Skip to main content
NCBI home page
Antimicrobial Agents and Chemotherapy logoLink to Antimicrobial Agents and Chemotherapy
. 1999 Jan;43(1):141–147. doi: 10.1128/aac.43.1.141

Characterization of Dihydrofolate Reductase Genes from Trimethoprim-Susceptible and Trimethoprim-Resistant Strains of Enterococcus faecalis

Teresa M Coque 1,2,†, Kavindra V Singh 1,2, George M Weinstock 1,3, Barbara E Murray 1,2,3,*
PMCID: PMC89034  PMID: 9869579

Abstract

Enterococci are usually susceptible in vitro to trimethoprim; however, high-level resistance (HLR) (MICs, >1,024 μg/ml) has been reported. We studied Enterococcus faecalis DEL, for which the trimethoprim MIC was >1,024 μg/ml. No transfer of resistance was achieved by broth or filter matings. Two different genes that conferred trimethoprim resistance when they were cloned in Escherichia coli (MICs, 128 and >1,024 μg/ml) were studied. One gene that coded for a polypeptide of 165 amino acids (MIC, 128 μg/ml for E. coli) was identical to dfr homologs that we cloned from a trimethoprim-susceptible E. faecalis strain, and it is presumed to be the intrinsic E. faecalis dfr gene (which causes resistance in E. coli when cloned in multiple copies); this gene was designated dfrE. The nucleotide sequence 5′ to this dfr gene showed similarity to thymidylate synthetase genes, suggesting that the dfr and thy genes from E. faecalis are located in tandem. The E. faecalis gene that conferred HLR to trimethoprim in E. coli, designated dfrF, codes for a predicted polypeptide of 165 amino acids with 38 to 64% similarity with other dihydrofolate reductases from gram-positive and gram-negative organisms. The nucleotide sequence 5′ to dfrF did not show similarity to the thy sequences. A DNA probe for dfrF hybridized under high-stringency conditions only to colony lysates of enterococci for which the trimethoprim MIC was >1,024 μg/ml; there was no hybridization to plasmid DNA from the strain of origin. To confirm that this gene causes trimethoprim resistance in enterococci, we cloned it into the integrative vector pAT113 and electroporated it into RH110 (E. faecalis OG1RF::Tn916ΔEm) (trimethoprim MIC, 0.5 μg/ml), which resulted in RH110 derivatives for which the trimethoprim MIC was >1,024 μg/ml. These results indicate that dfrF is an acquired but probably chromosomally located gene which is responsible for in vitro HLR to trimethoprim in E. faecalis.

Trimethoprim is a broad-spectrum antimicrobial agent used extensively worldwide alone or in combination with sulfamethoxazole for the treatment of enteric, respiratory, skin, and urinary tract infections (25). Enterococci are usually susceptible in vitro to trimethoprim (9); however, the role of trimethoprim as a therapeutic option for enterococcal infections is controversial (21, 31) and has largely been dismissed on the basis of reports of clinical failures (18), lack of efficacy in experimental animal models of endocarditis and peritonitis (6, 19), the ability of some enterococcal species to use preformed folates (20, 21, 52), the dependence of the bactericidal activity on the assay conditions (36), and, overall, the lack of studies addressing the clinical efficacy of this antibiotic (31). Renewed interest in old and new antimicrobial agents active against multidrug-resistant microorganisms has emerged in recent years. Recent reports showed a bactericidal effect of trimethoprim-sulfamethoxazole in combination with ciprofloxacin against Enterococcus faecalis and Enterococcus casseliflavus strains (48). Moreover, such a combination has been reported to cure two patients with endocarditis caused by highly gentamicin-resistant E. faecalis strains (39, 49). In addition, trimethoprim analogs developed in recent years have demonstrated good activity against enterococci except for those highly resistant to trimethoprim (28). Thus, trimethoprim or its analogs might be reconsidered as alternative agents with activity against enterococci resistant to multiple antibiotics.

Trimethoprim inhibits the enzyme dihydrofolate reductase (DHFR), which catalyzes the reduction of dihydrofolate to tetrahydrofolate in prokaryotic and eukaryotic cells (23). The most frequent mechanism of resistance to trimethoprim is the production of an additional drug-resistant DHFR, often found on mobile genetic elements (plasmids, transposons, cassettes) (2, 25). While more than 15 DHFRs conferring high-level resistance (HLR) to trimethoprim have been identified in gram-negative bacteria, only two are known in gram-positive organisms: S1DHFR encoded by dfrA (found in Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus hominis, and Staphylococcus haemolyticus) and S2DHFR encoded by dfrD (found in Staphylococcus haemolyticus and, recently, in Listeria monocytogenes) (5, 12, 13, 25, 40). Chromosomally encoded resistance to trimethoprim due to mutational changes in the intrinsic DHFR gene has been reported in Escherichia coli, Haemophilus influenzae, Streptococcus pneumoniae, and S. aureus (1, 11, 25). This mechanism typically confers low or intermediate levels of resistance, although higher MICs for the organism can be achieved when regulatory mutations, which lead to enzyme overproduction, are present (11, 25). Other mechanisms of bacterial resistance to trimethoprim that have been described are impermeability (found in isolates of Serratia, Enterobacter, Klebsiella, Pseudomonas, and Clostridium) and mutational changes in the thymidylate synthase gene (2, 24, 25, 45). Intrinsic resistance to trimethoprim occurs in Pediococcus cerevisiae, Bacteroides fragilis, Clostridium spp., Moraxella catarrhalis, a Neisseria sp., a Nocardia sp., and a Lactobacillus sp. (24, 25).

Clinical isolates of E. faecalis and occasional isolates of Enterococcus faecium and Enterococcus hirae with different levels of susceptibility to trimethoprim have been reported in Europe and the United States since the beginning of the 1980s (9, 17, 22, 29, 52). However, the frequency of trimethoprim resistance is unknown since this antibiotic is not usually tested against these microorganisms and the validity of the results have been questioned (31). Furthermore, the molecular nature of trimethoprim resistance has not been investigated, although previous studies suggested that a modified DHFR might be responsible for the HLR in E. faecalis (22) and showed that DNA from trimethoprim-resistant E. faecalis strains did not hybridize to a probe for dfrA, indicating that other determinants may be involved in the trimethoprim resistance in these organisms (17). This study reports the characterization of a gene responsible for HLR to trimethoprim in an E. faecalis strain and a gene which is suggested to code for the intrinsic species-specific DHFR of E. faecalis.

MATERIALS AND METHODS

Bacterial strains and plasmids.

The bacterial strains and plasmids used for the cloning of enterococcal DNA are listed in Table 1. Twenty-three clinical isolates from different hospitals in the United States and Lebanon were also studied, including 12 E. faecalis isolates with different levels of susceptibility to trimethoprim and 11 E. faecium isolates highly resistant to this antibiotic (kindly provided by R. Then). E. coli cells were grown in Luria-Bertani broth or agar, and enterococcal cells were grown in brain heart infusion medium, Mueller-Hinton (MH) medium, or Todd-Hewitt medium (Difco, Detroit, Mich.). When necessary, these media were supplemented with the following antibiotics at the indicated concentrations: chloramphenicol, 20 μg/ml; tetracycline, 10 μg/ml; ampicillin, 50 μg/ml; erythromycin, 8 μg/ml; and trimethoprim, 2 to >1,028 μg/ml. Antibiotic susceptibility was determined by the agar dilution method with MH agar supplemented with 0.1 U of thymidine phosphorylase following the guidelines of the National Committee for Clinical Laboratory Standards (37). Pulsed-field gel electrophoresis was performed to determine clonal relatedness among strains (33).

TABLE 1.

Bacterial strains and plasmidsa

Strain or plasmid Relevant properties Reference or source
Bacteria
 E. coli
  DH5α endA F−gyrA96 hsdR17 recA1 relA1 thi-1 supE44 φ80 dlacZΔM15 Gibco BRL, Grand Island, N.Y.
  XL1 Blue MRF′ endA1 gyrA96 hsdSMR lac (F′ proAB lacIqZΔM15 Tn10) (Tetr) recA1 relA1 supE44 thi-1 Stratagene
 E. faecalis
  RH110 OG1RF::Tn916ΔEm Emr Fusr Rifr 34, 41
  DEL (TX0638) Clinical isolate; Bla+ Emr Gmr Kmr Smr Tmpr 17
  BEI (TX0787) Clinical isolate; Bla+ Emr Gmr Kmr Smr Tmpr 17
  E47 (TX0608) (clonally   related to DEL) Clinical isolate; Bla+ Cmr Emr Gmr Kmr Tetr Tmps 17, 34
Plasmids
 pBeloBAC11 Cosmid vector; Cmr 43, 51
 pUC18 Plasmid vector; AmprlacZα
 pBlueScript SK(−) 2,958-bp phagemid derived from pUC19; Ampr Stratagene
 pLAFRx Cosmid vector with mob site and oriT of RK2; Tetr 27
 pAT113 Emr KmrlacZα oriR(pACYC184) oriT(RK2) att Tn1545 47
Recombinants
 pBEM240 pBeloBAC11 containing ∼30-kb fragment from E. faecalis DEL; Tmpr in E. coli (MIC, >1,024 μg/ml) This study
 pBEM241 An ∼1.4-kb EcoRV fragment of pBEM240 subcloned into pBlueScript SK(−); Tmpr in E. coli (MIC, >1,024 μg/ml) This study
 pKV66 pLAFRx containing ∼28-kb fragment from E. faecalis OG1RF (Tmps); Tmpr in E. coli (MIC, 64 μg/ml) 35
 pBEM170 pUC18 containing a 1.5-kb Sau3AI fragment from E. faecalis E47 (TMPs); Tmpr in E. coli (MIC, 128 μg/ml) This study
 pBEM173 An 800-bp HincII fragment of pBEM170 subcloned into pUC18; Tmpr in E. coli (MIC, 128 μg/ml) This study
 pBEM244 pBlueScript SK(−) containing an 800-bp HincII fragment from E. faecalis DEL; Tmpr in E. coli (MIC, 128 μg/ml) This study
 pBEM248 pAT113 (48) containing an ∼1.2 kb SalI-XbaI fragment of pBEM241; Tmpr in E. coli (MIC, >1,024 μg/ml) This study
Open in a new tab
a

Abbreviations: Bla+, β-lactamase producer; Cm, chloramphenicol; Em, erythromycin; Fus, fusidic acid; Gm, gentamicin; Km, kanamycin; Rif, rifampin; Sm, streptomycin; Tet, tetracycline; Tmp, trimethoprim; r, resistant; s, susceptible; Tn916ΔEm, Tn916 with the tetracycline resistance gene interrupted by a gene encoding erythromycin resistance (41). 

Enzymes and chemicals.

Restriction enzymes, T4 ligase, isopropylthio-β-d-galactoside (IPTG), 5-bromo-4-chloro-3-indolyl-β-d-galactoside (X-Gal), and RNase-free DNase were purchased from Promega (Madison, Wis.). Klenow DNA polymerase was obtained from Boehringer Mannheim (Indianapolis, Ind.), shrimp alkaline phosphatase was obtained from United States Biochemicals (Cleveland, Ohio), and Incert agarose and SeaKem Gold agarose were provided by FMC Bioproducts (Rockland, Maine). [α-32P]dCTP was purchased from Amersham (Arlington Heights, Ill.). Antibiotics, thymidine phosphorylase, lysozyme, mutanolysin, proteinase K, and other reagents were obtained from Sigma Chemical Co. (St. Louis, Mo.).

DNA manipulation.

Total DNA from the E. faecalis strains was prepared by a protocol described by Murray et al. (35). The protocol was modified as needed for specific strains (e.g., the addition of 0.5 ml of 10% deoxycholic acid and 0.25 ml of Brij 58) in the lysis step. Lysis was achieved by treatment with 5 ml of 20% Sarkosyl at room temperature for 15 min and additional incubation in a 37°C water bath for 1 h. After the addition of 0.5 ml of 10% deoxycholic acid and 0.25 ml of Brij 58, the samples were incubated at 50°C with gentle shaking overnight.

Plasmid DNA from E. coli was prepared by the alkaline sodium dodecyl sulfate (SDS) method (42) or by using a commercial kit (Promega). Plasmid DNA from enterococci was prepared as described previously (42, 50). DNA digested with restriction enzymes was analyzed either by 0.8% agarose gel electrophoresis or by pulsed-field gel electrophoresis by standard methods (42). DNA fragments were extracted from the gels by eluting the agarose pieces with Supelco columns (Supelco, Bellefonte, Pa.) and were further purified with phenol-chloroform and extracted with 95% ethanol.

Preparation of competent cells and transformation of DNA into E. coli were performed as described previously (7). Competent E. faecalis cells were prepared by the protocol described by Friesenegger et al. (16). E. faecalis RH110 was transformed with 2 μg of plasmid DNA by electroporation (10) with a Gene Pulser device (Bio-Rad, LaJolla, Calif.) and the following conditions: 1.8 kV, 25 mF capacitance, and 200 Ω resistance. Colony lysis and Southern hybridizations were performed under high-stringency conditions (50% formamide, 5× Denhardt’s solution, 5× SSC [1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate; pH 7.2], 0.1% SDS) at 42°C overnight, followed by three washes in 2× SSC–0.1% SDS at room temperature (5 min each) and two washes in 0.1× SSC–0.1% SDS at 50°C (15 min each time) by standard methods (42). Radiolabeled probes were prepared by incorporation of [α-32P]dCTP-labeled deoxyribonucleotides with a randomly primed DNA labeling kit in accordance with the manufacturer’s instructions (Boehringer Mannheim). The membranes were exposed to X-ray film, which was then processed in an automated film developer.

Cloning strategy.

Total DNA (about 1 mg) from E. faecalis DEL was partially digested with Sau3AI and size fractionated in a sucrose density gradient. Fractions containing fragments of 18 to 30 kb were ligated to pBeloBAC11 digested with BamHI and were treated with shrimp alkaline phosphatase (43, 51). The ligation mixture was packaged with the Gigapack III Gold Packaging Extract kit (Stratagene, LaJolla, Calif.), transduced into E. coli DH5α cells, and plated onto MH agar supplemented with 32 μg of trimethoprim per ml and onto Luria-Bertani agar plus chloramphenicol with IPTG and X-Gal to verify the cloning efficiency. Subcloning was performed by a DNase I digestion-based method (14). One milligram of DNA from initial cosmid clones was treated with 0.01 U of RNase-free DNase in 10 mM MnCl2–50 mM Tris-Cl (pH 7.5) for 5 min at room temperature, and the reaction was stopped by the addition of EDTA at a final concentration of 12 mM. Purification of DNA fragments was done by phenol-chloroform extraction and ethanol precipitation. The ends of the fragments were blunted with the Klenow enzyme. After a new phenol-chloroform extraction and ethanol precipitation, DNA was digested with EcoRV, ligated to digested pBlueScript SK(−) and introduced into E. coli XL1Blue by electroporation. An additional library was prepared with total DNA (about 1 mg) from E. faecalis DEL that had been partially digested with HincII, ligated with pBlueScript SK(−), transformed into E. coli XL1-Blue MRF′, and plated onto MH agar with 8 μg of trimethoprim per ml.

E. coli DNA libraries were also prepared from two Tmps E. faecalis strains, OG1RF (into pLAFRx) (35) and E47 (into pUC18) (34), and were screened for the presence of dfr genes by plating onto MH agar plates containing 4 or 8 μg of trimethoprim per ml. Recombinant clones and subclones showing resistance to trimethoprim were stored for further analysis.

Conjugation experiments.

Enterococcal matings were done by incubation of donor and recipient strains on membrane filters, in broth cultures, and after cross-streaking on agar as described previously (17, 32). E. faecalis DEL and Beirut were used as donors, and E. faecalis OG1RF was used as the recipient. Selection was done on MH agar supplemented with 32 μg of trimethoprim per ml.

DNA sequencing and computer analysis of sequencing data.

Double-stranded DNA sequencing was performed by the Taq-Dye-deoxy terminator method (Applied Biosystems, Foster City, Calif.) with a 373A DNA sequencing system (Applied Biosystems). The primers used were primers for the T3 and T7 promoter regions of pBlueScript SK(−), the universal primers for pUC18, and primers specifically selected from the studied sequences. Nucleotide and amino acid sequences were analyzed with the BLAST program by use of GenBank, EMBL, and Swiss-Prot with GCG software (Genetics Computer Group, Madison, Wis.).

Integration of dfrF into E. faecalis RH110.

The integrative vector pAT113 (kindly provided by P. Courvalin [47]) was used to introduce an enterococcal DNA fragment encoding trimethoprim resistance back into the chromosome of E. faecalis. An approximately 1.2-kb SalI-XbaI fragment of pBEM241 (Table 1) was cloned into the corresponding sites of pAT113. This recombinant plasmid was introduced into E. faecalis RH110 (which is E. faecalis OG1RF::Tn916ΔEm [41]) and plated onto MH agar plus 32 μg of trimethoprim per ml.

Nucleotide sequence accession numbers.

The sequence data reported in this study have been submitted to the GenBank/EMBL database and have the accession nos. AF028811 (dfrE) and AF028812 (dfrF).

RESULTS

Cloning and expression of the enterococcal DHFR genes in E. coli.

Results obtained from the cloning and subcloning experiments are summarized in Table 1. Tmpr clones obtained from the DNA library (in pBeloBAC11) from E. faecalis DEL (trimethoprim resistant [Tmpr]) partially digested with Sau3AI yielded recombinants with ∼30-kb inserts (e.g., for pBEM240, the MIC was >1,024 μg/ml). The smallest Tmpr EcoRV-digested subclone of pBEM240 contained a 1.4-kb insert and was named pBEM241; this subclone conferred HLR to E. coli (trimethoprim MIC, >1,024 μg/ml). The smallest inserts obtained from the HincII libraries from E. faecalis DEL (Tmpr) and E47 (trimethoprim susceptible [Tmps]) that caused moderate resistance in E. coli (in the multicopy plasmids pBlueScript SK(−) and pUC18) contained inserts of ∼800 bp and were designated pBEM244 and pBEM173, respectively (Table 1). The MICs of trimethoprim for these recombinant plasmids were 128 μg/ml.

Characterization of the enterococcal DHFR genes.

The 800-bp HincII fragments cloned from the Tmps strain E. faecalis E47 and the Tmpr strain E. faecalis DEL had identical nucleotide sequences (Fig. 1). This common fragment contained an open reading frame (ORF) of 495 bp that we provisionally have named dfrE. The G+C content of the gene is 42 mol%, similar to that of the E. faecalis chromosome. The nucleotide sequence 5′ to dfrE had a high degree of similarity to those of the thymidylate synthase (thy) genes from other bacterial species. The organization and the spacing between both genetic determinants (17 bp) suggest that dfr and thy from E. faecalis are located in tandem in an operon (Fig. 1). Comparison of the deduced amino acid sequence corresponding to dfrE (designated DHFREfs) with other DHFR sequences is shown in Fig. 2. An interesting feature of DHFREfs, which is shared with the intrinsic DHFR from Lactococcus lactis, S. pneumoniae, and Streptococcus (Enterococcus) faecium, is that position 49 is occupied by a glycine residue instead of the conserved serine residue in all Tmpr and Tmps DHFRs from gram-negative and gram-positive organisms except the aforementioned species.

FIG. 1.

FIG. 1

Open in a new tab

Nucleotide and deduced amino acid sequences of dfrE and thy from E. faecalis DEL and E47. Putative Shine-Dalgarno sequences are underlined. The asterisk indicates a stop codon. The vector sequence is double underlined.

FIG. 2.

FIG. 2

FIG. 2

FIG. 2

Open in a new tab

Alignment of the sequences of the chromosomal DHFR and E1DHFR from E. faecalis DEL with those of types S1, S2, I, and V and the chromosomal DHFRs from E. coli, Enterobacter aerogenes, Neisseria gonorrhoeae, Bacillus subtilis, S. pneumoniae, L. lactis, and Lactobacillus casei. The positions involved in the binding of trimethoprim (T), methotrexate (T and t), and the cofactor NADPH (n) have been taken from the studies with the E. coli K-12 DHFR; and the features of the secondary structures, β sheets and α helices, correspond to those previously described for other DHFRs (1, 4, 30, 40).

The ∼1.4-kb EcoRV fragment of pBEM241 cloned from the highly trimethoprim-resistant strain E. faecalis DEL (which in turn confers to E. coli HLR to trimethoprim) contains an ORF of 495 bp that encodes a putative 165-amino-acid polypeptide that we tentatively have named dfrF and E1DHFR, respectively (Fig. 3). The sequence 5′ to this dfr did not show similarity with thy sequences. The G+C content of this gene is 32 mol%. Analysis of the deduced amino acid sequence of E1DHFR with known DHFRs showed 38 to 64% similarity with other DHFRs from gram-positive and gram-negative organisms. Of the 12 residues involved in the trimethoprim binding region, 6 are conserved in most DHFRs (4, 30, 40). Of these six, four are present in E1DHFR (at positions 7, 31, 54, and 113) (Fig. 2). From the nonconserved residues, some interesting features can be observed. There is a glutamate instead of an aspartate residue at position 27, a change that is associated with some Tmpr DHFRs (3, 25). In addition, the residue at position 28 is occupied by a Gln in the E1DHFR. The substitution Gln28-Leu appears in some Tmpr DHFRs from gram-negative organisms (family 1 and DHFR type XII, which are highly resistant to trimethoprim) and from L. lactis. Interestingly, this change is always associated with the aforementioned Glu27-Asp. Finally, at position 49 E1DHFR has a glutamate residue instead of the glycine residue that is found in the chromosomal DHFRs from S. pneumoniae, S. (E.) faecium, E. faecalis, and L. lactis.

FIG. 3.

FIG. 3

Open in a new tab

Nucleotide and deduced amino acid sequences of dfrF conferring HLR to trimethoprim in E. faecalis. The putative −35 and −10 sequences are underlined. The putative Shine-Dalgarno sequence is double underlined. The asterisk indicates a TAG stop codon.

Transfer of trimethoprim resistance from strain DEL was not achieved by broth, filter, or cross-streak matings (17) or in this study. Plasmid DNA from this strain did not hybridized with a probe (1.4-kb XhoI-NotI fragment from pBEM241) for dfrF, but total genomic DNA did hybridize with this probe.

Integration of dfrF into the Tmps strain E. faecalis RH110.

The presence of multiple resistance markers in E. faecalis DEL made selection after genetic manipulation of this strain difficult. Efforts to interrupt the gene in the original host (E. faecalis DEL) were performed by cloning an internal fragment of the dfrF gene into pTEX5235 (44), a mobilizable derivative of pBlueScript SK(−) and introducing this suicide recombinant plasmid into DEL by using spectinomycin for selection; however, spectinomycin-resistant (Spr) mutants of this strain were too numerous (data not shown). To prove that dfrF was responsible for HLR to trimethoprim, pBEM248 was generated, electroporated into the Tmps strain E. faecalis RH110 (41), and plated onto MH agar plus trimethoprim. The MICs for eight Tmpr colonies were determined, and trimethoprim MICs for all colonies were >1,028 μg/ml. Southern analysis of these electrotransformants showed that the gene conferring HLR to trimethoprim in E. faecalis DEL had inserted at different positions of the RH110 chromosome and as a single copy in strains from six of eight colonies.

Distribution of the dfr genes among enterococci.

By using colony lysates and high-stringency conditions, a probe for dfrF hybridized only to DNA from E. faecalis isolates for which the trimethoprim MIC was ≥1,028 μg/ml and to the DNAs of 2 of 11 E. faecium isolates highly resistant to trimethoprim. These two E. faecium isolates were felt to be the same strain after they were analyzed by pulsed-field gel electrophoresis.

A probe for the dfrE gene from Tmps E. faecalis hybridized to the DNAs of all 12 E. faecalis isolates tested, and 4 of these were susceptible to trimethoprim, 5 were moderately resistant to trimethoprim, and 3 were highly resistant to trimethoprim, suggesting that this gene codes for the intrinsic DHFR of E. faecalis.

DISCUSSION

Trimethoprim-resistant bacteria, including enterococci, have been isolated worldwide since shortly after the introduction for clinical use in 1969 of the combination of trimethoprim and sulfamethoxazole (9, 17, 22, 25, 29). During the last two decades, trimethoprim-resistant gram-negative bacteria and staphylococci have been widely reported, and a variety of mechanisms responsible for acquired resistance to trimethoprim or trimethoprim-sulfamethoxazole have been characterized (11, 24, 25, 45). Among the trimethoprim-resistant bacteria, enterococci remain the least studied, probably due to the controversial role of trimethoprim as a therapeutic option (31).

HLR to trimethoprim among clinical isolates of other species is often due to the presence of an additional drug-resistant DHFR (24, 25, 45). Our results demonstrate the coexistence of two dfr genes in the highly trimethoprim-resistant strain E. faecalis DEL and suggest that high-level resistance in this strain is also due to the presence of an additional drug-resistant DHFR. Our data corroborate those of a previously published study by Hamilton-Miller and Stewart (22) in which the determination of the specific activities of DHFRs from different Tmpr and Tmps E. faecalis isolates showed that highly trimethoprim-resistant E. faecalis isolates possessed a DHFR that was 1,000-fold less sensitive than those from E. faecalis Tmps strains, suggesting that an alternative DHFR might be responsible for the high-level resistance in these microorganisms.

The dfrE gene that caused resistance in E. coli when it was cloned into a multicopy plasmid is believed to code for an intrinsic E. faecalis DHFR because it was found in all E. faecalis isolates studied, despite their level of susceptibility to trimethoprim. Analysis of the sequence of the DNA fragments containing dfrE revealed the presence of a thy homolog (potentially coding for a thymidylate synthase [TS]) 5′ to the dfrE gene, suggesting that they are located in tandem in a operon. DHFR and TS enzymes are part of a three-enzyme cycle of pyrimidine metabolism, together with serine hydroxy methyltransferase, and these genes are linked in some bacterial species and protozoa (15). Our results suggest that the organization of the genes coding for DHFR and TS in E. faecalis is similar to those in Bacillus subtilis, L. casei, S. epidermidis, and E. coli (12, 26, 38, 40, 46).

Several studies have demonstrated that HLR to trimethoprim in gram-negative bacteria can be attributed to more than one mechanism of resistance (45). In E. faecalis DEL, the presence of an altered chromosomal DHFR is unlikely since the sequences of the putative chromosomal gene from this strain and that of the Tmps E. faecalis strain studied were identical. However, the possibility of other mechanisms, such as impaired penetration, efflux, or enzyme overproduction, cannot be ruled out.

DHFRs share a number of residues involved in the quaternary structure and in protein-cofactor-substrate interactions (1, 25, 40). Some of these positions have been demonstrated to play a role in trimethoprim resistance in previous studies (1, 3, 11). In the E1DHFR, the presence of an aspartate instead of a glutamate residue at position 27 has previously been related to highly trimethoprim-resistant DHFRs in vertebrates and some DHFRs highly resistant to trimethoprim from gram-negative bacteria (family 1 and DHFR type XII) (3, 25). The chromosomal DHFRs from L. lactis and S. pneumoniae also carry this substitution, and in the latter species, this substitution has been proposed to explain the modest susceptibility to trimethoprim (1). In addition, E1DHFR has other amino acid differences at the trimethoprim binding positions which have not previously been linked to trimethoprim resistance. The substitution Gln28-Leu appears in highly trimethoprim-resistant DHFRs from gram-negative organisms (family 1 and DHFR type XII) and in the DHFR from L. lactis. Interestingly, this alteration is always associated with the aforementioned Glu27-Asp except in the case of S. pneumoniae. Finally, the amino acid residue at position 49 of the E1DHFR is occupied by a serine residue in most Tmpr and Tmps DHFRs except those of L. lactis, S. (E.) faecium, S. pneumoniae, and E. faecalis, in which he corresponding amino acid is a glycine. Whether these amino acids have the same role in the activity of the E. faecalis E1DHFR enzyme against trimethoprim as has been suggested for other DHFRs remains to be elucidated by direct mutagenesis experiments and/or analysis of the crystal structure of this protein.

The diversity of genetic determinants of trimethoprim resistance has led to the suggestion that they might have originated from different housekeeping DHFR genes and afterward spread geographically (25). This hypothesis has been demonstrated for the trimethoprim-resistant DHFR from S. aureus, which appears to have been derived from the chromosomal enzyme of S. epidermidis (12), but the origins of other acquired enzymes remain unknown. The E1DHFR is not derived from DHFR genes known to date. However, the low G+C content of dfrF gene (32 mol%) and the substitutions found at the trimethoprim binding positions common to the chromosomal DHFR from S. pneumoniae, S. (Enterococcus) faecium, and E. faecalis might suggest a streptococcal origin. Studies of chromosomally encoded DHFRs from other gram-positive organisms might provide important insights into the origin and evolution of this DHFR.

Many of the genetic determinants coding for acquired DHFRs are located on mobile elements (plasmids, transposons, or cassettes) that participate in the dissemination of these genes. On the basis of conjugation and hybridization studies, the dfrF in E. faecalis DEL appears to be located on the chromosome rather than a plasmid. Our findings are in agreement with those of Hamilton-Miller and Stewart (22), who demonstrated that HLR to trimethoprim in E. faecalis is not always transferable and observed that some E. faecalis strains with this antibiotic resistance pattern are plasmid free. The possibility that the determinant causing HLR to trimethoprim in E. faecalis may be on a transposable element, which is a common feature of enterococci (8), cannot be eliminated. Indeed, the two E. faecalis isolates DEL and E47, which are Tmpr and Tmps, respectively, have similar genomic SmaI digestion patterns (34), suggesting the possibility that this resistance was acquired via a mobile determinant.

Horizontal transfer of trimethoprim resistance genes can also occur among bacteria of different species and genera (5, 12, 40). The sequence of a PCR-amplified DNA product from a highly trimethoprim resistant E. faecium strain obtained by using intragenic primers for dfrF revealed the presence of an identical sequence (unpublished observation). The presence of dfrF in this E. faecium strain (two isolates that had identical patterns by pulsed-field gel electrophoresis) illustrates the interspecies occurrence of this resistance determinant. In addition, the fact that not all Tmpr E. faecium isolates hybridized with a probe for the trimethoprim resistance gene dfrF suggests that other determinants may be involved in this resistance.

The mechanism of resistance to trimethoprim has been the subject of intense research in recent years; however, there is still much to learn about the origin and evolution of these genes and the molecular nature of the resistance. We present evidence that HLR to trimethoprim in E. faecalis DEL and other enterococci is caused by the production of an additional DHFR; but more studies addressing the mechanism of resistance of E. faecalis isolates for which MICs range from 16 to 512 μg/ml, the mechanism of resistance of other enterococcal species, and overall, the role of the amino acid substitutions in the E1DHFR are necessary and might lead to a rational approach for the design of new folate synthesis inhibitors (27, 35, 51).

ACKNOWLEDGMENTS

We are grateful to Yi Xu and Xiang Qin for technical assistance in constructing DNase I random libraries, advice on sequence analysis, and helpful discussions. We also thank Rudolph L. Then of Hoffmann-La Roche Ltd. (Basel, Switzerland), who kindly provided to us the trimethoprim-resistant E. faecium isolates included in this study.

REFERENCES

  • 1.Adrian P V, Klugman K P. Mutations in the dihydrofolate reductase gene of trimethoprim-resistant isolates of Streptococcus pneumoniae. Antimicrob Agents Chemother. 1997;41:2406–2413. doi: 10.1128/aac.41.11.2406. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Amyes S G B, Towner K J. Trimethoprim resistance: epidemiology and molecular aspects. J Med Microbiol. 1990;31:1–19. doi: 10.1099/00222615-31-1-1. [DOI] [PubMed] [Google Scholar]
  • 3.Appleman J R, Howell E E, Kraut J, Blakley R L. Role of the aspartate 27 of dihydrofolate reductase from Escherichia coli in interconversion of active and inactive conformers and binding of NADPDH. J Biol Chem. 1990;265:5579–5584. [PubMed] [Google Scholar]
  • 4.Bolin J T, Filman D J, Mathews D A, Hamlin R C, Kraut J. Crystal structures of Escherichia coli and Lactobacillus casei dihydrofolate reductase refined at 1.7Å. I. General features and binding of methotrexate. J Biol Chem. 1982;257:13650–13662. [PubMed] [Google Scholar]
  • 5.Charpentier E, Courvalin P. Emergence of a new class of trimethoprim resistance gene dfrD in Listeria monocytogenes BM4293. Antimicrob Agents Chemother. 1997;41:1134–1136. doi: 10.1128/aac.41.5.1134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Chenoweth C E, Robinson K A, Schaberg D R. Efficacy of ampicillin versus trimethoprim-sulfamethoxazole in a mouse model of lethal enterococcal peritonitis. Antimicrob Agents Chemother. 1990;34:1800–1802. doi: 10.1128/aac.34.9.1800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chung C T, Niemela S Z, Miller R H. One-step preparation of competent Escherichia coli: transformation and storage of bacterial cells in the same solution. Proc Natl Acad Sci USA. 1989;86:2172–2175. doi: 10.1073/pnas.86.7.2172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Clewell D B. Movable genetic elements and antibiotic resistance in enterococci. Eur J Clin Microbiol Infect Dis. 1990;9:90–102. doi: 10.1007/BF01963632. [DOI] [PubMed] [Google Scholar]
  • 9.Crider S R, Colby S D. Susceptibility of enterococci to trimethoprim and trimethoprim-sulfamethoxazole. Antimicrob Agents Chemother. 1985;27:71–75. doi: 10.1128/aac.27.1.71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Cruz-Rodz A L, Gilmore M S. High efficiency introduction of plasmid DNA into glycine treated Enterococcus faecalis by electroporation. Mol Gen Genet. 1990;224:152–154. doi: 10.1007/BF00259462. [DOI] [PubMed] [Google Scholar]
  • 11.Dale G E, Broger C, D’Arcy A, Hartman P G, DeHoogt R, Jolidon S, Kompis I L, M. A, Langen H, Locher H, Page M G P, Stuber D, Then R L, Wipf B, Oefner C. A single amino acid substitution in Staphylococcus aureus dihydrofolate reductase determines trimethoprim resistance. J Mol Biol. 1997;266:23–30. doi: 10.1006/jmbi.1996.0770. [DOI] [PubMed] [Google Scholar]
  • 12.Dale G E, Broger C, Hartman P G, Langen H, Page M G P, Then R L, Stuber D. Characterization of the gene for the chromosomal dihydrofolate reductase (DHFR) of Staphylococcus epidermidis ATCC 14990: the origin of the trimethoprim resistant S1DHFR from Staphylococcus aureus? J Bacteriol. 1995;177:2965–2970. doi: 10.1128/jb.177.11.2965-2970.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Dale G E, Langen H, Page M G P, Then R L, Stuber D. Cloning and characterization of a novel, plasmid-encoded trimethoprim-resistant dihydrofolate reductase from Staphylococcus haemolyticus MUR313. Antimicrob Agents Chemother. 1995;39:1920–1924. doi: 10.1128/aac.39.9.1920. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Demolis N, Mallet L, Bussereau F, Jacquet M. Improved strategy for large-scale DNA sequencing using DNase I cleavage for generating random subclones. BioTechniques. 1995;18:453–457. [PubMed] [Google Scholar]
  • 15.Ehrhardt Howell E, Foster P G, Foster L M. Construction of a dihydrofolate reductase-deficient mutant of Escherichia coli by gene replacement. J Bacteriol. 1988;170:3040–3045. doi: 10.1128/jb.170.7.3040-3045.1988. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Friesenegger A, Fiedler S, Devriese L A, Wirth R. Genetic transformation of various species of Enterococcus by electroporation. FEMS Microbiol Lett. 1991;79:323–328. doi: 10.1016/0378-1097(91)90106-k. [DOI] [PubMed] [Google Scholar]
  • 17.Frosolono M, Hodel-Christian S L, Murray B E. Lack of homology of enterococci which have high-level resistance to trimethoprim with the dfrA gene of Staphylococcus aureus. Antimicrob Agents Chemother. 1991;35:1928–1930. doi: 10.1128/aac.35.9.1928. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Goodhard G L. In vivo v in vitro susceptibility of enterococcus to trimethoprim-sulfamethoxazole. JAMA. 1984;252:2748–2749. [PubMed] [Google Scholar]
  • 19.Grayson M L, Thauvin-Eliopoulos C, Eliopoulos G M, Yao J D C, DeAngelis D V, Walton L, Woolley J L, Moellering R C., Jr Failure of trimethoprim-sulfamethoxazole therapy in experimental enterococcal endocarditis. Antimicrob Agents Chemother. 1990;34:1792–1794. doi: 10.1128/aac.34.9.1792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hamilton-Miller J, Purves D. Enterococci and antifolate antibiotics. Eur J Clin Microbiol. 1986;5:391–394. [PubMed] [Google Scholar]
  • 21.Hamilton-Miller J M T. Reversal of activity of trimethoprim against gram-positive cocci by thymidine, thymine and “folates.”. J Antimicrob Chemother. 1988;22:35–39. doi: 10.1093/jac/22.1.35. [DOI] [PubMed] [Google Scholar]
  • 22.Hamilton-Miller J M T, Stewart S. Trimethoprim resistance in enterococci: microbiological and biochemical aspects. Microbios. 1988;56:45–55. [PubMed] [Google Scholar]
  • 23.Hitchins, G. H. 1973. Mechanism of action of trimethoprim-sulfamethoxazole. J. Infect. Dis. 128(Suppl.):433–436. [DOI] [PubMed]
  • 24.Huovinen P. Trimethoprim resistance. Antimicrob Agents Chemother. 1987;31:1451–1456. doi: 10.1128/aac.31.10.1451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Huovinen P, Sundstrom L, Swedberg G, Skold O. Trimethoprim and sulfonamide resistance. Antimicrob Agents Chemother. 1995;39:279–289. doi: 10.1128/aac.39.2.279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Iwakura M, Kawata M, Tsuda K, Tanaka T. Nucleotide sequence of the thymidylate synthase B and dihydrofolate reductase genes contained in one Bacillus subtilis operon. Gene. 1988;64:9–20. doi: 10.1016/0378-1119(88)90476-3. [DOI] [PubMed] [Google Scholar]
  • 27.Li X, Weinstock G M, Murray B E. Generation of auxotrophic mutants of Enterococcus faecalis. J Bacteriol. 1995;177:6866–6873. doi: 10.1128/jb.177.23.6866-6873.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Locher H H, Schlunegger H, Hartman P G, Angehrn P, Then R L. Antibacterial activities of epiroprim, a new dihydrofolate reductase inhibitor, alone and in combination with dapsone. Antimicrob Agents Chemother. 1996;40:1376–1381. doi: 10.1128/aac.40.6.1376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Mangan M W, McNamara E B, Smyth E G, Storrs M J. Molecular genetic analysis of high-level gentamicin resistance in Enterococcus hirae. J Antimicrob Chemother. 1997;40:377–382. doi: 10.1093/jac/40.3.377. [DOI] [PubMed] [Google Scholar]
  • 30.Mathews D A, Bolin J T, Burridge J M, Filman D J, Volz K W, Kaufman B T, Beddel C R, Champness J N, Stammers D K, Kraut J. Refined crystal structures of Escherichia coli and chicken liver dihydrofolate reductase containing bound trimethoprim. J Biol Chem. 1985;260:381–391. [PubMed] [Google Scholar]
  • 31.Murray B E. Antibiotic treatment of enterococcal infection. Antimicrob Agents Chemother. 1989;33:1411. doi: 10.1128/aac.33.8.1411. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Murray B E, An F Y, Clewell D B. Plasmids and pheromone response of the β-lactamase producer Streptococcus (Enterococcus) faecalis HH22. Antimicrob Agents Chemother. 1988;32:547–551. doi: 10.1128/aac.32.4.547. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Murray B E, Singh K V, Heath J D, Sharma B R, Weinstock G M. Comparison of genomic DNAs of different enterococcal isolates using restriction endonucleases with infrequent recognition sites. J Clin Microbiol. 1990;28:2059–2063. doi: 10.1128/jcm.28.9.2059-2063.1990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Murray B E, Singh K V, Markowitz S M, Lopardo H A, Patterson J E, Zervos M J, Rubeglio E, Eliopoulos G M, Rice L B, Goldstein F W, Jenkins S G, Caputo G M, Nasnass R, Moore L S, Wong E S, Weinstock G. Evidence for clonal spread of a single strain of β-lactamase-producing Enterococcus faecalis to six hospitals in five states. J Infect Dis. 1991;163:780–785. doi: 10.1093/infdis/163.4.780. [DOI] [PubMed] [Google Scholar]
  • 35.Murray B E, Singh K V, Ross R P, Heath J D, Dunny G M, Weinstock G M. Generation of restriction map of Enterococcus faecalis OG1 and investigation of growth requirements and regions encoding biosynthetic function. J Bacteriol. 1993;175:5216–5223. doi: 10.1128/jb.175.16.5216-5223.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Najjar A, Murray B E. Failure to demonstrate a consistent in vitro bactericidal effect of trimethoprim-sulfamethoxazole against enterococci. Antimicrob Agents Chemother. 1987;31:808–810. doi: 10.1128/aac.31.5.808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.National Committee for Clinical Laboratory Standards. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. 3rd ed. Approved standard. NCCLS document M7-A3. Wayne, Pa: National Committee for Clinical Laboratory Standards; 1993. [Google Scholar]
  • 38.Pinter K, Davisson V J, Santi D V. Cloning and expression of the Lactobacillus casei gene. DNA. 1988;7:235–241. doi: 10.1089/dna.1988.7.235. [DOI] [PubMed] [Google Scholar]
  • 39.Rambaldi M, Ambrosone L, Migliaresi S, Rambaldi A. Combination of co-trimoxazole and ciprofloxacin as therapy of a patient with infective endocarditis caused by an enterococcus highly resistant to gentamicin. J Antimicrob Chemother. 1997;40:737–738. doi: 10.1093/jac/40.5.737. [DOI] [PubMed] [Google Scholar]
  • 40.Rouch D A, Messerotti L J, Loo L S L, Jackson C A, Skurray L A. Trimethoprim resistance transposon Tn4003 from Staphylococcus aureus encodes genes for a dihydrofolate reductase and thymidylate synthetase flanked by three copies of IS257. Mol Microbiol. 1989;3:161–175. doi: 10.1111/j.1365-2958.1989.tb01805.x. [DOI] [PubMed] [Google Scholar]
  • 41.Rubens C E, Heggen L M. Tn916ΔE: A Tn916 transposon derivative expressing erythromycin resistance. Plasmid. 1988;20:137–142. doi: 10.1016/0147-619x(88)90016-9. [DOI] [PubMed] [Google Scholar]
  • 42.Sambrook J, Fritsch E F, Maniatis T. Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory Press; 1989. [Google Scholar]
  • 43.Shiyuza H, Birren B, Kim U-J, Mancino S T, Slepak Y, Tachiiri Y, Simon M. Cloning and stable maintenance of 300-kilobase-pair fragments of human DNA in Escherichia coli using an F-factor-based vector. Proc Natl Acad Sci USA. 1992;89:8794–8797. doi: 10.1073/pnas.89.18.8794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Teng F, Murray B E, Weinstock G M. Conjugal transfer of plasmid DNA from Escherichia coli to enterococci: a method to make insertion mutations. Plasmid. 1998;39:182–186. doi: 10.1006/plas.1998.1336. [DOI] [PubMed] [Google Scholar]
  • 45.Then R L. Mechanisms of resistance to trimethoprim, the sulfonamides, and trimethoprim-sulfamethoxazole. Rev Infect Dis. 1982;4:261–269. doi: 10.1093/clinids/4.2.261. [DOI] [PubMed] [Google Scholar]
  • 46.Tolleson W H, Ahmed F, Dunlap R B. Cloning of the linked thymidylate synthase and dihydrofolate reductase genes from aminopterin-resistant Lactobacillus casei. Fed Proc. 1987;46:2073. . (Abstract 862.) [Google Scholar]
  • 47.Trieu-Cuot P, Carlier C, Poyart-Salmeron C, Courvalin P. An integrative vector exploiting the transposition properties of Tn1545 for insertional mutagenesis and cloning of genes from gram-positive bacteria. Gene. 1991;106:21–27. doi: 10.1016/0378-1119(91)90561-o. [DOI] [PubMed] [Google Scholar]
  • 48.Tripodi M F, Utili R, Locatelli A, Rosario P, Florio A, Ruggiero G. Unorthodox antibiotic combinations including ciprofloxacin against high level gentamicin resistant enterococci. J Antimicrob Chemother. 1996;37:727–736. doi: 10.1093/jac/37.4.727. [DOI] [PubMed] [Google Scholar]
  • 49.Tripodi M F, Rambaldi A, Utili R, Rosario P, Attanasio V, Locatelli A, Adinolfi L E, Andreana A, Florio A, Ruggiero G. Resistance to aminoglycosides and other antibiotics among clinical isolates of Enterococcus spp. New Microbiol. 1995;18:319–323. [PubMed] [Google Scholar]
  • 50.Woodford N, Morrison D, Cookson B, George R C. Comparison of high-level gentamicin-resistant Enterococcus faecium isolates from different continents. Antimicrob Agents Chemother. 1993;37:681–684. doi: 10.1128/aac.37.4.681. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Xu Y, Jiang L, Murray B E, Weinstock G M. Enterococcus faecalis antigens in human infections. J Bacteriol. 1995;65:4207–4215. doi: 10.1128/iai.65.10.4207-4215.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Zervos M J, Schaberg D S. Reversal of the in vitro susceptibility of enterococci to trimethoprim-sulfamethoxazole by folinic acid. Antimicrob Agents Chemother. 1985;28:446–448. doi: 10.1128/aac.28.3.446. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Antimicrobial Agents and Chemotherapy are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES