Molecular detection of plasmid-mediated quinolone resistance genes among clinical isolates of Klebsiella pneumoniae during Covid-19 pandemic

Plasmid-mediated quinolone resistance (PMQR) genes confer low resistance to Fluoroquinolones (FQs). This study aims to detect five PMQR genes among FQs-resistant Klebsiella pneumoniae isolated from various clinical specimens. Out of 120 K. pneumoniae isolates, 68 FQs-resistance K. pneumoniae were included in a molecular study. Standard microbiological tests were used for identification and antimicrobial susceptibility. For the detection of PMQR genes, conventional polymerase chain reaction was used. A molecular study revealed that (73.5%) of samples harbored PMQR genes, and among them, 58% were co-carriages of PMQR gene variants. Aac ( 6’ )- Ib - cr gene was predominant (47.1%) among samples, and qepA had the lowest percentage (11.8%), qnr genes were (32.4%) (29.4%) (20.6%) qnrS , qnrB , and qnrA respectively. Overall, high percentages of PMQR genes were detected, and almost all of samples were phenotypically resistant to ciprofloxacin. As well, there was a significant statistical relationship between phenotypically ESBL-producers and qnrB and qepA genes.

. Second: aac (6')-Ib-cr is a variant of aac (6')-Ib conferring resistance to aminoglycosides (e.g., amikacin), the -cr include two mutations (Asp-179Tyr and Trp102Arg) this is a bifunctional variant able to acetylate (confer resistance) FQs with a piperazinyl substituent (ciprofloxacin and norfloxacin) while FQs that do not have this substitution (i.e., levofloxacin) are unaffected by this enzyme. The most common PMQR gene in FQ-susceptible and resistant clinical isolates is the aac (6')-Ib-cr gene . Third: OqxAB and qepA genes. QepA is a plasmid-mediated efflux pump gene that reduces susceptibility to hydrophilic FQs, particularly enrofloxacin, norfloxacin, and ciprofloxacin (Rahman et al. 2017).
In 2018 a novel mechanism of PMQR genes was identified on a pUM505 plasmid named the CrpP gene, which codes an enzyme-modifies ciprofloxacin (Chávez-Jacobo et al. 2018). Soon after, other studies defined that the CrpP gene is not always related to ciprofloxacin resistance (Hernández-García et al. 2021), and a recent study claimed that CrpP is not a fluoroquinolone-inactivating enzyme (Zubyk and Wright 2021). PMQR genes have changed the resistance pattern to FQs, especially among Enterobacteriales (Vieira et al. 2020). Although the PMQR genes confer low-level resistance that does not surpass the clinical susceptibility level, they accelerate the selection of higher-level resistance and make infection with pathogens carrying PMQR more challenging to treat (Rahman et al. 2017;Ayobola et al. 2021). Besides, a study in 2018 claimed that PMQR confirms sufficient resistance to ciprofloxacin in Salmonella typhimurium without any chromosomal mutations (Kuang et al. 2018).
Resistance to FQs is widespread in many Gram-negative bacteria that produce extended-spectrum beta-lactamase (ESBLs), including K. pneumoniae (Higashino et al. 2017). Moreover, PMQR genes are frequently co-located with ESBL genes on plasmids; resistance plasmids containing genes for ESBLs can be transferred, which facilitates the spread of PMQR determinants among Enterobacteriaceae species (Azargun et al. 2018;Xiong et al. 2021).
The importance of this study, according to our knowledge, is the first study in Iraqi Kurdistan to detect PMQR genes among ESBL-producers and non ESBL-producers FQs-resistant K. pneumoniae from clinical isolates, remarkably during covid-19 pandemic FQs extensively used among inpatient and outpatients. In addition, this study may provide information about the prevalence of PMQR genes among K. pneumoniae in the Duhok Province Kurdistan/Iraq region.

Study setting and bacterial isolates
A cross-sectional study was conducted between September 2021 and January 2022. During this period, 120 isolates of K. pneumoniae were collected from different clinical samples, including urine (73.3%), sputum (9.2%), pus (6.7%), blood (3.3%), vaginal swab (2.5%), vaginal discharge (2.5%), bronchial lavage (0.8%), pleural fluid (0.8%), and wound (0.8%). Samples were taken from outpatient and inpatient from several public and private hospitals in different districts of Duhok Province. Ages 14 years and above and not took antibiotics in the past three days from both sexes were included. All isolates were identified using conventional biochemical tests (Mahon et al. 2019) then isolates were preserved in brain heart infusion and 15% Glycerol at -20 °C until use (Addgene 2021).

Ethical considerations
The ethics committee permitted the study proposal and informed consent of the Duhok Polytechnic University and Duhok General Health Directorate, Kurdistan Region, Iraq.

Selecting isolates and DNA extraction
Among 120 isolates, only 68 FQs-resistance K. pneumoniae were selected for molecular study. DNA extraction was carried out with an optimized heat shock method, as described by (Higashino et al. 2017). The purity of samples was 1.75-1.9 (260/280), and concentrations were between 20-80 (ng/ul) by using the NanoDrop 2000 (Thermofisher, USA). The template DNA was stored at -20 °C for PCR applications.

Statistical analysis
Frequencies, crosstab, P value, and percentages were used to describe the data analysis of the study sample. SPSS v23.0 (SPSSInc, Chicago, IL, USA).

Results
Sixty-eight FQs-resistance K. pneumoniae from clinical isolates were chosen for molecular study; 42 were urine and mainly from female patients, while pleural fluid and bronchial lavage contained the lowest percentage. Generally, outpatients were higher than inpatients Table 2.
Ciprofloxacin showed the highest rate of resistance, 92.6%, among FQs. At the same time, norfloxacin and ofloxacin showed a lesser resistance rate among FQs. In addition, 32 samples were ESBL-producers Table 3.
Among 32 carriages of aac (6')-Ib-cr, 31 samples were resistant to ciprofloxacin, and among eight samples harboring qepA, all were resistant to ciprofloxacin. In addition, there was a significant relationship between ESBL-producers and harboring QepA and QnrB genes Table 4.
In a total of 50 K. pneumoniae harboring PMQR genes, 27 samples harbored more than one PMQR gene variants and only one sample was with all targeted PMQR genes Table 5.

Discussion
PMQR genes consist of three different mechanisms, and about nine genes are responsible for those mechanisms (Rahman et al. 2017). However, in the present study only following five genes (qnrA, qnrB, qnrS, aac (6')-Ib-cr and qepA) included. Plasmids carrying PMQR have been identified mainly in Enterobacteriales, including Klebsiella, Enterobacter, Salmonella species and Escherichia coli (Jacoby, 2018). Since K. pneumoniae is the major cause of hospital-acquired pneumonia, urinary tract infections, septicemia, and soft tissue infections are also considered risk factors for severe community-acquired infections ). In addition, the first PMQR gene was detected in K. pneumoniae, and most studies indicate a greater frequency of PMQR genes in K. pneumoniae than in E. coli (Eftekhar and Seyedpour 2015;Ferjani et al. 2015). Therefore, it was worth studying K. pneumoniae. Notably, there are no studies on PMQR among K. pneumoniae or other bacteria in Iraqi Kurdistan.
Since most of the samples in this study came from urine, most of the K. pneumoniae isolates were found in urine. Our result shows that out of 68 samples, urine samples were the most common among other samples 42/68, which is near to the study done in Iran by (Eftekhar and Seyedpour 2015), and among those 42 urine samples, 35 were from females. These results are near to the study done in Iraq by (Mohammed et al. 2019;Ahmed Hasan et al. 2021).
A molecular study of this study found that 47.1% of samples harbored the aac (6')-Ib-cr gene, which it appears that the aac (6')-Ib-cr gene is more prevalent than any other qnr genes; this result agrees with similar studies conducted in the south of Iraq, in Baghdad, Kareem et al. (2018) found out of 27 K. pneumoniae 92.5% harbored aac (6')-Ib-cr and in Najaf, in a total of 74 K. pneumoniae 52.7% has been reported (Al-Hilali et al. 2021). In Tehran, Iran was 53.2% (Eftekhar and Seyedpour 2015), Kashan-Iran (70.1%) (Shams et al. 2015), Saudi Arabia, 51.6% (Al-Agamy et al. 2018), in Jordan, 50% (Idris et al. 2017). Syria, out of 123 E. coli and K. pneumoniae, 75.6% had aac (6')-Ib-cr (Alheib et al. 2015). High percentages of this gene could be back to the overuse of aminoglycosides or hydrophilic FQs in that region and time, as it is known bacteria, to survive could transfer antibiotic resistance genes among other bacteria (Jacoby et al. 2014). The aac (6')-Ib-cr gene plays a crucial role in the emergence of clinical ciprofloxacin resistance. Hence, this is especially concerning at the nosocomial level, where strains carrying aac (6')-Ib-cr should be rapidly identified and treated with non-hydrophilic FQs, such as ofloxacin or levofloxacin (Frasson et al. 2011).
Our results found that the percentage of qnrS was 32.4%. This result near to Nourozi et al. (2020) in Tehran, Iran, found that in a total of 80 K. pneumoniae isolates 34% had qnrS, and Kareem et al. (2018) found that in all 27 K. pneumoniae, 37% had qnrS in Baghdad, Iraq.
In Turkey, the percentage was higher than in our study; out of 22 K. pneumoniae, 86.4% harbored qnrS (Unlu and Demirci 2020). On the other hand, some studies revealed fewer percentages; a study in Mashhad, Iran, found that among 130 K. pneumoniae (15.4%) of isolates harbored qnrS (Izadi et al. 2017), and another study in Tehran, Iran revealed that 79 isolates 2.5% harbored qnrS (Eftekhar and Seyedpour 2015). Azadpour et al. (2014) in Lorestan, Iran, found only one qnrS (5.5%) among 18 K. pneumoniae. Al-Hilali et al. (2021) detected only one qnrS. Similarly, Hadi et al. (2016) found that out of 20 K. pneumoniae, only one sample had the qnrS gene. While in Mosul, Mahmood and Abdullah (2015), among 57 isolates, did not find any qnrS gene. The reason behind these variations could be back to the sample size.
We found that 29.4% of isolates had qnrB. Our result is in accordance with the two studies done in Tehran, Iran, in which about 79 K. pneumoniae, 30.4% carried qnrB (Eftekhar and Seyedpour 2015), and out of 88 K. pneumoniae, the qnrB was present in 43% of isolates (Nourozi et al. 2020). In Togo, West Africa found out of 64 Klebsiella spp. 30.8% had qnrB. While in the south of Iraq, different percentages have been reported. Kareem et al. (2018) in Baghdad reported 51.8%, Mosul 15.8% (Mahmood and Abdullah 2015) Najaf 23% (Al-Hilali et al. 2021). Another study in Najaf by Alshammari and Al-Skhattat (2015) found that out of 89 K. pneumoniae isolates from clinical and environmental specimens, only one sample had qnrB. These variations in the frequency of qnrS and qnrB genes may be due to the geographic distribution of the PMQR genes or criteria of selection of FQs-resistance bacteria from clinical samples in that time and region.
A 20.6% of our samples harbored qnrA. This percentage has not previously been reported in Iraq among K. pneumoniae or other bacteria from clinical isolates. In their isolates, Hadi et al. (2016) and Al-Hilali et al. (2021) and found qnrA only in two samples. Alshammari and Al-Skhattat (2015) isolate one. Mahmood and Abdullah (2015) did not find qnrA. Similarly, some studies in Iran did not find qnrA (Azadpour et al. 2014;Eftekhar and Seyedpour 2015;Abbasi and Ranjbar 2018). In comparison, our result is near to Mahdane et al. (2022), which found that out of 80 E. coli and K. pneumoniae samples, 26.8% harbored qnrA. In general, qnrB seems somewhat more common than qnrA (Jacoby et al. 2014).
During the present study, qepA had the lowest percentage of other PMQR genes, which was 11.8% in a total of 68 isolates. However, a high percentage has been reported by Kareem et al. (2021) in Iraq, in a total of 27 K. pneumoniae (40.7%) harbored qepA, while Alshammari and Al-Skhattat (2015) found 4.5% in total of 89 K. pneumoniae had qepA. Some studies from other countries, such as Syria and Jordan, did not find qepA (Alheib et al. 2015;Idris et al. 2017). In Iran, out of 247 K. pneumoniae, 2% harbored qepA (Goudarzi et al. 2015).
In a total of 68 samples, 50 samples harbored PMQR genes, 46% had one gene, and 54% had more than one gene. Higher levels of FQs resistance have been noticed when the plasmid carries two or more FQs gene resistance (Jacoby et al. 2014). High frequency of qnrB and aac (6')-Ib-cr has been reported in K. pneumoniae (Eftekhar and Seyedpour 2015), but in our study, co-carriage of qnrA and aac (6')-Ib-cr found in 20% of samples, and 14 isolates had qnrB and aac (6')-Ib-cr. It is uncommon for a single isolate to possess more than two PMQR genes (Vinué et al. 2020). While in the current study, 26% of samples had three or more PMQR genes, which consider a high percentage. High co-carriage of PMQR genes in our study could be back to the over and misusing of FQs, improper hygiene and not strict guidelines for infection control leading to bacteria becoming resistant in the community and hospital settings.
Beta-lactam and FQs antibiotics are the most significant antibiotic agents used to treat K. pneumoniae infections (Goudarzi et al. 2015). Numerous studies have proved the link between PMQR determinants and the expression of ESBLs (Azargun et al. 2018). In the current study, there was a statically significant relationship among phonetically ESBL-producers and qepA and QnrB, while for other genes, statistically was not significant; this may be because the ESBL production was confirmed only by DDST, although this test considered the most accurate than other phenotypic confirmatory tests (Iqbal et al. 2017) however, the coexistence of AmpC gene may result in false negatives (Kaur and Chopra 2013). In comparison, some studies were confirmed genotypically. For example, a study in Azerbaijan and Iran revealed that genotypically confirmed Enterobacteriales that are ESBL-producing in UTI isolates had high FQs resistance and PMQR genes (Azargun et al. 2018). Another study in which DDST confirmed ESBL-producers found PMQR was more frequent in ESBL-producing K. pneumoniae (Izadi et al. 2017). However, FQ resistance in ESBL-producers is an urgent health concern in countries where ESBL-producing K. pneumoniae is prevalent (Higashino et al. 2017). Therefore, Monitoring and epidemiological studies are required to stop the spreading FQs resistance among ES-BL-producers.

Conclusion
During this study, most samples were resistant to ciprofloxacin; nearly half of the samples were ESBL-producers. A molecular study detected a high percentage of PMQR genes. The aac (6')-Ib-cr gene was more prevalent than any other PMQR gene. Among PMQR-harbored samples, 54% had more than one PMQR gene.