Abstract
Of the 176 randomly selected, commensal, gram-negative bacteria isolated from healthy children with low exposure to antibiotics, 138 (78%) carried one or more of the seven macrolide resistance genes tested in this study. These isolates included 79 (91%) isolates from the oral cavity and 59 (66%) isolates from urine samples. The mef(A) gene, coding for an efflux protein, was found in 73 isolates (41%) and was the most frequently carried gene. The mef(A) gene could be transferred from the donors into a gram-positive E. faecalis recipient and a gram-negative Escherichia coli recipient. The erm(B) gene transferred and was maintained in the E. coli transconjugants but was found in 0 to 100% of the E. faecalis transconjugants tested, while the other five genes could be transferred only into the E. coli recipient. The individual macrolide resistance genes were identified in 3 to 12 new genera. Eight (10%) of the oral isolates and 30 (34%) of the urine isolates for which the MICs were 2 to >500 μg of erythromycin per ml did not hybridize with any of the seven genes and may carry novel macrolide resistance genes.
The use of macrolide and related antibiotics (ketolides, oxazolidinones, streptogramins, and lincosamide) has increased dramatically over the last 15 years. A number of different mechanisms of macrolide resistance have been reported for gram-negative bacteria. These mechanisms include two esterase genes [ere(A) and ere(B)] found in Escherichia coli and Enterobacter, Klebsiella, Citrobacter, and Proteus species (1) and more recently, in Providencia stuartii, Pseudomonas species, and Vibrio cholerae (5, 17, 22). These mechanisms also include three phosphorylase genes [mph(A), mph(B), and mph(D)] found in E. coli (14, 15) and Pseudomonas (12) and one rRNA methylase gene [erm(B)] previously found in E. coli and Actinobacillus, Klebsiella, Neisseria, and Wolinella species (1, 4, 18-21). The strains described above were principally clinical isolates from hospital settings and/or from patients with clinical disease (17, 18, 20, 21), with members of the family Enterobacteriaceae and Pseudomonas species primarily isolated from France and Japan (1, 5, 11, 12, 14, 15). In contrast, relatively little is known about the presence of these six macrolide resistance genes in gram-negative bacteria from healthy individuals.
Recently, Cousin et al. found the efflux gene mef(A) in Neisseria gonorrhoeae and Acinetobacter junii (3, 7). This gene has been transferred, by conjugation in the laboratory, into a variety of gram-negative species, including Eikenella corrodens, Haemophilus influenzae, Kingella denitrificans, Moraxella catarrhalis, commensal Neisseria, and Neisseria meningitidis recipients using both gram-negative and gram-positive donors (7). These results suggested that the mef(A) gene may also be widespread among gram-negative species, so the mef(A) gene was included in the study of a group of randomly selected, commensal, gram-negative bacterial strains from oral and urine samples, collected from healthy children with low exposure to antibiotics from Lisbon, Portugal, for the presence of these seven acquired genes coding for macrolide resistance.
MATERIALS AND METHODS
Population and bacterial strains.
A total of 176 randomly chosen, commensal, gram-negative bacterial strains were isolated from oral and urine samples collected from healthy children in Lisbon, Portugal, who were participating in a randomized study designed to assess the safety of low-level mercury exposure from dental amalgam restorations. Children were 8 to 11 years of age during the recruitment period of February 1997 through April 1998, while isolates were from cultures obtained between December 1997 and March 1999 (Table 1). From the records, we found that during the collection period, five or six children per year received some type of medication from the doctors and included both antibiotics and nonantibiotic drugs. The bacteria represented 13 different genera (Table 1) and included 87 isolates from the oral cavity and 89 isolates from urine. The isolates were identified using CHROMagar orientation medium (DRG International, Inc., Mountainside, N.J.), standard biochemicals (10), and API kits according to the manufacturer's instructions (Biomerieux, Hazelwood, Mo.).
TABLE 1.
Distribution of macrolide resistance genes found in gram-negative isolates from the oral cavity and urine samples
| Genus | Oral isolates
|
Urine isolates
|
||
|---|---|---|---|---|
| No. of isolates | Gene(s) carrieda | No. of isolates | Gene(s) carried | |
| Acinetobacter | 3 | 1 mef(A) | ||
| 1 ere(A) | ||||
| 1 ere(B), erm(B) | ||||
| Citrobacter | 4 | 1 ere(A) | 1 | 1 mef(A) |
| 1 mef(A), erm(B) | ||||
| 1 mef(A), mph(A) | ||||
| 1 mef(A), ere(A), ere(B), mph(A) | ||||
| Enterobacter | 6 | 2 mef(A) | ||
| 1 erm(B) | ||||
| 1 ere(B) | ||||
| 1 mef(A), ere(B), mph(A) | ||||
| 1 mef(A), ere(A), ere(B), mph(A), mph(B) | ||||
| Escherichia | 7 | 3 mef(A) | 19 | 2 mef(A) |
| 2 erm(B) | 12 mph(A) | |||
| 1 ere(a) | 1 mph(B) | |||
| 1 mph(A), mph(B), mph (D) | 1 mph(A), mph(B) | |||
| 1 mef(A), mph(B) | ||||
| 1 mef(A), mph(D) | ||||
| 1 mef(A), ere(B), mph(A), mph(B) mph(D) | ||||
| Klebsiella | 19 | 5 mef(A) | 6 | 1 mef(A) |
| 4 ere(A) | 2 erm(B) | |||
| 2 erm(B) | 1 mph(A) | |||
| 2 mef(A), erm(B) | 2 mph(A), mph(B), mph(D) | |||
| 1 mef(A), mph(A) | ||||
| 1 ere(A), ere(B), mph(D) | ||||
| 1 mef(A), ere(A), mph(A), mph(B) | ||||
| 1 mef(A), ere(A), mph(A) | ||||
| 1 mef(A), ere(B), mph(D) | ||||
| 1 mef(A), ere(A), ere(B), mph(A), mph(D), erm(B) | ||||
| Morganella | 1 | 1 mef(A) | ||
| Pantoeae | 1 | 1 mef(A), ere(A), mph(A), mph(D), erm(B) | ||
| Providencia | 1 | 1 mef(A) | ||
| Pseudomonas | 31 | 14 mef(A) | 19 | 3 erm(B) |
| 3 ere(A) | 4 mef(A) | |||
| 2 mef(A), ere(A) | 1 mph(D) | |||
| 1 mef(A), erm(B) | 2 mph(A) | |||
| 1 mef(A), mph(B) | 1 mef(A), mph(B) | |||
| 1 mph(A), mph(B), mph(D) | 3 mef(A), erm(B) | |||
| 2 erm(B), ere(A), ere(B) | 1 mph(A), mph(B) | |||
| 1 mef(A), ere(A), mph(D) | 1 mef(A), ere(A), ere(B), mph(A), mph(B), mph(D) | |||
| 1 mef(A), ere(A), mph(D), erm(B) | 2 mef(A), ere(B), mph(A), mph(B), mph(D) | |||
| 1 mef(A), ere(B), mph(A), mph(D) | 1 mef(A), ere(B), mph(A), mph(B), mph(D), erm(B) | |||
| 2 ere(A), ere(B), mph(B), mph(D) | ||||
| 1 ere(A), mph(A), mph(B), mph(D) | ||||
| 1 ere(A), mph(A), mph(B), erm(B) | ||||
| Proteus | 13 | 1 mef(A) | ||
| 1 erm(B) | ||||
| 7 mph(D) | ||||
| 1 mph(A), mph(D) | ||||
| 2 mph(A), mph(B) | ||||
| 1 mef(A), erm(B) | ||||
| Ralstonia | 1 | 1 mef(A), mph(A) | ||
| Serratia | 3 | 1 mef(A), erm(B) | ||
| 1 mef(A), | ||||
| 1 mef(A), ere(A), ere(B) | ||||
| Stenotrophomonas | 3 | 1 mph(A) | ||
| 1 mef(A), ere(A), mph(A) | ||||
| 1 mef(A), mph(A), mph(D) | ||||
| Total | 79 | 59 | ||
Gene(s) carried on the isolates. The number is the number of isolates carrying the gene(s) shown.
Media.
Luria-Bertani (LB) agar (Difco Laboratories, Division of Becton Dickinson & Co., Sparks, Md.) unsupplemented or supplemented with ≥25 μg of erythromycin per ml was used to grow the clinical isolates. The E. coli strain HB101 and Enterococcus faecalis JH2-2 recipients were grown without antibiotic. All isolates were incubated at 36.5°C.
Agar dilution susceptibility testing.
Erythromycin MICs were determined by the agar dilution method as described by the National Committee of Clinical Laboratory Standards (13) for all isolates. MIC breakpoints are not available for macrolides for most gram-negative species. We did not attempt to distinguish resistance from susceptibility and used MICs.
Detection of acquired genes.
Isolates were initially screened by DNA-DNA hybridization of whole-cell dot blots and/or DNA dot blots as previously described (2-4, 6-9, 18). PCR was performed on selected isolates to confirm the presence of the various macrolide genes (3, 4, 8). Oligonucleotide probes for PCR and DNA-DNA hybridization are listed in Table 2. Radiolabeled probes were used as previously described (8).
TABLE 2.
Primers used in this study
| Gene | Primer | Sequence (5′→3′) | Reference(s) |
|---|---|---|---|
| mef(A) | mefF | TGT GCA TAT TTC TAT TAC G | This study |
| mefR | CCA ATT GGC ATA GCA AG | This study | |
| mefI | GCT GTG CAA TAA TGG GGC | This study | |
| ere(A) | ereAF | GCC GGT GCT CAT GAA CTT GAG | This study |
| ereAR | CGA CTC TAT TCG ATC AGAGGC | This study | |
| ereAI | TCA CTG GCT AGA GCT AGT CTT | This study | |
| ere(B) | ereBF | GCC TTG AAG CTA TGG CTC C | This study |
| ereBR | GGC CCA TTG GTA GGC AAC | This study | |
| ereBI | TTG GAG ATA CCC GAG TTG TAG | This study | |
| erm(B) | ermBF | GAA AAG GTA CTC AAC CAA ATA | 7, 8 |
| ermBR | AGT AAC GGT ACT TAA ATT GTT TAC | 7, 8 | |
| ermBI | AGC CAT GCG TCT GAC ATC TAT | 7, 8 | |
| mph(A) | mphAF | GTG AGG AGG AGC TTC GCG AG | This study |
| mphAR | TGC CGC AGG ACT CGG AGG TC | This study | |
| mphAI | GAT ACC TCC CAA CTG TAC GCA | This study | |
| mph(B) | mphBF | TTA AAC AAG TAA TCG AGA TAG C | This study |
| mphBR | CCT TGT ACT TCC AAT GCT T G | This study | |
| mphBI | GCG TAT GGA TGC AGT AAG AGC | This study | |
| mph(D) | mphDF | GTG TTC TTG CTT GGC TCG TAA | This study |
| mphDR | ATC TGG TCG GGG TTG ATA A | This study | |
| mphDI | GCG GAT CTC CTC CCA GAG TG | This study |
Mating.
Fifteen donors and one transconjugant donor were selected to be mated with erythromycin-susceptible E. faecalis JH2-2 and/or erythromycin-susceptible E. coli HB101. The erythromycin MIC for E. faecalis JH2-2 was <0.5 μg/ml, and the erythromycin MIC for E. coli HB101 was 16 μg/ml. Matings were performed on agar plates, and transconjugants were identified as previously described because they expressed erythromycin resistance (6-8, 18). The transconjugants were selected with 5 μg of erythromycin per ml for E. faecalis JH2-2 and 25 to 50 μg of erythromycin per ml for E. coli HB101. The presence of acquired macrolide resistance genes in 5 to 10 of the transconjugants from each mating pair was determined by DNA-DNA hybridization and PCR for each of the macrolide resistance genes as previously described (2-4, 7, 8, 18). Positive controls were used in each assay.
DNA-DNA hybridization.
DNA-DNA hybridization of Southern blots, whole-cell bacterial dot blots, whole-cell DNA dot blots, and/or PCR dot blots was performed as previously described. DNA was hybridized with the appropriate 32P-labeled probe as previously described (2).
PCR.
Seven different PCRs were performed to detect each of the seven genes separately. PCR assays for the erm(B) and mef(A) genes were conducted as previously described (2-4, 7, 8, 17). For the other five genes, new assays were developed using 2 U of Taq polymerase (Perkin-Elmer Cetus, Norwalk, Conn.), 200 μM (each) deoxynucleotide triphosphates, 1× PCR buffer (1.5 mM MgCl2), 100 ng of each primer, and ≥200 ng of whole DNA as the template. To detect the ere(A) and ere(B) genes, PCR was performed as follows: an initial denaturation step (96°C for 3 min); followed by 35 cycles of PCR, with 1 cycle consisting of denaturation (96°C for 30 s), annealing (56°C for 1 min), and elongation (72°C for 2 min). For the mph(A), mph(B), and mph(D) genes, the initial denaturation, denaturation, and elongation times and temperatures were the same, but annealing was done at 57°C for 1 min. For all assays, the final step was 72°C for 10 min, followed by incubation at 4°C. Positive and negative controls were included in each run. The PCR products were visualized on a 1.5% agarose gel as previously described. Southern blots and/or dot blots of the PCR products were hybridized using an internal 32P-labeled probe to verify the PCR products as previously described (2-4, 7, 8, 18).
RESULTS
Macrolide susceptibility and detection of the seven macrolide resistance genes.
Of the 13 genera, four were found in both the oral and urine samples (Table 1). The erythromycin MICs for the isolates ranged from 2 to >500 μg/ml; the erythromycin MICs for most of the isolates were ≥64 μg/ml. The MICs for isolates within a species varied, with Pseudomonas (MICs, 4 to >500 μg/ml) and Klebsiella (MICs, 16 to >500 μg/ml) having the widest range (data not shown). Of the 176 isolates, 138 (78%) isolates, including 79 (91%) isolates from the oral cavity and 59 (66%) isolates from urine samples, hybridized with one or more of the seven gene probes used (Table 1). Forty-four (56%) oral and 40 (68%) urine isolates carried one of the seven macrolide genes, while 35 (44%) oral and 19 (32%) urine isolates carried two or more of the seven macrolide genes examined. The mef(A) gene was found in 74 (54%) of the isolates, including 51 (65%) of the oral isolates and 22 (37%) of the urine isolates and was the most common gene found in this population. The mph(A) gene was the second most commonly found gene; it was found in 45 (33%) isolates, which included 18 (23%) oral isolates and 27 (46%) urine isolates.
The other five genes, erm(B), ere(A), ere(B), mph(B), and mph(D), were found in 14 to 22% of the total isolates (Table 1). The distribution of the erm(B) gene was similar in the isolates from the oral cavity (22%) and urine samples (19%), while the ere(A) gene was found more commonly in the oral isolates (38%) than in the urine isolates (1%). In contrast, the mph(B), mph(D), and/or ere(B) genes were more common in the urine isolates than in the oral isolates (Table 1).
The mef(A) gene has previously been found in A. junii (7), but this is the first report of this gene in the other 12 genera examined (Table 3). The mph(A) gene was identified in eight new genera, the mph(D) gene was identified in seven new genera, the erm(B) gene was found in six new genera, the ere(B) gene was found in five new genera, and the ere(A) gene and the mph(B) genes were found in three new genera (Table 3).
TABLE 3.
Macrolide resistance genes found in new genera
| Type of gene | Gene | No. of new genera | New genera |
|---|---|---|---|
| rRNA methylase | erm(B) | 6 | Acinetobacter, Enterobacter, Pantoeae, Proteus, Pseudomonas, Serratia |
| Efflux (major facilitator) | mef(A) | 12 | Citrobacter, Enterobacter, Escherichia, Klebsiella, Morganella, Pantoeae, Proteus, Providencia, Pseudomonas, Ralstonia, Serratia, Stenotrophomonas |
| Esterase | ere(A) | 3 | Pantoeae, Pseudomonas, Stenotrophomonas |
| ere(B) | 5 | Acinetobacter, Citrobacter, Enterobacter, Pseudomonas, Stenotrophomonas | |
| Phosphorylase | mph(A) | 8 | Citrobacter, Enterobacter, Klebsiella, Pantoeae, Pseudomonas, Proteus, Serratia, Stenotrophomonas |
| mph(B) | 3 | Enterobacter, Pseudomonas, Proteus | |
| mph(D) | 7 | Escherichia, Klebsiella, Pantoeae, Proteus, Pseudomonas, Serratia, Stenotrophomonas |
Eight isolates (10%) from the oral cavity (two Acinetobacter isolates, three Enterobacter isolates, two Klebsiella isolates, and one Pseudomonas isolate) were negative for all seven macrolide resistance genes. Thirty (34%) isolates from urine samples (one Acinetobacter isolate, eight Escherichia isolates, five Klebsiella isolates, four Morganella isolates, four Proteus isolates, and eight Pseudomonas isolates) were negative for all seven macrolide resistance genes. These isolates, representing seven genera, did not hybridize with any of the gene probes used [ere(A), ere(B), mph(A), mph(B), mph(D), mef(A), or erm(B)]. The erythromycin MICs for oral isolates ranged from 2 to 256 μg/ml; the erythromycin MICs for five isolates were ≥64 μg/ml. The erythromycin MICs for urine isolates ranged from 2 to >500 μg/ml; the erythromycin MICs for 24 isolates were ≥64 μg/ml.
Characterization of specific isolates.
Fifteen isolates from 10 genera were chosen for further characterization (Table 4). There was no apparent correlation between MIC and the number or type of macrolide resistance gene(s) carried (Table 4). Mating experiments were done using the 15 isolates as donors and E. faecalis and/or E. coli as the recipient(s) (Table 4). A. junii 329 has been included in Table 4 for comparison (7). The presence of acquired macrolide resistance genes were determined in 5 to 10 transconjugants, from each mating pair. The mef(A) gene was transferred to the E. faecalis recipient at frequencies from 10−5 to 10−9/recipient. The mef(A) gene also transferred to the E. coli recipient at frequencies from 10−5 to 10−9/recipient. One transconjugant, E. coli 11 donor (HB101 transconjugant) carrying the mef(A) gene, was used as a donor and mated with the E. faecalis recipient. Transconjugants from this mating were detected at low frequencies, indicating that the HB101 transconjugant maintained its ability to retransfer the mef(A) gene (Table 4).
TABLE 4.
Transfer of macrolide resistance genes
| Donor | ERYa MIC (μg/ml) | Gene(s) carried | Recipient | Frequencyb | Gene(s) transferred |
|---|---|---|---|---|---|
| Oral isolates | |||||
| Acinetobacter junii 329c | 2 | mef(A) | E. faecalis | 9.5 × 10−6 | mef(A) |
| Citrobacter freundii 16 | 256 | mef(A), erm(B) | E. faecalis | 1.6 × 10−8 | mef(A), erm(B)d |
| Enterobacter cloacae 240 | 256 | mef(A) | E. faecalis | 5.3 × 10−9 | mef(A) |
| Escherichia coli 11 | 64 | mef(A) | E. faecalis | 6.3 × 10−7 | mef(A) |
| mef(A) | E. coli | 3.9 × 10−9 | mef(A) | ||
| E. coli HB101 transconjugante | mef(A) | E. faecalis | 2.0 × 10−10 | mef(A) | |
| Klebsiella sp. 7 | 128 | mef(A) | E. faecalis | 3.0 × 10−7 | mef(A) |
| Klebsiella sp. 8 | 256 | mef(A), erm(B) | E. faecalis | 2.9 × 10−8 | mef(A), erm(B)f |
| mef(A), erm(B) | E. coli | 4.5 × 10−8 | mef(A), erm(B) | ||
| Klebsiella sp. 9 | 256 | mef(A), ere(B), mph(D) | E. faecalis | 5.1 × 10−6 | mef(A) |
| Klebsiella sp. 106 | 128 | mef(A) | E. faecalis | 2.4 × 10−8 | mef(A) |
| K. oxytoca 561 | 128 | mef(A), erm(B), ere(A), ere(B), mph(A), mph(B) | E. faecalis | 2.0 × 10−7 | mef(A), erm(B)g |
| E. coli | 3.8 × 10−5 | mef(A), erm(B), ere(A), ere(B), mph(A), mph(B) | |||
| Pantoeae agglomerans 323 | 256 | mef(A), ere(A), mph(A), mph(D), erm(B) | E. faecalis | 1.3 × 10−7 | mef(A), erm(B)h |
| E. coli | 4.7 × 10−9 | mef(A), erm(B), ere(A), mph(A), mph(D) | |||
| Pseudomonas sp. 203 | >500 | mef(A), ere(A), ere(B), mph(A), mph(D), erm(B) | E. faecalis | 4.0 × 10−8 | mef(A), erm(B)i |
| E. coli | 2.2 × 10−7 | mef(A), ere(A), ere(B), mph(A), mph(D), erm(B) | |||
| Pseudomonas sp. 333 | 128 | mef(A), erm(B) | E. faecalis | 1.0 × 10−5 | mef(A), erm(B)j |
| mef(A), erm(B) | E. coli | 4.0 × 10−9 | mef(A), erm(B) | ||
| Pseudomonas putida 366 | >500 | ere(A), mph(A), mph(D), mef(A) | E. faecalis | 2.5 × 10−9 | mef(A) |
| Serratia liquefaciens 136 | 64 | mef(A), ere(A), ere(B) | E. faecalis | 7.2 × 10−8 | mef(A) |
| Urine isolates | |||||
| Morganella morganii 236 | 500 | mef(A) | E. faecalis | 6.7 × 10−8 | mef(A) |
| Proteus sp. 21 | >500 | mef(A), mph(A) | E. faecalis | 1.0 × 10−9 | mef(A) |
ERY, erythromycin.
Number of transconjugants per recipient.
Data from reference 7.
33% of E. faecalis transconjugants carried both the mef(A) and erm(B) genes, all carried mef(A).
Transconjugant from E. coli 11 mated with E. faecalis JH2-2 and selected on erythromycin.
50% of E. faecalis transconjugants carried both the mef(A) and erm(B) genes; all carried mef(A).
No erm(B) gene was detected in the E. faecalis transconjugants with the donor K. oxytoca 561.
100% of the E. faecalis transconjugants carried both the mef(A) and erm(B) genes.
50% of the E. faecalis transconjugants carried both the mef(A) and erm(B) genes; all carried mef(A).
25% of the E. faecalis transconjugants carried both the mef(A) and erm(B) genes; all carried mef(A).
In six of the gram-negative donors, both the erm(B) and mef(A) genes were present, and the overall transfer of macrolide resistance genes varied from 10−5 to 10−9/recipient (Table 4). In five of the matings, the mef(A) gene was transferred and detected in the E. faecalis transconjugants, while the erm(B) gene was detected in 25 to 100% of the E. faecalis transconjugants. With the Klebsiella oxytoca 561 donor, the erm(B) gene was not detected in the E. faecalis transconjugants even after multiple mating experiments (Table 4). The esterase and phosphorylase genes were not detected in the E. faecalis transconjugants from any of the matings.
Seven of the donors carried macrolide resistance genes in addition to the erm(B) and/or mef(A) genes, with two carrying a total of six different macrolide resistance genes and one carrying a total of five different macrolide resistance genes. These donors were mated with the E. coli recipient, and the frequency of transfer ranged from 10−5 to 10−9 per recipient (Table 4). All the macrolide resistance genes carried in the donors were identified in the E. coli transconjugants tested.
DISCUSSION
In this study, 78% of the randomly selected, commensal, gram-negative bacteria from a healthy population with low exposure to antibiotics were found to carry one or more of the seven macrolide resistance genes tested. Unfortunately, there is little previously published data from other countries available for comparison of data, because most studies have focused on macrolide resistance in pathogenic bacteria from clinical settings and/or diseased hosts (1, 3, 4, 6, 7, 9, 11, 12, 14-18, 22). However, we can compare these results with those of an earlier study by Luna et al. (8) on 615 randomly selected, commensal, gram-positive isolates from the same group of Lisbon children as the current study over the same time period. In that study, 222 (36%) isolates carried one or more of four different rRNA methylases and mef(A) efflux gene, which is less than 50% of the rate we found in the current gram-negative study, indicating that in this Lisbon population, acquisition of macrolide resistance genes is more prevalent in commensal, gram-negative isolates than in commensal, gram-positive isolates. The mef(A) and erm(B) genes were examined in both groups of bacteria, and the percentage in each population was compared. For the mef(A) and erm(B) genes, 9 and 60% were found in the gram-positive bacteria, respectively, while 54 and 19% were found in the gram-negative isolates, respectively.
The distribution of the seven macrolide resistance genes in the gram-negative bacteria varied by genus and location. However, with only four genera found in both the oral cavity and urine samples, it was not surprising that these seven different genes were found in different percentages of the oral and urine isolates. The mef(A) gene was identified in at least one isolate from each of the 13 genera examined, while the mph(A) gene was the second most commonly found gene overall (Table 1). Each of the seven macrolide resistance genes was identified in a number of new genera, suggesting that the host range of these genes is larger than the literature has suggested (Table 3).
We had 8 isolates from the oral cavity and 30 isolates from urine samples that did not carry any of the known genes. The erythromycin MICs for the majority of the isolates were ≥64 μg/ml. Susceptibility to clindamycin, determined by agar dilution, was examined for these strains. The clindamycin MICs for most of the isolates was the same or plus or minus up to eightfold of their erythromycin MICs (data not shown). Other erm genes have been identified in gram-negative Actinobacillus spp. (21), Neisseria spp. (18), and a variety of anaerobic gram-negative genera (2, 19-21). Thus, it is possible that some of the other known rRNA methylase genes are present in the 38 gram-negative strains found in this study. We are currently testing these isolates for the presence of the erm(A), erm(C), erm(F), erm(G), and erm(Q) genes, which are the most widespread genes found other than the erm(B) gene. Three interesting isolates were identified. For one Acinetobacter isolate, the erythromycin MIC was 16 μg/ml, and the clindamycin MIC was >500 μg/ml. For one E. coli isolate, the erythromycin MIC was 64 μg/ml, and the clindamycin MIC was >500 μg/ml. For one Pseudomonas isolate, the erythromycin MIC was 8 μg/ml, and the clindamycin MIC was 500 μg/ml. These three isolates may carry a lincosamide resistance gene. These isolates and all of the nonreactive isolates will be examined for the presence of the lnu(A) and lnu(B) transferase genes, which have previously been found only in Staphylococcus and Enterococcus, respectively (19).
The mef(A) gene transferred to both the E. coli and E. faecalis recipients, with all transconjugants receiving and maintaining this gene. In contrast, the erm(B) gene, a gene of gram-positive bacterial origin, transferred and was maintained in some but not all E. faecalis transconjugants examined (Table 4). The variability in transfer and maintenance appears to be donor specific, since the same recipient was used with each mating. The reason for this variability is currently being examined but could be due to the location of the erm(B) gene (plasmid versus conjugative transposon). However, all seven genes were associated with mobile elements in most of the donors examined and were able to conjugally transfer these genes to unrelated recipients in the laboratory. It is also likely that these strains would be able to conjugally transfer these genes to unrelated bacteria in their human hosts as well.
Few studies have examined erythromycin MICs for members of the family Enterobacteriaceae or for Pseudomonas species. Both groups have been thought to be innately nonsusceptible to erythromycin due to innate multidrug-resistant transporters which confer resistance to 14-membered macrolides (16). Clearly, more work needs to be done to determine whether the data from this group of bacteria can be generalized to bacteria from other geographic locations, isolates collected during different time periods, and populations of bacteria from humans, animals, and the environment.
Acknowledgments
This study was supported in part by grant U01 DE-1189 and contract N01 DE-72623 from the National Institute of Dental and Craniofacial Research of the National Institutes of Health. K. K. Ojo was supported by grant U24 AI-50139 from the National Institute of Allergy and Infectious Diseases of the National Institutes of Health.
REFERENCES
- 1.Arthur, M., C. Molinas, C. Mabilat, and P. Courvalin. 1990. Detection of erythromycin resistance by the polymerase chain reaction using primers in conserved regions of erm rRNA methylase genes. Antimicrob. Agents Chemother. 34:2024-2026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chung, W. O., C. Werckenthin, S. Schwarz, and M. C. Roberts. 1999. Host range of the ermF rRNA methylase gene in human and animal bacteria. J. Antimicrob. Chemother. 43:5-14. [DOI] [PubMed] [Google Scholar]
- 3.Cousin, S. L., Jr., W. L. Whittington, and M. C. Roberts. 2003. Acquired macrolide resistance genes and the 1 bp deletion in the mtr(R) promoter in Neisseria gonorrhoeae. J. Antimicrob. Chemother. 51:131-133. [DOI] [PubMed] [Google Scholar]
- 4.Cousin, S. L., Jr., W. L. Whittington, and M. C. Roberts. 2003. Acquired macrolide resistance genes in pathogenic Neisseria spp. isolated between 1940 and 1987. Antimicrob. Agents Chemother. 47:3877-3880. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Kim, Y.-H., C.-J. Cha, and C. E. Cerniglia. 2002. Purification and characterization of an erythromycin esterase from an erythromycin-resistant Pseudomonas sp. FEMS Microbiol. Lett. 210:239-244. [DOI] [PubMed] [Google Scholar]
- 6.Luna, V. A., P. Coates, J. H. Cove, A. E. Eady, T. T. H. Nguyen, and M. C. Roberts. 1999. A variety of gram-positive bacteria carry mobile mef genes. J. Antimicrob. Chemother. 44:19-25. [DOI] [PubMed] [Google Scholar]
- 7.Luna, V. A., S. Cousin, Jr., W. L. H. Whittington, and M. C. Roberts. 2000. Identification of the conjugative mef gene in clinical Acinetobacter junii and Neisseria gonorrhoeae isolates. Antimicrob. Agents Chemother. 44:2503-2506. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Luna, V. A., M. Heiken, K. Judge, C. Uelp, N. Van Kirk, H. Luis, M. Bernardo, J. Leitao, and M. C. Roberts. 2002. Distribution of the mef(A) gene in gram-positive bacteria from healthy Portuguese children. Antimicrob. Agents Chemother. 46:2513-2517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Luna, V. A., and M. C. Roberts. 1998. The presence of the tetO gene in a variety of tetracycline resistant Streptococcus pneumoniae serotypes from Washington State. J. Antimicrob. Chemother. 42:613-619. [DOI] [PubMed] [Google Scholar]
- 10.Murray, P. R., E. J. Baron, J. H. Jorgensen, M. A. Pfaller, and R. H. Yolken. 2003. Manual of clinical microbiology, 8th ed. ASM Press, Washington, D.C.
- 11.Nakamura, A., I. Miyakozawa, K. Nakazawa, K. O'Hara, and T. Sawai. 2000. Detection and characterization of a macrolide 2′-phosphotransferase from a Pseudomonas aeruginosa clinical isolate. Antimicrob. Agents Chemother. 44:3241-3242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Nakamura, A., K. Nakazawa, I. Miyakozawa, S. Mizukoshi, K. Tsurubuchi, M. Nakagawa, K. O'Hara, and T. Sawai. 2000. Macrolide esterase-producing Escherichia coli clinically isolated in Japan. J. Antibiot. 53:516-524. [DOI] [PubMed] [Google Scholar]
- 13.National Committee for Clinical Laboratory Standards. 1995. Approved standard. Methods for dilution antimicrobial susceptibility tests for bacteria that grow aerobically. NCCLS document M7-A3. National Committee for Clinical Laboratory Standards, Villanova, Pa.
- 14.Noguchi, N., J. Katayama, and K. O'Hara. 1996. Cloning and nucleotide sequence of the mphB gene for macrolide 2′-phosphotransferase II in Escherichia coli. FEMS Microbiol. Lett. 144:197-202. [DOI] [PubMed] [Google Scholar]
- 15.Noguchi, N., A. Emura, H. Matsuyama, K. O'Hara, M. Sasatsu, and M. Kono. 1995. Nucleotide sequence and characterization of erythromycin resistance determinant that encodes macrolide 2′-phosphotransferase I in Escherichia coli. Antimicrob. Agents Chemother. 39:2359-2363. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Plante, I., D. Centron, and P. H. Roy. 2003. An integron cassette encoding erythromycin esterase, ere(A), from Providencia stuartii. J. Antimicrob. Chemother. 51:787-790. [DOI] [PubMed] [Google Scholar]
- 17.Putman, M., H. W. van Veen, and W. N. Konings. 2000. Molecular properties of bacterial multidrug transporters. Antimicrob. Agents Chemother. 64:672-693. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Roberts, M. C., W. O. Chung, D. Roe, M. Xia, C. Marquez, G. Borthagaray, W. L. Whittington, and K. K. Holmes. 1999. Erythromycin-resistant Neisseria gonorrhoeae and oral commensal Neisseria spp. carry known rRNA methylase genes. Antimicrob. Agents Chemother. 43:1367-1372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Roberts, M. C., J. Sutcliffe, P. Courvalin, L. B. Jensen, J. Rood, and H. Seppala. 1999. Nomenclature for macrolide and macrolide-lincosamide-streptogramin B antibiotic resistance determinants. Antimicrob. Agents Chemother. 43:2823-2830. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Roe, D. E., A. Weinberg, and M. C. Roberts. 1995. Characterization of erythromycin resistance in Campylobacter (Wolinella) rectus. Clin. Infect. Dis. 20:S370-S371. [DOI] [PubMed] [Google Scholar]
- 21.Roe, D. E., A. Weinberg, and M. C. Roberts. 1996. Mobile rRNA methylase genes in Actinobacillus actinomycetemcomitans. J. Antimicrob. Chemother. 37:457-464. [DOI] [PubMed] [Google Scholar]
- 22.Thungapathra, M., Amita, K. K. Sinha, S. R. Chaudhuri, P. Garg, T. Ramamurthy, G. B. Nair, and A. Ghosh. 2002. Occurrence of antibiotic resistance gene cassettes aac(6′)-Ib, dfrA5, dfrA12, and ereA2 in class I integrons in non-O1, non-O139 Vibrio cholerae strains in India. Antimicrob. Agents Chemother. 46:2948-2955. [DOI] [PMC free article] [PubMed] [Google Scholar]