| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, August 2007, p. 2720-2725, Vol. 51, No. 8
0066-4804/07/$08.00+0 doi:10.1128/AAC.00026-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Dipartimento di Biologia Molecolare, Sezione di Microbiologia, Università di Siena, Siena, Italy,1 Dipartimento Area Critica Medico Chirurgica, Clinica Malattie Infettive, Università di Firenze, Florence, Italy,2 Distrito de Salud Cordillera, Departamento Santa Cruz, Camiri, Bolivia,3 Instituto de Medicina Tropical Alexander von Humboldt, Universidad Peruana Cayetano Heredia, Lima, Peru,4 Department of Microbiology and Tumor Biology-MTC, Clinical Microbiology, Karolinska Institut, Stockholm, Sweden5
Received 9 January 2007/ Returned for modification 1 March 2007/ Accepted 24 May 2007
| ABSTRACT |
|---|
|
|
|---|
| INTRODUCTION |
|---|
|
|
|---|
The CTX-M-type ß-lactamases are members of a lineage of ESBLs which exhibited dramatic dissemination in the past decade; in several clinical settings, these enzymes currently outnumber the classical TEM- and SHV-type ESBLs (5, 9, 21). Based on their amino-acidic sequence diversity, the vast number (more than 50) of CTX-M variants identified thus far have been classified into five major phylogenetic groups: CTX-M-1, CTX-M-2, CTX-M-8, CTX-M-9, and CTX-M-25 (http://www.lahey.org/studies/webt.stm) (5, 9). Strains producing CTX-M-type ESBLs have been reported as important causes of both nosocomial and community-acquired infections (5, 9). CTX-M determinants have been detected not only in clinical isolates but also in commensal bacteria from humans and animals and in isolates from products of the food chain and sewage, revealing a wide distribution and suggesting the presence of environmental reservoirs of these resistance determinants (8, 9). In fact, recent studies have addressed the importance of the influx into the hospitals of CTX-M-producing isolates from the community (3).
ANTRES is a collaborative research project dealing with antibiotic use and resistance in two countries of Latin America, Bolivia, and Peru (http://www.unifi.it/infdis/antres/default.htm). Within this project, a large-scale survey of fecal carriage of antibiotic-resistant Escherichia coli strains in preschool-age healthy children was carried out in 2002 (1) and was repeated in 2005. Comparative analysis of results from the two surveys revealed that, in this three-year period, a remarkable increase in the rates of resistance to expanded-spectrum cephalosporins had occurred (0.1% in 2002 versus 1.7% in 2005) (1; A. Bartoloni, L. Pallecchi, C. Fiorelli, T. Di Maggio, C. Fernandez, A. L. Villagran, A. Mantella, F. Bartalesi, M. Strohmeyer, A. Bechini, M. Gamboa, M. Rodriguez, C. Kristiansson, G. Kronvall, E. Gotuzzo, F. Paradisi, and G. M. Rossolini, unpublished data). In this report, we demonstrate that this phenomenon was mainly the consequence of the dissemination of CTX-M-type determinants among commensal E. coli strains.
(These results were presented in part at the 46th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, CA, 2006.)
| MATERIALS AND METHODS |
|---|
|
|
|---|
In vitro susceptibility testing. Antimicrobial susceptibility was determined by the disk diffusion test, according to the methods and interpretative criteria proposed by CLSI (12). Antibiotic disks were from Oxoid (Oxoid, Milan, Italy). E. coli ATCC 25922 was used for quality control purposes. The double-disk synergy test for confirmation of ESBL activity was carried out as described previously (16) by screening isolates for synergism between clavulanate (represented by a disk with amoxicillin-clavulanate) and cefotaxime, ceftazidime, and aztreonam.
Molecular analysis techniques. Basic procedures for DNA extraction and analysis were performed as described by Sambrook and Russell (29). Nucleotide sequences were determined on both strands of PCR amplification products at the Macrogen sequencing facility (Macrogen Inc., Seoul, Korea). Analysis and comparisons of nucleotide sequence were carried out using programs available at the NCBI web interface (http://www.ncbi.nlm.nih.gov). The presence of CTX-M-related sequences was investigated by colony blot hybridization with a blaCTX-M probe, as described previously (19). Primers used in this work for PCR amplification of antibiotic resistance genes are reported in Table 1. PCR conditions were as described in the corresponding references or, for PCR amplification of blaCTX-M alleles belonging to groups 1 and 9 and of blaSHV genes, as follows: initial denaturation at 94°C for 5 min; denaturation at 94°C for 45 s, annealing at 50°C for 45 s, and elongation at 72°C for 5 min, repeated for 35 cycles; and final extension at 72°C for 10 min. Forward primers used for amplification of group 2 or for amplification of group 1 and 9 CTX-M genes targeted ISCR1 and ISEcp1 elements, respectively. Phylogenetic grouping of the E. coli isolates was determined by the multiplex PCR-based method developed by Clermont et al., which allows identification of the four major phylogenetic groups (A, B1, B2, and D) (11). Randomly amplified polymorphic DNA (RAPD) genotyping was performed as previously described (17) using, separately, the decamer primers 1290 and 1254. RAPD patterns were considered to be different when the profiles differed by at least one band. Analysis of the RAPD patterns was performed with Diversity Database fingerprinting software, version 2 (Bio-Rad Laboratories, Hercules, CA).
|
5 x 10–8 transconjugants/recipient. Mobilization of nonconjugative blaCTX-M-harboring plasmids was carried out by electroporation using E. coli DH5
and MH agar containing cefotaxime (2 µg/ml) for selection of transformants. The presence of blaCTX-M genes in the transconjugants and transformants was always confirmed by PCR. Plasmid fingerprinting analysis after digestion with PstI was carried out with all transferable A/C-, I1-, and F-type plasmids carrying group 2, 9, and 1 CTX-M determinants, respectively. Nucleotide sequence accession numbers. The sequences of blaCTX-M-56 and the new variant allele of blaCTX-M-24 have been deposited in GenBank and assigned accession numbers EF374097 and EF374096, respectively.
| RESULTS |
|---|
|
|
|---|
Overall, 54 isolates showing reduced ceftriaxone susceptibility were collected in the 2005 survey, and 50 of them were confirmed as ESBL producers in a double-disk synergy test. The four isolates that did not exhibit ESBL production showed a resistance phenotype suggestive of production of an AmpC-type ß-lactamase (resistance to cephamycins and to ß-lactam-inhibitor combinations, susceptibility to cefepime) and were not further investigated in this work.
The 50 ESBL-producing isolates were tested for the presence of CTX-M-like determinants by colony-blot hybridization. Results showed that 44 isolates (88%) carried blaCTX-M-related sequences. Most (84%) of the CTX-M-producing isolates also carried a blaTEM-1b gene, while none of them carried blaSHV-related sequences. Prevalences of fecal carriers of CTX-M-positive E. coli strains were 2.4% in Camiri and 0.9% in Villa Montes, Bolivia, and 1.1% in Yurimaguas and 1.1% in Moyobamba, Peru. The six ESBL-producing blaCTX-M-negative isolates were shown to harbor SHV-type ESBL determinants, either blaSHV-2 (four isolates, two from Bolivia and two from Peru) or blaSHV-12 (two isolates from Peru). Two of these isolates (one SHV-2-producing isolate and one SHV-12-producing isolate, both from Peru) also carried a blaTEM-1b gene.
Characterization and distribution of the CTX-M determinants. Sequencing of the CTX-M-encoding genes identified variants belonging to CTX-M-2 (41%), CTX-M-1 (32%), and CTX-M-9 (27%) groups. CTX-M-2 group determinants included blaCTX-M-2 and a new variant named blaCTX-M-56 (GenBank accession number EF374097) that differed from blaCTX-M-2 (2) by a single nucleotide mutation (G824A), resulting in a replacement of serine by asparagine at position 275. CTX-M-1 group determinants included only blaCTX-M-15. CTX-M-9 group determinants included blaCTX-M-14 and a new allele of blaCTX-M-24 (GenBank accession number EF374096), showing two synonymous transversions (C823A and C825A) compared to blaCTX-M-24 as described by Eckert et al. (13). This new blaCTX-M-24 allele differed from blaCTX-M-14 (18) by a single transversion (C825A) that was responsible for the replacement of a serine by an arginine at position 275 and was likely derived from blaCTX-M-14. As far as the 5'-flanking regions of CTX-M genes are concerned, all group 2 alleles were found to be associated with an ISCR1 element, while all group 1 and 9 alleles were found to be associated with an ISEcp1 element.
Concerning the geographic distribution, group 2 CTX-M determinants were the most prevalent in Bolivia (62%) and the least prevalent in Peru (11%), group 9 CTX-M determinants were the most prevalent in Peru (67%) and absent in Bolivia, while group 1 CTX-M determinants were the second most prevalent in both countries (38% in Bolivia and 22% in Peru). In Bolivia, the most prevalent CTX-M variant was blaCTX-M-2 (54%) followed by blaCTX-M-15 (38%) and blaCTX-M-56 (8%). In Peru, the most prevalent CTX-M variant was blaCTX-M-14 (45%) followed by blaCTX-M-24 (22%), blaCTX-M-15 (22%), and blaCTX-M-2 (11%) (Fig. 1).
|
|
RAPD genotyping identified 31 different types, of which 25 were unique and 6 included more than one isolate (Table 2). All the isolates harboring CTX-M-2 group determinants (blaCTX-M-2 and blaCTX-M-56) showed unique RAPD genotypes, while clonal expansion was observed for some isolates harboring CTX-M determinants of the other groups (Table 2). In particular, dissemination of CTX-M-15 in Bolivia (not observed in 2002) appeared to be the consequence of the spread of two clones (RAPD types 30 and 31) circulating in both Camiri and Villa Montes (Table 2). Clonal expansion of CTX-M-producing isolates was also observed in Peru for CTX-M-14-, CTX-M-15-, and CTX-M-24-producing isolates (RAPD types 20, 29, and 25 through 26, respectively), even though the spreading of each clone was limited to a single city (Table 2). None of the CTX-M-producing isolates collected in 2005 showed genomic similarity (based on RAPD analysis) to those collected from the same areas in the 2002 survey (data not shown).
Transferability of CTX-M determinants and typing of the blaCTX-M-harboring plasmids. Conjugal transfer of CTX-M determinants was observed with 33 isolates (75%) representative of 24 different RAPD types (Table 2). All the CTX-M-2 group determinants were carried by conjugative plasmids, mainly of the A/C type (72%), which always carried other resistance genes (cotransfer of gentamicin resistance occurred in 94% of cases) (Table 2). Plasmid replicon typing of transconjugants obtained with the three CTX-M-2-producing isolates from the 2002 survey (19) identified in all cases plasmids of the A/C type, underlining the role of these mobile genetic elements in the dissemination of CTX-M-2 group determinants in the study areas. Conjugal transfer of CTX-M-1 and CTX-M-9 group determinants was observed less frequently (57% and 58%, respectively), and in some cases nonconjugative CTX-M-9 group determinants could be mobilized by electroporation (Table 2). Transfer of CTX-M-9 group determinants was related to I1-type plasmids and was never associated with cotransfer of other resistance traits. Transfer of CTX-M-1 group determinants was related to conjugative plasmids harboring none of the replicon types investigated (RAPD type 31 from Bolivia) or to F-type conjugative plasmids which also carried determinants responsible for fluoroquinolone resistance (RAPD types 28 and 29 from Peru) (Table 2).
Plasmid fingerprinting analysis was carried out with all transferable A/C-, I1-, and F-type plasmids carrying group 2, 9, and 1 CTX-M determinants, respectively. Different plasmid profiles were observed within plasmids carrying each replicon type (Table 2). However, in most cases, plasmids carrying CTX-M genes of the same group shared some bands of the same size, suggesting a common ancestry (data not shown). In particular, the plasmids carrying blaCTX-M-56 showed a profile identical to that of one of the plasmids carrying blaCTX-M-2, further supporting the hypothesis of a local evolution of the new CTX-M variant (Table 2 and data not shown). Moreover, the same profile (profile A) was observed with blaCTX-M-2-harboring plasmids from both Peru and Bolivia and was apparently identical to that observed with the A/C-type blaCTX-M-2-harboring plasmid detected in Peru during the 2002 survey (19), demonstrating the presence of an epidemic plasmid responsible for the dissemination of CTX-M-2 ESBL in these geographic areas (Table 2 and data not shown).
Characterization of the plasmid-mediated fluoroquinolone resistance determinants. The three transconjugants showing decreased susceptibility to ciprofloxacin were investigated for the presence of known plasmid-mediated fluoroquinolone resistance genes (24). None of the qnr-type determinants investigated (qnrA, qnrB, and qnrS) was detected. In contrast, all three transconjugants gave positive results in a PCR experiment using primers designed for amplification of aac(6')-Ib variants (20). Sequence analysis of the amplicons identified, in all cases, an aac(6')-Ib-cr variant (100% identity with aac(6')-Ib-cr of GenBank entry DQ303918) (26) that could account for the decreased susceptibility to ciprofloxacin expressed by the transconjugants. To our best knowledge, this is the first detection of aac(6')-Ib-cr in Latin America.
The F-type plasmids harboring blaCTX-M-15 and aac(6')-Ib-cr showed identical patterns of F replicons (FIA, FIB, and FII) and similar plasmid profiles (Table 2 and data not shown). Sequence analysis of the FII-regulatory antisense RNA CopA, which is involved in the control of replication and in the incompatibility behavior of the FII plasmids (4), evidenced, in all cases, identical CopA sequences (100% identical to that of plasmid pRSB107 of GenBank entry AJ851089), further supporting the idea of a common origin of these three plasmids.
| DISCUSSION |
|---|
|
|
|---|
Interestingly, the increased prevalence of ESBL-producing isolates was also associated with an increased diversity of circulating ESBL determinants, although the use of ceftriaxone as the only ESBL screening agent might have limited the detection of ESBLs with specific ceftazidimase activity. Compared to the baseline survey carried out in 2002 (19), members of the CTX-M-1 group (CTX-M-15) emerged in Bolivia also, while members of the CTX-M-9 group (CTX-M-14 and CTX-M-24) emerged in Peru. Moreover, SHV-type ESBLs emerged in both countries.
Molecular characterization of the CTX-M-producing isolates suggested different mechanisms of spreading for determinants belonging to different CTX-M groups. Concerning CTX-M-2 group determinants, which are widely disseminated in Latin America (5, 9, 22, 23) and have been present in the study areas since 2002 (19), their exclusive location on conjugative plasmids and the high genotypic diversity of bacterial hosts suggest that plasmid dissemination played a major role in their diffusion. This hypothesis is further supported by plasmid replicon typing and fingerprinting analysis results, which revealed that one or a few related A/C-type conjugative plasmids could be involved in their dissemination in these areas. In contrast, a role in clonal expansion, along with plasmid mobilization, was evidenced for dissemination of CTX-M determinants more recently introduced in the study areas (the CTX-M-1 group in Bolivia and the CTX-M-9 group in Peru).
Two new CTX-M alleles were identified in this study: a new CTX-M-2 group variant named blaCTX-M-56 (in Bolivia) and a new blaCTX-M-24 allele (in Peru). Interestingly, the two differed by a single nucleotide mutation from the most prevalent CTX-M-type gene detected in each country (CTX-M-2 in Bolivia and CTX-M-14 in Peru), suggesting a local evolution of these resistance determinants. This hypothesis was further supported by results of replicon typing and plasmid fingerprinting, which showed a common genetic context for these new blaCTX-M alleles and their likely precursors (A/C type plasmids for blaCTX-M-2 and blaCTX-M-56; I1-type plasmids for blaCTX-M-14 and blaCTX-M-24).
CTX-M-encoding plasmids are known to often carry additional determinants of resistance to non-ß-lactam agents, including quinolones (9). In this study, we detected the presence of plasmid-mediated quinolone resistance determinants in three blaCTX-M-15-harboring conjugative plasmids from Peruvian isolates. These plasmids carried the aac(6')-Ib-cr gene and showed identical patterns of replicon types (FIA, FIB, and FII) and similar plasmid profiles, suggesting a common origin. Moreover, sequence analysis of the FII replicons evidenced an IncFII plasmid backbone identical to that of pRSB107, which was recently also found in blaCTX-M-15-harboring plasmids from E. coli clinical isolates in the United Kingdom (14). These findings support the evidence of the dissemination of FII-type blaCTX-M-15-harboring plasmids in different countries, a result that has recently also been observed for a different lineage of IncFII plasmids detected in CTX-M-15-producing E. coli strains from Canada and United Kingdom (6, 14). Furthermore, to our best knowledge this is the first detection of an aac(6')-Ib-cr gene in commensal bacteria from healthy humans and also the first report of the presence of this resistance determinant in Latin America.
| ACKNOWLEDGMENTS |
|---|
We are grateful to Patrice Nordmann and Laurent Poirel (Université Paris-Sud, K.-Bicetre, France) for supplying controls for qnr genes. We wish to thank all the members of the ANTRES Study Group for their support in this study.
| FOOTNOTES |
|---|
Published ahead of print on 4 June 2007. ![]()
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| Clin. Vaccine Immunol. | Clin. Microbiol. Rev. |
|---|---|
| J. Clin. Microbiol. | ALL ASM JOURNALS |