1. Background
Quinolones and fluoroquinolones are broad-spectrum antimicrobial agents that are used extensively in both medical and veterinary practice. Therefore, residues from these antibiotics are present in the environment (1). This widespread use is associated with an increased level of quinolone resistance, particularly within the last 10 years (2). Bacterial resistance to quinolones commonly results from chromosomal mutations (3, 4). However, several recent studies showed that quinolone resistance can also be mediated by plasmid-carried genes, so called plasmid-mediated quinolone resistance (PMQR) genes (2), which encode the Qnr proteins, which protect the quinolone targets (5); the Aac (6’)-Ib-cr enzyme, which acetylates not only aminoglycosides but also ciprofloxacin and norfloxacin (5); and QepA, a plasmid-encoded efflux pump (5).
The first plasmid-mediated quinolone resistance gene, named qnr, was reported in 1998. This gene encodes a 218-amino acid protein, which was later renamed QnrA; it is a member of the pentapeptide-repeat family. More recently, four additional proteins, QnrB, QnrS, QnrC, and QnrD, have been identified in several enterobacterial species (6, 7). These proteins interact with quinolones, topoisomerases, and DNA, and act by limiting the binding of quinolones to their targets (8). By itself, the qnr gene confers low-level resistance to quinolones. However, the presence of this gene facilitates the acquisition of high-level resistance among initially susceptible strains (9, 10).
In 2005, a second plasmid-mediated mechanism that independently contributes to quinolone resistance through modification of the antibiotic molecule was described; the encoded protein, Aac (6’)-Ib-cr, is a variant of 6’acetyl transferase, which is known to modify the chemical structure of aminoglycosides, and it confers broad spectrum resistance to ciprofloxacin and norfloxacin (11).
The third PMQR mechanism is mediated by QepA, an efflux pump belonging to the major facilitator subfamily, which pumps fluoroquinolones out of bacterial cells (12, 13).
PMQR determinants have been identified worldwide, with varying prevalence rates (9). Their presence is associated with resistance to other antimicrobial agents, particularly to β-lactams (2).
In Aleppo, there have been no studies published on the incidence of ciprofloxacin resistance and its association with PMQR genes.
2. Objectives
The aim of the current study is to determine the prevalence of plasmid-mediated quinolone resistance genes in a collection of ESBL-producing isolates of E. coli and K. pneumonia in Aleppo, Syria.
3. Materials and Methods
3.1. Bacterial Isolates
A total of 123 non-duplicate ESBL-producing isolates (99 E. coli and 24 K. pneumoniae isolates) were obtained from three university hospitals offering tertiary medical care in Aleppo city between October 2010 and June 2011 (14). These isolates were stored in microvials (Microbank™) at -80°C.
3.2. Antimicrobial Susceptibility Testing
A broth microdilution method was used to determine the minimum inhibitory concentration (MIC), in accordance with the Clinical and Laboratory Standards Institute (CLSI; formerly NCCLS) guidelines (15). Each bacterial suspension was adjusted to 1 McFarland standard, which contains approximately 1.5 × 108 cfu/mL, and then diluted with broth to a final density of 5 × 105 cfu/well, and inoculated onto microplates containing the test drug at various concentrations. Each plate was incubated at 35 - 37°C for 18 hours. The reference strains E. coli ATCC25922 and K. pneumoniae ATCC700603 were included in each run as controls. The MIC breakpoints used for susceptibility and resistance to ciprofloxacin were ≤ 1 μg/mL and ≥ 4 μg/mL, respectively.
3.3. Screening Procedures
Plasmids were extracted using the QIAprep Miniprep Kit (Qiagen), according to the manufacturer’s instructions. Screening for qnrA, qnrB, qnrS, aac (6’)-Ib, and qepA was performed by simplex polymerase chain reaction (PCR) with specific primers and PCR conditions that have been previously described (16-19).
Negative controls (without DNA template) were included in each run. Amplification products were identified by their sizes on 1.5% agarose gels after electrophoresis at 130 V for 25 min and staining with ethidium bromide.
All PCR products positive for aac (6’)-Ib were further analyzed by digestion with BtsCI (Thermo scientific) to identify aac (6’)-Ib-cr, which lacks the BtsCI restriction site that is present in the wild-type gene (16). The wild-type aac (6’)-Ib PCR product yielded 210-bp and 272-bp fragments after restriction (18). Fragments were extracted from the gel by using the QIA prep Miniprep Kit (Qiagen).
4. Results
4.1. Phenotype Confirmation
The MIC of ciprofloxacin ranged from < 0.125 μg/mL to > 128 μg/mL. We found that 65.81% of tested isolates were resistant to ciprofloxacin. The MICs of the tested antimicrobial agents for the isolates are shown in Table 1.
Distribution of the Minimum Inhibitory Concentrations of Ciprofloxacin for Escherichia coli and Klebsiella pneumoniae Isolates
4.2. Prevalence of PMQR Genes
In total, 123 isolates were included in this study. Only 34.14% (42/123) were confirmed to have at least one of the three qnr genes. The prevalence of each PMQR gene is shown in Table 2.
| Species | Total No. of Isolates | No. of Positive Isolates for PMQR Genes | |||||
|---|---|---|---|---|---|---|---|
| qnrA | qnrB | qnrS | aac (6’)-Ib | aacv (6ʹ)-Ib-cr | qepA | ||
| Escherichia coli | 99 (80.5) | 0 | 18 (18.1) | 11 (11.1) | 80 (80.8) | 78 (78.78) | 0 |
| Klebsiella pneumoniae | 24 (19.5) | 0 | 12 (50) | 9 (37.5) | 18 (75) | 15 (62.5) | 0 |
| Total | 123 | 0 | 30 (24.4) | 20 (16.3) | 98 (79.7) | 93 (75.6) | 0 |
Prevalence of the Five Plasmid-Mediated Quinolone Resistance Determinants a
Qnr genes were detected more frequently in K. pneumoniae (62.5%, 15/24) than in E. coli (27.27%, 27/99). Both qnrB and qnrS were significantly more prevalent in K. pneumoniae isolates than in E. coli isolates. Aac (6’)-Ib-cr was the most prevalent PMQR gene in our isolate pool (75.6%, 93/123), and aac (6’)-Ib-cr accounted for 94% (93/98) of all aac (6’)-Ib genes detected. Neither qnrA nor qepA was detected.
The distribution of PMQR determinants among the tested isolates and their corresponding MICs are shown in Table 3.
| Genes | Isolates (n) | MIC μg/mL (%) | |||
|---|---|---|---|---|---|
| < 1/8 - 1/2 | 1 - 4 | 8 - 32 | 64 - 128 < | ||
| qnrB | |||||
| Escherichia coli | 18 | 4 (22.2) | 2 (11.1) | 3 (16.6) | 9 (50) |
| Klebsiella pneumoniae | 12 | 2 (16.6) | 3 (25) | 3 (25) | 4 (33.3) |
| qnrS | |||||
| Escherichia coli | 11 | 1 (9) | 2 (18.1) | 1 (9) | 7 (63.6) |
| Klebsiella pneumoniae | 9 | 1 (11.1) | 3 (33.3) | 0 | 5 (55.5) |
| aac (6’)-Ib-cr | |||||
| Escherichia coli | 78 | 18 (23) | 13 16.6) | 14 (17.9) | 33 (42.3) |
| Klebsiella pneumoniae | 15 | 1 (6.6) | 3 (20) | 2 (13.3) | 9 (60) |
| qnrB + qnrS | |||||
| Escherichia coli | 2 | 0 | 0 | 1 (50) | 1 (50) |
| Klebsiella pneumoniae | 6 | 1 (16.6) | 2 (33.3) | 0 | 3 (50) |
| qnrB + aac (6’)-Ib-cr | |||||
| Escherichia coli | 15 | 3 (20) | 2 (13.3) | 3 (20) | 7 (46.6) |
| Klebsiella pneumoniae | 8 | 0 | 2 (25) | 2 (25) | 4 (50) |
| qnrS + aac (6’)-Ib-cr | |||||
| Escherichia coli | 9 | 1 (11.1) | 1 (11.1) | 1 (11.1) | 6 (66.6) |
| Klebsiella pneumoniae | 6 | 0 | 2 (33.3) | 0 | 4 (66.6) |
| qnrB + qnrS + aac (6’)-Ib-cr | |||||
| Escherichia coli | 2 | 0 | 0 | 1 (50) | 1 (50) |
| Klebsiella pneumoniae | 4 | 0 | 1 (25) | 0 | 3 (75) |
Minimum Inhibitory Concentration (MIC) Values of Plasmid-Mediated Quinolone Resistance-Positive Isolates
5. Discussion
It has been more than 30 years since fluoroquinolones were first introduced in Syria. A Syrian study suggested that the resistance rate against ciprofloxacin has reached 39.1%, with upward trends in the use of fluoroquinolones in community and hospital settings (14). In the present study, 65.81% of the ESBL-producing isolates from Aleppo University Hospitals were resistant to ciprofloxacin. Therefore, we investigated the prevalence of PMQR determinants and analyzed their association with phenotypic ciprofloxacin resistance.
Previous studies reported that qnr genes were rare (20); however, in the present study, we found that the prevalence of qnr genes in our study was higher (34.14%) than that reported in other studies (21-23). Although the prevalence of each PMQR gene varied by species, in general, qnr genes were more prevalent in K. pneumoniae (62.50%, 15/24) than in E. coli (27.27%, 27/99), as was previously described in studies conducted in France (24), the United States (17), Spain (21), and China (25). The most frequently detected qnr gene was qnrB (24.4%), as has been reported in in other studies (23, 26), and we also noted an absence of qnrA, which has also been reported previously (1, 18, 23, 26-28).
Notably, there was no statistically significant association between qnr and ciprofloxacin resistance, and qnr genes were common among both ciprofloxacin-sensitive/intermediate isolates (28.5%, 12/42) and resistant isolates (37%, 30/81). Our findings agree with other reports demonstrating that qnr alone does not confer resistance to fluoroquinolones. However, its presence may facilitate the selection of additional chromosomal mechanisms, such as changes in DNA gyrase (gyrA) and/or topoisomerase IV (parC) genes (2, 10, 17, 26, 29), and the presence of qnr does not necessarily lead to MICs above the CLSI breakpoints for resistance to ciprofloxacin (30). Furthermore, using ciprofloxacin breakpoints as markers for detection may underestimate the prevalence of qnr genes, which raises concern for the undetected spread of these genes (23). Consequently, infections caused by qnr-positive isolates might be treated with quinolones, thus enhancing the selection of resistant mutants (2) and increasing the risk of therapeutic failure (23).
We noted the absence of qepA among the studied strains. The QepA efflux pump, first described in 2007 in two E. coli clinical isolates from Japan and Belgium (12, 13), has already been detected in France, with a new variant QepA2 (31). However, qepA is still very rare, except in China where two recent studies underlined the predominance of the qepA gene in enterobacterial strains isolated from food-producing animals. The most surprising finding of our study was the wide penetration of the aac (6’)-Ib-cr allele, which was more prevalent (75.6%) than the qnr genes (43.14%). Notably, aac (6’)-Ib-cr accounted for 94% (93/98) of the aac (6’)-Ib genes detected.
This high proportion of aac (6’)-Ib-cr/aac (6’)-Ib has also been reported in other studies (11, 25), and it probably reflects an extended emergence and ongoing dissemination of under detected aac (6’)-Ib-cr. Moreover, its presence as part of an integron cassette (11, 32) suggests that it could be widely mobile among plasmids.
Although the qnr genes were predominant in K. pneumoniae, aac (6’)-Ib-cr was the most prevalent PMQR gene in E. coli (78/99, 63.4%), and it was much less prevalent in K. pneumonia (12.20%, 15/24). These differences are in agreement with previous observations (16, 18, 23); however, the reason for these differences is not yet understood, since it is known that some plasmids can carry both aac (6’)-Ib-cr and qnrA genes (17).
To investigate the contribution of the aac (6’)-Ib-cr gene to ciprofloxacin resistance, we analyzed the relationship between the presence of aac (6’)-Ib-cr and resistance to ciprofloxacin. Our resistant isolates were significantly more frequently aac (6’)-Ib-cr–positive (81.8%, 66/81) than our sensitive/intermediate isolates (64.28%, 27/42). Thus, aac (6’)-Ib-cr was significantly associated with phenotypic ciprofloxacin resistance (P = 0.02).
There was no relationship between the presence of qnrA, qnrB, or qnrS and aac (6’)-Ib-cr; qnr genes were present in 33.33% (31/93) of the aac (6’)-Ib-cr-positive strains and 28.57% (10/35) of the aac (6’)-Ib-cr-negative strains, indicating that the qnr genes and aac (6’)-Ib-cr can circulate independently. This result is consistent with some previous results (16), but is in contrast to other results from China where the aac (6’)-Ib-cr variant was detected in 55.2% of qnr-positive in E. coli and K. pneumoniae isolates but in only 6% of qnr-negative isolates (25). In conclusion, our study showed that the prevalence of plasmid-mediated qnr and aac (6’)-Ib-cr quinolone resistance genes was high among Syrian clinical ESBL-producing isolates of E. coli and K. pneumonia, and that this association should be further studied in the future. However, the distribution of the aac (6’)-Ib-cr variant differed between the two species; it was detected more often in E. coli isolates than in K. pneumonia isolates, which is the reverse of that for qnr genes. Conversely, the qnr genes were more prevalent in K. pneumoniae than in E. coli, and qnrB was more prevalent than qnrA or qnrS, and the prevalence of ciprofloxacin resistance in our isolates was associated with the prevalence of the aac (6’)-Ib-cr variant.
Finally, it seems likely that the increasing use of fluoroquinolones over the last 10 years created an opportunity for the emergence of ciprofloxacin-resistant clinical isolates with PMQR determinants. Additional regional epidemiological data on antimicrobial resistance throughout Syria is needed to promote appropriate antimicrobial therapy and effective infection control. In addition, ensuring the use of antibiotics that are not substrates for aac (6’)-Ib-cr might reduce selection pressure for this variant but not for the qnr genes.
