Abstract
The aim of this study was to describe the distribution of the trimethoprim–sulfamethoxazole resistance genes and their association with class 1 integrons in a collection of clinical isolates of Enterobacteriaceae recovered at the University Hospital Sahloul in Tunisia. A total of 80 isolates of Enterobacteriaceae were studied, including six different species. There were 35 extended-spectrum beta-lactamases (ESBL)-producing isolates. Resistance to trimethoprim–sulfamethoxazole was assessed by the disk diffusion method. Polymerase chain reaction (PCR) with primers specific for sul1, sul2, and sul3 was used to detect the three known sulphonamide resistance genes. The presence of class 1 integrons in the studied isolates was detected using PCR and the resistance gene cassettes were characterized by directly sequencing the PCR products obtained with 5′conserved segment (5′CS) and 3′conserved segment (3′CS) primers. The int1 gene was found in 68 out of 80 enterobacterial isolates. The sul1 gene was found in 22 isolates (27.5%), sul2 gene in 5 isolates (6.25%), and both genes in 49 isolates (61.25%). Eight of the studied isolates had no dfr alleles, and in the remaining 72 isolates, 7 dfr genes were identified. The most prevalent were dfrA7 (40%) and dfrA17 (33%). Class 1 integrons were found to be an important genetic element of resistance to trimethoprim–sulfamethoxazole among the clinical isolates of Enterobacteriaceae. The types, combinations, and frequency of the gene cassettes in integrons provide useful data for the surveillance of antimicrobial resistance in our hospital and for the prescription practice of cotrimoxazole.
Introduction
Very few data are available regarding the resistance of Enterobacteriaceae to sulphonamides and trimethoprims in North Africa, and particularly in Tunisia. Therefore, the aim of our study was to investigate the molecular basis of resistance to those compounds and to characterize the gene cassettes associated with class 1 integrons in a collection of clinical isolates of Enterobacteriaceae recovered at the University Hospital Sahloul in Tunisia.
Materials and Methods
Bacterial strains
During the period of study from September 2004 to May 2006, a total of 1,100 nonduplicate enterobacterial isolates had been identified at the Sahloul Hospital (Sousse, Tunisia), a 550-bed University hospital. A total of 440 isolates (40%) were trimethoprim–sulfamethoxazole resistant and 80 of them were retained for that study.
Isolates were associated with urinary tract infections (n = 75), bacteremia (n = 3), and wound infections (n = 2) and belonged to six bacterial species: Escherichia coli, Klebsiella pneumoniae, Enterobacter cloacae, Citrobacter freundii, Providencia rettgeri, and Morganella morganii. There were 35 extended-spectrum beta-lactamases (ESBL)-producing isolates including all bacterial species previously mentioned. The ESBL isolates were characterized as CTX-M-15 and SHV-12. 5 The ESBL-negative isolates included all bacterial species except M. morganii.
Antibiotic susceptibility testing
All isolates were screened using the disk diffusion method on Muller–Hinton agar (Bio-Rad, Marnes-La-Coquette, France) to study their resistance to the following antibiotics: trimethoprim, sulfamethoxazole, the combination of both compounds, streptomycin, gentamycin, kanamycin, tobramycin, netilmicin, and chloramphenicol. The results of antibiotic resistance testing were interpreted according to the recommendations of the Comité de l'Antibiogramme de la Société Française de Microbiologie (CA-SFM at www.sfm.asso.fr). E. coli ATCC 25922 was used as a control.
Integron polymerase chain reaction detection
All isolates were screened for the presence of type 1 integrons by polymerase chain reaction (PCR), with the primers int1 for and int1 rev (Table 1), as previously described by Frank et al. 7
PCR for detection of sulphonamide resistance genes
Screening for sul1 and sul2 genes was done by PCR using the primers and conditions previously described. 18 Specific primers were used to detect the sul3 gene at an annealing temperature of 53°C (Table 1).
Three E. coli clinical isolates (strains 02-57295, 02-58161, and 03-709) obtained during December 2002 and January 2003 in the Tenon Hospital (France) and harboring the int1 gene and the sul1 gene, sul2 gene, and the sul3 gene were, respectively, used as positive controls (A. Doloy, G. Arlet, personal data). Water was used as negative control.
Characterization of class 1 integron resistance gene cassettes
Following the amplification of the variable region of class 1 integrons by PCR (with 5′CS and 3′CS primers), PCR products were subjected to DNA sequencing using PCR primers and additional sequencing primers (Table 1) which were designed using Oligo4 software.
DNA sequencing was performed with an ABI Prism 310 DNA sequencer (Applied Biosystems, Foster City, CA). Nucleotide sequences were analyzed with the BLAST program of the National Center for Biotechnology Information (www.ncbi.nlm.nih.gov).
Results
All tested isolates (80) were resistant to trimethoprim, sulfamethoxazole, and the combination of both compounds (Table 2).
ESBL, extended-spectrum beta-lactamases; SXT, trimethoprim + sulfamethoxazole; TMP, trimethoprim; SMX, sulfamethoxazole; S, streptomycin; GM, gentamycin; K, kanamycin; TM, tobramycin; NET, netilmicin; C, chloramphenicol.
Concerning sulfamethoxazole resistance, the majority of isolates contained one or more sul alleles. PCR analysis indicated that 71 isolates carried the sul1 gene, 54 the sul2 gene, and 49 carried both sul1 and sul2 genes while sul3 was not found (Table 2).
All the isolates that were studied were also resistant to trimethoprim. Eight of these isolates had no dfr alleles. Out of those eight isolates, three contained no int1 gene and five had type 1 integrons. In the remaining 72 isolates, seven dfr genes encoding dihydrofolate reductase were identified. They were distributed as follows: 3 (4%) dfrA1, 8 (9%) dfrA5, 29 (40%) dfrA7, 1 (1.5%) dfrA14, 8 (11%) dfrA15, 24 (33%) dfrA17, and 1 (1.5%) dfrA21. Nine out of the 74 dfr alleles containing isolates were not associated with class 1 integron structures: one isolate contained the dfrA21, one isolate contained the dfrA1 allele, two contained the dfrA17 allele, and five isolates contained the dfrA5 allele (Table 2).
Among the studied strains, 68 isolates (85%) contained the type 1 integron while 12 isolates (15%) were negative for the int1 integrase gene. Seven of these 12 isolates were positive for sul1 but negative for sul2. One K. pneumoniae isolate was positive for sul1 and sul2 genes and negative for int1 gene, and finally four ESBL-producing E. coli isolates were negative for the class 1 integron gene and sul1 and sul2 genes; these isolates carried dfr allele dfrA5 associated with erythromycin esterase gene ere(A) (Table 2).
We detected 10 different cassette arrays. Four genetic cassette arrangements composed of aminoglycoside/streptomycin resistance genes and dfr genes, two cassette arrangements composed of erythromycin esterase (ere (A)) and dfr genes, two composed of chloramphenicol resistance genes (CmlA6/CatA1) and aminoglycosides resistance genes (aadA1/aadB), one cassette arrangement composed of the streptothricin resistance gene (estX) and the aminoglycoside resistance gene aadA1, and finally one cassette arrangement composed of the gene encoding the oxacillinase OXA-30 and the aminoglycoside resistance gene aadA1 (Table 2).
Discussion
Our results generally corroborate the integron prevalence observed in Enterobacteriaceae in Asia and Europe.4,9,10,20,30,31 In these studies, class 1 integrons were found in about 70% of trimethoprim-resistant isolates. All studies indicated that integrons of class 1 are very rare in trimethoprim-susceptible isolates.
In our study, there were strains that have a resistant phenotype to an antibiotic (streptomycin, trimethoprim, etc.) without the presence of an adequate resistance gene cassette (Table 2). This resistance could be directly related to the presence of resistance genes within the integron, 22 or the resistance may be conferred via chromosomal mutations that alter the ribosomal binding site of the antibiotic (like streptomycin). 22
The trends we observed in the distribution of sulfamethoxazole-resistant allele distribution (sul1 > sul2) are not in accordance with previous studies.3,7,14,18 The high prevalence of the sul1 gene could be explained by its frequent association with aadA gene cassettes encoding aminoglycoside adenyltransferases in type 1 integrons 6 ; this gene cassette was often found in our isolates, whereas the sul2 gene was often found adjacent to the resistance genes strA 6 (not described in this study).
The majority of studied strains (85%) contained the type 1 integron, whereas 15% were negative for the int1 integrase gene. Seven of these 12 isolates were positive for sul1 but negative for sul2 as described in the Central African Republic for E. coli isolates. 7 Previously, one plasmid has been described in an island carrying a class 1 integron and a truncated int1 gene which could have given similar results. 15
Our results also confirm the high prevalence of the aadA and dfrA gene cassettes, particularly the high prevalence of the dfrA7 followed by dfrA17. This correlates with previous studies of African Enterobacteriaceae isolates.7,8 However, our results do not correlate with European studies where dfrA1 was found to be the most common allele.3,13,18,28 Studies performed in Korea30,31 and Australia 29 found dfrA17 and dfrA12 to be the most common alleles.
Approximately 10 linkages between antibiotic resistance gene cassettes have been observed in this study, of which 9 of them have been previously described.1,9,16,17,19,23,25–27 The linkage with erythromycin esterase (ere(A)) and dfr genes (dfrA5 and dfrA7) was not described. To our knowledge, this is the first report of the presence of ereA2, dfrA5, and dfrA7 cassettes in class 1 integrons of Enterobacteriaceae. Only one association between ereA2 and dfrA5 was reported in Vibrio cholerae. 2
Trimethoprim–sulfamethoxazole is an effective and inexpensive antimicrobial combination used to treat a host of diseases. However, in the last few years, we have observed an important increase in resistance to this compound. There have been few studies of the genetic distribution underlying trimethoprim and sulfamethoxazole resistance in Africa but no studies were performed in Tunisia. To date, this is the first report describing the molecular characterization of class 1 integron in Enterobacteriaceae isolates from Tunisia.
Footnotes
Acknowledgment
This work was partially funded by grants from the Ministry of Higher Education, Scientific Research and Technology, Tunisia.
Disclosure Statement
All authors disclose no commercial associations that might create a conflict of interest in connection with this study.
