Abstract
Sulfonamide resistance remains prevalent among clinical isolates of Escherichia coli in the United Kingdom, despite a dramatic (>97%) national decline in the rate of prescription of sulfonamides in the 1990s. To investigate potential mechanisms accounting for this persistence, we characterized plasmids carrying sul2, the most prevalent determinant of sulfonamide resistance. Among 33 conjugative and 5 nonconjugative plasmids carrying sul2, resistance to other antimicrobial agents was common, but the spectrum of resistance profiles was diverse: 82%, 74%, and 45% carried resistance to ampicillin, streptomycin, and trimethoprim, respectively. Resistance to mercury was carried by 33% of the plasmids, but none conferred significant resistance to silver or to any of three disinfectants tested. The potential virulence genes iutA (aerobactin system) and traT (serum survival) were carried by 21% and 36% of the plasmids, respectively. The 33 conjugative plasmids belonged to five different incompatibility groups, FIB, B/O, I1, K/B, and P (42%, 33%, 9%, 3% and 3%, respectively), with 3 plasmids being unassigned, and to 19 similarity groups on the basis of their restriction profiles. The sequences flanking sul2 were diverse and suggested more than one mechanism of genetic mobility. The five nonconjugative plasmids were all related to p9123 (pBP1), which was previously found to confer a fitness advantage on its host. We propose that the persistence of sul2, despite the reduced rate of prescription of sulfonamides, is due to a combination of coselection by antibiotics still in common use, a lack of a selective disadvantage in sul2 carriage, and the genetic mobility of sul2.
Co-trimoxazole, a combination of sulfamethoxazole and trimethoprim, was introduced in the United Kingdom in 1969 and was widely used, becoming in particular a mainstay of treatment for urinary tract infections. In 1979, trimethoprim alone became available for the treatment of urinary tract infections, and in 1995, the licensed indications for co-trimoxazole were restricted in favor of the use of trimethoprim alone, primarily owing to concern about the rare side effects of sulfonamides (1). Despite the massive reduction in the rate of sulfonamide use that accompanied the switch in prescribing from co-trimoxazole to trimethoprim, resistance to sulfonamides has persisted at high rates among clinical isolates of Escherichia coli (2, 6).
Three acquired genes imparting sulfonamide resistance have been described in E. coli; of these, only two (sul1 and sul2) are prevalent in human isolates, including those from east London, United Kingdom (2, 6). The sul2 gene has remained the more prevalent over time and is found both alone and in combination with sul1. While sul1 is typically associated with class 1 integrons (26), sul2 is generally not considered part of a distinct genetic element, although it is frequently found adjacent to the streptomycin resistance gene pair strAB (8, 23, 25). We have previously demonstrated that clinical isolates of E. coli resistant to sulfonamides frequently carried genes for resistance to other antimicrobial agents but had not established whether the genes for resistance to multiple compounds were carried on plasmids that were the same as or separate from those that carried the sul determinants (2, 6). We also determined that a small nonconjugative plasmid carrying sul2, p9123, conferred a fitness advantage on its host, whereas three other plasmids carrying sul2 had a fitness cost (5). In the study described here, we have further characterized plasmids carrying sul2 among clinical isolates of E. coli. We investigated the genetic linkage of sul2 with resistance to other antimicrobials and xenobiotics and assessed the genetic diversity of these plasmids as a marker for the mobility of the sul2 gene.
(The results of this study were presented in part at the 45th Interscience Conference on Antimicrobial Agents and Chemotherapy, Washington, DC, 2005.)
MATERIALS AND METHODS
Bacterial isolates and antimicrobial susceptibility testing.
This study used a collection of E. coli isolates collected at the Royal London Hospital in 1999 and described previously (6). Antibiotic susceptibility was determined by the disc diffusion assay, in accordance with the guidelines of the British Society for Antimicrobial Chemotherapy (3). Antibiotic discs were obtained from Oxoid, Basingstoke, United Kingdom. The MICs of heavy metals (mercury chloride and silver nitrate; Sigma, Poole, United Kingdom) and detergents (triclosan, chlorhexidine, and cetrimide; Sigma) were determined by broth microdilution, according to the guidelines of EUCAST for antimicrobial susceptibility testing (7).
Plasmid characterization and manipulation.
Plasmid DNA was extracted by the method of Kado and Liu (12) (for visualization only) or the alkaline lysis method (24). SacI-digested plasmid DNA was separated with a CHEF DRII apparatus (Bio-Rad, Hemel Hempstead, United Kingdom) at 6 V/cm and with ramping from 0.1 s to 1.0 s over 8 h. Conjugation was by the plate-mating protocol of Livermore and Williams (16) with E. coli J62-1 (nalidixic acid resistant, β-galactosidase negative) as the recipient. Nonconjugative plasmids were introduced into E. coli J62-1 by electrotransformation (24). The transconjugants and transformants were selected on Iso-Sensitest agar (Oxoid) with sulfamethoxazole (500 μg/ml; Sigma) and nalidixic acid (100 μg/ml; Sigma). For the cloning of sul2, plasmid DNA was digested with BamHI, ligated into pGEM-3Z (Promega, Southampton, United Kingdom), transformed into E. coli JM109 competent cells (Promega), and selected on Iso-Sensitest agar (Oxoid) with sulfamethoxazole (500 μg/ml; Sigma) and ampicillin (100 μg/ml; Sigma). Southern transfer was performed by standard techniques (24). Probes were labeled and detected with the PCR DIG probe synthesis kit and the DIG DNA labeling and detection kit (Roche Diagnostics, Lewes, United Kingdom), respectively.
PCR amplification and sequencing.
PCR was performed with boiled cell lysates by using the ReddyMix master mixture (Abgene, Epsom, United Kingdom). Sequencing was conducted at the Genome Centre of Barts and The London Hospital. Plasmids were assigned to incompatibility groups by using the multiplex PCR described by Johnson and colleagues (11), and the assignments were confirmed by using single primer pairs (4). Virulence genes were detected as described by Johnson and Stell (10). The other primers used for PCR and sequencing are listed here: sulfonamide resistance gene primers sul1-F (5′-CCGATATTGCTGAGGCGG-3′; GenBank accession no. U42226; nucleotides 435 to 453), sul1-R (5′-CCAACGCCGACTTCAGCT-3′; GenBank accession no. U42226; nucleotides 684 to 702), sul2-F (5′-TCGTCAACATAACCTCGGACAG-3′ [6]), sul2-R (5′-GTTGCGTTTGATACCGGCAC-3′ [6]), sul3-F (5′-GAGCAAGATTTTTGGAATCG-3′ [19]), and sul3-R (5′-CTAACCTAGGGCTTTGGA-3′ [19]); streptomycin resistance gene primers strA-F (5′-CAACTGGCAGGAGGAACA-3′ [15]), strA-R (5′-CGCAGATAGAAGGCAAGG-3′ [15]), strB-F (5′-TTCTCATTGCGGACACCT-3′ [15]), strB-R (5′-GGCATTGCTCATCATTTG-3′ [15]), aadA1-F (5′-TATTGATTTGCTGGTTACGG-3′ [15]), and aadA1-R (5′-CGCTATGTTCTCTTGCTTTT-3′ [15]); upstream sul2 primers sul2 out (5′-CCTGCGCAATGGCTGCGTCTGG-3′; GenBank accession no. AY360321; nucleotides 102 to 124), sul2 up Type 1 (5′-TCTTCCGAACATGACACCAGTC-3′; GenBank accession no. AY360321; nucleotides 5980 to 6002), sul2 up Type 2 (5′-ACAGCTCCATAGGCCGCTTTCC-3′; GenBank accession no. M28829; nucleotides 7712 to 7734), sul2 up Type 2A (5′-GCCATTTCAGCGGCTGTGATGC-3′; GenBank accession no. AB109805; nucleotides 3555 to 3577); sul2 up Type 2B (5′-AGTGGTTCGACCATGCGCAAGC-3′; GenBank accession no. M28829; nucleotides 7435 to 7457), sul2 up Type 2C (5′-AAGTAGTAACCCACCTCCGCAG-3′; GenBank accession no. AY055428; nucleotides 21478 to 21500), sul2 up Type 2D (5′-CTGATGGTGCCCGAAAACTATA-3′; GenBank accession no. NC_007365; nucleotides 59902 to 59924), sul2 up Type 3 (5′-TGGTAAATATCGCCATGGGCAG-3′; GenBank accession no. AJ319822; nucleotides 1532 to 1554), repA-F (5′-GCTGACGGCCTGCTGATCCAGC-3′; GenBank accession no. M28829; nucleotides 6195 to 6217), repC-F (5′-ACGACCCGGCGCACTGTCTG-3′; GenBank accession no. M28829; nucleotides 6758 to 6778), repC-R (5′-TGCTGATGCGCACATGCTGG-3′; GenBank accession no. M28829; nucleotides 7281 to 7301), and trbC-out (5′-AGGTGGAAAACAAGAGAGTCG-3′; GenBank accession no. NC_007365; nucleotides 59623 to 59644); and miscellaneous primers intI1-F (5′-GCGTGATTGTATCTCACT-3′ [2]), intI1-R (5′-GACGCTCCTGTTGCTTCT-3′ [2]), blaTEM-F (5′-ATGATGATTCAACATTTCCG-3′ [15]), and blaTEM-R (5′-CCAATGCTTAATCAGTGACG-3′ [15]).
RESULTS AND DISCUSSION
Genetic linkage to other resistance determinants.
Sulfonamide resistance plasmids bearing sul2 were transferred into a susceptible (plasmid-free) host, E. coli J62-1, in order to explore the cocarriage of genes for resistance to other compounds in a standard genetic background. One hundred thirteen independent sul2-containing E. coli isolates were investigated, of which 32 (28%) also carried sul1 (6). Conjugative plasmids were successfully transferred from 41 (36%) isolates, but in 8 of these the plasmid carrying sul2 was consistently cotransferred with another plasmid so they were not investigated further. Among the 33 conjugative plasmids that transferred independently of other plasmids, 8 (24%) carried both sul1 and sul2 and 25 (76%) carried sul2 only; none carried sul3. The transfer of plasmids by transformation was attempted with DNA extracted from the isolates that failed to conjugate and was successful with the DNA from five isolates; all five transformed (nonconjugative) plasmids carried sul2 but not sul1 or sul3.
All 33 conjugative and 5 nonconjugative plasmids conferred at least one additional resistance trait besides sulfonamide resistance to the E. coli recipient, but the spectrum of resistance patterns was diverse (Table 1). The most frequently occurring coresistances were to ampicillin (82%), streptomycin (74%), and trimethoprim (45%). These results suggest that the continued extensive use of ampicillin (and related β-lactam antibiotics) and trimethoprim might coselect for sulfonamide resistance.
TABLE 1.
Properties of sul2-containing plasmids
| Plasmid type and name | Restriction profile | Incompatibility group | Plasmid size (kb) | sul2 fragment size (kb)a | Flanking sequence
|
Resistance phenotypeb
|
Resistance genotype
|
Presence of the following virulence factors
|
Presence of intI1 | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Upstream | Downstream | A | S | W | T | C | K | Hg2+ | TEM | strA/ strB | aad | cvaC | iutA | traT | ||||||
| Plasmids with sul2 only | ||||||||||||||||||||
| p99015 | Z1 | FIB | 118 | 36 | 2B, repCc | strA | R | R | − | R | R | − | R | + | +/+ | − | − | − | − | − |
| p99018 | Z1 | FIB | 118 | 36 | 2B, repC | strA | R | R | − | R | R | − | R | + | +/+ | − | − | − | − | − |
| p99019 | Z2 | FIB | 126 | NDd | 2B, repC | strA | R | R | − | R | R | − | R | + | +/+ | − | − | − | + | − |
| p99026 | Y | ND | 73 | ND | 2A | strA | R | R | − | − | − | − | − | + | +/+ | − | − | − | + | − |
| p99034 | X1 | FIB | 119 | ND | 2B, repC | strA | R | R | R | − | − | − | R | + | +/+ | − | − | + | + | + |
| p99272 | X2 | FIB | 119 | 33 | 2B, repC | strA | R | R | R | R | − | − | R | + | +/+ | − | + | + | + | + |
| p99290 | X2 | FIB | 119 | 33 | 2B, repC | strA | R | R | R | − | − | − | R | + | +/+ | − | − | + | + | + |
| p99339 | X3 | FIB | 137 | 33 | 2B, repC | strA | R | R | R | − | − | − | R | + | +/+ | − | − | + | + | + |
| p99135 | W1 | I1 | 76 | 22.5 | 2B | ND | R | − | − | − | − | − | − | + | −/− | − | − | − | − | − |
| p99136 | W2 | I1 | 100 | 22.5 | 2B | ND | R | − | − | − | − | − | − | + | −/− | − | − | − | − | − |
| p99332 | V | FIB | 107 | 35 | 2B | strA | R | R | − | R | R | − | R | + | +/+ | − | + | + | + | − |
| p99288 | U | FIB | 101 | ND | 2B | ND | R | R | − | − | − | − | − | + | −/− | + | + | − | + | + |
| p99054 | T1 | ND | 45 | 35 | 2B | ND | R | − | − | − | − | − | − | + | −/− | − | − | − | − | − |
| p99082 | T1 | ND | 45 | 35 | 2B | ND | R | − | − | − | − | − | − | + | −/− | − | − | − | − | − |
| p99009 | S1 | B/O | 97 | 5.1 | 2D, trbC | glmM | − | − | R | R | − | − | − | − | −/− | + | − | − | − | − |
| p99051 | S2 | B/O | 75 | 4.5 | 2D, trbC | strA | R | R | − | − | − | − | − | + | +/− | − | − | − | − | − |
| p99057 | S3 | B/O | 83 | 4.5 | 2D, trbC | strA | R | R | − | − | − | − | − | + | +/− | − | − | − | − | − |
| p99163 | S4 | B/O | 80 | 4.5 | 2D, trbC | strA | R | R | − | − | − | − | − | + | +/− | − | − | − | − | − |
| p99164 | S5 | B/O | 77 | 4.5 | 2D, trbC | strA | R | R | − | − | R | − | − | + | +/− | − | − | − | − | − |
| p99372 | S6 | B/O | 81 | 4.5 | 2D, trbC | strA | R | R | − | − | − | − | − | + | +/− | − | − | − | − | − |
| p99171 | R | B/O | 97 | 4.5 | 2D, trbC | strA | R | R | R | R | − | − | − | + | +/− | − | − | − | − | + |
| p99130 | Q | FIB | 72 | 6 | 2A | strA | R | R | R | R | R | − | − | + | +/+ | − | − | + | + | − |
| p99151 | P | FIB | 136 | 4.9 | 2B, repC | strA | R | R | − | R | R | − | R | + | +/+ | − | − | − | + | − |
| p99197 | O | B/O | 110 | 25 | 2B, repC | ND | R | − | R | − | − | R | R | + | −/+ | − | − | − | − | − |
| p99246 | N | K/B | 109 | 8 | 2D, trbC | ND | − | − | R | − | − | − | − | − | −/− | + | − | − | − | − |
| Plasmids with sul1 and sul2 | ||||||||||||||||||||
| p99030 | A1 | B/O | 104 | 3.8 | 2D, trbC | strA | R | R | R | R | − | − | − | + | +/+ | − | − | − | − | + |
| p99309 | A1 | B/O | 104 | 3.8 | 2D, trbC | strA | R | R | R | R | − | − | − | + | +/+ | − | − | − | − | + |
| p99145 | A2 | B/O | 94 | 3.8 | 2D, trbC | strA | − | R | R | − | − | − | − | − | +/+ | − | − | − | − | + |
| p99170 | B | I1 | 102 | 13.5 | 2B | ND | − | − | R | − | − | − | − | − | −/− | + | − | − | − | + |
| p99184 | C | FIB | 156 | 17.5 | 2B, repC | strA | R | R | − | R | − | − | − | + | +/+ | − | − | + | + | + |
| p99323 | D | FIB | 144 | 6.5 | 2B, repC | strA | R | R | R | − | R | − | − | + | +/+ | − | − | − | + | + |
| p99376 | E | P | 112 | 17.5 | 2B, repC | strA | R | R | R | R | − | − | R | + | +/+ | + | − | − | − | + |
| p99378 | F | FIB | 102 | ND | 2B, repC | strA | R | R | R | − | − | − | R | + | +/+ | − | + | − | + | + |
| Nonconjugative plasmids with sul2 | ||||||||||||||||||||
| p99249 | G1 | 6.2 | orf5 | strA | − | R | − | − | − | − | − | − | +/+ | − | − | − | − | − | ||
| p99257 | G1 | 6.2 | orf5 | strA | − | R | − | − | − | − | − | − | +/+ | − | − | − | − | − | ||
| p99314 | G1 | 6.2 | orf5 | strA | − | R | − | − | − | − | − | − | +/+ | − | − | − | − | − | ||
| p99187 | G2 | 7.0 | orf5 | strA | R | − | − | − | − | − | − | + | +/+ | − | − | − | − | − | ||
| p99253 | G3 | 9.0 | orf5 | ND | R | − | R | − | − | − | − | + | −/+ | − | − | − | − | − | ||
The size of the SacI restriction fragment to which the sul2-specific probe hybridized.
A, ampicillin; S, streptomycin; W, trimethoprim; T, tetracycline; C, chloramphenicol; K, kanamycin; R, resistant; −, no resistance.
Amplified with primer repC-F, in addition to primer 2B.
ND, not determined, but amplification with the strA-specific primer was not successful.
The plasmids were tested by PCR for the presence of blaTEM, a likely determinant of ampicillin resistance, and strAB and aadA1, the two principal determinants of streptomycin resistance in E. coli. There was a complete correlation between the detection of blaTEM and ampicillin resistance. The strA and strB genes were detected in 21 (55%) plasmids, all of which conferred streptomycin resistance. strA alone was detected in a further six plasmids; streptomycin resistance was still observed in the corresponding transconjugants, albeit with increased zone diameters in some cases (e.g., p99163 and p99164). strB alone was detected in two plasmids, neither of which imparted streptomycin resistance. The aadA1 gene was identified in five plasmids, yet three of these did not impart streptomycin resistance, as determined by disc diffusion. The plasmids were also tested for the presence of intI1; this gene encodes the integrase of type 1 integrons and is frequently associated with sul1, which forms part of the 3′ conserved region of these elements (26). It was detected in 14 plasmids, including all 8 sul1-containing plasmids. Six plasmids contained intI1 without sul1, suggesting a rearrangement of the typical class 1 integron, possibly comparable to previously described variants in which sul1 has been lost (18).
The transconjugants and the transformants were tested for decreased susceptibility, relative to the susceptibility for the plasmid-free recipient, to various xenobiotics. Twelve plasmids (32%) conferred increased resistance to mercuric chloride (MICs, 32 μg/ml and 4 μg/ml, respectively). None of the plasmids conferred resistance to silver nitrate (MICs, 4 μg/ml both with and without plasmids). Similarly, no appreciable resistance was detected to chlorhexidine, triclosan, or cetrimide, which represent the bisguanide, bisphenol, and quaternary ammonium classes of disinfectants, respectively (median MICs, 0.5, 0.5, and 12 μg/ml, respectively; the MICs did not differ by more than 1 twofold dilution in any transconjugant or transformant). Thus, among all these agents, only mercury might potentially have a role in coselection for sulfonamide resistance. An association between mercury resistance and multiple-antimicrobial resistance in clinical E. coli isolates has been demonstrated previously, although resistance to sulfonamide was not recorded (21). The installation of amalgam fillings in primates has been reported to result in the release of significant quantities of mercury and might select for increased mercury resistance among intestinal flora, although this is controversial and a study with children did not find any difference in the prevalence of mercury-resistant flora among those with and without amalgam fillings (20, 27).
Genetic linkage to other selective markers.
In addition to linkage to resistance, we considered the possibility that sul2 could be genetically linked to other determinants that might give isolates a selective advantage in the host. We therefore investigated whether a range of putative or established virulence genes that have previously been associated with plasmid carriage (10) could be detected on the 33 conjugative plasmids. The genes afa/draBC and bmaE were not detected on any of the plasmids. The genes cvaC (colicin V, used as a marker for ColV virulence plasmids [10]), iutA (aerobactin siderophore system), and traT (serum survival) were detected on 4, 7, and 13 (11%, 18% and 34%) of the plasmids, respectively. The prevalence of these five potential virulence genes among conjugative plasmids carrying sul2 was comparable to or lower than their overall prevalence reported previously among 74 urine isolates of E. coli (9), suggesting no coselection of sul2 with virulence factors.
Genetic background of sul2.
A further factor likely to affect the long-term persistence of sul2 is the frequency with which it transfers to new genetic backgrounds and, in particular, to new plasmids. sul2 has most commonly been found as part of a cassette structure adjacent to the streptomycin resistance genes strA and strB. This cassette is not recognized as a mobile element in its own right but has been associated with transposition mediated by other elements, such as SXT, IS26, and ISCR2 (17, 29, 30). As a marker for the extent of the long-term mobility of sul2, we investigated the diversity of plasmid carrying sul2 by PCR-based determination of their incompatibility groups and comparison of their restriction profiles.
Replicon typing (4, 11) of the 33 conjugative plasmids demonstrated five different incompatibility groups (Table 1). Groups FIB (14 plasmids) and B/O (11 plasmids) were the most common, while groups I1, P, and K/B were represented only rarely (1 to 3 plasmids each). No replicon product was obtained for three plasmids.
The SacI restriction profiles (Table 1) of the 33 conjugative plasmids gave 29 distinct profiles that differed by at least one band. The use of pulsed-field gel electrophoresis allowed the excellent separation of bands in the range of 2 to 50 kb (data not shown). These profiles formed 19 similarity groups (each profile within a group differed by no more than three bands from another one in the group). Plasmids belonging to the same restriction group also belonged to the same incompatibility group (as determined by PCR) and generally had the same or similar resistance profiles and virulence gene profiles. In contrast, there was wide diversity in the resistance and the virulence gene profiles between restriction groups, whether or not these shared the incompatibility group. Similarly, the size of the SacI restriction fragment that hybridized to the sul2 probe correlated closely with the restriction group but indicated diversity between groups in the sequences surrounding sul2.
The five nonconjugative plasmids belonged to a single similarity group; three were indistinguishable from previously described plasmid p9123, whose restriction profile corresponds to that of pBP1 (5), while the other two carried additional resistance genes that coincided with their larger sizes.
Analysis of regions flanking sul2.
The finding of five different incompatibility groups among conjugative plasmids carrying sul2, in addition to pBP1-like plasmids, indicates that sul2 has been able to transfer into multiple plasmid backbones. The restriction profile diversity of plasmids carrying sul2 even within the two most prevalent incompatibility groups further suggests either that sul2 has been mobilized into these groups on multiple occasions or that the plasmid backbones have diversified extensively since the acquisition of sul2. To investigate the variation in the genetic environment of sul2, the upstream sequences of 25 sul2 genes deposited in the GenBank database were aligned by using ClustalW software (28). Within 400 bp upstream of the sul2 start codon, 12 different sequences were found, and these diverged at multiple positions. On the basis of the most frequently represented sequences, PCR primers were developed to characterize regions flanking sul2 in our plasmids (Fig. 1). Upstream sequences were classified as types 1, 2 (2A, 2B, or 2C), and 3, with the subtypes of type 2 having a conserved sequence of 302 bp immediately upstream of sul2 and subsequently diverging.
FIG. 1.
Organization of genes adjacent to sul2. (A) BamHI fragments cloned from conjugative sul2-containing plasmids. The sequenced regions are indicated by bold lines and are filled in gray, with the positions of the sequencing primers indicated by small arrows. Homology to other genes was determined by the amplification of overlapping PCR products of the predicted length (repA, repC, sul2, strA, strB) and by partial restriction site conservation (trbA-trbC). Dotted lines represent uncharacterized DNA. The scale is in bp. B, BamHI restriction sites. (B) Representative GenBank submissions with different sul2 genes in different contexts. The positions of the diagnostic PCR products used to distinguish the upstream regions are indicated above the sequences. For pO113, the serrated line indicates the region that is replaced by a sequence that includes sul2 in plasmids of group (Grp) 2D.
All 5 of the nonconjugative plasmids (restriction profile G) and none of the 33 conjugative plasmids contained the type 1 sequences found upstream of sul2 in small plasmids typified by pBP1 (5). None of the plasmids contained the type 3 upstream sequences observed in pMVSCS1 and other small plasmids that have been identified in the members of the family Pasteurellaceae (13). However, in all conjugative plasmids, a primer pair designed to amplify the type 2 sequence 162 bp upstream of sul2 in small plasmids typified by RSF1010 (22) gave an amplification product of the expected size. Of these, 20 were further classified as type 2B, having homology to RSF1010 for at least 439 bp upstream of sul2, and two were classified as type 2A, having an alternative upstream sequence previously described in several plasmids from Vibrio spp. and the Pasteurellaceae. None of the plasmids had the type 2C sequence, as described in the Vibrio cholerae SXT resistance element (30), while in 11 plasmids of four restriction groups, the sequence upstream was not amplified by type 2A, 2B, or 2C primers.
In order to look more closely at the DNA flanking sul2, BamHI restriction fragments containing the gene were cloned from a subset of conjugative plasmids and the relevant sequences were determined (Fig. 1). In four plasmids having an upstream type 2B sequence but different restriction profiles and belonging to three incompatibility groups (groups FIB, B/O, and P), sequencing of about 700 bp upstream of sul2 confirmed approximately 99% identity to the corresponding sequence of RSF1010. Further analysis by amplification of the overlapping PCR products confirmed that homology to RSF1010 extended to the 5′ end of repA, while sequencing out from this gene identified an adjacent copy of the IS26 transposase in all cases (Fig. 1). In three of these plasmids, strA and strB were identified downstream of sul2 by overlapping PCR, which matches the organization of RSF1010, but the plasmids diverged from each other and from RSF1010 at the 3′end of strB. These three plasmids of different incompatibility groups therefore appear to represent independent insertions of a large segment of RSF1010 adjacent to (or together with) a copy of IS26. (In the fourth plasmid of this group, a BamHI site had arisen downstream of sul2, and the relative locations of strA and strB were not determined.) However, not all conjugative plasmids with a type 2B upstream sequence had the same extent of homology to RSF1010, as only 13 of 20 yielded a PCR amplification product with primers spanning from the middle of repC to sul2 (Fig. 1 and Table 1).
Subcloned BamHI fragments from six plasmids with unknown type 2 upstream sequences were characterized; these belonged to three different groups by restriction profile analysis, but all were incompatibility group B/O. All six plasmids revealed the same genetic arrangement upstream of sul2 (Fig. 1). Immediately upstream of sul2, a sequence of 302 bp was identical to the sequence of RSF1010. (An apparent recombination site at 302 bp upstream of sul2 was also observed in several other sequences in the GenBank database compared to the sequence of RSF1010.) This was followed by a 274-bp sequence with no significant homology to any sequence in the GenBank database. Then, from 583 bp upstream of sul2 lay a sequence with 97% identity to trbC in the transfer region of E. coli plasmid pO113, which in turn shares homology with IncI plasmid R64; neither pO113 nor R64 carries sul2 at this position (14). PCR amplification with primers spanning from the 3′ end of trbC to the 5′ end of sul2 demonstrated that the same upstream sequence (designated type 2D) was present in all 11 plasmids with an unknown type 2 region (Table 1). Downstream of sul2 there was considerable diversity among the group B/O (type 2D) plasmids, including disruption of the conserved strA-strB gene arrangement in some cases (Table 1 and Fig. 1). We speculate that diversification subsequent to the acquisition of sul2 may explain the restriction fragment polymorphism among many of the incompatibility group B/O plasmids.
Upstream sequences among plasmids not transferred.
Since we had been unable to transfer plasmids from a number of clinical isolates, we considered whether they might contain sul2 in similar or different genetic contexts. We therefore examined flanking sequences upstream of sul2 in 35 clinical isolates from which plasmids had not been transferred. Among these isolates, 3 (8%) had type 1, pBP1-like sequences; 30 (86%) had type 2 sequences, as detected in all conjugative plasmids; and in 2 (6%), neither of these upstream sequences could be detected. Of the isolates with type 2 sequences, 17 isolates (57% of 30 isolates) had the type 2B upstream sequence indicating homology to RSF1010, 6 isolates (20%) had the type 2D upstream sequence, 1 isolate (3%) had the type 2A upstream sequence, and 6 isolates (20%) gave positive results with the type 2 primer but did not amplify with the type 2A, 2B, and 2D primers (the type 2C primer was not tested with these isolates). We conclude that the upstream sequences observed among the 38 nonconjugative and conjugative plasmids that we have characterized in detail are broadly representative of those in the E. coli population investigated but that there is likely to be additional diversity in the population among plasmids that we have not been successful in transferring.
Factors involved in the persistence of sul2.
It is likely that the carriage of sul2 on plasmids that also determine resistance to antimicrobials still in common use (such as ampicillin) contributes to the maintenance of sulfonamide resistance in the United Kingdom E. coli population via coselection. However, the findings of this study do not support the hypothesis that sulfonamide resistance is maintained through coselection by other xenobiotics, such as disinfectants or heavy metals, nor is it maintained through linkage with known virulence factors.
The diversity of restriction profiles and sul2 flanking sequences among the plasmids characterized in the present study demonstrates that sul2 has a history of genetic mobility. In some cases, mobility appears to be associated with IS26, while the presence of a common recombination site at 302 bp upstream of sul2 may indicate another mechanism of genetic transfer that is currently unclear. The persistence of sul2 over time represents a balance between the rate of loss of the gene by selective disadvantage, deletion from plasmids, and the loss of plasmids carrying sul2 versus the rate of gain by selective advantage (including fitness advantage, as observed with p9123 [5]), transfer to new plasmids, and the transfer of plasmids carrying sul2. Although we can put no time scale on the rate of sul2 transfer to new plasmids, we suggest that its evident mobility will contribute to its persistence in the bacterial population. We would predict that other resistance genes that have become very widely established within their host populations, particularly where resistance is carried on diverse mobile elements, will be similarly persistent unless their carriage represents a significant selective disadvantage.
Acknowledgments
This study was funded by the Department of Health (Resistance to Antibiotics and Other Antimicrobial Agents Programme, grant no. 91).
Footnotes
Published ahead of print on 15 December 2008.
REFERENCES
- 1.Anonymous. 1995. Co-trimoxazole use restricted. Drug Ther. Bull. 33:92-93. [DOI] [PubMed] [Google Scholar]
- 2.Bean, D. C., D. M. Livermore, I. Papa, and L. M. Hall. 2005. Resistance among Escherichia coli to sulphonamides and other antimicrobials now little used in man. J. Antimicrob. Chemother. 56:962-964. [DOI] [PubMed] [Google Scholar]
- 3.British Society for Antimicrobial Chemotherapy. 2004. BSAC disc diffusion method for antimicrobial susceptibility testing. British Society for Antimicrobial Chemotherapy, Birmingham, United Kingdom. [DOI] [PubMed]
- 4.Carattoli, A., A. Bertini, L. Villa, V. Falbo, K. L. Hopkins, and E. J. Threlfall. 2005. Identification of plasmids by PCR-based replicon typing. J. Microbiol. Methods 63:219-228. [DOI] [PubMed] [Google Scholar]
- 5.Enne, V. I., P. M. Bennett, D. M. Livermore, and L. M. Hall. 2004. Enhancement of host fitness by the sul2-coding plasmid p9123 in the absence of selective pressure. J. Antimicrob. Chemother. 53:958-963. [DOI] [PubMed] [Google Scholar]
- 6.Enne, V. I., D. M. Livermore, P. Stephens, and L. M. Hall. 2001. Persistence of sulphonamide resistance in Escherichia coli in the UK despite national prescribing restriction. Lancet 357:1325-1328. [DOI] [PubMed] [Google Scholar]
- 7.European Committee for Antimicrobial Susceptibility Testing (EUCAST) of the European Society of Clinical Microbiology and Infectious Diseases (ESCMID). 2003. Determination of minimum inhibitory concentrations (MICs) of antibacterial agents by broth dilution. Clin. Microbiol. Infect. 9:ix-xv. [Google Scholar]
- 8.Ito, H., H. Ishii, and M. Akiba. 2004. Analysis of the complete nucleotide sequence of an Actinobacillus pleuropneumoniae streptomycin-sulfonamide resistance plasmid, pMS260. Plasmid 51:41-47. [DOI] [PubMed] [Google Scholar]
- 9.Johnson, J. R., T. T. O'Bryan, P. Delavari, M. Kuskowski, A. Stapleton, U. Carlino, and T. A. Russo. 2001. Clonal relationships and extended virulence genotypes among Escherichia coli isolates from women with a first or recurrent episode of cystitis. J. Infect. Dis. 183:1508-1517. [DOI] [PubMed] [Google Scholar]
- 10.Johnson, J. R., and A. L. Stell. 2000. Extended virulence genotypes of Escherichia coli strains from patients with urosepsis in relation to phylogeny and host compromise. J. Infect. Dis. 181:261-272. [DOI] [PubMed] [Google Scholar]
- 11.Johnson, T. J., Y. M. Wannemuehler, S. J. Johnson, C. M. Logue, D. G. White, C. Doetkott, and L. K. Nolan. 2007. Plasmid replicon typing of commensal and pathogenic Escherichia coli isolates. Appl. Environ. Microbiol. 73:1976-1983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Kado, C. I., and S. T. Liu. 1981. Rapid procedure for detection and isolation of large and small plasmids. J. Bacteriol. 145:1365-1373. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kehrenberg, C., and S. Schwarz. 2002. Nucleotide sequence and organization of plasmid pMVSCS1 from Mannheimia varigena: identification of a multiresistance gene cluster. J. Antimicrob. Chemother. 49:383-386. [DOI] [PubMed] [Google Scholar]
- 14.Leyton, D. L., J. Sloan, R. E. Hill, S. Doughty, and E. L. Hartland. 2003. Transfer region of pO113 from enterohemorrhagic Escherichia coli: similarity with R64 and identification of a novel plasmid-encoded autotransporter, EpeA. Infect. Immun. 71:6307-6319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Livermore, D. M., M. Warner, L. M. Hall, V. I. Enne, S. J. Projan, P. M. Dunman, S. L. Wooster, and G. Harrison. 2001. Antibiotic resistance in bacteria from magpies (Pica pica) and rabbits (Oryctolagus cuniculus) from west Wales. Environ. Microbiol. 3:658-661. [DOI] [PubMed] [Google Scholar]
- 16.Livermore, D. M., and J. D. Williams. 1996. ß-Lactams: mode of action and mechanisms of bacterial resistance, p. 561. In V. Lorian (ed.), Antibiotics in laboratory practice, vol. 4. The Williams & Wilkins Co., Baltimore, MD. [Google Scholar]
- 17.Parkhill, J., G. Dougan, K. D. James, N. R. Thomson, D. Pickard, J. Wain, C. Churcher, K. L. Mungall, S. D. Bentley, M. T. Holden, M. Sebaihia, S. Baker, D. Basham, K. Brooks, T. Chillingworth, P. Connerton, A. Cronin, P. Davis, R. M. Davies, L. Dowd, N. White, J. Farrar, T. Feltwell, N. Hamlin, A. Haque, T. T. Hien, S. Holroyd, K. Jagels, A. Krogh, T. S. Larsen, S. Leather, S. Moule, P. O'Gaora, C. Parry, M. Quail, K. Rutherford, M. Simmonds, J. Skelton, K. Stevens, S. Whitehead, and B. G. Barrell. 2001. Complete genome sequence of a multiple drug resistant Salmonella enterica serovar Typhi CT18. Nature 413:848-852. [DOI] [PubMed] [Google Scholar]
- 18.Partridge, S. R., G. D. Recchia, H. W. Stokes, and R. M. Hall. 2001. Family of class 1 integrons related to In4 from Tn1696. Antimicrob. Agents Chemother. 45:3014-3020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Perreten, V., and P. Boerlin. 2003. A new sulfonamide resistance gene (sul3) in Escherichia coli is widespread in the pig population of Switzerland. Antimicrob. Agents Chemother. 47:1169-1172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Pike, R., V. Lucas, P. Stapleton, M. S. Gilthorpe, G. Roberts, R. Rowbury, H. Richards, P. Mullany, and M. Wilson. 2002. Prevalence and antibiotic resistance profile of mercury-resistant oral bacteria from children with and without mercury amalgam fillings. J. Antimicrob. Chemother. 49:777-783. [DOI] [PubMed] [Google Scholar]
- 21.Poiata, A., I. Badicut, M. Indres, M. Biro, and D. Buiuc. 2000. Mercury resistance among clinical isolates of Escherichia coli. Roum. Arch. Microbiol. Immunol. 59:71-79. [PubMed] [Google Scholar]
- 22.Radstrom, P., and G. Swedberg. 1988. RSF1010 and a conjugative plasmid contain sulII, one of two known genes for plasmid-borne sulfonamide resistance dihydropteroate synthase. Antimicrob. Agents Chemother. 32:1684-1692. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Radstrom, P., G. Swedberg, and O. Skold. 1991. Genetic analyses of sulfonamide resistance and its dissemination in gram-negative bacteria illustrate new aspects of R plasmid evolution. Antimicrob. Agents Chemother. 35:1840-1848. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
- 25.Scholz, P., V. Haring, B. Wittmann-Liebold, K. Ashman, M. Bagdasarian, and E. Scherzinger. 1989. Complete nucleotide sequence and gene organization of the broad-host-range plasmid RSF1010. Gene 75:271-288. [DOI] [PubMed] [Google Scholar]
- 26.Stokes, H. W., and R. M. Hall. 1989. A novel family of potentially mobile DNA elements encoding site-specific gene-integration functions: integrons. Mol. Microbiol. 3:1669-1683. [DOI] [PubMed] [Google Scholar]
- 27.Summers, A. O., J. Wireman, M. J. Vimy, F. L. Lorscheider, B. Marshall, S. B. Levy, S. Bennett, and L. Billard. 1993. Mercury released from dental “silver” fillings provokes an increase in mercury- and antibiotic-resistant bacteria in oral and intestinal floras of primates. Antimicrob. Agents Chemother. 37:825-834. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Toleman, M. A., P. M. Bennett, and T. R. Walsh. 2006. ISCR elements: novel gene-capturing systems of the 21st century? Microbiol. Mol. Biol. Rev. 70:296-316. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Waldor, M. K., H. Tschape, and J. J. Mekalanos. 1996. A new type of conjugative transposon encodes resistance to sulfamethoxazole, trimethoprim, and streptomycin in Vibrio cholerae O139. J. Bacteriol. 178:4157-4165. [DOI] [PMC free article] [PubMed] [Google Scholar]