Occurrence of plasmid-mediated quinolone resistance genes in Pseudomonas aeruginosa strains isolated from clinical specimens in southwest Iran: a multicentral study

This study aimed to assess the presence of qnrA, qnrB, qnrC, qnrD, qnrS, qepA, and aac(6′)-Ib-cr determinants as well as quinolone resistance pattern of clinical isolates of P. aeruginosa in Ahvaz, southwest Iran. A total of 185 clinical isolates of P. aeruginosa were collected from 5 university-affiliated hospitals in Ahvaz, southwest Iran. The disk diffusion method was applied to assess the quinolone resistance pattern. The presence of qnrA, qnrB, qnrC, qnrD, qnrS, qepA, and aac(6′)-Ib-cr genes was investigated by the polymerase chain reaction (PCR) method. Overall, 120 (64.9%) isolates were non-susceptible to quinolones. The most and the less quinolone resistance rates were observed against ciprofloxacin (59.4%) and ofloxacin (45.9%), respectively. The prevalence rates of qnr genes were as follows: qnrA (25.8%), qnrB (29.2%), and qnrS (20.8%). The qnrB gene was the most common type of qnr genes. The qnr genes were occurred in 37.5% (n = 45/120) of quinolne-resistant isolates, simultaneously. The qnrC, qnrD, qepA, and aac(6′)-Ib-cr genes were not recognized in any isolates. In conclusion, the ofloxacin was the most effective quinolone. This study was the first to shed light on the prevalence of PMQR genes among P. aeruginosa isolates in southwest Iran.

The qnr genes counteract with the blockage effects of quinolone antibiotics on the microbial enzymes such as topoisomerase II (DNA gyrase) and topoisomerase IV 10 . Other probable quinolone resistance mechanisms in P. aeruginosa include chromosomal mutations in quinolone resistance determining region (QRDR) of topoisomerase (parC/parE) and DNA gyrase (gyrA/gyrB) encoding genes, and mobile efflux pumps such as OqxAB [10][11][12][13][14] . Also, mutations of the regulatory genes that affect the permeability or efflux process are among the contributed quinolone resistance mechanisms 10,11 .
To the best of our knowledge, there are scarce data available on the prevalence of PMQR genes among clinical isolates of P. aeruginosa worldwide, especially in Iran. Thus, considering the importance of P. aeruginosa and the qnr genes, the present study aimed to assess the prevalence of qnrA, qnrB, qnrS, qnrC, qnrD, qepA, and aac(6′)-Ib-cr genes as well as the antibiotic resistance pattern of fluoroquinolone-resistant P. aeruginosa strains isolated from clinical specimens in Ahvaz city, southwest of Iran.
Resistance to quinolone compounds. Overall, 120 (64.9%) isolates were non-susceptible to quinolone compounds used in this study. The antibacterial susceptibility testing revealed that ofloxacin (54.1% susceptible) was the most effective drug compared to the other fluoroquinolone antibiotics. The highest resistance rates of isolates were against ciprofloxacin (59.4%) followed by levofloxacin (56.8%), norfloxacin and gatifloxacin (each 48.6%) ( Table 2). All 120 quinolone-resistant isolates were simultaneously resistant against 2 or more quinolones (Table 3).
Antibiotic resistance rates of quinolone-susceptible and quinolone-resistant isolates. The P.

Discussion
In the Iran, few studies on the incidence of quinolone-resistant P. aeruginosa and its plasmid-mediated mechanisms have been reported. In this regard, we designed the present study to the better understanding of molecular and epidemiological aspects of quinolone resistance pattern of clinical P. aeruginosa isolates in Ahvaz, southwest Iran. In this study, 185 clinical P. aeruginosa isolates were investigated. The most strains were isolated from wound (37.3%), urine (21.6%), and blood (19.4%). Also, the majority of them were collected from Burn, Urology, and ICU ward. In agreement with these findings, a previous report from Iran by Izadi Pour Jahromi et al. 15 from Iran, isolated the most P. aeruginosa strains from wound and urine. Also, they showed the more incidence of P. aeruginosa in Burn, Pediatric, and ICU wards. Also, Shahraki Zahedani et al. 16 reported the most prevalent P. aeruginosa isolates in urine (17.44%), wound (24.41 %), and blood (33.72%) samples.
This research revealed a total quinolone resistance rate of 64.9% among clinical P. aeruginosa isolates that was higher than the previous reports from Saudi Arabia (42.4%) and Egypt (57.2%) 13,17 . In this study, the resistance rates against five tested quinolones including gatifloxacin, norfloxacin, ciprofloxacin, ofloxacin, and levofloxacin varied from 45.9% to 59.4%. The antimicrobial susceptibility testing revealed that the most effective quinolone in the present study was ofloxacin, while more of our isolates were resistant to ciprofloxacin. In contrast to the current findings, Rajaei et al. 18 from Iran reported the ciprofloxacin as the most effective antibiotic against clinical P. aeruginosa isolates. In another study by Adwan et al. 14 from Palestine who investigated 11 clinical P. aeruginosa isolates, 100.0 % of them were resistant against norfloxacin and ciprofloxacin. El-Badawy et al. 13  www.nature.com/scientificreports/ gemifloxacin, and moxifloxacin against P. aeruginosa isolates, disclosed resistance rates ranging from 28.3% to 41.3%. In their study, the gemifloxacin with 28.3% and nalidixic acid with 41.3% resistance rates were the most and the less effective quinolones, respectively that were not evaluated in the current research.
In this study, although the resistance to all tested antibiotics was more than 50.0%, aztreonam, tobramycin, cefepime, and imipenem were among the most effective drugs against both quinolone-resistant and quinolone-susceptible strains. However, our isolates showed a high resistance rates (more than 70.0%) against other  www.nature.com/scientificreports/ aminoglycosides (gentamycin, amikacin), cephalosporins (ceftazidime, ceftriaxone), and penicillins (piperacillin, piperacillin/tazobactam). Previous studies from Iran have approved the high resistance rates of the P. aeruginosa to a wide range of antibiotics, which was similar to current results 19,20 . In contrast to these findings, Brzozowski et al. 21 from Poland reported a lower resistance rates for ciprofloxacin (39.1%), amikacin (30.7%), cefepime (42.6%), ceftazidime (33.2%), gentamycin (37.6%), piperacillin/tazobactam (39.6%), tobramycin (38.1%), and imipenem (67.8%) in clinical P. aeruginosa. These discrepancies could be due to differences in the geographical regions and diversity of antibiotic prescription patterns, as well as the lack of a comprehensive and organized monitoring program for the proper use of antibiotics in several countries. In recent years, the prevalence of carbapenem-resistant Gram-negative bacteria has increased worldwide. This is a unique clinical problem, as these drugs are long regarded to be the most active and powerful treatment for the infections caused by MDR bacteria. In the current study, 68.1% of P. aeruginosa isolates were resistant to imipenem. In line with our findings, previous reports by Farajzadeh Sheikh et al. 19 (90.7%) and Tarafdar et al. 20 (95.8%) from Iran, stated a high resistance rate against carbapenems in the clinical P. aeruginosa isolates. This may be due to the presence of metallo-β-lactamase or carbapenemase enzymes and upregulation of different efflux pumps in these strains 22 .
Another remarkable result of the current study was the high occurrence of MDR P. aeruginosa (78.4%) that was greater than previous indicated statistics from Iran (31.4%), Poland (48%) 28 , and Egypt (66.6%) 23 . However, the XDR rate (8.1%) was lower than a report by Shahraki Zahedani et al. 16 from Iran (12.3%). No PDR isolate was detected in this study. In another study by Pérez et al. 24 , who investigated 53 P. aeruginosa isolates from Greece, Italy, and Spain, a total of 30.2% MDR, 35.8% XDR, and 3.8% PDR strains were reported. In our region, incorrect antibiotic prescriptions might be a contributing factor to this higher prevalence of MDR isolates.
This study was the first report on the presence of the qnr genes in quinolone-resistant clinical P. aeruginosa isolates in patients from southwest Iran. According to our findings, 38.3% of isolates were positive for the qnr genes, from which 29.2%, 25.8%, and 20.8% isolates had the qnrB, qnrA, and qnrS genes, respectively. So far, despite the importance of the subject, few studies have addressed this issue.
In a study by Michalska et al. 25 from Poland, the qnrB with a frequency of 20.0% was the only detected gene among clinical P. aeruginosa isolates. In contrast to the current research, they did not find the qnrA and qnrS genes. Also, our findings significantly differed from the results reported by Molapour et al. 26 from Iran who did not find any qnr gene in 149 quinolone-resistant P. aeruginosa that were isolated from burn patients. In another study by Cayci et al. 27 from Turkey, qnrA, qnrB, and qnrS genes were not detected in P. aeruginosa isolates.
The qnrB gene was the most common qnr type in our region. However, a previous reports by Rajaei et al. 18 from Kerman city, Iran indicated the qnrA gene as the predominant PMQR (16.6%). Also, in comparison with our findings, they showed a lower prevalence rates for qnrB (13.3%) and qnrS (11.6%). Moreover, Saleh et al. 17 from Egypt, showed a lower prevalence rates for qnr genes in comparison with our findings. In their study, the total prevalence rate for qnr genes was equal to 4.5%. They detected the qnrB and qnrS in 1.8% and 2.7% of P. aeruginosa, respectively. El-Badawy et al. 13 from Saudi Arabia reported a much higher occurrence rate for qnrS (79.5%) than our study. Also, in the previous studies from Egypt, Iraq, and China, the qnrS has been recorded as the major quinolone resistant gene among MDR P. aeruginosa strains 17,28,29 .
In this study, the qnrC, qnrD, qepA, and aac(6′)-Ib-cr genes were not detected in studied isolates. In contrast to our report, El-Badawy et al. 13 and Jiang et al. 29 reported a prevalence of 79.5% and 0.9% for qnrD, respectively. However, previous studies from Saudi Arabia 13 , Egypt 17 , and China 29 were not detected the qepA gene that was in line with the current study. The aac(6′)-Ib-cr gene was first identified in 2003, and it has subsequently been found in a variety of regions and sources 30 . This gene has been reported in P. aeruginosa strains in previous studies from Saudi Arabia (71.8%) 13 , Iran (8.3%) 18 , and China (1.9%) 29 . Also, in a report by Molapour et al. 26 from Iran, all 149 P. aeruginosa isolates harbored aac(6′)-Ib-cr gene that was inconsistent with our result.
In the current study, the qnr genes were occurred in 37.5% (n = 45/120) of isolates, simultaneously. In line with these results, the coexistence of PMQR genes has been reported previously in P. aeruginosa isolates from various countries including Saudi Arabia and China 13,29 .
We found that 74 (61.7%) fluoroquinolone-resistant clinical isolates were negative for the qnrA, qnrB, , qnrC, qnrD, qnrS, qepA, and aac(6′)-Ib-cr genes. The quinolone resistance phenomenon in these isolates may be due to the other qnr genes like qnrE and qnrVC, or recently introduced crpP gene that encodes a ciprofloxacin modifying enzyme CrpP that were not investigated here 12,31 . After being discovered in a Brazilian Vibrio cholerae strain in 1998, qnrVC gene is now more often linked with bacteria that live in the aquatic environment 31,32 . This gene has different alleles. In two recent studies by Khan et al. 31 and Lin et al. 33 the prevalence rates of 12.0% and 2.3% were reported for this gene in clinical P. aeruginosa. Also, Khan et al. 31 reported a prevalence of 63.0% for crpP gene in corneal P. aeruginosa isolates. So far, these genes have not been investigated and reported in any bacteria from Iran.
As the majority of quinolone-resistant P. aeruginosa strains in this study lack the PMQR genes, it is recommended to investigate the other aforesaid mechanisms to shed light on the precise molecular epidemiology of these isolates.
In conclusion, considering that the antibiotic resistance profiles constantly differ in each area and hospital setting, the periodic surveillance program is very crucial for each country to determine the most appropriate treatment choices. Based on our results, the ofloxacin was the best quinolone for the treatment of P. aeruginosa infections. Also, aztreonam, cefepime, and tobramycin could be suitable alternative treatment when there are restrictions or inhibitions on the use of quinolones. Also, the high frequency of MDR P. aeruginosa justifies the need to develop a monitoring program to reduce this occurrence and control the more spread of these strains in our region. This research was the first work in southwest Iran which adds to our knowledge of how P. aeruginosa withstand quinolones. The qnrB was the most PMQR determinant in clinical P. aeruginosa isolates from southwest Iran, while the qnrC, qnrD, qepA, and aac(6′)-Ib-cr genes were not detected. Other possible mechanisms of www.nature.com/scientificreports/ resistance should also be studied for better characterization of quinolone-resistant P. aeruginosa isolates. Lack of evaluation of chromosomal mutations in the QRDR region and failure to use whole-genome sequencing to provide more in-depth data on high-risk clones and other resistance genes (such as extended spectrum betalactamases and carbapenemases) often associated with PMQR, were some limitations of this study. Bacterial isolation and identification. All studied samples were inoculated on 5% sheep blood agar and MacConkey agar plates and incubated aerobically at 37 °C for 24 h. The P. aeruginosa isolates were identified by standard bacteriological analyses such as Gram staining, catalase, oxidase, biochemical reaction on sulfurindole-motility (SIM) agar, triple sugar iron (TSI) agar, lysine iron agar (LIA), oxidative fermentative (OF) test, growth at 42 °C, and growth on cetrimide agar 34 . All culture media were purchased from Merck Co., Darmstadt, Germany. The suspected P. aeruginosa isolates were confirmed by polymerase chain reaction (PCR) using specific primers for ecfX gene as described by Sands et al. (Table 6) 35 .
DNA extraction. DNA was extracted using boiling method as previously described 38 . Briefly, few colonies were picked up from an overnight growth (24 h) on a Mueller-Hinton agar and suspended into 500 μl of distilled water. The mixture was vortexed for 20 s and then boiled for 10 min. In the end, all samples were centrifuged at 14,000 rpm for 10 min and the supernatants were stored at − 20 °C as DNA template for polymerase chain reaction (PCR) assay. The quality and quantity of DNA (ng/µl) were evaluated by measuring the absorbance of A260 and A280 nm with a NanoDrop spectrophotometer (Thermofisher Scientific, USA). The DNA samples that had a concentration of at least 50 ng/µl, were selected for PCR.