Characterization of Trimethoprim-Sulfamethoxazole Resistance Genes and Their Relatedness to Class 1 Integron and Insertion Sequence Common Region in Gram-Negative Bacilli

Abstract

Trimethoprim-sulfamethoxazole (TMP-SMX) has been used for the treatment of urinary tract infections, but increasing resistance to TMP-SMX has been reported. In this study, we analyzed TMP-SMX resistance genes and their relatedness with integrons and insertion sequence common regions (ISCRs) in uropathogenic gram-negative bacilli. Consecutive nonduplicate TMP-SMX nonsusceptible clinical isolates of E. coli, K. pneumoniae, Acinetobacter spp., and P. aeruginosa were collected from urine. The minimal inhibitory concentration was determined by Etest. TMP-SMX resistance genes (sul and dfr), integrons, and ISCRs were analyzed by PCR and sequencing. A total of 45 E. coli (37.8%), 15 K. pneumoniae (18.5%), 12 Acinetobacter spp. (70.6%), and 9 Pseudomonas aeruginosa (30.0%) isolates were found to be resistant to TMP-SMX. Their MICs were all over 640. In E. coli and K. pneumoniae, sul1 and dfr genes were highly prevalent in relation with integron1. The sul3 gene was detected in E. coli. However, in P. aeruginosa and Acinetobacter spp., only sul1 was prevalent in relation with class 1 integron; however, dfr was not detected and sul2 was less prevalent than in Enterobacteriaceae. ISCR1 and/or ISCR2 were detected in E. coli, K. pneumoniae, and Acinetobacter spp. but the relatedness with TMP-SMX resistance genes was not prominent. ISCR14 was detected in six isolates of E. coli. In conclusion, resistance mechanisms for TMP-SMX were different between Enterobacteriaceae and glucose non-fermenting gram-negative bacilli. Class 1 integron was widely disseminated in uropathogenic gram-negative baciili, so the adoption of prudent use of antimicrobial agents and the establishment of a surveillance system are needed.

Introduction

Urinary tract infections (UTIs) have been reported to be the most common hospital acquired infection, which are associated with significant morbidity and mortality [9,10]. The predominant causative pathogen of UTIs is Escherichia coli; however, there have been increasing reports of other Enterobacteriaceae such as Klebsiella pneumoniae and gram-negative non-fermenters such as Acinetobacter spp., and Pseudomonas aeruginosa as causes of UTIs [5,7,15]. Trimethoprim-sulfamethoxazole (TMP-SMX) has been used for several decades as efficient antibiotics for the treatment of UTIs [11]. In many countries, however, the presence of resistance to TMP-SMX can lead to treatment failure in cases of UTIs [9]. Sulfonamide resistance in gram-negative bacilli generally arises from the acquisition of dihydropteroate synthase (DHPS) genes in integrons that are not inhibited by the drug [11]. Currently, three different types of DHPS genes (sul1, sul2, and sul3) are known [9]. The sul1 gene is found linked to other resistance genes in class 1 integrons and on large conjugative plasmids [25], while sul2 is usually located on small nonconjugative plasmids [21], large transmissible multiresistance plasmids [9], or through insertion element common region (ISCR2) element [24]. Although rare, sul3, a plasmid-borne sulfonamide resistance gene, is also present [25].

TMP affects bacterial folic acid synthesis by the inhibition of dihydrofolate reductase (DHFR), which catalyzes the reduction of dihydrofolate to tetrahydrofolate [11]. There are several mechanisms of TMP resistance, such as development of permeability barriers, efflux pumps, existence of naturally insensitive target DHFR enzymes, mutational and regulation changes in target enzymes, and the acquirement of drug-resistant target enzymes [11]. Among them, the acquirement of DHFR variants encoded by dfr genes is the most common mechanism for TMP resistance, which results in high-level resistance in various bacteria [20]. To date, more than 30 different dfr genes are known, which are usually found in gene cassettes within integrons [3,20] and are also associated with ISCR1 [12].

Although several literatures studied sul and/or dfr genes in relation to class 1 integron in E. coli [9,13,19,20], there are limited reports investigating the prevalence of TMP-SMX resistance genes in relation to integrons and ISCRs in other Enterobacteriaceae such as Klebsiella pneumoniae and gram-negative glucose non-fermenters such as Acinetobacter spp., and P. aeruginosa in Korea. Therefore, in this present study, we investigated the prevalence of TMP-SMX resistance and sul genes and dfr genes in various uropathogenic gram-negative bacilli and their association with class 1 integrons and ISCRs.

 

Materials and Methods

Bacterial Isolates

Consecutive non-duplicate TMP-SMX nonsusceptible clinical isolates of E. coli, K. pneumoniae, Acinetobacter spp., and P. aeruginosa were obtained from urine specimens collected at Chungnam National University Hospital from September 2011 to September 2013. All isolates were identified using the Vitek 2automated ID system (bioMèrieux Vitek Inc., Hazelwood, MO, USA).

Antimicrobial Susceptibility Testing

The MIC for trimethoprim-sulfamethoxazole was determined by the Etest, conducted in accordance with the Clinical and Laboratory Standards Institute (CLSI) guidelines [16]. E. coli ATCC 25923 and Pseudomonas aeruginosa ATCC 27853 were used as control strains.

PCR Amplification and Sequencing for sul Genes and ISCRs

Bacterial genomic DNA was obtained from each target strain by using a genomic DNA extraction kit (Bioneer, Daejeon, Korea) according to the manufacturer’s instructions.

Bacterial genomic DNA was amplified by PCR. The PCR was performed using 50 ng of bacterial whole DNA, 2.5 µl of 10× Taq buffer, 0.5 µl of 10 mM dNTP mix, 20 pmol of each primer, and 0.7 U of Taq DNA polymerase (SolGent, Daejeon, Korea), in a total volume of 25 µl. Each target site was amplified in a SEEAMP (Seegene, Seoul, Korea). The amplified products were separated by electrophoresis on 1.5% (w/v) agarose gels containing ethidium bromide, and visualized using a BioDoc-14 imaging system (UVP, Cambridge, UK). The amplicons were purified with a PCR purification kit (SolGent), and sequenced using a BigDye Terminator Cycle Sequencing kit (PE Applied Biosystems, Foster City, CA, USA) and an ABI PRISM 3730XL DNA analyzer (PE Applied Biosystems).

The following primers were used for the PCR and sequencing of sul genes: sul1F (5’-CTTCGATGAGAGCCGGCGGC-3’) and sul1R (5’-GCAAGGCGGAAACCCGCGCC-3’) [1], sul2F (5’-GCGCTCAAGGCAGATGGCATT-3’) and sul2R (5’-GCGTTTGATACCGGCACCCGT-3’) [1], and sul3F (5’-GAGCAAGATTTTTGGAATCG-3’) and sul3R (5’-CTAACCTAGGGCTTTGGATAT-3’) [18]. For detection of all ISCRs, the following primers were used: CRF (5’-CACTWCCACATGCTGTKKC-3’) and CRFF-r (5’-CGCTTGAGSCGTTGCRYCC-3’) [24].

Identification of Integrons and dfr Genes

Class 1 integrons were amplified using the primers hep58 (5’-TCATGGCTTGTTATGACTGT-3’) and hep59 (5’-GTAGGGCTT ATTATGCACGC-3’). Class 2 integrons were amplified using the primers hep51 (5’-GATGCCATCGCAAGTACGAG-3’) and hep74 (5’-CGGGATCCCGGACGGCATGCACGATTTGTA-3’) [26]. Class 3 integrons were amplified using the primers Int3F (5’-GCCTCCGGCAGCGACTTTCAG-3’) and Int3R (5’-ACGGATCTGCCA AACCTGACT-3’) [22].

Only trimethoprim resistance genes, dfr, located inside class 1 integrons were characterized. Sequencing of purified class 1 integron amplicons were performed using the forward primer hep58. When sequencing with the forward primer revealed no dfr alleles, samples were resubmitted with the reverse primer to provide a complete sequence. Class 1 integron amplicons sequence were compared with published dfr allele sequences from GenBank.

 

Results

Bacterial Isolates and Antimicrobial Susceptibility Testing

During the study period, 119 E. coli, 81 K. pneumoniae, 17 Acinetobacter spp. isolates, and 29 P. aeruginosa isolates were collected from urine specimens. Among these isolates, 45 E. coli (37.8%), 15 K. pneumoniae (18.5%), 12 Acinetobacter spp. (70.6%), and 9 Pseudomonas aeruginosa (30.0%) isolates were found to be resistant to TMP-SMX. MIC90 values were >640 in all of the isolates.

Characterization of TMP-SMX Resistance Genes and Their Relatedness to Class 1 Integrons and ISCRs

In 45 E. coli isolates, 10 isolates contained only the sul1 gene and 10 other isolates contained only the sul2 gene. Twenty-four isolates carried both the sul1 and sul2 genes (Table 1). Class 1 gene cassettes were detected in 34 (75.6%) of 45 isolates. Various gene cassette arrays of dfrA17–aadA5 (27 isolates), dfrA12-aadA2 (3 isolates), dfrA1-aadA1 (2 isolates), aacA4-arr3-dfrA27 (1 isolate), aadA2 (1 isolate) were found and are shown in Table 2. Among 34 isolates with the sul1 gene, 33 isolates (97.1%) were found to carry class 1 integrons. Thirty-four isolates (75.6%) carried dfr genes, most of which were found within gene cassette arrays of class 1 integron. An ISCR was detected in 4 isolates with sul1, 1 isolate with sul2, and 5 isolates with sul1 and sul2. Among 10 isolates with an ISCR, one isolate with ISCR1 had class 1 integron, and 2 isolates with ISCR2 had class 1 integron and sul1 and/or sul2. One isolate had ISCR3, and 6 isolates contained ISCR14 (Table 3).

Table 1.Prevalence of sul and dfr genes among trimethoprim-sulfamethoxazole resistant isolates from urine specimens.

Table 2.Gene cassette arrays in class 1 integrons from trimethoprim-sulfamethoxazole resistant isolates from urine specimens.

Table 3.Distribution of ISCR, integrons, and sul genes among trimethoprim-sulfamethoxazole resistant isolates from urine specimens.

In 15 K. pneumoniae isolates, sul1 and sul2 were detected in 2 and in 5 isolates, respectively, and 8 isolates contained both sul1 and sul2 (Table 1). All isolates with sul1 had class 1 integron with various gene cassette arrays of dfrA1-ofrC (6), aadA2 (2), aac(6')-Ib-oxa-1-aadA2 (1), and dfrA1-aadA1 (1) (Table 2). Eight isolates contained dfr genes, all of which were within the gene cassette array of class 1 integron. ISCR2 was detected in one isolate with the sul2 gene.

In 13 Acinetobacter spp., 9 and 3 isolates had sul1 and sul2, respectively, and one isolate had both sul1 and sul2 (Table 1). All 9 isolates with sul1 carried class 1 integron with gene cassette arrays with aacA4-catB8-aadA1 (6) and aacA44-IMP1-oxa-2 (3) (Table 2). Three and 2 isolates had ISCR1 and ISCR2, respectively, and 2 isolates with ISCR1 contained the class 1 integon and sul1 gene, and 1 isolate with ISCR2 contained the sul2 gene.

In 9 P. aeruginosa isolates, 9 isolates contained sul1 (Table 1). Among those isolates, 8 carried the class 1 integron with cassette arrays of aadB–cmlA-oxa-10-aadA1 (7) and aadA2 (1) (Table 2). No isolates were found to carry the sul2 gene and ISCR.

The dfr gene was not found in the any Acinetobacter spp. and P. aeruginosa isolates.

 

Discussion

Recent rise in TMP-SMX resistance is thought to be due to horizontal gene transfer and clonal expansion [2]. In this study, we evaluated genes related to TMP-SMX resistance and their relatedness with the class 1 integron and ISCR in uropathogenic gram-negative bacilli.

According to our study, the prevalence of the class 1 integron in TMP-SMX-resistant E. coli isolates was 73.3% (34 out of 45 isolates), which was increased in comparison with previous reports in Korea [13,28]. The most prevalent class 1 gene cassette array was drfA17-aadA5 (79.4%), which was found to be the most prevalent one in Korea [13,28] and also in China [23], but different from the gene cassette array found in other parts of the world [2]. Class 1 integrons carrying a single gene cassette were prevalent in clinical E. coli isolates from the 1980s, whereas class 1 integrons carrying multigene cassettes were prevalent in clinical isolates from the 1990s and 2000s in Korea [13,28], indicating that class 1 integrons in E. coli facilitated acquisition of resistance to a broad spectrum of antibiotic agents [23]. Therefore, it was found that the class 1 integron with multiple resistance genes was more widely distributed in Korea, which directly contributed to the resistance not only to trimethoprim and sulfamethoxazole, but also to other antibiotics. The TMP-SMX resistance genes sul1, sul2, and dfr were found to be widely disseminated in E. coli (Table 1). Isolates with sul1 and dfr genes showed significant relatedness with class 1 integron in accordance with other studies [2,8,13,20,23], whereas sul2 did not show any relationship with class 1 integron, suggesting other transfer mechanisms. Interestingly, in most of the E. coli isolates in our study, the allele frequencies of the sul1 and sul2 genes were exactly the same. This finding was somewhat different from the previous literature in which the gene frequency distribution was reported as sul2 > sul1 > sul3 [2]. However, in recent study, the pattern of increased sul1 frequency over sul2 was observed [9]. This might be caused by wide dissemination of the class 1 integron, which is in close relationship with the sul1 gene. The sul3 gene was also detected in association with class 1 integron (aadA2), which was the first report in South Korea.

With K. pneumoniae, the prevalence of sul1 and sul2 genes was similar to that in E. coli. The class 1 integron was also highly prevalent in K. pneumoniae (73.3%) and was strongly associated with the sul1 and dfr genes as in E. coli, while different gene cassette arrays were shown. The dominant gene cassette array was dfrA1-ofrC, which was similar in other literature [14]. Thus, class 1 integrons with various gene cassette arrays in association with sul1 and dfr genes were highly prevalent in Enterobacteriaceae, and the variation of the gene cassettes in class 1 integrons may reflect the horizontal transfer of integrons among members of the Enterobacteriaceae family [14].

In contrast to Enterobacteriaceae, the prevalence of sul1 was much higher than that of sul2 in glucose non-fermenting gram-negative bacilli. Class 1 integron was also widely disseminated with different gene cassette arrays among different species as in Enterobacteriaceae, but dfr genes were not detected in any of the isolates. Thus, the mechanism of resistance to TMP-SMX in gram-negative non-fermenting bacilli was significantly different from Enterobacteriaceae.

The ISCR has been found in numerous gram-negative organisms and identified as being closely associated with the spread of many antibiotic resistance genes. ISCR elements in relation to TMP-SMX resistance genes are ISCR1 in relation to dfrA1, 12, and 17 gene cassettes in class 1 integrons, and ISCR2 in relation to the sul2 gene [26]. In our study, the prevalence of ISCRs in E. coli was 22.2% (10 out of 45 isolates), 13.3% (2 out of 15) in K. pneumoniae, and 38.5% (5 out of 13) in Acinetobacter spp. Among 6 isolates with ISCR2, 3 isolates harbored sul2, but among 3 isolates with ISCR1, we could not find related gene cassettes of dfrA1, 12, and 17. Therefore, in terms of TMP-SMX resistance, ISCRs did not show a strong association than class 1 integron did in our study. Interestingly, ISCR14, known to be related to rmtD and erm(B), was detected in 6 E. coli isolates [17]. In K. pneumoniae and P. aeruginosa, rmtD, a 16S ribosomal RNA methyltransferase gene conferring high level of resistance to aminoglycosides, was found [4,27]. The existence of ISCR14 raises the possibility of the presence of rmtD in E. coli. Consequently, further study is warranted to uncover the possibility of ISCRs for carrier of resistance genes, especially in E. coli and other gram-negative bacilli.

In conclusion, resistance to TMP-SMX in Enterobacteriaceae, E. coli, and K. pneumoniae was explained by the acquisition of sul1, sul2, and dfr genes. In most of the cases, class 1 integrons with various multigene cassette arrays in association with sul1 and dfr genes were widely disseminated in uropathogenic gram-negative bacilli. Dominant gene cassette arrays were different between E. coli and K. pneumoniae. On the other hand, resistance to TMP-SMX in glucose non-fermenting gram-negative bacilli, P. aeruginosa and Acinetobacter spp., was explained by either the sul1 gene in P. aeruginosa or the sul1 and sul2 genes in Acinetobacter spp. Class 1 integrons were also widely distributed among these isolates in relation to the sul1 gene. Dominant gene cassette arrays were also different among species. Interestingly, the dfr gene, which was observed in relation to the class 1 integron in Enterobacteriaceae, was not observed in any isolates of glucose non-fermenting gram-negative bacilli. Several ISCRs were observed in Enterobacteriaceae and glucose non-fermenting gram-negative bacilli, but we could not find strong relatedness between TMP-SMX resistance genes.

In summary, the class 1 integron carrying multigene was highly prevalent among gram-negative bacilli, but mechanisms of TMP-SMX resistance were different between Enterobacteriaceae and glucose non-fermenting gram-negative bacilli. The wide dissemination of integrons may be because of the horizontal transfer of antibiotic resistance gene cassettes, so the adoption of prudent use of antimicrobial agents and the establishment of a surveillance system are needed for further dissemination of the class 1 integron in gram-negative bacilli.