Association of qnr Genes and OqxAB Efflux Pump in Fluoroquinolone-Resistant Klebsiella pneumoniae Strains

Background The aim of this study was to investigate the frequency and relationship between plasmid-mediated quinolone resistance genes and OqxAB pump genes, as well as the genetic linkage in K. pneumoniae strains isolated from Hamadan hospitals in the west of Iran. Materials and Methods In this study, 100 K. pneumoniae clinical strains were isolated from clinical samples of inpatients at Hamadan Hospital in 2021. The antimicrobial susceptibility testing was performed using the disk diffusion method. The frequencies of genes encoding OqxAB efflux pumps and qnr were investigated by PCR. Molecular typing of qnr-positive K. pneumoniae isolates was assessed by ERIC-PCR. Results Antibiotic susceptibility testing showed high resistance (>80%) to fluoroquinolones. The gene encoding the OqxAB efflux pump was detected in more than 90% of K. pneumomiae strains. All K. pneumoniae isolates were negative for qnrA, and 20% and 9% of the isolates were positive for qnrB and qnrS, respectively. The genes encoding oqxA and oqxB were detected in 96% of qnr-positive strains. A qnrB + /qnrS + profile was observed in 16% of qnr-positive K. pneumoniae strains. Ciprofloxacin MIC ≥ 256 μg/ml was detected in 20% of qnr-positive strains. Genetic association analysis by ERIC-PCR revealed genetic diversity among 25 different qnr-positive strains of K. pneumonia. Conclusion However, no significant correlation was found between the qnr and the OqxAB efflux pump genes in this study. The high rate of fluoroquinolone resistance and determinants of antibiotic resistance among diverse K. pneumoniae strains increase the risk of fluoroquinolone-resistance transmission by K. pneumoniae strains in hospitals.


Introduction
One of the causes of pneumonia, sepsis, and urinary tract infections in hospitalized patients is known to be Klebsiella pneumoniae, a Gram-negative bacterium in the Enterobacteriaceae family. Tis bacterium is more likely to infect people whose immune systems are weak. Te majority of antibiotics used to treat infections caused by this bacterium are beta-lactam and fuoroquinolones [1].
Controlling antibiotic resistance in multidrug-resistant K. pneumoniae (MDR-KP) is a major challenge. Optimal treatment options for MDR-KP infections are not yet well established. New antimicrobial agents against MDR-KP have been developed over the past decades and are currently in various stages of clinical research [2]. Fluoroquinolones such as ciprofoxacin are among the antibiotics used to treat infections associated with K. pneumoniae [3]. Quinolones and fuoroquinolones represent a wide range of antibiotics whose mode of action is to inhibit bacterial DNA gyrase enzymes [4]. In recent years, increasing resistance to these antibiotics has led not only to problems in the treatment procedure but also to increased treatment costs and longer hospital stays. Terefore, there is a need to assess resistance levels and mechanisms of resistance to these antibiotics in nosocomial pathogens such as K. pneumonia infections [5].
Resistance to ciprofoxacin can develop through a variety of mechanisms. Ciprofoxacin resistance is primarily associated with chromosomal mutations that alter ciprofoxacin target proteins (DNA gyrase and topoisomerase IV). Quinolone resistance can also be induced by mutations in the Efux pump gene regulators. In addition to mutations in susceptible targets, ciprofoxacin resistance can also be mediated by plasmid-mediated quinolone resistance (PMQR) genes and efux pumps such as OqxAB [4]. Te OqxAB is expressed on plasmids in clinical isolates of E. coli and K. pneumoniae [6]. Te qnr plasmid protects ciprofoxacin targets from inhibition and causes low-level resistance to quinolone. Diferent qnr genes have been found in bacterial strains from various regions of the world. Many qnr genes have been discovered including qnrA, qnrS, qnrB, qnrC, and qnrD and more recently qnrVC and qnrT [6,7].
Te aim of this study was to investigate the frequency and relationship between plasmid-mediated quinolone resistance (qnr) genes and OqxAB pump genes (oqxA and oqxB), as well as the genetic linkage in K. pneumoniae strains isolated from Hamadan hospitals in the west of Iran.

Bacterial Isolation and Identifcation.
A total of 100 clinical isolates of K. pneumoniae were randomly collected in 2021 from patients at three major hospitals in Hamadan. K. pneumoniae isolates were identifed by routine microbiological testing. Lactose-fermenting mucoid (pink-colored) colonies on MacConkey agar were selected. K. pneumoniae isolates were identifed by IMVIC test (Indole production, Methyl Red (MR), Voges Proskauer (VP), and Simmons citrate), as well as by SIM, TSI, Urea Hydrolysis, and Lysine Decarboxylase tests. Te results of the biochemical assay were confrmed by PCR using speciesspecifc primers for the ureD gene responsible for urea hydrolysis in K. pneumoniae [8,9].

ERIC-PCR.
All qnr-positive K. pneumoniae strains were selected for molecular typing by ERIC-PCR. Tis procedure was performed using the primers as described previously [12] and the program in Table 1. Te ERIC profles were analyzed by an online database analysis service (insilico.ehu.es). A cutof similarity of ≥95% was considered.

Data
Analysis. Data were analyzed using SPSS software version 22.0 (IBM Co., Armonk, NY, USA). Te relationships between diferent qnr genes, OqxAB efux pump genes, antibiotic resistance, sample source, and hospital wards were analyzed using the Pearson chi-squared or fsher exact test. Te relationship between qnr genes and the ciprofoxacin MIC values was analyzed by the t-test. P values less than 0.05 were considered statically signifcant.

Phenotypic Characteristics of K. pneumoniae Isolates.
Klebsiella pneumoniae colonies were observed as pink mucoid colonies on McConkey agar. According to biochemical tests, K. pneumoniae colonies were indole and MR negative, and VP, Simmons citrate, urea hydrolysis, and lysine decarboxylase tests were positive. Te presence of the ureD gene in K. pneumonia isolates were confrmed by PCR.

Prevalence of qnr and Efux Pump
Genes. Te results of qnr-genes detection by PCR showed that all K. pneumoniae isolates were negative for qnrA, and 20% and 9% of the isolates were positive for qnrB and qnrS, respectively ( Figure 1). A qnrB+/qnrS + profle was observed in 16% of qnr-positive K. pneumomiae strains. Te oqxA and oqxB genes were detected at 95% and 98% of K. pneumoniae isolates, respectively ( Figure 2). According to the results, the gene encoding oqxA and oqxB were detected in 24 (96%) of qnr-positive strains. Table 2 provides information on isolates containing the qnr genes. Static analysis of the results showed that there were no signifcant correlations between qnr genes, efux pumps, ciprofoxacin resistance, hospital wards, sample sources, ciprofoxacin MICs, or patterns of antibiotic resistance (P value ≥0.05). However, efux pump genes and ciprofoxacin resistance were found to be signifcantly correlated (P value ≤0.04).
ERIC-PCR analysis of genetic linkage revealed genetic diversity among 25 diferent K. pneumonia qnr-positive strains. Te results of the ERIC-PCR revealed the presence of 18 diferent ERIC profles, including 5 common types and 13 unique types (including one isolate). Tere were two to three isolates in the common types ( Figure 3). Our fndings revealed the distribution of MDR and fuoroquinoloneresistant K. pneumoniae strains in some Hamadan hospital wards, as well as the distribution of related K. pneumoniae clones (ERIC type) in some hospital wards (ERIC types A, B, and E in the intensive care units). Strains of a common type showed diferent patterns of antibiotic resistance and different profles of efux pumps and qnr genes, and no signifcant relationship was observed between them ( Table 2). Tere were no signifcant correlations between qnr genes and ERIC types.

Discussion
Fluoroquinolone resistance is an important problem associated with K. pneumoniae. Fluoroquinolone resistance has been suggested to play an important role in the successful evolution of K. pneumoniae strains [14]. Tis study showed a high level (≥80%) of resistance to ciprofoxacin in K. pneumoniae strains. Most fuoroquinolone-resistant strains also display the MDR phenotype. Tis fnding highlights the problem of management of K. pneumoniae infections in hospitals and poses signifcant challenges for clinicians. Many of the isolates in this study were isolated from patients admitted to the ICU, which may be one of the main reasons for the high resistance to ciprofoxacin in this study. Te rate of fuoroquinolone resistance in our study is higher than that in the study in Iran and some other regions [15][16][17][18][19][20]. However, a recent study of Hamadan reported a high rate of resistance (≥80%) to fuoroquinolones [21]. In total, our fndings indicate that the qnr genes were present in 25% of K. pneumoniae isolates. None of the isolates contained qnrA, and qnrB was identifed as the most prevalent qnr gene. Te prevalence of qnrB and qnrS was 20% and 9%, respectively, in the isolates. On the distribution of the qnr gene in K. pneumoniae, there are various reports, including those from Iran. Nourozi et al. reported that qnrB (43% of isolates) was the most frequently detected qnr gene, followed by qnrS (34% of isolates) and qnrA (23% of isolates) in K. pneumoniae isolates from hospitals in Tehran [22]. Te study results of Salimbahrami et al. showed that 47 (52%), 22 (25%), and 21 (23%) K. pneumoniae isolated from hospitals in Sari (northern Iran) contains the qnrB, qnrA, and qnrS genes, respectively [16]. In another study from southwestern of Iran by jomehzadeh et al., ciprofoxacin resistance was lower than our results while qnrA, qnrB, and qnrS were detected in (12%), (24%), and 17% of K. pneumonia isolates [23]. In the following, studies that reported results similar to ours are reviewed. In a study conducted by Malek Jamshidi et al., at Yazd central laboratory, the qnrA gene was not detected in any of the K. pneumoniae strains, and the qnrB gene (25%) was identifed as dominant [23]. In a study from India, 22% of isolates possessed both qnrB and qnrS genes, while the qnrA gene was not detected in any strains [19]. Studies in China, Singapore, and Malaysia have demonstrated that qnrB is the major qnr gene, which is consistent with our fndings and the majority of Iranian studies [24][25][26].
Te diferences in qnr genes frequency distribution of the of K. pneumoniae strains in diferent studies may be due to diferences in geographical distribution of fuoroquinoloneresistant strains, sampling locations, and infection control strategies in the hospitals. Most qnr-positive strains had increased MICs for ciprofoxacin (28% and 20% of qnrpositive strains, MIC ≥ 128, and MIC ≥ 256, respectively). Tese results indicate that the presence of the qnr gene may play a role in reducing susceptibility to fuoroquinolones. It has also been shown that the qnr genes products can protect fuoroquinolone targets from antibiotic action. Tese genes are widely distributed in Enterobacteriaceae. Te qnr gene is thought to induce low to moderate resistance to quinolones. Factors other than PMQR, such as gyrA and gyrB mutations and efux pumps, may also play a role in the emergence of fuoroquinolone resistance. Plasmid-mediated qnr gene transfer between nosocomial pathogens increases the risk of transmitting resistance to fuoroquinolones and reduces the susceptibility of these pathogens to these antibiotics.
Another important factor in resistance to fuoroquinolones is the efux pump. In this study, we investigated the association of the OqxAB efux pump with resistance to fuoroquinolones. In this study, >95% of the isolates carried the OqxAB efux pump genes. Te OqxAB gene has been repeatedly reported in quinolone-resistant Enterobacteriaceae [27]. Expression of OqxAB has been  Table  2: Characteristics of qnr-positive K. pneumoniae strains based on source, hospital wards, antibiotic resistance, qnrs and oqxAB genes profles, and ERIC types. shown to be associated with reduced susceptibility to quinolones in K. pneumonia [28,29]. Te present study found a signifcant relationship between ciprofoxacin resistance and the presence of OqxAB efux pumps. Te percentage of the OqxAB genes in our study is higher than that in other Iranian studies [22,30,31]. In our study, 96% of the qnrpositive strains harbored the efux pump genes and exhibited the MDR phenotype. High ciprofoxacin resistance may be related to the coexistence of the efux pump OqxAB and the qnr genes. For molecular typing of K. pneumoniae isolates, our study used ERIC-PCR, which is a simple, inexpensive, and accessible method with sufcient reliability and reproducibility. Such as many studies, genetic diversity has been distinguished among K. pneumoniae strains [12,[32][33][34]. Our results demonstrated the prevalence of related K. pneumoniae strains (ERIC type) in some hospital wards, as well as the distribution of MDR and fuoroquinolone-resistant K. pneumoniae strains in some hospital wards, suggesting that these strains need further investigation. In this study, antibiotic resistance profles and qnr gene profles of strains belonging to the same ERIC class were diferent. To learn more about resistant strains, we recommend further investigation of the diferent mechanisms that infuence the genetic diferentiation of K. pneumoniae. OqxAB efux pump genes and antibiotic resistance patterns were not signifcantly correlated with the ERIC groups, as shown in our results. Our results also show that widespread K. pneumoniae colons with diferent antibiotic resistance patterns and high proportions of antibiotic  International Journal of Microbiology 5 resistance determinants such as oqxAB and qnr genes can interfere with the control of nosocomial infections. In addition, high-risk fuoroquinolone-resistant K. pneumoniae clones have been shown to maintain ftness, facilitating their spread in hospital settings [35].

Conclusion
However, the results of this study suggested that there is no signifcance association between PMQR and OqxAB efux pump genes but the alarming prevalence of MDR phenotypes, presence of OqxAB efux pump, and PMQR determinants in heterogeneous fuoroquinolone-resistant K. pneumoniae strains in hospitals increase the risk of fuoroquinoloneresistance transmission among hospital-adapted pathogens and cause challenges for clinicians in hospitals. More bacterial strains should be tested and a wider area sampled for more accurate and better results. All of this requires research funding and more collaboration between hospitals, research centers, and universities.

Data Availability
Te data used to support the fndings of this study are included in the article.

Ethical Approval
Tis study was approved by the Ethics Committee of Hamadan University of Medical Sciences (IR.UMSHA.REC.1400.185).  Figure 3: Dendrogram of ERIC-PCR types of qnr-positive K. pneumoniae strains, compared by dice method and grouped by UPGMA method. 6 International Journal of Microbiology