic acid that is present in humans and animals. Acquired, high-level resistance is also seen in some enterococcal isolates.
VANCOMYCIN RESISTANCE — Both high- and low-level resistance to glycopeptides can occur in enterococci. The definitions of vancomycin susceptibility and resistance and the epidemiology of and infection control measures against vancomycin-resistant enterococci are discussed separately. (See "Vancomycin-resistant enterococci: Epidemiology, prevention, and control".)
Low-level resistance to vancomycin — An intriguing intrinsic feature of two enterococcal species (E. gallinarum and E. casseliflavus) is their capacity to express low-level resistance to vancomycin (minimum inhibitory concentrations [MICs] 8 to 16 mcg/mL). This is due to the presence of a cluster of chromosomal genes involved in the synthesis of serine-ending peptidoglycan precursors (vanC and vanT genes, encoding a D-alanyl:D-serine ligase and a serine racemase, respectively) as well as destruction of the "normal" D-alanine–ending peptidoglycan (the vanXYc gene encoding a bifunctional D,D-peptidase/D,D-carboxypeptidase enzyme) [41-43].
The D-alanyl:D-serine ligase genes in both E. gallinarum and in E. casseliflavus (vanC-1 and vanC-2) are specific for these species and can be used to differentiate them from each other and from other enterococcal species [44,45]. A third putative species, E. flavescens, has a gene that is almost identical to vanC-2, and many consider it the same species as E. casseliflavus [45].
Many isolates of these species test susceptible to vancomycin in vitro and even those that express resistance more fully are usually inhibited by 32 mcg/mL of vancomycin. This low level of resistance is explained by the fact that the D-alanyl-D-serine terminus of peptidoglycan precursors has moderately reduced affinity for vancomycin [46]. Some acquired van gene clusters can also confer low-level resistance to vancomycin by synthesizing peptidoglycan precursors ending in D-serine. These particular phenotypes include VanE, VanG, VanL, and VanN (table 2) [47-50].
High-level resistance to vancomycin — High-level vancomycin resistance is the most problematic resistance of enterococci because it often appears in strains already highly resistant to ampicillin (primarily E. faecium). (See "Treatment of enterococcal infections".)
Despite our awareness of the propensity of enterococci to acquire antibiotic resistance, the emergence of vancomycin resistance in the late 1980s was unanticipated. The surprise derived largely from the fact that vancomycin had been commercially available for over thirty years without high-level resistance ever being seen. However, the drug did not begin to receive substantial clinical use until the late 1970s and early 1980s when methicillin-resistant S. aureus, methicillin-resistant coagulase-negative staphylococci, and Clostridioides difficile became pathogens of major concern in the United States.
When one examines the complex nature of the mechanism of vancomycin resistance and the diverse number of vancomycin-resistance gene clusters, it seems most likely that resistance arose in the distant past, perhaps in response to glycopeptides (which are natural products) present in the environment, and has now been selected for in clinically relevant organisms by the commercial use of glycopeptide antibiotics, including the use of avoparcin in the past in animal feeds in the European Union and elsewhere. Consistent with this hypothesis is the identification of soil organisms (Paenibacillus thiaminolyticus and P. apiarius) that harbor vancomycin-resistance genes that are almost identical to those of enterococci [51].
Mechanism of high-level resistance to vancomycin — Vancomycin inhibits enterococci by binding to the D-alanyl-D-alanine (D-Ala-D-Ala) terminus of cell wall precursors, compromising the subsequent enzymatic steps in the synthesis of the cell wall. High-level resistance to vancomycin is encoded by different clusters of genes referred to as the vancomycin-resistance gene clusters (eg, vanA, B, D, and M gene clusters).
The end result is the replacement of D-Ala-D-Ala–ending peptidoglycan precursors with D-alanyl-D-lactate termini [46]. The replacement of D-alanine by D-lactate, which disrupts one of the five hydrogen bonds required for the interaction of vancomycin with its target, decreases the binding affinity of this antibiotic almost 1000-fold.
The three major phenotypes, VanA, VanB, and VanD, can sometimes be differentiated by the level of vancomycin resistance, susceptibility to the glycopeptide antibiotic teicoplanin, and whether the resistance is induced by exposure to teicoplanin (table 2) [52]. Accurate discrimination of these phenotypes can be achieved by identification of the corresponding genes using polymerase chain reaction (PCR) or hybridization techniques [53].
VanF is a phenotype described in Paenibacillus popilliae, a biopesticidal agent used in agriculture. The vancomycin resistance gene cluster (vanF) has 77 percent identity with the vanA gene cluster, and it has been postulated that these genes were originally transferred from P. popilliae to enterococci [54]. Other members of the Paenibacillus and Rhodococcus family also harbor genes homologous to vanA [55].
VanA is the most common type of vancomycin resistance, usually mediates higher levels of resistance than other types, and causes cross-resistance to teicoplanin. The vanA gene cluster is typically found on a transposon identical or related to Tn1546, which, in turn, is often found within a plasmid [52]. The vanA cluster has disseminated to other bacterial species, including clinical isolates of methicillin-resistant S. aureus (MRSA). Several isolates of these vancomycin-resistant S. aureus have been reported in various countries. (See "Staphylococcus aureus bacteremia with reduced susceptibility to vancomycin", section on 'Vancomycin-resistant S. aureus'.)
These findings emphasize the fact that vancomycin-resistance gene clusters can overcome the species barrier and disseminate to several bacterial species, limiting the therapeutic options even further.
VanA phenotype — The genes encoding the VanA phenotype (vanS, R, H, A, X, Y, Z) have been best studied [52,56].
●The vanS and vanR genes encode a two-component regulatory system that is involved in the induction of expression of resistance. VanS is likely to sense an alteration of the cell envelope caused by vancomycin and then signals to the response regulator, VanR, to turn on the expression of three genes, vanA, vanH, and vanX, which are necessary and sufficient for resistance.
●VanH (encoded by the gene vanH) is a dehydrogenase that generates significant quantities of D-lactate for peptidoglycan synthesis.
●VanA (encoded by the vanA gene) is a ligase that catalyzes the binding of D-alanine to D-lactate (the corresponding ligases in other types are VanB, VanD, VanF, and VanM), resulting in the formation of D-alanyl-D-lactate. This dipeptide is then added by endogenous enterococcal enzymes to a tripeptide cell wall precursor. As a consequence, a pentapeptide precursor ending in D-Ala-D-Lac (pentadepsipeptide) is synthesized.
●The formation of a D-Ala-D-Lac–containing precursor is not sufficient to cause resistance to vancomycin if the cell continues to produce the normal cell wall precursor ending in D-Ala-D-Ala. VanX and VanY complete the resistance phenotype. VanX is a D,D-dipeptidase that cleaves D-Ala-D-Ala [57]. Any D-Ala-D-Ala dipeptide that escapes this dipeptidase activity and is incorporated into peptidoglycan precursors is removed by VanY. VanY is a D,D-carboxypeptidase, which cleaves the terminal D-Ala from the cell wall precursor, leaving a tetrapeptide precursor that can be utilized for cell wall synthesis but to which vancomycin does not bind [46].
●The function encoded by vanZ is not known, although, when cloned alone, it confers teicoplanin resistance onto the host strain, and has been associated with increases in the MIC of other lipoglycopeptide antibiotics including dalbavancin and oritavancin [58-60].
VanB phenotype — The VanB phenotype, the second most common type, is less frequently encountered than VanA. Some enterococci with this phenotype will be missed if concentrations of vancomycin used for screening are too high (>20 mcg/mL) (table 2).
The major phenotypic difference between VanA- and VanB-producing strains is that teicoplanin is not a good inducer of expression of the genes in the vanB cluster; as a result, vanB-containing bacteria usually test teicoplanin susceptible. However, mutations can and do occur spontaneously within the vanB cluster that result in teicoplanin-resistant derivatives.
Most of the genes in the vanB gene cluster have homologs to genes in the vanA cluster and appear to function in a similar manner to produce a cell wall precursor ending in D-Ala-D-Lac. However, the vanB gene cluster has a distinct gene, vanW, it has no vanZ, and its vanY gene is in a different location. Although this phenotype was originally reported as nontransferable, there are now a number of reports of strain to strain transfer of the vanB gene cluster as part of large (90 to 250 kb) chromosomally located mobile elements and as part of plasmids. The vanB gene cluster has been observed in Streptococcus gallolyticus [61] and in anaerobic bacteria (Eggerthella lenta and Clostridium inoculum) [62].
Some isolates of E. faecium that carry the vanA gene cluster but express the VanB phenotype have been described in Southeast Asia [63]. These isolates usually harbor insertions or deletions in the vanY, vanX, vanZ, or vanR genes.
Vancomycin dependence — An interesting variation of vancomycin resistance is that of vancomycin dependence. Vancomycin-dependent enterococci (VDE) evolve from vancomycin-resistant strains, usually from those that contain the vanB gene cluster. These organisms have lost the ability to make their native cell wall, which contains D-Ala-D-Ala termini; as a result, they are dependent upon the continued production of peptidoglycan termini containing D-Ala-D-Lac. Because the vancomycin-resistance genes are normally inducible, the D-Ala-D-Lac–ending peptidoglycan is synthesized only when vancomycin is present (eg, in a patient receiving vancomycin therapy or when vancomycin is added to the in vitro culture medium).
A number of patients have been described with VDE that appear to have negative cultures, unless the samples are cultivated in the presence of vancomycin [64]. In one instance, blood cultures, drawn while the patient was on vancomycin, grew in the initial blood culture bottle but failed to grow on subculture, except around a vancomycin antibiotic disk. Although stopping vancomycin therapy may be of temporary benefit, VDE can revert to non-dependence but are still vancomycin resistant. They can revert by either becoming constitutive producers of the D-Ala-D-Lac–containing cell wall by a mutation that restores the activity of the intrinsic D-Ala:D-Ala ligase or by mutations in the transcriptional terminator region of the regulatory genes, resulting in expression of the resistance genes in the absence of vancomycin [65,66].
LINEZOLID RESISTANCE — Linezolid is active against gram-positive bacteria, including enterococci and staphylococci. Its mechanism of action involves the inhibition of protein synthesis by interactions within the 50S ribosomal subunit [67]. Linezolid interferes with the positioning of the aminoacyl-tRNA in the A site of the bacterial ribosomes [68].
Linezolid resistance has been reported in clinical isolates of both staphylococci and enterococci [69-72]. In enterococci, the G2576U (Escherichia coli–numbering scheme) mutation in domain V of the 23S rRNA is the most common mutation associated with resistance [72-74]. Since bacteria carry several copies of the 23S rRNA gene, the number of rRNA genes mutated appears to be an important determinant of resistance (gene dosage effect); the minimum inhibitory concentration (MIC) increases in proportion to the number of mutated genes [74,75]. The linezolid dose and length of treatment influences the number of mutated genes and the resistance phenotype [74].
Subsequently, plasmid-mediated resistance to linezolid was described in United States human clinical isolates of staphylococci [76]. The mechanism of resistance involves the methylation of an adenine at position 2503 (A2503) of the 23S rRNA. This modification is carried out by a methyltransferase (designated cfr, for chloramphenicol-florfenicol resistance) [77,78]. The cfr gene was previously found in the chromosome of a methicillin-resistant S. aureus (MRSA) strain recovered from the sputum of a patient who died of a nosocomial pneumonia in Colombia [79]; analysis of DNA from the cfr region of the Colombian isolate suggested that the MRSA strain may have acquired this gene through an enterococcal donor [77]. Furthermore, the cfr gene has been described in a clinical isolate of E. faecalis from animal origin in China and on a transferable plasmid from a human isolate of E. faecalis recovered from a patient in Thailand [80,81]. Two variants of cfr (A and B) have been described in enterococci.
A transferable gene encoding a protein with an adenosine triphosphate (ATP)-binding cassette motif, optrA, has been found to confer resistance to linezolid. The gene was found in animal and human isolates of enterococci (E. faecalis and E. faecium) from China. The optrA gene belongs to the family of ATP-binding proteins; these proteins confer resistance by a ribosomal protection mechanism, interacting with the ribosome and displacing antibiotic from its target [38,82,83]. A somewhat related gene reported in enterococci, poxtA, also appears to mediate ribosomal protection against oxazolidinones.
Oxazolidinone resistance among enterococcal isolates has been increasingly documented [84-86]. Linezolid resistance was initially described sporadically, and it was associated with prolonged use of the antibiotic [70]. In a study from Chicago, linezolid-resistant vancomycin-resistant enterococci (VRE) was described in five patients (four renal transplant recipients), usually after more than 21 days of therapy and often in patients who had received several courses of other antimicrobials [69].
Horizontal spread of specific linezolid-resistant outbreak strains has been described [85,86]. In one study, the risk factors associated with the acquisition of linezolid-resistant enterococci (LRE) included administration of piperacillin-tazobactam and/or cefepime, peripheral vascular disease, total parenteral nutrition, and receipt of a solid organ transplant [84]. Although it is evident that the emergence of linezolid resistance is associated with the heavy use of this antibiotic, LRE have been isolated from patients without previous exposure to the antibiotic [87].
DAPTOMYCIN RESISTANCE — Daptomycin is a cyclic semisynthetic lipopeptide antibiotic that is active against a wide variety of gram-positive bacteria. It has potent bactericidal activity mainly due to its ability to penetrate the bacterial cell wall and insert into the cytoplasmic membranes in a calcium-dependent manner. Daptomycin appears to target the membrane (interacting with phosphatidylglycerol [PG], a cell-membrane phospholipid) with preference for the bacterial septum causing alterations in cell division, cell membrane architecture, and function [88,89]. Data suggest that daptomycin interacts with bactoprenol intermediates of lipid II in the presence of PG at the membrane level, altering peptidoglycan synthesis [90]. Further, daptomycin appears to directly affect enzymes involved in cell membrane and cell wall metabolism, displacing them from the inner leaflet of the cell membrane [91].
Daptomycin has been used successfully for the treatment of a variety of VRE infections, including bone, joint, and soft tissue infections [92], bacteremia [93], and endocarditis [94]. Retrospective studies suggest that daptomycin may be associated with better clinical outcomes than linezolid in the treatment of VRE bacteremia [95,96].
Resistance to daptomycin has been reported increasingly in enterococci, including isolates from patients who have never received the drug [97-100]. Several mutations have been associated with development of resistance, suggesting that distinct genetic pathways are responsible for this phenotype. Initial insights into the mechanisms of daptomycin resistance in enterococci have been provided by studies using whole genome sequencing of both E. faecalis and E. faecium isolates recovered from patients before and after daptomycin therapy [101,102]. Enterococci employ two main strategies to become resistant to daptomycin, namely, diversion of the antibiotic from the septal target of the antibiotic (characteristic of E. faecalis) and "repulsion" of the antibiotic molecule from the cell surface impairing the interaction of the drug with the cell membrane (commonly observed with E. faecium) [103,104].
Several genes that have been associated with development of daptomycin resistance may be grouped in two categories: genes that code for regulatory systems that control the cell envelope stress response to antibiotics and antimicrobial peptides and genes encoding enzymes involved in phospholipid metabolism. One protein named LiaX, a component of the LiaFSR regulatory system, plays an important role in sensing and activating the stress-response system in the presence of daptomycin and antimicrobial peptides [105].
Failure of daptomycin monotherapy in the treatment of VRE endocarditis without evident development of resistance (ie, lack of increase in minimum inhibitory concentration) has been well documented [106]. Two strategies have been attempted to prevent emergence of resistance:
●The first strategy involves the use of higher doses of daptomycin than 6 mg/kg intravenously (IV) every 24 hours (which is the dose approved by the US Food and Drug Administration for treatment of S. aureus bacteremia). Since enterococci are less susceptible to daptomycin than S. aureus, some experts recommend the use of higher doses (10 to 12 mg/kg IV every 24 hours) to treat severe deep-seated enterococcal infection [95].
●The second strategy involves combination of daptomycin with beta-lactams, taking advantage of the see-saw effect (increased susceptibility to beta-lactams when cells become resistant to daptomycin), or other antimicrobial agents. Persistent bacteremia due vancomycin-resistant E. faecalis and E. faecium (which failed daptomycin monotherapy at high dose) has been cleared with the combination of daptomycin plus ceftaroline (E. faecalis) and daptomycin plus ampicillin (E. faecium) [94,107]. Furthermore, in vitro data suggest that the beta-lactam may increase daptomycin binding to the cell membrane and further clinical data are needed.
RESISTANCE TO TETRACYCLINE AND NEWER GENERATION TETRACYCLINE DERIVATIVES — Resistance to tetracycline antibiotics in enterococci is common, and is mediated by drug efflux pumps encoded by tet(K) and tet(L) or ribosomal protection factors which recycle the stalled ribosome complex encoded by tet(M), tet(O), and tet(S) [108]. Several other tetracycline derivatives are now available in clinical practice, including tigecycline, omadacycline, and eravacycline, which typically retain activity against tetracycline resistant bacterial isolates, including enterococci. These compounds are typically poor substrates for tetracycline efflux pumps and ribosomal protection factors. However, reports of tet(L) and tet(M) gene copy number expansion and subsequent hyperproduction of the efflux pump and ribosomal protection factor leading to tigecycline resistance have been reported in clinical isolates of E. faecium [109].
Although reports of resistance are still uncommon, it appears that modification of the binding site is the primary mechanism of resistance for these other tetracycline derivatives. This can occur either by mutations in the genes encoding the ribosomal RNA, or those arising in the rpsJ gene, which encodes the S10 protein of the 30S ribosomal subunit [110,111]. Changes in the S10 protein cluster at the end of a long peptide loop whose end is near the drug binding site on the 16S rRNA [112]. These changes are likely to affect the binding pocket and alter the affinity of the drug for the target site.
DETECTION OF ANTIBIOTIC RESISTANCE — Because of the difficulty in detecting some resistances of enterococci using standard laboratory methods, several special tests are recommended for these organisms.
Beta-lactam resistance — Nitrocefin, a specific beta-lactamase test, is recommended for detecting the rare beta-lactamase-producing Enterococcus among isolates from endocarditis and other serious infections [113]. Susceptibility testing at a high inoculum is also effective at detecting penicillin and ampicillin resistance since more enzymes will be present; as noted above, these organisms will generally not appear resistant when tested at the standard inoculum. (See 'Beta-lactam resistance' above.)
Vancomycin resistance — To detect vancomycin resistance, enterococci can be inoculated onto brain-heart infusion (BHI) agar containing vancomycin 6 mcg/mL, followed by incubation for 24 hours at 35ºC. This screening test should detect vanA- and most vanB-containing enterococci but may also detect some organisms with the vanC phenotype. If minimum inhibitory concentrations (MICs) are determined, incubation for a full 24 hours is recommended (rather than 16 to 20 hours, which is commonly used) with careful examination for even faint growth [113]. For isolates intermediate in susceptibility (MIC 8 to 16 mcg/mL), identification to the species level is suggested by the Clinical Laboratory Standards Institute to identify the intrinsically resistant species, E. casseliflavus and E. gallinarum.
Vancomycin dependence — Vancomycin-dependent enterococci (VDE) may be detected by incorporating vancomycin (eg, 6 mcg/mL) into BHI or agar or by adding a disk to an agar surface. Commercial agar for detection of vancomycin-resistant enterococci (VRE) will suffice.
Aminoglycoside resistance — For aminoglycoside testing, BHI agar with 500 mcg/mL of gentamicin or 2000 mcg/mL of streptomycin or BHI broth with 500 mcg/mL of gentamicin or 1000 mcg/mL of streptomycin is recommended to detect resistance to synergism. For streptomycin-containing media, if there is no growth at 24 hours, plates or wells should be re-incubated for an additional 24 hours. Paper disks containing large quantities of aminoglycosides are commercially available in Europe but have not yet been approved in the United States.
Most enterococci with high-level resistance to gentamicin grow just as well on 2000 mcg/mL as on 500 mcg/mL, but 500 mcg/mL is used because it is sufficient to differentiate intrinsically resistant enterococci (MICs <64 mcg/mL) from strains with acquired resistance, and it is less costly since it uses less reagent.
For streptomycin, MICs of intrinsically resistant strains can be as high as 256 to 512 mcg/mL; as a result, higher concentrations must be used to distinguish acquired from intrinsic resistance to this drug. However, 2000 mcg/mL in broth has failed to support adequate growth of some resistant strains due to production of streptomycin adenylyltransferase; this observation has led to the use of 1000 mcg/mL in this medium. BHI medium is recommended because it better supports growth than Mueller-Hinton medium.
Aminoglycosides other than streptomycin or gentamicin need not be tested. Furthermore, the results may be misleading. As an example, MICs of amikacin cannot distinguish intrinsically resistant strains from strains that have acquired the kanamycin-neomycin phosphotransferase, which causes resistance to synergism with amikacin. In addition, difficulties with E. faecium and tobramycin were described above; strains of this species produce tobramycin 6'-acetyltransferase, which modifies tobramycin but not gentamicin [29] and should be considered resistant to synergism with tobramycin. (See 'Aminoglycoside resistance' above.)
One word of caution relates to strains with intermediate gentamicin resistance mediated by the newly described gene aph(2'')-Ic, which confers resistance to synergism with gentamicin [35]. Strains described to date showed growth of 1 to 15 colonies on agar plates with 500 mcg/mL of gentamicin. This is far less than the confluent growth usually observed with strains with high-level resistance to gentamicin and may therefore be overlooked. In addition, this enzyme did not seem to interfere with synergism with netilmicin, unlike the bifunctional enzyme. Similarly, aph(2")-Id is not active against amikacin [34].
Resistance differences related to species — A characteristic of E. faecium that is not detected by routine antimicrobial testing is the intrinsic resistance of members of this species to synergism between cell wall–active agents and tobramycin. There is no established clinical laboratory test to identify this property except for identification of the species or performance of time-kill curves.
An expanded-range MIC determination might be helpful. However, concentrations of tobramycin would have to be higher than those used for other organisms, and it would not be practical for clinical laboratories to have such a test available, particularly since alternative aminoglycosides can be used and their resistance can be detected by routine methods. A gene probe to detect the gene [aac(6')-Ii] responsible for resistance to tobramycin synergism [29] has been used as a reliable method for identifying E. faecium with this property [114].
Another characteristic species difference is that higher levels of resistance to penicillins and carbapenems are typically seen in E. faecium and E. raffinosus (which is very similar biochemically to E. avium) compared with other enterococcal species. This resistance is almost never due to beta-lactamase production, which is uncommon in general in enterococci and is extremely rare in non–E. faecalis.
E. gallinarum and E. casseliflavus — The potential of E. gallinarum and E. casseliflavus to express low-level resistance to vancomycin is another characteristic difference among enterococcal species. Whether members of these species that test susceptible to vancomycin can be adequately treated by this antibiotic is not known; however, since they are typically susceptible to penicillin and ampicillin, the use of one of these agents would be more prudent.
The recovery of E. gallinarum and E. casseliflavus strains with low-level vancomycin resistance is not an indication for isolation of the patient, since the intrinsic resistance is not found on plasmids or transposons that might spread to other bacteria. Nonetheless, the description of a hospital-wide outbreak caused by a strain of E. gallinarum suggests that isolation may be useful when dissemination of a virulent strain is suspected [115].
Although vanA, which has propensity to be on transferable elements, has appeared in both E. gallinarum and E. casseliflavus, transmission of infection has been rare. The presence of vanA in E. gallinarum causes higher levels of resistance than seen with the VanC phenotype, as well as resistance to teicoplanin [116]. As a result, the presence of vanA can be distinguished from the intrinsic low-level VanC-resistant phenotype.
VanB, which can also cause low-level vancomycin resistance, has not been reported in either of these two species.
Daptomycin resistance — Daptomycin requires the presence of physiologic levels of calcium ions (50 mcg/mL) for maximum activity. As a result, the Clinical Laboratory Standards Institute recommends only the broth dilution method with Mueller-Hinton broth adjusted to a final calcium ion concentration of 50 mg/L [113]. Calcium-supplemented daptomycin Etest strips also appear to be an accurate method to detect daptomycin resistance, although there are discrepancies with results obtained by broth microdilution [117].
Nonetheless, data suggest that MIC determination for daptomycin by both broth microdilution and Etest is unreliable, with poor interlaboratory reproducibility [118]. Furthermore, controversies exist in the breakpoint for daptomycin. For E. faecium, susceptibility (broth microdilution) should be interpreted as follows: (1) susceptible dose-dependent: MIC <4 mcg/mL and (2) resistant: MIC >8 mcg/mL (note that there is no “susceptible” category). For E. faecalis the breakpoints are <2, 4, and >8 mcg/ml, for susceptible, intermediate, and resistant, respectively. Despite these changes and due to the inaccuracy of the tests, clinical judgment should be exercised when treating "daptomycin-susceptible" enterococcal bacteremia and other deep-seated infections with this antibiotic.
SUMMARY
●Genetics of resistance in enterococci – Enterococci can develop resistance through mutations as well as acquisition of new genes on plasmids and transposons that can cross species and genera. Many of the acquired resistances likely emerged among enterococci colonizing humans or animals that were given antibiotics for other reasons. (See 'Genetics of resistance in enterococci' above.)
●Beta-lactam resistance – Enterococci have intrinsic resistance to many beta-lactam compounds, particularly cephalosporins. Ampicillin resistance in Enterococcus faecalis, albeit very rare, can be caused by penicillinase, which can be overcome with the addition of a beta-lactamase inhibitor. Enterococcus faecium is intrinsically resistant to ampicillin due to the presence of a low-affinity penicillin-binding protein, can become more highly resistance via amino acid changes and increased expression of pbp5, and thus is not susceptible to beta-lactamase-inhibitor combinations. (See 'Beta-lactam resistance' above.)
●Aminoglycoside resistance – Enterococci have intrinsic low to moderate level resistance to aminoglycosides, but gentamicin and streptomycin retain synergistic activity against enterococci when combined with a cell wall agent. When high-level resistance to one of these aminoglycosides is present (eg, through ribosomal mutations or acquisition of modifying enzymes), the synergistic effect of that agent is eliminated. (See 'Aminoglycoside resistance' above.)
●Trimethoprim-sulfamethoxazole and clindamycin resistance – Although enterococci may be susceptible in vitro to trimethoprim-sulfamethoxazole, they seem to be resistant in vivo due to their capacity to utilize preformed folic acid. (See 'Trimethoprim-sulfamethoxazole and clindamycin resistance' above.)
●Vancomycin resistance – Resistance to vancomycin develops through alteration of its binding site, the D-alanyl-D-alanine terminus of peptidoglycan precursors. Production of D-alanyl-D-lactate–ending peptidoglycans, for which vancomycin has significantly reduced affinity, confers high-level vancomycin resistance, primarily seen in E. faecium. The vancomycin-resistance gene clusters that encode high-level resistance can disseminate to other bacterial species. (See 'Vancomycin resistance' above.)
●Linezolid resistance – Enterococcal resistance to linezolid is acquired through mutations and modifications of ribosomal RNA and proteins. The emergence of resistance is generally associated with the use of linezolid, although resistant enterococci have been isolated from patients without previous exposure to the antibiotic. The emergence of linezolid resistance in enterococci through transferable plasmids carrying cfr and other genes (optrA and poxtA) raises concern of spreading linezolid resistance to other strains and species. (See 'Linezolid resistance' above.)
●Daptomycin resistance – Resistance to daptomycin has been reported both after antibiotic exposure and in the absence of drug administration. The mechanism of resistance involves mutations in genes associated with the regulation of the cell envelope response to antibiotics and cell membrane phospholipid metabolism, among others. Clinical failure of daptomycin in the setting of infection with enterococci exhibiting apparent in vitro susceptibility to the drug has been described; results of minimum inhibitory concentration tests should be interpreted with caution due to poor reproducibility of the test. (See 'Daptomycin resistance' above.)
●Resistance to tetracyclines and other tetracycline derivatives – Resistance to tetracyclines is common in enterococci and occurs due to tetracycline efflux pumps or ribosomal protection factors. Resistance to newer generation tetracycline derivatives such as tigecycline, omadacycline, and eravacycline, is associated with changes in the rpsJ gene encoding the S10 protein of the 30S ribosomal subunit. These changes are likely to alter the drug binding pocket and may act in concert with overexpression of traditional resistance determinants such as drug efflux pumps.
آیا می خواهید مدیلیب را به صفحه اصلی خود اضافه کنید؟
