Klebsiella pneumoniae Isolates
In a cross-sectional study, in total, 100 K. pneumoniae isolates were randomly isolated from clinical samples, including wounds (4%), blood (3%), trachea (15%), and urine (78%) samples of patients hospitalized in different (ICU, pediatric, emergency, and internal) wards in three hospitals of Borujerd, in the west of Iran.

Antimicrobial  Susceptibility Test
The antibiotic susceptibility of isolated K. pneumoniae to quinolone antibiotics, including Ciprofloxacin (30 µg), Norfloxacin (30 µg), Nalidixic acid (10 µg), and Ofloxacin (5 µg) was tested using a standard disk diffusion method. Then the determination of Ciprofloxacin MICs by E-test was carried out according to CLSI criteria. Standard strain K. pneumoniae ATCC 12022 was considered as a control.

E-test Procedure
To change the susceptibility or resistance of the isolated bacteria from Enterobacteriaceae family to Ciprofloxacin based on CLSI criteria for MIC, values of 1 μg/mL were considered as sensitive while the values of 2 μg/mL and ≥4 μg/mL were considered as semi-susceptible and resistant, respectively.

DNA Extrcation and PCR Assay
PCR and sequencing were performed to amplify and detect mutations in gyrA and parC genes in quinolone and also fluoroquinolone-resistant K. pneumoniae isolates. Generally, the genomic DNAs of the mentioned isolates were extracted by a commercial DNA extraction kit (Cinapure DNA, CinaClon, Iran) according to the manufacturer’s instructions and were amplified by PCR assay. The oligonucleotide primers used in the PCR reactions were gyrA-F (5'‐CGACCTTGCGAGAGAAAT‐3'); gyrA-R (5'-GTTCCATCAGCCCTTCAA‐3'), and primer parC-F (5'‐TACGTCATCATGGACAGG‐3'); parC-R (5'‐GCCACTTCACGCAGGTTG‐3') (Metabion, Germany). For all amplification reactions, a PCR mixture was used which contained 12.5 µL of 2X Master Mix (Ampliqon, Denmark) in the total volume of 50 µL. DNA amplification was performed in a thermocycler (PEQ STAR, Germany). For gyrA, the PCR program was: an initial denaturation at 98°C for 5 minutes, 35 cycles of denaturation at 94°C for 1 minute, 1 minute of annealing at 64°C, and 2 minutes of extension at 72°C, and final extension at 72°C for 5 minutes. The temperature profile for parC gene was comprised of 95°C for 2 minutes, followed by 35 cycles of 1 minute at 95°C, 1 min of annealing at 60°C, 2 minutes of extension at 72°C, and final extension of 10 minutes at 72°C. DNA fragments were analyzed by agarose gel electrophoresis (1%, wt/vol) containing 0.5 mg of ethidium bromide per liter and visualized by ultraviolet illumination.
The sequencing of PCR products was carried out using the Applied Biosystems DNA analyzers sequencing system (Bioneer Co., Korea). To detect the mutation in gyrA and parC genes, the sequences of these genes in Ciprofloxacin-resistant strains were copied in FASTA format in online pairwise alignment tools (Emboss/needle), then they were aligned with the sequences of gyrA and parC genes of the wild-type strain of K. pneumoniae.

Statistical Analysis
SPSS version 21 (SPSS Inc., Chicago, IL., USA) was employed for statistical analysis. Descriptive statistics and Pearson’s chi‐square tests were used to evaluate the correlation between the mutation and Ciprofloxacin resistance. P-values less than 0.05 were considered as statistically significant.

 

Resistance to Fluoroquinolones
The results demonstrated that among 100 K. pneumoniae isolates, 38 of them were resistant to Nalidixic acid and Ciprofloxacin, simultaneously, while 18 and 15 isolates were resistant to Ofloxacin and Norfloxacin, respectively. Furthermore, 15 isolates were resistant to Nalidixic acid, Ciprofloxacin, Ofloxacin, and Norfloxacin, simultaneously. The presence of mutations in gyrA and gyrB genes was evaluated in 38 quinolone-resistant isolates (Figure 1). Mutations were detected in 37% (n=14) of them. The frequencies of mutations in gyrA, parC, and both gyrA and parC were 64% (n=9), 57% (n=8), and 21% (n=3) in strains with point mutations (Figure 2). In 63% (n=24) of quinolone-resistant strains, no mutations were detected at all. According to the results of our E-test, MIC values ranged from 2 μg/mL to 128 μg/mL, whereas 20 isolates (86%) resistant to Ciprofloxacin with MIC= 128 μg/mL were determined. The MIC values including 64 μg/mL and 2 μg/mL were detected in one (4.5%) and two (8.5%) of the Ciprofloxacin-resistant isolates, respectively.

Seqeuencing of the parC and gyrA QRD
The results of sequencing data were acquired and analyzed using Chromas software. Then the nucleotide sequences were compared and the protein sequences of the mutations created in the gyrA and parC genes were evaluated. The results are presented in Table 1. Data sequences were examined to detect the presence of amino acid substitution in certain positions. Fourteen amino acid substitutions in GyrA and 16 amino acid substitutions in ParC proteins were observed in quinolone-resistant isolates. These results are presented in Table 1 and Figure 2.
 In the case of GyrA protein, 14 point mutations were detected at position numbers: Arg232 →Gln, Val437→Ala, His194 →Gln, Ala204→Ser, Arg207→Trp, Leu210→Tyr, and Arg214 →Cys, Pro216 →Ser and Ser83, 217, 21→Phe, Asp232 →Ala, Arg502 →Met. Five- and two-point mutations in GyrA were detected in two isolates by number 7 and 11, respectively (Table 1). In the case of ParC protein, 16 amino acid substitutions were detected at position numbers: Trp10, 208 →Arg, Trp11→Cys, Met46→Thr, Ser141, 143 and 80 →Ile, 139 and Gln149 →His, Lys155 →Glu, Ser196→Pro, Gly205→Trp, Ser211→Ala, Leu245 →Ser. Two-point mutations in ParC were detected in 4 isolates (isolates by numbers 8, 11, 29, 61) and five-point mutations were detected in one isolate (by number 90). Also, a double mutation in gyrA and parC genes was identified in 3 strains (strains number 6, 8, and 11) and most of the quinolone-resistant strains exhibited a high MIC value (128µg/mL).

Figure 1. Gel electrophoresis of PCR products for gyrA and parC genes. The 1^st lane represents the positive control for gyrA, the 2^nd lane shows the negative control, the 3^rd lane is a sample containing gyrA and lane 4 represents a 100 bp DNA size marker; the 5^th and 6^th lanes show the positive and negative controls of parC gene, respectively;  the 7^th lane shows the parC sample.

Figure 2. Substitution of amino acids and number of mutations in parC gene (A) and gyrA gene (B) in K. pneumoniae isolates.


Table 1. Substitution of amino acids in  GyrA and ParC  proteins in K. pneumoniae isolates



 

Discussion

In this study, a relatively high resistance (38%) to fluoroquinolones including Ciprofloxacin was observed among K. pneumoniae isolates in Borujurd hospitals. In general, the findings of this study indicate that mutations in certain positions of gyrA and parC genes may have a profound impact on developing a high-level resistance to quinolone in K. pneumoniae. However, other mechanisms of resistance to quinolones should be considered.  According to the results, the frequency of point mutations in gyrA and parC genes was 37% in quinolone-resistant strains. This suggests that other mechanisms can have a role in resistance to quinolone, along with the mutations in gyrA and parC genes. Bacteria gain resistance to quinolones in several ways such as changes in target enzymes including DNA gyrase and topoisomerase IV, changes in the endocytic pathways like penetrating the antimicrobial agent, increasing efflux pumps, and decreasing  porin-expressions (9-11), and obtaining Qnr plasmids (2, 3). In addition to the plasmid-mediated efflux systems, chromosomal multidrug efflux pumps are able to actively remove fluoroquinolones and other agents from the bacterial cell (6). It has been proved that the main mechanism for resistance (which is dependent on the chromosome) to quinolones is the mutations in the genes encoding subunits A and B in DNA gyrase and topoisomerase IV. These enzymes are the main targets of quinolones.
The permeability of cell membrane and the ability of antibiotics in penetrating into the cell are key indexes in the efficacy of antibiotics such as fluoroquinolones with intracellular targets (12). The inner and outer membranes in the cell walls of gram-negative bacteria are considered as a particular barrier. Mutations created in down-regulation of these proteins could increase the MIC of fluoroquinolones (13).
The real contribution of plasmid mediated quinolone resistance to the MIC of Ciprofloxacin is not obvious in Ciprofloxacin-resistant isolates. Thus, there is not a significant difference between the MICs of Ciprofloxacin-resistant isolates with or without qnrB genes (6). A study showed that among the K. pneumoniae strains producing ESBL, resistance to Ciprofloxacin is higher. Resistant strains that have mutations in the gyrA and parC genes, or the plasmid aac (6') -Ib-cr gene, along with the qnr genes reveal a high resistance to Ciprofloxacin (14).
Compared to our research, some studies have reported a lower resistance to quinolones as well as mutation rates in K. pneumoniae. In 2018, Mirzaii et al. reported that 15 (6.8%) of the 220 K. pneumoniae isolates were resistant to at least one of the quinolones and only 1 (6.7%) of the 15 K. pneumoniae isolates was found to carry Ser83Leu substitution in gyrA protein in Tehran (15). In a study by Alijani et al. it was revealed that the Ciprofloxacin resistance rate was lower than that of our study (16). However, Huang reported much more resistance to Ciprofloxacin (more than 60%) and also point mutations in gyrA (more than 95%) in China hospitals (17). Thus, based on the results, the double mutation Ser83→Leu + Asp87→Asn was the most prevalent type of gyrA changes in antimicrobial-resistant isolates (15-18). In 2014, Norouzi et al. reported some K. pneumoniae strains substituting Ser 83 →Ile and Ser 83 →Phe in gyrA gene (8). Marquz et al. reported a resistance rate (37%) to Ciprofloxacin (similar to our study) in Malaysia’s hospitals (6). However, they reported that 33 isolates had multiple mutations either in gyrA alone (n=1) or in both gyrA and parC regions (n=32). There are some differences among reports because the pattern of antibiotic resistance is different in various regions and this resistance profile is constantly changing in time with the increasing use of antibiotics.
According to the results of our study and others in Iran and Asian countries (Malaysia, Taiwan, Japan, and Singapore), the most similar findings in amino acid substitutions Ser 83→ Phe in gyrA region and ser80→Ile in parC region were those regarding K. pneumoniae isolates reported from Malaysia (6).
Some researchers believe that at least 3 mutations in the gyrA gene are required to produce a high level of resistance to Ciprofloxacin (18). Fu et al. identified 7 mutation patterns without any effects on drug resistance in the isolates including the mutations outside the QRDR region and the conversion of alanine 171 to serine and valine 198 to isoleucine (19).
Generally, the present study reveals that regardless of mutation position, several types of mutations in gyrA and parC genes may result in resistant strains. In 2014, Farah al-Marzooq reported that the rise of Ciprofloxacin MICs strongly correlates with the rise in the total number of mutations in gyrA and/or parC genes (6). On the other hand, our findings are in contrast with those of Fu. Y et al. (20) and indicate that only one mutation could cause a high rate of resistance with MIC 128 µg/mL.
In our study, quinolone-resistant strains did not show any significant mutations on specific positions of the chromosome which suggests that other mechanisms such as plasmid resistance or alteration in the efflux pumps (21) may create quinolone resistance in these strains while chromosomal resistance is not involved. In the relation between gyrA gene mutations and fluoroquinolones resistance, various results have been obtained. In a study in Brazil in which gene mutations in gyrA and parC genes were investigated in intestinal bacteria isolated from patients, 33% of resistant isolates had mutations in codons 21 or 23 or both, and none of the isolates susceptible to Ciprofloxacin had gene mutation in topoisomerase (3).
In general, there are some suggestions to increase knowledge about resistance to fluoroquinolones including investigating the mechanism of resistance to quinolones especially Ciprofloxacin using molecular methods and the evaluation of resistance profile to quinolones especially in hospitals. Complete antibiotic treatment can be effective in preventing microbial resistance and in implementing care plans. These inhibitive measures can also reduce antibiotic resistance. Moreover, periodic and systematic studies on resistance patterns are necessary in order to obtain an appropriate treatment regimen for patients (22, 23).

 

Conclusion

Based on our findings, the relatively high resistance to fluoroquinolones including Ciprofloxacin is significant particularly in hospitalized patients. The same amount of fluoroquinolone resistance is vital for treating patients especially for those who are hospitalized or have a weakened immune system because they should be treated by antibiotics for a long time. Resistance to Ciprofloxacin is significant in K. pneumoniae and it is feared that it will increase. Hence, it is better to use more effective quinolone antibiotics or a drug with effects similar to Ciprofloxacin such as Levofloxacin. The concomitant mutations in the QRD regions are estimated to contribute meaningfully to high-level fluoroquinolone resistance and to the presence of different substitutions at amino acids in GyrA and ParC proteins. This suggests that different sites have noticeable roles in resistance to quinolones. However, the mutations in gyrA and gyrB genes, other mechanisms of resistance to quinolones, their effects on resistance, and MIC changes should be investigated. Further investigations of molecular techniques so as to fully examine the genetic pattern of resistance, the existence of precise care, monitoring programs, and serious preventive measures are recommended.

 

Acknowledgements

We would like to express our gratitude to all members of the microbiology laboratories of hospitals in Borujerd, Iran.

 

Conflicts of Interest

Authors declared no conflict of interests.