- Top of page
A hundred and eleven Gram-negative bacilli from community-acquired infections were characterized by antimicrobial susceptibility testing, screened for class 1 and 2 integrons, and statistically evaluated for the association between antibiotic profile and the presence of integrons. The frequency with which integrons were harbored was 28.8%. Three E. coli strains contained a dfrA17 variant inserted in a class 1 integron. Results of PFGE indicated that some E. coli strains carrying integrons were clonally related. Carriage of gene cassettes was significantly associated with resistance to certain antibiotics (P < 0.05).
Acquisition of antibiotic resistance by pathogenic bacteria has been a growing problem worldwide and, in particular, multidrug resistance in such bacteria is a major problem in the treatment of infectious diseases. The acquisition of resistance genes by horizontal transfer via plasmids and transposons is considered to play a role in the occurrence of this multiresistance. Class 1 integrons comprise a site-specific recombination system capable of integrating and expressing those genes contained in cassette-like structures (1). Integrons harbored by human clinical isolates are disseminated worldwide. The class 1 integron is probably the most prevalent type of integron carried by such isolates. It has been also shown that Tn7 carrying a class 2 integron can be present in clinical pathogens (2). Integrons have been shown to be associated with multidrug resistance in Gram-negative bacilli in hospital settings. However, there is little data on their prevalence in the community. Therefore, in Turkey, limited studies have been performed on the prevalence of integron-associated antibiotic resistance in Gram-negative pathogens of clinical (3) and environmental (4, 5) origin.
During the six months from early February to the end of July 2007, 111 clinical isolates of Gram-negative bacilli including E. coli (n= 72), Enterobacter sp. (n= 11), P. aeruginosa (n= 12), Klebsiella pneumoniae (n= 5), Burkholderia cepacia (n= 2), Acinetobacter baumannii (n= 2), Salmonella spp. (n= 1), Serratia marcescens (n= 1), Proteus mirabilis (n= 1), Cedecea dovisae (n= 1) and nonidentified Gram negative bacilli (n= 3) were isolated from clinical specimens (including urine, sputum, bronchoalveolar lavage and wounds) of unrelated outpatients with various infections attending various clinics at the 200-bed 82nd Year of Government General Hospital, Rize, Turkey. All bacterial strains were identified to the species level by the biochemical reactions described by Brenner (6).
Strains were tested for antimicrobial resistance by the disc diffusion method (7) using discs supplied by Oxoid (Basingstoke, UK) including AMP (10 μg), CFZ (30 μg), CXM (30 μg), CAZ (30 μg), FEP (30 μg), IPM (10 μg), STR (10 μg), GEN (10 μg), CIP (5 μg), SXT (1.25 μg/23.75 μg), TET (10 μg), CHL (30 μg). The results were separately interpreted by using the breakpoints from the CLSI guidelines for the family Enterobacteriaceae and non-fermenters such as P. aeruginosa and Acinetobacter spp. (7).
To prepare DNA templates for PCR, 1.5 ml of the overnight culture of bacterial isolates was harvested by micro-centrifugation. After decanting the supernatant, the pellet was re-suspended in deionized water. The cells were lysed by boiling, and the debris removed by centrifugation. One μl of supernatant was used as template for PCR amplifications. All PCR reactions were carried out in a Mastercycler Personal thermal cycler (Eppendorf, Hauppauge, NY, USA) using Taq polymerase, nucleotides and buffers purchased from MBI Fermentas (Vilnius, Lithuania). All isolates were screened for integrons with the specific oligonucleotide primers 5′-CS; 5′-GGCATCCAAGCAGCAAG-3′ and 3′-CS; 5′-AAGCAGACTTGACCTGA-3′, amplifying the variable regions of class 1 (8) and class 2 integrons with the primers hep51; 5′-GATGCCATCGCAAGTACGAG-3′ and hep74; 5′-CGGGATCCCGGACGGATGCACGATTTGTA-3′ (9). Reaction compositions and cycling parameters were performed as previously described (8, 9). The PCR amplicons were then electrophoresed on 1% agarose gel stained with 0.5 μg/ml ethidium bromide and visualized under ultraviolet light.
After the PCR products had been purified from the agarose gel by using QIAQuick Purification Kits (Qiagen, Cologne, Germany) prior to sequencing, they were cloned into the pGEMT Easy vector according to the manufacturer's instructions (Promega, Madison, WI, USA). Recombinant plasmids carrying amplicons of integrons were sent to Macrogen (Seoul, Korea) for sequencing by using two promoter primers (SP6 and T7) complementary to the sequence of the plasmid vector pGEM-T Easy. Data from sequencing was compared with those available in the GenBank database by using the alignment search tool, BLAST, accessible from the National Center for Biotechnology Information website (http://www.ncbi.nlm.nih.gov/BLAST), and by the multiple sequence alignment program, CLUSTAL W, accessible from the European Bioinformatics Institute website (http://www.ebi.ac.uk/clustalw).
For PFGE, genomic DNA was digested with XbaI for 2 hr at 37°C and separated in a 1% agarose gel with a contour-clamped homogeneous electric field apparatus (CHEF-DR2; Bio-Rad, Hercules, CA, USA). The conditions for electrophoresis were 6 volts/cm at 14°C for17 hr, with the switch time ramped from 2.2 to 63.8 s. Pulsed-field gel electrophoresis patterns were interpreted by using the criteria established by Tenover et al. (10).
The χ2 test was performed for comparing susceptibility results integron-positive and –negative isolates by using SPSS for Windows (Version 129 11.0 Software).
The frequency of resistance of the 111 clinical isolates to each of the antibiotics tested can be observed in Table 1. Resistance to aminoglycosides such as STR (50.4%) and GEN (24.3%) was frequent. The lowest resistance rates were for IPM (1.8%) and CHL (10.8%). Resistance to TET (40.5%), SXT (54.9%) and CIP (19.8%) was also observed. In Turkey, carbapenems (e.g. IPM), extended-spectrum β-lactams (e.g. CAZ and FEP), aminoglycosides (e.g. GEN) are often used in the treatment of hospital- and community-acquired infections due to Enterobacteriaceae and non-fermenters (11, 12).
Table 1. Association between antibiotic profile and integrons in 111 Gram-negative bacilli isolated from clinical specimens
|Antibiotic||Antibiotic susceptibility (n= 111)||Integron-positive isolates (n= 32)||Integron-negative isolates (n= 79)||P valueb|
|% R (no.)||% I (no.)||% S (no.)||% R (no.)||% I (no.)||% S (no.)||% R (no.)||% I (no.)||% S (no.)|
|CFZ||40.5 (45)|| 3.6 (4)||55.8 (62)||40.6 (13)|| 6.2 (2)|| 53.1 (17)||40.5 (32)|| 2.5 (2)||56.9 (45)||0.565|
|CXM||38.7 (43)|| 1.8 (2)||59.4 (66)||40.6 (13)|| 0.0 (0)|| 59.3 (19)||37.9 (30)|| 2.5 (2)||59.4 (47)||0.211|
|CAZ||17.1 (19)||14.4 (16)||68.4 (76)||18.7 (6)|| 6.2 (2)|| 75.0 (24)||16.4 (13)||17.2 (14)||65.8 (52)||0.048|
|FEP||12.6 (14)|| 2.7 (3)||84.6 (94)||12.5 (4)|| 0.0 (0)|| 87.5 (28)||12.6 (10)|| 3.7 (3)||83.5 (66)||0.129|
|IPM|| 1.8 (2)|| 0.9 (1)||97.2 (108)|| 0.0 (0)|| 0.0 (0)||100.0 (0)|| 2.5 (2)|| 1.2 (1)||96.2 (76)||0.130|
|STR||50.4 (56)||10.8 (12)||61.2 (68)||84.3 (27)|| 6.2 (2)|| 9.3 (3)||36.7 (29)||11.3 (9)||51.8 (41)||0.001|
|GEN||24.3 (27)|| 4.5 (5)||71.1 (79)||31.2 (10)|| 6.2 (2)|| 62.5 (20)||21.5 (17)|| 3.7 (3)||74.6 (59)||0.227|
|CIP||19.8 (22)|| 0.9 (1)||79.2 (88)||25.0 (8)|| 3.1 (1)|| 71.8 (23)||17.7 (14)|| 0.0 (0)||82.2 (65)||0.091|
|SXT||54.9 (61)|| 0.0 (0)||45.0 (50)||93.7 (30)|| 0.0 (0)|| 2.5 (2)||39.2 (31)|| 0.0 (0)||60.7 (48)||0.001|
|TET||40.5 (45)|| 9.9 (11)||49.5 (55)||50.0 (16)||12.5 (4)|| 40.6 (13)||36.7 (29)|| 8.8 (7)||54.4 (43)||0.112|
|CHL||10.8 (12)|| 6.3 (7)||82.8 (92)||12.5 (4)||15.6 (5)|| 71.8 (23)||10.1 (8)|| 2.5 (2)||87.3 (69)||0.005|
Carriage rates of class 1 integron in 32 (28.8%) and class 2 integron in 1 (0.9%) of 111 clinical isolates were significantly associated with resistance to the following antibiotics: extended-spectrum β-lactam (e.g. CAZ) (P < 0.05), aminoglycosides (e.g. STR) (P < 0.01), sulfonamides (e.g. SXT) (P < 0.01) and CHL (P < 0.01). There were no significant differences (P > 0.05) in terms of integron-positive and –negative isolates for the other antibiotic groups listed in Table 1. Multiple antibiotic resistance, defined here as resistance to five or more antibiotics, has been reported to be correlated with the presence of integrons, and many antibiotic resistance gene cassettes encoding resistance to a wide range of antibiotics have previously been reported (13, 14).
Our results are in general agreement with those reported by Severino and Magalhaes (15). They found that integron-positive isolates were significantly more often resistant to aminoglycosides, quinolones and β-lactams than were integron-negative isolates. Although we found an association between frequency of resistance to CAZ and integron carriage (P < 0.05) (Table 1), no extended-spectrum β-lactamase-producing strains were detected amongst CAZ-resistant organisms by using the double-disc synergy test (16). These prospective results raise the hypothesis that much antibiotic resistance to a wide range of antibiotics is closely related to carriage of integrons. In this study, we detected many gene cassettes, including aadA1, aadA2 and aadA5, that encode the enzyme aminoglycoside adenyltransferase, which confers resistance to STR and spectinomycin. We also detected the dfrA1, dfrA5, dfrA7 and dfrA17 genes that encode the enzyme dihydrofolate reductase, which confers resistance to trimethoprim. All gene cassette arrays were inserted in class 1 and 2 integrons carried at different frequencies by clinical Enterobacteriaceae and P. aeruginosa of community origin. However, it has been reported that healthy humans can also carry pathogens harboring class 1 integrons (17).
Herein, we describe a new integron-associated gene cassette allele, dfrA17-like, (GenBank accession numbers FJ001844, FJ001850 and FJ001851) which was identified in three E. coli strains (D2, D39 and D44), two of them in urine from patients with urinary tract infection, while the third specimen was of unknown clinical origin. All three strains were isolated from unrelated outpatients attending the pediatric clinic (Table 2). We found that this dfrA17-like gene encodes a putative dihydrofolate reductase-like enzyme conferring resistance to the antibiotic trimethoprim, but differs in its amino acid sequence from the other DHFRXVII protein in the GenBank database, as shown in Figure 1. Furthermore, according to the PFGE results (PFGE types VI, VII and VIII), this new allele inserted in a class 1 integron was harbored in clonally different E. coli isolates (Table 2). In the same region in which the current study was conducted, another novel gene cassette allele, aadA7-like (GenBank accession number DQ899757) has been identified in clonally spreading P. aeruginosa strains isolated from patients hospitalized in intensive care units in a university hospital (3).
Table 2. Epidemiological data for 32 integron-bearing clinical isolates from outpatients attending clinics
|Strain||Organism||R type||Policlinic||Specimen||Date of isolationb||PFGE group||Integron or gene cassette array/size (bp)||GenBank. Accession No.|
|D2||E. coli||AMP STR SXT||Pediatrics||Urine||06/25/2007||VI||dfrA17-like/767||FJ001844|
|D3||E. coli||AMP TET STR GEN SXT||Pediatrics||Urine||07/24/2007||II||dfrA17-aadA5/1664||FJ001843|
|D4||E. coli||AMP STR CAZ||Gynecology||Urine||04/30/2007||III||dfrA17-aadA5/1664||FJ001845|
|D8||E. coli||AMP CXM TET STR GEN SXT CHL CAZ CFZ||Urology||Urine||07/14/2007||ND||aadA2/593||FJ001852|
|D14||E. coli||AMP STR SXT CAZ||Pediatrics||Urine||07/30/2007||I||dfrA17-aadA5/1664||FJ001846|
|D18||P. aeruginosa||AMP CXM TET STR GEN SXT C CFZ||Thoracic||Sputum||06/20/2007||ND||aadA2/593||FJ001854|
|D21||E. coli||AMP CXM STR SXT||Pediatrics||Urine||02/13/2007||I||dfrA17-aadA5/1664||FJ001847|
|D23||E. coli||AMP TET STR SXT||Thoracic||Sputum||11/22/2006||ND||dfrA17-aadA5/1664||FJ001848|
|D26||E. coli||AMP STR SXT CZ||Gynecology||Urine||02/15/2007||I||dfrA17-aadA5/1664||FJ001849|
|D28||E. coli||AMP TET STR SXT||Unknown||Unknown||Unknown||V||dfrA1- aadA1/1586||FJ001874|
|D36||E. coli||AMP TET STR SXT CIP CAZ CFZ||Pediatrics||Urine||05/25/2007||III||dfrA1- aadA1/1586||FJ001855|
|D39||E. coli||AMP FEP CXM TET SXT CIP CAZ CFZ||Unknown||Unknown||Unknown||VII||dfrA17-like/767||FJ001850|
|D44||E. coli||AMP TET SXT||Gynecology||Urine||05/04/2007||VIII||dfrA17-like/767||FJ001851|
|D47||E. coli||AMP CXM TET STR GEN SXT CFZ||Pediatrics||Urine||07/13/2007||IX||dfrA5/721||FJ001870|
|D48||E. coli||AMP CXM STR SXT CIP CFZ||Gynecology||Urine||06/19/2007||X||dfrA1- aadA1/1586||FJ001856|
|D49||E. coli||AMP TET STR SXT CIP||Pediatrics||Urine||07/12/2007||IV||dfrA1- aadA1/1586||FJ001857|
|D51||E. coli||AMP STR SXT CHL||Pediatrics||Urine||05/02/2007||I||dfrA1- aadA1/1586||FJ001858|
|D56||P. aeruginosa||AMP CXM STR GEN SXT CHL CFZ||Urology||Urine||06/04/2007||ND||aadA2/593||FJ001869|
|D61||E. coli||AMP FEP CXM TET STR GEN SXT CFZ||Pediatrics||Urine||05/21/2007||II||dfrA1- aadA1/1586||FJ001859|
|D64||E. coli||AMP STR SXT||Gynecology||Urine||03/20/2007||ND||aadA1/1009||FJ001871|
|D66||E. coli||AMP CXM STR SXT CFZ||Pediatrics||Urine||02/15/2007||I||dfrA1- aadA1/1586||FJ001860|
|D69||E. coli||AMP FEP CXM TET STR GEN SXT CAZ CFZ||Pediatrics||Urine||06/15/2007||XI||dfrA1- aadA1/1586||FJ001861|
|D76||Salmonella sp.||AMP TET STR SXT CHL||Pediatrics||Feces||02/21/2007||ND||dhfr1-sat1-aadA/2224||FJ001872|
|D90||E. coli||AMP CXM TET STR SXT CIP CFZ||Unknown||unknown||06/18/2007||XII||dfrA1- aadA1/1586||FJ001862|
|D92||E. coli||AMP TET STR GEN SXT||Pediatrics||Urine||06/12/2007||II||dfrA1- aadA1/1586||FJ001863|
|D94||E. coli||AMP TET STR GEN SXT CIP CAZ||Urology||Urine||05/09/2007||IV||dfrA1- aadA1/1586||FJ001864|
|D95||E. coli||AMP STR SXT CHL CFZ||Pediatrics||Urine||05/02/2007||XIII||dfrA1- aadA1/1586||FJ001865|
|D100||P. aeruginosa||AMP CXM TET STR GEN SXT CHL CFZ||Thoracic||BAL||06/07/2007||ND||aadA2/593||FJ001853|
|D101||E. coli||AMP TET STR SXT CIP||Pediatrics||Urine||06/07/2007||ND||dfrA1- aadA1/1586||FJ001866|
|D104||Enterobacter sp.||AMP TE STR GEN SXT CIP||Thoracic||Urine||05/28/2007||ND||dfrA1- aadA1/1586||FJ001867|
|D110||E. coli||AMP FEP CXM TET STR GEN SXT CIP CHL CAZ CFZ||Surgery||Wound||07/08/2007||XIV||dfrA1- aadA1/1586||FJ001868|
|D113||E. coli||AMP STR SXT CHL||Pediatrics||Urine||05/02/2007||XV||dfrA7/769||FJ001873|
Figure 1. Alignment of the putative dihydrofolate reductase-like gene, dfrA17-like inserted in a 767-bp class 1 integron (GenBank accession number FJ001844) with the derived amino acid sequences of various DHFRXVII proteins under the GenBank accession numbers AM412236, AM937244, EU081911, DQ838665, and AF475280. Shaded boxes indicate conserved amino acids.
Download figure to PowerPoint
We analyzed 23 of the 27 E. coli strains carrying class 1 integrons and detected 12 PFGE types (Fig. 2): type I (n= 5), type II (n= 3), type III (n= 2) and type IV (n= 2). The types, V to XV (n= 11) patterns were indistinguishable. It is noteworthy that most of epidemiologically related E. coli strains carrying class 1 integrons were urinary isolates and cultured from outpatients attending pediatric or gynecology clinics (Table 2). Moreover, there have been several reports (18, 19) that class 1 and 2 integrons are widespread in uropathogenic E. coli isolates from urinary samples of outpatients.
Figure 2. PFGE profiles of XbaI digestion of whole-cell DNA of epidemic and non-epidemic strains of E. coli isolates. Lane MWM, molecular weight marker (50 Kbp Ladder, Bio-Rad).
Download figure to PowerPoint
Consequently, integrons will continue to threaten the usefulness of antibiotics in hospital- and community-acquired infections by capturing and collecting new cassette alleles.